2015.131996.introduction To Geomorphology Text

Alistair F.
Pitty
Introduction to
Geomorphology
UNIVERSITY OF JODHPUR LIBRARY
METHUEN & CO LTD

First published 1971 by
Methuen & Co Ltd, 11 New Fetter Lane, London EC4
©1971 Alistair F. Pitty
Filmset in Photon Times 11 on 12jpt. by
Richard Clay {The Chaucer Press), Ltd, Bungay, Suffolk
and printed in Great Britain by
Fletcher& Son Ltd, Norwich, Norfolk
SBN 416 11730 9 hardback

416 29760 9 paperback
This book is available in both hardbound and paperback

editions. The paperback edition is sold subject to the
condition that it shall not, by way of trade or otherwise,
be lent, re-sold, hired out, or otherwise circulated without
the publisher’s prior consent in any form of binding or
cover other than that in which published and without
it is
-ti similar condition being imposed on the subsequent purchaser.
Distributed in the USA by Barnes & Noble Inc

1
Contents
Acknowledgements
Preface
I. Definitions, nature and basic postulates i
A. Definitions 1
B. The Nature of Geomorphology 1
1. Description and interpretation 1
2. Process and form 1
3. Artistic and scientific elements 16

4. Qualitative and quantitative aspects 20
5. Laboratory and work
field 29
6. The role of geomorphology 33
C. Basic Postulates 37
1. Catastrophism and uniformitarianism 37
2. Stillstands and the mobility of earth structures 42
3. The Cycle of Erosion 48
4. Morphoclimatic zones 54
5. Structure, process and stage 60
6. The necessity for simplification of geomorphological 66
complexity
D. Some Conclusions about Geomorphology 72
II. Landforms and Structure 79
A. Geophysical Considerations 79
1. Earth movements and mountain building
79
2. Continental drift 84
3. Vulcanicity 86
4. Hydrothermal activity and meteorite craters 89
'ONIVmSlTY ©F JOD-SPUR LIBRAS^

9
7
1
11
vi Introduction to Geomorphology
B. Geological Considerations 91
1. Resistance to erosion 9
2. Rock disposition 98
3. Distinctive lithologies 106
4. Variations within an individual rock 1 08
1.
III. Physical, chemical and biological basis of 112

geomorphological processes
A. Regional Climate 112
Temperature 112
B. Amounts and Motions of Water and Ice 1 14

1. Hydrological considerations 1 14
2. Hydraulic considerations 1 1
3. Waves and offshore currents 1 1
4. Ice and snow 1 2
C. Mechanical and Frictional Forces 125
D. Sediments and Mechanical Characteristics of Soils 127

1. Particle sizes 127
2. Particle shape 1 3
3. Surface appearance of particles 133
4. Disposition of particles . 134
5. Soil fabrics 136
6. Soil hydrology 136
7. Mechanical properties of clays 1 37
8. Frozen soils 139
E. Geochemical Considerations I 4I
1 . Physicochemical factors 1 4
2. Weathering reactions 142
3. Physicochemical characteristics of weathering environments 144
4. Mineral solubilities 145
F. Biological Activity I 47
1. Acceleration of mechanical processes 148
2. Retardation of mechanical processes 153
3. Biochemical weathering 160
4. Acceleration of chemical weathering 160
1
Contents vii
1 ^^
5 . Retardation of chemical weathering
6. Organisms as aids in landform study
168
G. Human Activity
IV. Inter-relationships between processes and

landforms
A. Rock Breakdown
1. Erosion
2. Frost action
3. Other mechanical processes of rock breakdown 186
4. Chemical weathering
5. Weathering forms 193
B. Transportation 199
1. Transportation of particles in a fluid 201
2. Amounts of sediment transport 209
3 . Mass movement 216
4. Transportation of dissolved material 222
C. Deposition 224
1. Depositional landforms on slopes 225
2. Depositional landforms in fluvial environments 228
3. Deltas 233
4. Estuaries 236
5. Depositional landforms along coasts 237
6. Depositional landforms in eolian environments 244
7. Landforms due to glacial deposition 248
D. 10.
inter-relationships between Rock Breakdown, Transport,
and Deposition 254
1. Time lags 254
2. Influence of source 25 7
3. Influence of transportation on rate of supply 260'
4. Covariances between phenomena 26
5. Reworking of sediments in stream channels 274
6. Dynamic depositional landforms 279
7. Reworking of sediments into distinctive patterns 282
8. Recombination of weathered chemical elements 285
9. Influence of weathered material on subsequent weathering 292
The concept of quasi-equilibrium 295
viii Introduction to Geomorphology
E. Inter-relationships between Form and Process 298
F. The Significance of Changes in Weather and in Climate 307

1. Local changes due to differing aspects 307
2. Changes in the effectiveness of processes with time 313
3. Examples of past climates 32
V. Landforms and time 327
A. Dating 327
B. The Initial Form 342
C. Stages in Slope Evolution 344
D. Stages in Drainage System Evolution 355
E. Stages in Glaciated Areas 372
F. Stages in Coastline Evolution 380
G. Stages involving more than one Process and in the
Evolution of the Land Surface as a whole 391
Appendix; Some simple methods of field

measurement 402
A. Slope Measurement 402
B. Soil Moisture Content 403
Procedure 405
C. Estimation of Stream-Flow 406
Procedure 406
D. Particle Size Analysis of Sands and Pebbles 408
Procedure 408
E. Heavy Minerals 410
F. Soil pH 410
Procedure 411
G. Dissolved Solids 412
Procedure 4 1
H. Suspended Sediment 412

Procedure
Selected Bibliography 414

Indexes
489
Acknowledgements
A book is that of the

great deal of the effort behind the presentation of this
technical and secretarial staff in the Department of Geography in the
University of Hull, especially that of Mr Keith Scurr and Miss Wendy A.
Wilkinson who drew the figures. Mr R. R. Dean arranged the schedule for
drawing at a time when illustrations for a wide range of other publications
were being requested. Mr A. Key assisted with the drawing at a final stage.
Mr J. B. Fisher and Mr S. Moran prepared photographic prints of the

drawings. The by Miss C. Hayward, who
typing was done with great care
typed most of the text, and by Mrs B. Smith, assisted by Miss H. Coburn,
who typed the appendix and the captions.
At the outset, when an attempt to span the breadth of geomorphology
seemed scarcely possible was particularly grateful to the following for their
I
assistance in compiling information; when demands on the Map Library were

not too heavy, the Curator, Miss A. M. Ferrar, arranged for her assistant.
Miss B. Myers, to help with compilation; one of my tutorial groups, com-
prising Misses J. A. Cavanagh, S. V. Lloyd, C. A. Luland, M. D. V.
Raybould, and Mr G. D. Sanderson; and Mrs P. A. Pitty, who also prepared
abstracts used in parts of Chapter II. Research students who, by the
thoroughness and enthusiasm of their assistance as demonstrators, helped to
make my teaching commitments less exacting at the time was writing when I
the book include Mr I. Reid (practicals), Mr and Mr

D. W. John (tutorials),
K. A. Falconer (field work). Previously Dr A. C. Imeson had assisted on
these occasions. I am grateful to the publishers for their help and kind
editorial assistance and advice. I am also grateful to many other members,
past and present, of the University of Hull, for their continued interest in the
progress of the work, particularly to Dr J. R. Dennett, Mr M. J. Hutchinson,
Mr M. J. Burgess, Mr F. G.
Gray, Miss P. Zalasinski, and Miss A. Holloway,
and to many more geomorphologists than those listed in the bibliography and
text, who, by their absorbing studies, have made a wide reading of the
subject
a most rewarding experience.
In addition to those who have helped in the immediate task of preparing
this book, I would like also to acknowledge those who, at earlier stages,
encouraged me to seek the rewards of studying geography in general and
X Introduction to Geomorphology
landforms in particular; my parents for encouraging me to admire and

explore landscapes and for teaching me to use, value, and to draw maps; my
sixth-form geography master, Mr Ian H. Watts, for showing me how to start
learning for oneself, and the Department of Geology in the University of

Manchester, particularly Dr F. M. Broadhurst, for most informative extra-
mural courses; as an undergraduate the formation of first notions about
geomorphology were influenced by the teaching of my tutor. Dr J. M.
Houston, by Dr R. P. Beckinsale, by Miss M. Marshall’s special option on the
history of the cycle of erosion concept, and by my post-graduate supervisor.
Dr M. M. Sweeting.
I hope that I can make these acknowledgements without associating any of
the above with the shortcomings of this book.
The author and publisher would like to thank the following for permission
to reproduce copyright figures and tables.
Journals! Editors
American Journal of Science for figs: I.4.E (Vol. 260), 1.7 (Vol. 260), II.6
(Vol. 265), II.IO.A (Vol. 256), III.l (Vol. 257), III.17 (Vol. 266), IV. 15.B
(Vol. 268), IV.36.A (Vol. 263), IV. 36.C (Vol. 258), IV. 43.B (Vol. 264),
IV.49.B (Vol. 262), V.l (Vol. 263), V.3.C (Vol. 258), V.5 (Vol. 261), V.7
(Vol. 265), V.27 (Vol. 255); Australian Geographical Studies for figs. I.3.C,
IV.31.A; The Canadian Geographer for fig. IV.24.C; Economic Geology for
figs. IV. 42, V.13; Geografski Glasnik for fig. V.18; Geological Magazine for
figs. I.5.A, I.5.B, I.6.A, IV.30.A, IV.45.D; Nature for figs. III.l, IV.26.B;
Rassegna Speleologica Italiana for fig. V.l 7. A; Science, Volume 132, 1960,
fig. 5. Copyright 1960 by the American Association for the Advancement of
Science, for fig. II.5.A: Scientific American, August 1960, page 88 for fig.
IV. 34.J.
Publishers
Archiv fiir Wissenschaftliche Geographie for figs. IV.45.E, V.6, V.12.D;

Blackwell Scientific Publications for figs. III.IO, III.15.A; Clarendon Press
for figs. 1.8, 1.9, III.12.B, III.29, IV.9.D, IV.26.C, IV.39.B, IV.44.B, V.4,
V. 30, Table III, 2, Table IV.2, Table V.3; Elsevier, Amsterdam, for figs.
1.12, IV.13.A, IV.13.B, IV. 18.D, IV.38; Eyrolles, Paris, for fig. V.25.C from
Plages et cotes de sable by J. Larras; figs. I.4.B, IV.28.A, IV.34.A-D,
IV.36.B adapted from Fluvial Processes in Geomorphology by Luna B.
Leopold,M. Gordon Wolman, and John P. Miller. W. H. Freeman and
Company. Copyright 1964; Generalstabens Litografiska Anstalt for figs.
Acknowledgements xi
11.15.
A, II.15.B, III.8.B, III.13, IV, IV.l, IV.5.A, IV.IO, IV.13.D, IV.15.A,
IV.24.D, IV.24.F, IV.30.B, IV.47.A, IV.47.D, IV.51, IV.53, V.3.A, V.3.B,
Table IV.l, Table IV.9, Table App. 2; Gebruder Borntraeger for figs. I.4.A,
11.15.
fig. App. l.C,
E, II.15.F, IV.7.A, IV.7.B, IV.18.A, IV.33.A, IV.33.B,
fig. App. l.D, Table App. 1; Heinemann Educational
Books Ltd for fig.
IV.6.D redrawn from the original diagram by Mr A. Bartlett of the
Geography Department of the University of Sydney; Liverpool University
Press for fig. V.22; Longman Group Ltd for fig. V.ll.A; fig. IV. 19 from
McGraw-Hill Encyclopedia of Science and Technology 1962, used with per-
mission of McGraw-Hill Book Company; Masson et Cie for figs. III.6.D,
III. 9.A, III.9.B, IV.26.D, IV.26.E, IV.39.A, V.24, Table 1.1; Methuen and
Co for figs. I.6.B, III.3, IV.18.B; Thomas Nelson and Sons Ltd for fig.
IV. 43.A, part of IV. 6.D; Oliver and Boyd for fig. V.3 l.E from Morphology of
theEarth by L. C. King, and Table IV.8 from Principles ofLithogenesis, I by
N. M. Strakhov; Panstowowe Wydawnictwo Naukowe for fig. IV.8.C (from
Przeglad Geograficzny) and fig. IV.9.A (from Geographia Polonica); Prince-
ton University Press for fig. 11. 1; Redakcja Biuletun Peryglacjalny for figs.
IV.8.A, IV.8.B, and IV.39.C; University of Chicago Press for the following
figs, taken from Journal of Geology: 1.2 (and Marie Morisawa), I.4.C (and H.
J. Finkel), II.7 (and R. F. Flint), III.6.B (and E. D. Sneed and R. L. Folk),
III. 6.C (and P. Kuenen), III.12.D (and D. C. Rhoads), III. 18 (and J. S.
Olson), IV.l l.C (and W. O. Kupsch), IV.15.D (and J. C. Harms, et al.),
IV. 20 and IV.40 (and V. P. Zenkovitch), IV.45.C (and M. Demorest), IV.50

(and P. McKenzie), V.21 (and E. Buffington, D. G. Moore), fig. App. 1, Table
IV. 11 (and C. B. Beaty), Table V.2 (and R. Le B. Hooke); Universitets
Forlaget for figs. IV.3.A, IV.12.B, IV.12.C, IV.49.C, V.31.B, Table IV.9;
Van Nostrand Reinhold Co for figs. 11.12, IV.3.B, IV.7.C, Table IV. 12, and
Table IV.13; John Wiley and Sons Inc, for fig. IV.39.E; Wiley-Inter science
for fig. V.19.A.
Learned Societies
American Association of Petroleum Geologists for figs. II.8.B-E, IV.23.A,

IV.23.B, V.l l.B, V.l l.C; American Geophysical Union (and W. B. Langbein
and S. A. Schumm) for fig. I.6.D, and
from Soviet Hydrology, figs. IV.36.D,
IV.44.C, IV.52; figs. IV.18.C and V.12.A reproduced by permission from the
Annals of the Association of American Geographers', Cave Research Group
of Great Britain for fig. IV.26.A; Drzavna zalozbu Slovenije, Ljubljana
for
fig. V.8; Field Studies Council for fig. IV.6.B; Geografisko drustvo Slovenje
for figs. III. 11 and V.l 8;
fig. IV.47.C redrawn from an illustration in the
Journal of Glaciology by permission of the Glaciological

Society; The
Geological Society of America for figs. I.3.B, II.4, II.5.B, IL5.C,
II.9, II.IO.B,
xii Introduction to Geomorphology
II.IO.C, II.I1, III.6.E, III.7.A, III.8.A, III.16.A, IV.2, IV.5, IV.14, IV.21,
IV.25, IV.27.B, IV.28.E, IV.28.F, IV.34.F-H, IV.43.C, IV.46, IV.48.B,
IV.53, V.14, V.23, Table V.l; The Geological Society of London for
figs. I.5.C, I.5.D, III.8.A, III.I2.E, IV.24.E, V.26, V.31.A; Geologiska
Foreningens I Stockholm IV.24A; The Geo-
Forhandlingar for figs.
logists Association for figs. I.IO, IV.6.D, IV.27.A, V.25.B; Houston
Geological Society for figs. IV. 17 (and E. H. Rainwater) IV.28.G (and A. S.

Naidu); Institute of British Geographers for figs. 11.14, IV.8.D, V.15, V.25.A;
International Association of Scientific Hydrology for Table IV.7; Libraire
des Comptes rendus de I’Academie des Sciences, for Tables IV.4 and IV.5;
Macaulay Institute for Soil Research for fig. IV.48.C; New Zealand
Geographical Society for fig. IV.34.E; Royal Geographical Society for the
following taken from Geographical Journal: IV. 12. A (and W. H. Ward),
IV. I3.C (and C. A. Lewis, G. M. Lass), IV.39.D (and M. S. Longuet-Higgs,
D. W. Parkin), V.12.B, V.12.C (and J. C. Doornkamp, P. H. Temple), V.3 l.C
(and D. E. Sugden, B. S. John); Royal Scottish Geographical Society for fig.
IV.3.C; The Royal Society for figs. II.2, II.3; The Royal Society of New
South Wales for fig. IV.31.C; Schweizerische Geologische Gesellschaft for
fig.V.31.D; Society of Economic Paleontologists and Mineralogists for the
following from the Journal of Sedimentary Petrology: II. 8. A (and M.
Gipson), IV.1 LA (and B. J. Bluck), IV. 1 LB (and S. Sengupta), IV.22 (and E.
D. McKee, G. C. Tibbits), IV.23.C (and E. D. McKee and G. C. Tibbits),

IV.32, IV.45.B (and B. J. Bluck), IV.37 (and W. F. Tanner), IV.41 (and D. H.
Yaalon), IV.45.A (and J. P. Allen), IV.49.A (and C. E, Weaver), V.l 3 (and P.
F. Williams, B. R. Rust); Societe Geologique de France for fig. III.16.B; Soil
Science Society of America for figs. I.6.C, 1.13, III.12.A, III.12.C, IV.48.A,
IV.55.B, Table III. 1, Table III. 3, Table IV. 10, Table App. 2; The Yorkshire
Geological Society (and R. Agar) for fig. V. 1 0.
Universities
Bonn University Institute of Geography for figs. IV.6.A, IV.3LD; Florida

State University for figs. III. 18, IV.27.C, IV.28.D, V.2; University of Hull
for figs. III.4.B, III.15.B, IV.9.C, fig. App. 1; Louisiana State University,
Coastal Institute (and F. A. Welder) for fig. I V.3 5; University of Lyon for fig.
IV.7.D; University of Strasbourg for figs. III.7.B, IV.44.A.
Miscellaneous
Figs. II.15.C. Based on Crown copyright Geological Survey diagram. Repro-

duced by permission of the Controller, H.M. Stationery Office; Centre
National de la Recherche Scientifique, Paris for figs. IV.9.B, IV. 16 and
IV.44.D; La Commission geologique de Finlande for figs. IV.24.B and V.20;
Acknowledgements xiii
Department of Mines and Technical Surveys, Ottawa for figs. I.4.D, III.4.C,
IV.47.B from Geographical Bulletin reproduced with permission of
Information Canada; Service des Etudes Scientifique for fig. IV.9; U.S. De-
partment of Agriculture for fig. IV.55.A; Yugoslav Ministry of Defence for
fig. IV.6.C; Professor R. Coque for figs. 11.12, 11.13, III. 14, V.28, V.29;
Professor A. Jahn for figs. IV.33.C, IV.33.D.
Preface
In the last twenty-five years there has been considerable progress in all
branches of science. On a modest scale the part-time science of geomorpho-

logy has achieved a great deal during this time, and the geographers, geo-
logists, hydrologists, and specialists from other fields who all contribute to the
study of geomorphology have become aware of the comparative suddenness
of change within the context of landform studies. Commenting from a geo-
logist’s point of view on the activity in the actual dating of Pleistocene deposits,
F. W. Shotton (1967) found it difficult to realize that almost all this work has
been done in less than two decades. On the geographical side, O. H. K. Spate
(1960) observed that many had been startled by the suddenness of the rise of
quantification. Since 1945 many new techniques in addition to radio-carbon
dating have been utilized. Vertical air photographs gave penetrating insights
into tectonics and led to a demand for certain revisions of traditional geomor-
phological concepts. In 1947 the first systematic offshore acoustic study was
made and since 1955 the profiling of continuous depth surveys of the bottom
topography offshore has become widespread and detailed. Information from
little-known areas has come in. For instance, in the past twenty years much
research on tropical soils has modified profoundly an understanding of
weathering processes. Periglacial features, from being picturesque curiosities
have become accepted as a fundamental phenomenon recognizable in areas
covering more than a third of the earth’s land surface. After 1957-8 the first
reliable estimate of the volume of the Antarctic ice-sheet was the product of
the co-operative investigations of the International Geophysical year. It was
as recently as 1960 that the various glacial stages on the Arctic borders were
worked out, and it is only in the last seven years that a body of measurements
of postglacial submergence and rebound have emerged. Different approaches
to landform study have developed, partly because workers in related sciences
have found the study of geomorphological problems as rewarding as their
own. In 1 945 the engineer R. E. Horton discussed in detail his hydrophysical
approach to drainage basin study which was considered in Europe by P.
Pinchemel and taken up with vigour in the United States by A. N. Strahler
and his students as a possible alternative to W. M. Davis’s approach. In 1943
Preface xv
J.Tricart began his geomorphological investigations but with the deep feeling
that insufficiencies of Davisian geomorphology left him unarmed for the
task.
Today W. M. Davis, still the father of geomorphology for many has become
an Aunt Sally for others. Criticisms and attacks on Davis have taken a variety
of forms, including L. B. Leopold and W. B. Langbein’s attempt to run Davis
down with a thermodynamic engine. In 1959 C. A. M. King’s book on coasts
and beaches appeared, emphasizing the engineers’ use of mathematical theory,

scale models and wave tank experiments to further the study of wave action
and shore processes. It has become increasingly apparent that interest in
landform study stems from more than one source. For those at one extreme,
landforms are essentially a sporting challenge, at the other, a problem in
theoretical physics. Also apparent in much of geomorphological work since
1945 are changes in emphasis. Within this time there has been a rapid
development in the study of slopes, and studies developing from F.
Hjulstrom’s thesis (1935) on the morphological activity of rivers concentrate
on investigations of geomorphological processes. A
traditional concern with
erosional forms increasingly includes the study of mass movements and

depositional features. Recently a new factor, government interests and finan-
cial support from a variety of sources, has benefited geomorphology. For

instance, government-sponsored surveys in Canada and Australia have not
only brought back factual information from little-known areas but have also
provided some geomorphologists with the opportunity of full-time employment
in the study of landforms and with the stimulus of working in close associa-
tion with scientists in related disciplines. Another recent factor has been the
intensification of interest in practical problems. The tackling of coastal engin-
eering problems or the military problems of crossing rough terrain has
provided the geomorphologist with useful information. The investigation of
ny other practical problems from which geomorphology has benefited
include the possibility of influencing climate artificially which has heightened
interest in glaciers which store three-quarters of the Earth’s fresh water. The
surge of interest and major advances in the study of the transport of solutes
and sediment in streams are partly due to increased public awareness of
problems of water supply and water-quality management. In the last two

decades the range of literature available to the geomorphologist has increased.
After the lapse of two geomorphological journals during the Second World
War, the Revue de geomorphologie dynamique first appeared in 1950 and the
Zeitschrift fur Geomorphologie reappeared in 1957. More specialized jour-
nals, the Journal of Glaciology in 1947 and the Biuletyn Peryglacjalny in
1954, have appeared. A most significant recent trend for geomorphologists
depending on sources written in English has been the degree to which journals
published in many countries include an increasingly large number of articles
or long abstracts in English. Also translation services
have made available a
xvi Introduction to Geomorphology
huge store of information on hydrology, soils, and geography compiled by

scientists in the Soviet Union.
Scientific progress has had several repercussions within geomorphology.
The main one. accentuated by the very smallness of the number of scientists
who can devote all their work to landform study, is the tendency for a
multitude of specialized fields of interest to emerge. For instance, from the
geological point of view, the word geomorphology has been prefaced by
words like structural or morphotectonic. Studies orientated towards ‘process
geomorphology’ may include prefixes such as ‘climatic’ or more specifically
‘periglacial’ or ‘tropical’. Studies of past conditions or of relict features
absorb the attention of Pleistocene, stratigraphical, or historical geomorpho-

logists:and some petroleum geologists are increasingly attracted to the study
of paleogeomorphology. Approaches to geomorphology have also been de-
scribed by a variety of terms, sometimes as a slogan by some or as an
implied criticism by others, and include the adjectives quantitative, qualita-
tive, descriptive, dynamic, mathematical, theoretical, geological, geographi-
cal, regional, systematie, and, not least, Davisian. Also, as conflicting con-
clusions of geomorphologists working in different areas may reflect their
differing field experiences and schools of thought, geomorphology may be

linked, for example, with Poland or with Scandinavia. The variety of these
diverse themes raises the questions of their validity, their definition, of where
they might lead, and to what extent and in what way they are complementary,
opposed, or merely tortologous.
Optimistically, it is hoped that the first part of this book will serve as a
guide to the reader in placing his further reading, including the subsequent
sections of the present book, in the context of the rapidly developing and
expanding field of geomorphology. An optimistic approach in attempting to
present a balanced sketch of geomorphology is justified because the very
reasons of pace and variety which militate against the feasibility of this task
make it equally important that some coherent attempt should be made to
search for and to illustrate the underlying themes of landform study. Al-
though the intention is to present a guide-book rather than a textbook, it is
hoped that sufficient detail is become familiar with

included for the reader to
some examples of the findings of more recent work. However, with one
exception it is not intended deliberately to promote certain themes nor syste-
matically to oppose more traditional studies. It is felt that the pace of progress
makes the uncritical adoption of innovations as inadvisable as the dogmatic
adherence to outdated ideas. The sole conscious deliberation has been to
illustrate, if only with a restricted number of examples, the nature of the
actual, observed facts of landform study in preference to summarizing the

stockpile of hypotheses which have yet to be tested against facts yet to be
observed.
I
Definitions, nature and basic

postulates
A. Definitions
Geomorphology is the study of landforms. This obvious statement is difficult
to extend without stating a partial view of the subject. Traditionally the study
was essentially that of the origin and evolution of landforms. Many would still
agree with N. M. Fexuiejiian that ‘the study and interpretation of the records
leftby erosion constitute the larger part of the science of geomorphology’,
and see their work, as S. W. iWooldride e did, as answerable for the last brief
chapter of the geological record. The historical element in geomorphology
is evident from the need to interpret still observable traces of events that
once took place on the earth’s surface. On the other hand, workers like
L. B. Leopold, M._ G. Wolm an. and J. P. Miller (1964), while observing that
‘much of geomorphology is stratigraphic geology’, have made the study

of contemporary processes a conspicuous part of their own approach to
the subject. Much recent work tends increasingly to reflect the definition of
F. Hjulstrom, that geomorphology is the science of landforms and of land-
ferming” processes. Yet another view sees geomorphology not merely as a
study of landforms and processes but stresses the importance of the inter-
actions at the contact surface between the terrestrial parts of the lithosphere
and the liquid and gaseous envelopes which surround it, the ever-changing
inter-adjustments between different forces of nature, and the changes in
appearance of the land surface as the dynamic balance between these forces
is modified.
B. The Nature of Geomorphology

1 . Description and interpretation
Any science starts with observations of natural phenomena. Basic descrip-

tions in g eomorphology are of three kir^ s. being
observational, classificatory.
2 Introduction to Geomorphology
or explanatory-descri ptive in nature. The traditions of German geomorpho-

Jogy emphasize the importance of observation, illustrate t he desire to provide
as detadsd-deseriptions as possible of relie f farm land dateba^ to the work

of von Richthofen ^the 1880s. Secondly, 7ome la ndforms have suffic ient
symmetry and distinctiveness of form to be classified as one of a certain type.

Thus a term lil{e~river meander conveysTmmediately to the reader a large
amount of descriptive information. A third approach is that of genetic de-
scription, which aims at further economy by describing in a word not only
information about the appearance of a form but also implies an interpretation
of its origin, e.g. peneplain. Geomorphology is unusual in the degree to which
this descriptive-explanatory or genetic approach has been employed. Since
1945 there have been changes ways in which observations are re-
in the
corded. with the growing recognition that some degree of schematization is

necessary. Increasingly there have been attempts to economize on lengthy
verbal descriptions, to sharpen the precision of observation with the develop-
ment of specialized mapping techniques, and to search for ways of reducing
observations of landform to numbers. These attempts have included the de-
scription of complicated terrains as well as of simpler symmetrical landforms.
There has also been a movement away from genetic description in an attempt
to separate clearly description from interpretation.
A. Cailleux and J. Tricart (1956) put an initial problem, that of scale, into
perspective by defining a series of size orders. A notion of scale not only
provides a framework in which to categorize observations but also influences
the methods used to collect the observations and which of the related special-
isms, from geophysics to soil science, will have most bearing on the inter-
pretation of the results (Table 1. 1). The actual size of the scale orders is less
important than the stress laid on the significance of scale. Within the size
range of features studied in geomorphology little in detail is known beyond
the general fact that some forms may vary with size, whereas others may not.
For instance, river meanders have the same dimensions in plan regardless
of scale. By contrast, in small-scale sand forms the coarsest material collects
on the crests, whereas the reverse is invariably the case for large-scale
dunes.
Landform maps have been prepared on a variety of scales, over some
larger areas where no detailed contour map is available, or from smaller areas
to record aspects of the geometry of the ground surface, such as discontin-
uities, which a contour map does not show. Fig. I.l.A illustrates morpholo-
gical mapping techniques used by R. A. G. Savigear. A similar approach can
be used in coastal studies, as the symbols used by M. A. Arber (1949) in
describing cliff profiles show (fig. I.I.B). For mapping purposes, coasts might
be divided between cliffed and non-cliffed, with both categories then sub-
divided according to whether their plan is regular or irregular in outline.
Table I.l Classification of geomorphological features [after Tricart, 1965).
Time-
Units of
earth's Characteristics Equivalent Basic mechanisms span of
climatic controlling the persis-
surface of units, with
examples units relief tence
Order in km^
differentiation of 10’
I 10’ continents, ocean large zonal systems
basins controlled by earth’s crust be- years
astronomical fac- tween sial and sima
tors
10^ crustal movements, 10*

II large structural broad climatic
entities (Scandina- types (influence of as in the formation years
vian Shield, Tethys, geographical fac- of geosynclines.
Congo basin) tors on astrono- Climatic influence
mical factors) on dissection
10‘ main structural subdivisions of the tectonic units hav- 10^

III
units (Paris basin. broad climatic ing a link with years
Jura, Massif types, but with little paleogeography;
Massif) significance for ero- erosion rates
sion influenced by litho-
logy
IV 10’ basic tectonic units: regional climates influenced pre- 10’

mountain massifs, influenced predomi- dominantly by years
horsts, fault nantly by geo- tectonic factors;
troughs graphical factors, secondarily by
especially in moun- lithology
tainous areas
limit of isostatic adjustments
V 10 tectonic irregular- local climate, predominance of 10^-

ities, anticlines, influenced by pat- lithology and static 10’
synclines, hills, tern of relief; adret, aspects of structure years
valleys ubac, altitudinal
effects
VI 10-’ landforms; ridges, mesoclimate, predominance of 10"

terraces, cirques, directly linked to processes, years
moraines, debris, the landform, e.g. influenced by litho-
etc. nivation hollow logy
VII 10-^ microforms; soli- microclimate, predominance of 10’

fiuction lobes, poly- directly linked with processes, years
gonal soils, nebka, the form. e.g. lapies influenced by litho-
badland gullies (karren) logy
VIII 10-* microscopic, e.g. micro-environment related to processes

details of solution and to rock texture
and polishing
Further subdivisions could be based on local conditions, such as bay shape,

rock type, and gradient.
Description by the measurement of landforms is for the most part difficult
because of the extremely complex details of form. Furthermore, where a

series of measurements is involved it is not easy to summarize these com-
SCOPE - PORM MAPPING SYMBOLS

'Trrr-fnt Angular convM breok oi slope »»•»«» Cliffs (40* or more)
Angular convex end concove breoks

•
vv tr y
-
Angular concave break of slope tttttttttt of slope too dose together to
ollow the use of separote symbols
.MICRO-
-’A- “A-* Smoothly convex chonge of slope ^SLOPES
Minor indefinite smoothly convex
tttttt concove changes of slope
too dose together to ollow the
- v"v-* Smoothly concove chonge of slope use of separate symbols
EXAMPLE or SYMBOLS FOR SEA -CLIFF TYPES

Hog's- bock _ Flot- topped ttttttttttt Bevelled Low coost
FigureI.l Examples of landform mapping symbols. (Above) Some slope-form
mapping symbols {from Savigear. 1965). (Below) Sea-cliff types (from Arber,
1949).
pactly in diagnostic indices, and because of discontinuities and irregularities,

equations are largely inapplicable. Efforts to describe landform in numbers
are usually based on ratios between various dimensions of the form under
consideration. For instance, R. E. Horton (1932) suggested a simple ratio of
length of drainage basin, L, to its width, W, length being the longest dimen-
Table 1.2 Indices of drainage basin shapes (from Morisawa. 1958).
Drainage basin A B C D
E 0-82 0-86 0-59
C 0-58 0-64 0-47 0-45
L/fV 1-32 0-97 0-50 2-17
area (sq miles) 25-7 18-7 86.0 106
sion from the mouth to the opposite side, with width measured normal to the
length. S. A. Schumm suggested the index E = dlLm^ where E is an elongation
ratio between
d, the diameter of a circle with the same area as the drainage
basin, and L^, the maximum length of the basin parallel to the main river. A
similar index, suggested by J. P. Miller, is C = AblAc, where C is a circularity
ratio between basin area, As, and the area of a circle Ac, having the same
perimeter. Table 1.2 lists calculations of these indices for the basin outlines
Definitions, nature and basic postulates 5
shown in fig. 1.2. However, it is apparent that the measurement of basin length
is problematic when the long axis of the basin trends away from the line
between basin mouth and the opposite side of the basin and that several
differently shaped basins might have similar indices due to irregularities in
their outline. The circularity ratio, for instance, is a description of increased

irregularity of the basin outline rather than an index of basin shape. The
description of landforms is full of dilemmas like these, and the efficiency and
Figure 1.2 Various drainage basin shapes {from Morisawa, 1958)-, these examples
are from the Appalachian plateau.
exact meaning of every measurement and index must be carefully and criti-
cally evaluated. Figs. 1.3 and 1.4 illustrate the definitions on which some other
measurements of landforms have been based. The terrain unit (fig. 1.3) used
by military engineers could have as much geomorphological significance as
the hydrologists’ drainage basin and can be described by ratios similar to
those used for basin shape. The range of examples with the
in fig. 1.4 starts
definition ofdepth of a weathering recess, the simplicity of the index belying
the infrequency with which such observations have been recorded. By
con-
trast the unparalleled regularity of the river meander has been
measured
frequently. However, even those forms which are striking in their geometric
Figure l.S Basic definitions of some descriptive schemes for categorizing land
surface components.
A. The military engineers’ terrain unit {after Krenkel and Hoadley, 1963,
Waterways Exp. Sta. Rep. No. 4-86).
B. Stream-ordering systems for describing drainage composition (a/rer A". L. Bow-
den and J. R. Wallis, 1964, Bull. geol. Soc. Am., Vol. 75).
C. A classification of land surface profiles based on interfluve and valley profiles

(from Ollier, 1 96 7).
regularity may show substantial departures from symmetry, like the con-
trasted lengths of barchan horns. Usually erosional forms, like the nivation
hollow, have greater irregularity still, and several measurements might be
made to avoid ambiguities. Fig. I.4.E illustrates the description of a different
aspect of landforms, relating to their spacing and consistency of orientation,
rather than to their actual size. Although the geometrical regularity of drum-
makes these forms an obvious example, spacing and
lins orientation studies of
more irregular forms and terrain units are feasible.
A type of description, coarser than measurement but easier to define, is the
ranking of features according to their The most widely used example of
size.
ranking in geomorphology is stream orders, by the use of which R. E. Horton

hoped to improve runoff predictions but although some modifications have
been suggested, certain basic difficulties remain in applying the scheme in
geomorphology. Many field workers report that it is sometimes difficult to
follow a definite channel in the field and to find its corresponding representa-
tion on a map where the restrictions of reduced scale limit the details of form
which can be described. Besides being rather coarse, with difficulties at the
outset in standardizing the concept of a first-order channel, the applicability

of the stream-ordering scheme to dry valleys or to relict Pleistocene wadis
indicates that it bears far less relation to contemporary hydrological charac-
teristics than might be supposed.
When describing landform in section, the equivalent to the contour map
which provides a plan view is the surveyed cross-profile, whether it be down-
slope, downstream, or a cross-section of smaller features. Attempts to convert
the graphical description into numbers and indices are perhaps fewer in
number and show less variety than descriptions of shapes in plan. Ratios can
bC'Used in a similar way, however, as between the height and length of a
same fundamental problem of complexity and irregular-
landslip area, but the
ity in the form remains. Distinctive and regular landforms which are easily
categorized in cross-section occupy only a small proportion of the land-

surface area. Recently C. D. Ollier has suggested a nine-fold classification of
relief cross-sections based on the recognition that both interfluves and valleys
might be classified as rounded, flat, or angular. Fig. I.3.C shows that many
diverse landscapes might be described, if not classified, by precise reference to
these simple subdivisions, particularly if the scale and any asymmetry is also
noted.
There are many difficulties in devising precise, objective, and unambiguous
descriptions of the intricacies of landform, and despite the time involved in
such attempts the results often appear to meet with limited success and may
afford little insight. These factors may help to explain why there has been a
reluctance to abandon completely the concise genetic descriptions in geomor-
phology. However, with growing realization of two fundamental limitations in
genetic description, schemes like the classification of coasts developed by
Davis, Gulliver, and Johnson, or the Cycle of Erosion itself are now less
widely favoured. Firstly, the incorporation of new information and theories
into existing interpretations may mean that the original terms are no longer
the same in a descriptive sense. A classic example is the Irish eskers which
gave their name to other long sinuous ridges of sand and gravel only to be
denied the right, in the light of subsequent research, to be called eskers.
Added to the difficulty of including new information in genetic schemes, is the
growing awareness of the danger of circular argument. For example, it was
formerly understood that with the attainment of ‘old age’ a subdued rolling
topography results, and that conversely the presence of subdued rolling ter-
rain implied that a state of old age had been reached in landscape evolution.
Nonetheless the vigorous work in geomorphological mapping in Europe,
Figuiel.4 Examples, from more symmetrical forms, of some measured dimen-
sions in landform study.
A. Depth of cavernous weathering {from Calkin and Cailleux. 1962).
B. Meanders (from Leopold et al„ 1964).
C. Barchan dune {after FinkeL 1959).

distinct from the aims of purely descriptive morphological mapping

advocated by R. A. G. Savigear, embodies decisions on the age, origin, and
sometimes on the erosional agent of the landforms mapped.
Explanation in landform studies poses formidable difficulties which in part
reflect the problems of achieving a precise basic description of even simple
forms. A major problem in searching for explanations of land form is the
intrinsic incompleteness of the geomorphological and geological evidence

which does not usually form a sufficient basis for conclusions deduced. There
is therefore no prospect of proof in the absolute sense as in mathematics.
Interpretations tend inevitably to be qualified by such phrases as ‘tends to’ or
‘possibly’. A further problem in the study of the inter-adjustments between

land form and natural forces changing continually through time is that
isolated causes are usually inseparable from distinct effects. All that can be
recognized is a distinction between the antecedent conditions which affect
subsequent conditions. In addition to the incompleteness of evidence, the
sheer variety and intricacy of inter-relationships between biological,
chemical, physical factors, and antecedent landforms mean that there are no
invariable laws, as in physics, on which to base interpretations of present-day
landforms.
Some geomorphologists who advocate the use of quantitative methods have
solved this problem of explanation to their own satisfaction, by measuring
their success in terms of the variability which they can account for in inter-
correlations between parameters of land form. The explanation thus achieved

is statistical rather than scientific, and in depending on the minimization of
antecedent and subsequent conditions in order to correlate parameters simul-
taneously, eliminates the very aspect of landform study which requires a
causal explanation. Sometimes these studies are clearly influenced by the
technologists’ responsibility to predict rather than the scientists’ interest to
understand, and any geomorphological explanations that emerge from these
studies may be superficial and quite unsurprising.
In geomorphological literature uncertainty, disagreement, and debate
about the origin of specific landforms abound compared with a dearth of
measurements of both form and process. Possibly the influential views of W.
M. Davis on the role of the imagination in geomorphological research may
have encouraged his successors to admire too readily the stimulating spec-
tacle of the highly assumptive reasoner juggling with scanty data. ‘We may
D. Nivation hollow (from F. A. Cook and KG. Raiche, 1962, Geog. Bull.,
No. 18). These dimensions were measured at Resolute, North West Territories,
Canada. DCE is the top of the backwall, DBE the base of the backwall, and DAE
the front of the hollow floor.
E. Drumlin spacing and orientation (.after Reed et al., 1962).

still give praise to those who apply themselves chiefly to gaining first-hand
knowledge of observable facts,’ be allowed, ‘but we have learned to give
greater praise to those who, on a good foundation of visible facts, employ a
well-trained constructive imagination in building ingenious and successful
Figure I.S An illustration of a principal mode of explanation in geomorphology -

the independent use of different methods pointing towards the same conclusion.
Direction of ice movement across East Anglia during the Lowestoft Advance (A)
and during the Gipping Advance (B) deduced from changes in lithological
composition of tills (from D. F. W. Baden-Powell, 1948, Geol. Mag., Vol. 85).
Similarly, direction of ice movement during the Lowestoft Advance (C) and
during the Gipping Advance (D). as indicated by preferred orientation in the
tills (from R. G. West and J. J. Donner, 1956, Quart. Jour. Geol. Soc.,
Vol. 112).
theories which shall bring to sight the invisible facts.’ However, Davis by
assuming that the facts of earth science had, by the end of the nineteenth
century, been ‘abundantly acquired and so thoroughly systematized’ was
unwittingly advocating praise for speculating thinking. If geomorphology has
taken Davis’s disdain for the unimaginative to heart, perhaps his lesson
should be revised so that the work of patient and painstaking description at
1
any level in geomorphology receives more approbation and the unsubstan-

tiated speculations of the lively mind a little less.
‘Judge not according to the appearance.’ St. John VII, 24
2. Process and form
Since 1945 there has been more detailed investigation of processes operative
in landscape development. This trend has led to some disagreement about the
degree to which process studies are an integral part of the study of land form.
This problem of definition can be posed by contrasting two quotations. R. A.
Bagnold (1941) did not think that he would see ‘t he geomorphologist re st
content in his study for its own sake of the shape and movement of sand
knows why sand collects into dunes at all, instead of
accumulations, until he
scattering evenly over the land as do tine grams ol du st, and how dunes
assume and maintain their own especial shapes’. In 1958 S. W. Wooldridge
wrote m difterent vein, for he regarded it ‘as quite fundamental that
Geomorphology is primarily concerned with the interpretation of forms, not
the study of processes. The latter can be left to Physical Geology.’ In follow-
ing this largely morphological approach it is necessary to infer the nature of
the processes from the forms they are believed to have created. However, in
many situations even the identity of the main process may be in doubt.
Increasingly the argument appears that any generalization about past condi-
tions and the possibility of suggesting evolutionary changes depends on the
initial systematic measurement of processes in order to establish and to
understand what is happening today. Together with an increasing emphasis
on the importance of process studies come criticisms of the work of earlier
geomorphologists, particularly that of W. M. Davis, for establishing an
approach to landform study which oversimplified or neglected process
studies. However, there is probably a limit to which W. M. Davis might be
held responsible for any alleged inadequacies which also may underlie a
possible weakness in S. W. Wooldridge’s belief. This is the degree to which
physical geology has not developed, in the interpretation of past processes
and ancient results, comparative studies of modern processes which can be
directly observed. As recently as 1966 W. H. Bradley in his address as the
retiring president of the Geological Society of America suggested that there
was a great deal more to be learnt about geology from observing present-day
processes than had yet been explored. Geology is unique in offering for
investigation, in most circumstances, only the results of natural processes,
like Plutonic igneous complexes. It is therefore not surprising that a strong
tradition of inferring the nature of processes has grown up in many branches
of geology, including geomorphology. However, geomorphologists are
not
alone in realizing that they can observe and measure the actual
processes
involved. For instance, in addition to the developments in the study of con-

temporary processes in sedimentology in the last two decades (J. R. L. Allen,
1968), there is currently an increased interest in studies of living foraminifera
by palaeontologists.
In geomorphology the need for more information on processes has
emerged in a variety of ways. Most significant is the growing appreciation of
uncertainty if not total ignorance of the nature of most processes and the
ways in which these processes and land form interact. For instance, little is
known about sub-glacial mechanisms, and the complex problems of ice move-
ment mechanics may still be regarded as incompletely solved, not to mention
the mechanisms by which ice might erode bedrock. Several important aspects
of fluvial dynamics, such as the nature of turbulence, are not yet well known,
and nor are the relationships between fluviatile processes and the reworking
of depositional landforms. Understanding and knowledge of the highly
dynamic processes on coasts is grossly deficient, and the climate of the soil is
still very poorly understood with the processes of rock decomposition and
weathering in relation to the formation of soils needing much further study. In
particular there is a lack of basic data concerning mineral solubilities and the
factors affecting the stability of the dissolved constituents, particularly in
tropical ground water and soil water.
In view of the rudimentary knowledge in studies of process, some workers
consider it too optimistic to assume that the nature of processes can be
inferred from the land form. There is also the risk of a circular argument
using the evidence of forms to build an hypothesis of their genesis and then to
explain the same forms by this hypothesis. There is also the neglected prob-
lem of ‘homologies’ or ‘convergence’ in landforms, where different processes
may produce similar forms. For example, small circular weathering pits
occur in varied environments and on a larger scale rounded ponds in formerly
glaciated areas might be either kettle holes or ancient pingos. Rounded sum-
mits with tor-like eminences also occur in varied environments, so that those
near Fairbanks in Alaska have to some observers appeared very reminiscent
of tors on Dartmoor. In coastal areas there appears, for example, to be more
than one genetic type of barrier island. In the absence of processes studies the
investigation of homologous forms can lead to insoluble disagreement be-
tween a variety of hypotheses of origin and even in the interpretation of the
field evidence on which these theories are built. Although many striking
processes are usually significant not all fundamental processes are necessarily
either conspicuous or even visible. Therefore, even the observation of pro-
made without measurements, is sometimes insufficient. For instance,
cesses, if
the outward characteristics of a river channel are not usually diagnostic of

whether a channel is in the process of aggrading or degrading.
Another reason for the increased interest in process studies stems from the
need, in addition to understanding how a form was produced, to use this

assumed knowledge as a step towards a broader interpretation or reconstruc-
tion. For instance, without understanding clearly the ways in which shore
platforms develop at present, conclusions drawn from them about relative

movements of land and sea are uncertain. Similarly, by studying the condi-
tionsunder which a present-day sediment develops, such as frost action

producing patterned features in frozen ground, the knowledge gained can then
be used to interpret fossil structures of similar origin in reconstructing past
environmental conditions.
There are innumerable difficulties in studying geomorphological processes
and in attempting to integrate the information acquired into the interpretation
of landform. The tackling of these difficulties would call for too great a
diversion from the essential purpose of studying actual land form, in the view
of some workers, whereas for others these difficulties merely re-emphasize the
unlikelihood of making adequate inferences about the mode of landform
development from a study of the forms alone. One difficulty is the slowness of
operation of some processes in relation to the time-span which one investi-
gator can expect to devote to their study. For example, one of the criteria in
C. F. S. Sharpe’s definition of soil creep is that it may be imperceptible except

to measurements of long duration. Frequently the complexity of the forces
involved is another major difficulty. It is the complexity, for instance, of
beach movements that makes the detailed nature of its processes so little
understood (C. A. M. King, 1959), for here there are four distinctly different
hydrodynamic zones which are overlapping and compressed into a relatively
narrow strip between the backshore and the nearshore. A third problem is
that while most processes are theoretically observable, in practice it is diffi-
cult, if not dangerous, to do so. Moreover, the physical risk tends inevitably
to increase at times when mechanical processes are at their most active and in
areas where biochemical activity is most vigorous. Thus a simple and observ-
able process like cut and fill in the zone of breaking waves and offshore, or in
streams, is not well-documented because the points are difficult to reach even
at low water.
There are several problems in evaluating the results of process studies. In
many areas there is the difficulty, if interpretations of the past are to be made,
of estimating the degree to which present-day erosion rates are due to arti-
ficially accelerated erosion since man has swung the scales of the natural
balance between the forces previously involved. In all situations there is the
problem of deciding which came first; whether the process essentially
preceded the form and thereby determined its characteristics, the converse, or
whether the two co-exist in a closely interdependent state of ‘dynamic equilib-
rium’. Many workers consider that there have been too many changes par-
ticularly in the last 2 million years, for the last suggestion to be anything
more than a theoretical, remote possibility. S. A. Schumm and R. W. Lichty

consider that the problem of antecedent and subsequent conditions and the
possibility of states of balance depend largely on the time-scale involved.
They take the example of helicoidal flow measured at river bends and suggest
that in the short term, the meander is an anterior condition and that the
observed helicoidal flow exists because of the meander. However, the mean-
der itself, if considered in a broader time context, may exist partly because of
different antecedent conditions of flow. Even where a given form is seen to be

the result of a known process, it may be difficult to ascertain how much the
form owes to the present intensity of the process and how much to slightly
different conditions in the past. An example of this problem is the size of the
solutional nip on limestone coasts in the tropics where the size might in part
be due to slight changes in past sea-levels. A similar problem is to decide

whether the characteristic features observed along a river valley are in fact
relicts of infrequent floods and not the result of day-to-day conditions of

normal flow. For slowly evolving micro-features in active periglacial areas it
is difficult to apportion the contribution of the present climate compared with

that of slightly different climates a century ago. On a larger scale, P. Birot has
reservations about supposing that the inselberg is typical of dry-season tropi-
cal areas because it cannot be assumed that these forms were created under
the same conditions as exist today.
In addition to the difficulty of being unable to infer processes from forms,

there is the restriction that forms cannot be inferred from processes. In
consequence the study of sediments has grown within the context of landform
studies to assist in bridging the gap between process and form. The study of
sediments involves not only depositional structures like gravel lenses but also
the measurement of statistical parameters describing characteristics of groups
of individual particles. The value of the study of sediment characteristics is
that they are in part a product of the same processes which operate on the
landforms, and these characteristics, in part reflecting the rock which makes
up the landform, also may accelerate or retard the rate at which processes
operate and may influence their mode of operation. For example, the
dynamics of the swash zone on a beach, although difficult to observe in
action, are revealed in part by grain-size distributions across the foreshore
slope. In turn the size may influence the rate of infiltration of the swash into
the sediment. Even in environments where mechanical action is much
reduced, the fact that many sedimentary particles will then have spent con-
siderable amounts of time in a weathering profile means that the importance
of process, climate, and time might be inferred from chemical variations.
Thus, just as the usefulness of a study of contemporary processes extends to
interpretations of the past, so also can the study of modern sediments, by
revealing characteristic properties developed in distinctive environments cast
light on the nature of past environments discernible from the characteristics

of fossil sediments. In Northamptonshire where sediments, recognized as fills
in former ice-wedges, descend to 12 m together with other disturbances struc-
tures, it is possible to suggest that permafrost affected the ground to depths in
the order of 30 m and to suspect that solifiuction could have been a factor
in the development of the present-day slope forms. Former terraces can be
correlated by the statistical characteristics of sediments and by primary

sedimentary structures in alluvial fills. In 1860 A. Geike, from his studies of
sedimentary deposits, made one of the first reconstructions of a former
process, by advocating that glacial deposition of boulder clay seemed prefer-
able to C. Lyell’s concept of deposition from icebergs. In fact it was W. M.
Davis who in 1890 was probably the first to recognize and describe backset
bedding in his studies of fiuvioglacial outwash deposits in New England, and
one may wonder whether the definition of geomorphology by the pure mor-
phologist may overlook the degree to which the study of sediments has always
been an integral part of landform study.
The study of processes and sediments in geomorphology raises two further
points. The first is the very real risk of not only venturing beyond what is
strictly required for landform interpretation but also in consequence claiming

for, or handing to, the geomorphologist studies that lie more appropriately in
other hands. In the study of sediment transport in rivers, for instance, the
hydrodynamicists and hydraulicists are often primarily concerned with
energy relationships between particles and the transporting fluid, whereas the
geomorphologist is more concerned simply with the amount of material
moved and with the problem of locating from where in the catchment the
material was eroded. Many geomorphologists will be happy to agree with a
leading expert on the transportation dynamics of sedimentary particles who,
while recognizing that the geomorphologist would inquire into how and why
processes produce distinctive forms, held the personal view that ‘the subject
of sand movement lies far more in the realm of physics than of geomorpho-
logy’ (R. A. Bagnold).
The limit is more drawn in the case of glaciology as the realization

easily
that glacier flow is one of the most useful illustrations of plastic flow of a
crystalline material has led to very intensive study well beyond the bounds of
the geomorphologist. The fact that the flow of ice is perhaps understood better
than the flow of any metal or rock material and that geomorphologists still do
not know how ice erodes is perhaps a significant demonstration of the degree
to which the study of land form depends on the study of ice-flow
physics.
A further point raised by the increased emphasis on process in landform
studies is the nature of ‘dynamic’ geomorphology which in a general way
relates to the interactions of form, process, and soil. However, there is a
substantial contrast in emphasis between the American and the European
view of ‘dynamic’ geomorphology. The dynamic basis of geomorphology,

according to A. N. Strahler, depends on basic physical properties of both
materials and processes and their varying responses to applied shear stress.
The interest is in how the physics of a process operates and in speculations on
how physics might be applied to geomorphology. In Europe the view of
‘dynamic’ geomorphology as is evident from the work of A. Cailleux, J.
Tricart, and others, is more confined to inquiring where given processes
operate and how much work they achieve at present and to what extent the
areas affected and the work done were different in the past. A further charac-
teristic of their approach is to stress the importance of the effects of chemical
and biological as well as mechanical processes.

It can be concluded that while landform is the subject of primary interest in
geomorphology and is the object of study unique to the subject, the-inter-

pretation of a given problem of regional geomorphology or the systematic
study of a distinctive form involves the geomorphologist in the study of
processes as well. Ideally the geomorphologist tries to infer processes from
the forms and also to consider what forms the processes studied might create.
Due to intrinsic limitations in the evidence available neither can be achieved

satisfactorily and the danger of circular argument always threatens when the
conclusions from one approach are confronted with those from the other.
Lending confidence to the geomorphologist during his hypothetical leaps
between form and process is the flimsy safety net provided by the study of
sediments.
3. Artistic and scientific elements
Geomorphology might be described as a part-time science not just because

nearly all its exponents have other commitments in science, education, or
technology but also because artistic elements of geomorphology itself and in
thetemperament of the investigator have been a significant factor in its study.

Fundamental to an understanding of geomorphology is an appreciation of the
widely held opinion, which is at times a loudly pronounced belief, that
familiarity with the countryside, coast, hills, and mountains initiates, nur-
tures. and develops an interest in land form. For many leading geomorpholo-
gists a deep sense of the beauty of the natural landscape has been the inspira-
tion of a life-time’s devotion to its study. Artistic sensitivity and ability is not,
however, confined to the subjective investigator who strides exhilarated to the
wind-swept summit to scan the horizon for evidence to fit his theory. Nor is
there much
evidence to suggest that recent advances in scientific aspects of the
subject have been at the expense of these artistic elements. A. Rapp begins the
presentation of his exhaustive records and measurements from northern
Lapland: ‘The valley of Karkevagge is rich in interesting problems and in
7
Definitions, nature and basic postdates 1
natural beauty. Nine summers have not been enough to make me feel tired of
it.’ An ardent concern for conservation was an expression of a love of nature
felt by J. P. Miller, a pioneer in the use of quantitative methods in landform
study, and A. N. Strahler shows in his textbook an outstanding artistic skill in
the diagrammatic drawing of landforms.

A literary sense is particularly important in geomorphology with workers
compiling a rapidly growing range of qualitative descriptions of unique fea-
tures from contrasted environments. In developing new interpretations, skill
is required in defining new terms with precision, clarity, and with due regard
for previously proposed terms. Partly because of the fundamental importance
of regional comparisons in landform study the linguistic ability of geomor-
phologists who have translated work from foreign languages is valuable.
Judging from the frequency with which their work is cited in bibliographies,
one of the most important contributions to geomorphology for many students

in the 1950s was Czech and Boswell’s translation into English of W. Penck’s
Die Morphologische Analyse. Despite the obvious need for literary skill in
recording the facts of a qualitative science, and style in presenting its argu-
ments, there may be some truth in criticisms that some geomorphologists
have been carried away by a flow of words. K. Bryan (1940) wrote of ‘mild
intoxication on the limpid prose of Davis’s remarkable essays’ influencing the
geomorphologist to consider the question of slopes in words, not knowledge,
and with phrases, not by critical observation. Certainly Davis wrote his
Essays, believed in the power of words and was fascinated by them. He con-
tributed in 1 926 to the journal Science an essay entitled ‘The value of outrageous
hypotheses in geology’, writing at length on the interesting etymology of the
word ‘outrageous’ in preference to using a less ambiguous synonym in his
title. Commonly used in the writings of W. M. Davis and in those who
similarly have strived for literary effect are two devices which, in appealing
directly to the imagination, require scientific restraint in their interpretation.
First, the appeal of brief, vivid, and perhaps poetic expression has led both
popular and technical writers to animate the landscape by the skilful use of
metaphor (LeGrand, 1960). The mental picture which these words portray
may, however, be quite unrelated to the dynamics of the process. A more
detailed inquiry into the mechanics of apparently self-explanatory headward
erosion reveals little beyond the fact that it breaks, bites, or even gnaws back
into a divide. Secondly, due to the intrinsic incompleteness of evidence for
geomorphological interpretations the analogy is used so frequently that its
well-known limitation is Sometimes, however, ill-chosen
easily overlooked.
analogies are unmistakable applicable in one or two senses only and the
imagination may then be overloaded with innumerable impressions of how
the geomorphological factor differs from analogous
the phenomenon.
Analogies extend beyond the written word into the domain of quantitative
Enalyscs, pErticulErly where some rcEsoned bEsis for e mEthemEticsI model is
sought. AgEin the suElogies require very csreful scrutiny. For instance, early
estimates of the viscosity of lava ‘streams’ were based on the analogy between
these flows and that of river water, and it was the mistaken assumption that
hydraulic equations could be applied to lava.

The metaphor and the analogy have influenced geomorphology profoundly.
The collection of drumlins into a basket of eggs may appear a harmless way
of describing a distinctive form, like many of the other kitchen-sink images in
geomorphology. Nonetheless R. J. Chorley (1958) was unable to resist the
temptation to liken the process of drumlin formation to that of egg-laying, a

suggestion incorporated into later serious studies.
Favouring the development of another artistic quality in geomorphology’s
character is the passage of time which makes the studies of those with an
historian’s expertise and objectivity increasingly useful. With the greater
interest in processes has come the realization of difficulties in studying
changes over a span of a few years, in artificially modified and scaled-down
environments, or in theoretical speculations on possible time-sequence
changes of form. There is therefore a growing interest in critical searches for
records of events, like delta growth, datable over spans of decades or even
centuries. For example, in the search for references to Spurn Head, a promi-
nent spit at the mouth of the much-sailed Humber. G. de Boer (1964) has
searched through records spanning more than a millennium. Historical
analysis of the patterns of development and decline of ideas within the study
of landforms also becomes increasingly valuable as the passage of time leaves
a growing length of history of geomorphology stretching back from the
present day. Detailed studies like that of R. J. Chorley, A. J. Dunn, and R. P.
Beckinsale (1964), or of G. L. Davies (1968). provide helpful aids to an
understanding of contemporary geomorphology and in contemplating a
future growth which might avoid past mistakes.
It seems that recent advances in more scientific aspects of geomorphology
have not diminished the need for some artistic flair among its exponents.
Their usefulness has developed also, but has been urged less than the scientific
element where the need for growth has been reiterated for more than two
generations. One of the first in the post- 1945 period was R. P. Sharp’s
suggestion that the practising geomorphologist, in addition to being a good
geologist, should be well-versed in soil science and acquainted with principles
of hydraulics, meteorology, oceanography, and botany. He also thought that
the application of physics and geoph 3'sics to geomorphological research
would offer the greatest promises for fundamental advance.
It would seem
would clash with the idiom of the Davisian

that inevitably this philosophj'
word-picture approach. However, Davis himself had said at the beginning of
the present century that the methods of mathematics and phj'sics were ‘so
9
much needed in all parts of earth science’ and I. C. Russell had added
astronomy and chemistry to the list of desirable assets. Therefore if Davis and
his followers appear to some contemporary critics to have had no
more than
‘a butterfly-catcher’s sort of interest in scenery’, the apparent absence of
scientific rigour in the past and the sensation of its impact at the present are
not easily explained without skilled and sympathetic historical analysis.
In understanding recent scientific progress in geomorphology it is helpful
to recognize several distinct themes which equally can be described as scien-
tific. First, scientific method, with its rigorous procedure from first-hand
observation to interpretation has always been more important than some
discussions, preoccupied with the Davisian approach, might either allow or
illustrate. Secondly, the application to geomorphological problems of scien-
tific techniques borrowed from neighbouring sciences, such as soil science
and hydrology, increases markedly. This trend leads to a third aspect of
science in geomorphology but one which is impossible to define in general
terms, and is, some extent, a matter of personal preference, with the
to
definitions of some proving unacceptable to others. This is the degree to
which the study of geomorphology by including the investigation of pro-
cesses. thereby necessitates also the study of basic mechanical and chemical
principles. The extreme view is that of some theorists who go so far as to
believe that most geomorphological generalizations should in fact or in prin-
ciple be deductively derivable from the laws of physics or of chemistry.

Others would deny this on the grounds of the complexity, significance of
changes with time and the importance of biological factors. It is only fair to
note that the first view is that of only a few, and includes most of the theorists
whose familiarity either with the work of those with broader definitions or
with actual field experience is conspicuously limited.
With the increased adoption of scientific techniques and the attempts to
apply the principles of certain sciences to the interpretation of land form, the
distinction between science and technology has to be emphasized as a fourth
aspect of scientific aspects in the study of landform. The geomorphologist has
to make a deliberate effort to recognize this distinction in order to avoid
claiming the engineer’s role or adopting techniques useful in engineering but
which have little relevance to the solution of his scientific problem. Compared
with science, which proceeds from observation to understanding, technology

proceeds quickly through observation and understanding and then on to meet
the practical demands for application and control. The technologist cannot
wait either for exhaustive measurements or for the development of
a com-
plete, substantiated theory. He has to economize on the first two stages which
intrigue the scientist and make do with modifications to readily available data
often already collected for other purposes,
or derived from scale-model
experiments, theoretical models, or trial-and-error methods.
With certain
simplifying assumptions, the engineer arrives at conclusions which, although

ina scientific sense are significantly dififerent from those existing in the actual
problem, are well within margins set by safety and economic considerations,
and allow work on the project to start on schedule.
Although there are many examples in recent geomorphological literature
of the blurring of science with technology, the problem of defining the objec-
tives of space research has recently provided a sharp and general reminder of
the fundamental distinction between scientific and technological endeavour.
Clearly there is a limit to what the geomorphologist might gain from the
experience and methods of technologists concerned with engineering in the
natural environment. For example, although for practical purposes in
hydraulic engineering river discharge is the most important parameter, the
hydraulic factors measured in conjunction with flow determination are not

sufficiently inclusive for geomorphological or even for hydrological analysis
of the river system (Leopold and Skibitzke, 1967). Conversely, where a
geomorphologist contributes to an engineering project there is an equally
clear limit on the degree to which he can claim that the technological achieve-
ment is relevant to geomorphology.
Geomorphologists also have to be cautious before using the term ‘science’
itself too liberally, because unlike the rigorous sciences of physics and
chemistry landform study knows no laws, if the word law is used in its correct
universal sense, as a concise expression of a group of inter-relationships that
have been repeatedly observed to be consistent. G. K. Gilbert’s ‘law’ of
divides or R. E. Horton’s ‘laws’ of drainage network arrangement, although
giving the superficial impression of scientific simplicity are no more than
useful generalizations. They admit many exceptions, and the areas to which
they might be relevant may require careful selection and delimitation to
exclude domains where immense complexity and infinite local variety pre-
vails. Unlike the biological sciences there are no intrinsic features like genes
with which to provide a known link for a series of forms through time or for a
division between species, and unlike the social sciences there is no possibility
of dialogue between investigator and investigated. Geomorphology certainly
has an important and growing scientific content, but to claim for it the full
status of a science would be not so much pretentious as premature and might
even misrepresent its true unique nature.
4. Qualitative and quantitative aspects
The complexity and irregularity of forms and processes complicated further

by changes with time, makes an essentially qualitative approach the only
realistic way of approaching many geomorphological problems. However,
despite the emphasis on the power of words laid by W. M. Davis and

1
reinforced by and on the accumulation of a vast store of

his followers,
qualitative records by European geomorphologists, since 1945 there has been

an increasing realization that in some instances forms and particularly pro-
cesses are sufficiently regular for more exact approaches and, in other
instances, that qualitative analysis has proceeded as far as it can go. The
clarificationbrought to time-scale problems, whose solution was impossible
to resolve on the basis of earlier vague controls, by the measuring of recent
geological time, provides a considerable incentive for searchers for more

exact methods.
Increasingly geomorphologists are defining their views on the desirability,
scope, and hopes of this trend. Extremists include a few who are too hopeful
in their claims, contrasting with disclaimers at the opposite extreme who

occasionally lapse into hostility. Many, however, recognize that in attempts to
retard or to promote quantification, the reactions provoked by extreme views

tend to balance each other out at any instant in time and merely provide a
scatter of points about an inexorable, upward trend. Most views therefore
range from enthusiastic involvement to sceptical interest.
Mackin’s words summarize the misgivings of many, particularly those of
the older, more experienced workers;
... the very act of making measurements, in a fixed pattern, provides a

solid sense of accomplishment. If the measurements are complicated,
involving unusual techniques and apparatus and a special jargon, they give
the investigator a good feeling of belonging to an elite group, and of
pushing back the frontiers. Presentation of the results is simplified by use
of mathematical shorthand, and even though nine out of ten interested
geologists do not read that shorthand with ease, the author can be sure that
seven out of the ten will at least be impressed. It is an advantage or
disadvantage of mathematical shorthand, depending on the point of view,
that things can be said in equations, impressively, even arrogantly, which
are so nonsensical that they would embarrass even the author if spelled out
in words.
An illustration of a more extreme, but still significant viewpoint comes

from a note entitled ‘Robot geology’ in which Link (1954) confesses that
‘Somehow I hate to see budding geologists feeding machines numbers. They
should be out in the field learning to get the right
numbers to put into the
infernal machines.’ From
a geographical viewpoint, O. H. K. Spate (1960)
feared that expertise in a fashionable technique might disguise
an essential
poverty of thought, and feared the possibilities of the dogged analysis
of trivia,
the smoke-screen of formulae and the elaborate
discovery of the well known.
The reader encountering the conclusion that ‘large rivers have large bends
and small rivers have small bends’ may appreciate these forewarnings. More
recently where computer package programs make intricate statistical analyses
available to anyone with a few numbers, pointless complications increase to

include the instant and enormous output of the unintelligible. In general there
is a suspicion that taking to quantitative techniques dulls one’s geomorpho-

logical common sense and perceptiveness. Clearly an understanding of contem-
porary geomorphology depends on some familiarity with the reasons which
justify this unease as well as on illustration of the possibilities of greater
precision which this approach affords. Before discussing quantification in
geomorphology several basic facts should be stated. First, any antithesis

between qualitative and quantitative aspects of geomorphology is largely false
due to their essential interdependence. Quantitative geomorphology indicates
an emphasis or even provides a slogan rather than a definition for a separate
specialism within the subject. Secondly, the quantitative element in geomor-
phology differs substantially depending on whether form, process, or time is
involved. In process studies long-established familiarity with quantitative

techniques already exists in subjects like hydrology, hydraulics, agricultural
engineering, soil chemistry, or sedimentology. In the measurement of time
many disciplines like archaeology, botany, and geochemistry similarly pro-
vide a range of techniques and experience in their use. By contrast the
quantification of land form is basically the responsibility of the geomorpho-
logist. Consequently there is little experience available in related subjects and
quantitative studies of form are more rudimentary and those of the functional
inter-relationships between form and process very few. Thirdly, since many of
the investigations made primarily to demonstrate that some geomorpholo-
gical problems are susceptible to a quantitative approach tackle problems that
are peripheral to the bulk of questions that the geomorphologist has tradition-
ally asked, their results are often of limited interest and difficult to integrate
into existing studies. Finally investigators may use the adjective ‘quantitative’
in widely differing senses. This is because quantification involves several
distinct stages and the term could be applied to any of one of these. It has
been used to describe work ranging from the careful measurements of an
experimenter in the field to the speculations of armchair theorists.
1. MEASUREMENT
The value of qualitative reasoning behind the collection of data is one of the
most important aspects of quantification, and the importance of design before
rather than after measuring cannot be overstressed.
(i) Definition of measurement. S. A. Schumm (1960) suggested that one of

the most difficult problems to resolve in studies of sediment types is the
selection of a parameter representative of the physical property of the sedi-
ment. The difficulty of expressing the erodibility of soils by single numerals
has also been stressed. In overcoming this initial difficulty, some arbitrary
decision, such as deciding at which point a cirque floor is judged to pass

upward into the steepened headwall, is necessary. Many geomorphologists
unaccustomed to experimental procedure have remained entirely qualitative
in their approach because they were unable to accept this arbitrary schemati-
zation, which would lead to the measurement of something that was neither
‘natural’ nor ‘real’. Attempts to overcome this problem by defining a feature
to be measured in some genetic sense, like the interpretation of erosion
surfaces as relics of a once continuous erosional plain, can, due to the

intrinsic uncertainty in geomorphological interpretations, usually be coun-
teredby arguments rejecting the genetic interpretation.
The fundamental problem of defining a parameter to measure is difficult
enough for static forms, but if the phenomena move, as in process and sedi-
ment studies, some changes elude definition by being too slow or too rapid.
(ii) Making a measurement. In addition to the definition of a measurement,

with its inevitable approximations, there is the problem of how to measure the
parameter adequately, with hopes of accuracy limited by the intrinsic varia-
bility of the phenomena. Also, for some variables, like the cohesiveness of a
sediment in relation to its resistance to erosion, no field method has been
developed. Ceaseless motion, as on the shore, makes it difficult to measure sea
state in any long-term consistent manner or to collect samples in the surf.
Movements in some less dynamic environments, like the bottom of a glacier

or river, are very difficult to measure. In fact, the most serious hindrance to a
complete understanding of sedimentary processes in general is probably the
lack of instrumentation by which short-term velocity and pressure variations
many others where existing
can be accurately recorded. In this situation, as in
methods appear crude, some workers prefer to await the development of more
refined methods. However, with more sophisticated apparatus the greater
delicacy requires precautions and even restrictions in its uê, may involve
patient calibration or the risk of expensive breakages or that equipment and
markers employed in continuous measurements may be accidentally
disturbed. It may also be impossible to tackle some problems geographically
with expensive apparatus as the number of fixed points at which records can
be made is inevitably limited.
Another problem in making measurements is the huge scale on which
measurements have to be collected in order to encompass the complexity of
inter-relationships in the natural environment and the range of geographi-
cal variability. For example, experiments on the scale of those with which
the engineer, R. E. Horton, was involved in 1939. relating to the effects
of
rain intensity, rain impact, and energy on infiltration capacity depending on
192 infiltrometer tests on 96 plots, could be very difficult to equal for
geomorphoiogists who are mostly restricted to part-time research. Another

example is the 1 962 resolution of an international commission on snow and
ice: ‘Our primary aim is to encourage all countries which have glaciers to
make very simple measurements on as many glaciers as possible. We wish to
emphasize this point; measurements should not be confined to a few glaciers
in each country, but should cover the maximum possible number.’ Whether
measurements are concentrated in a small area, or extend to include con-
trasted environments, longitudinal analysis through time is usually required
as well, the 1000 years of recorded water-levels on the River Nile being one
of the most sobering examples.
(iii) Second-hand measurements. One common way of making some progress

where sufficient first-hand measurements are unobtainable for practical or
financial reasons is to use measurements derived from information collected

for another purpose. The derivation from maps of numbers to describe land
form is a well-known example. Fluvial geomorphoiogists have used the data
collected by water engineers in preference to actual measurements of the
parameters which theoretically are most meaningful from a scientific point of
view because only information of first, approximate kind is available in large

volumes. Similarly nearly all geomorphology includes climatic data which
being influenced by certain meteorological abstractions may give only an
approximate indication of conditions in the soil where weathering processes
operate.
(iv) Criticisms of measurements. Any set of geomorphological measurements

may be open to the criticisms of vague definition, inaccuracy, triviality, or
irrelevance. They may be too small in number, collected from too few places,
and observed too infrequently. Critics will also correctly warn of the risk of
unsuspected artificial disturbances to the feature or to the recording device.
The collection of numbers may be biased unintentionally in one direction or
may subconsciously tend to support a preconceived view, or the omission of
concomitant measurements of possible related phenomena may reduce their
value.
In 1896 W. M. Davis wrote, ‘Verbal descriptions are so insufficient in
geographical teaching that supplementary illustrations, in the forms of maps,
views and models, must be employed as far as possible.’ Significantly, he did
not include measurements in this list nor in the publication of his research
during the subsequent forty years. Perhaps he knew better.
2. CORRELATION
The measurement of the degree of association between two geomorphological
variables involves many difficulties. A common problem is that within a range
between the largest and smallest observations of a sample, the bulk of the
measurements may bunch near one end of the scale. For instance, in a
drumlin field the smallest drumlin may only be half the size of that of the
majority, which the largest may exceed by several times. This raises statistical
problems, starting with the simple fact that an average value is not really
representative.
The relationship between two geomorphological phenomena, as is fre-
quently the case in nature, is rarely that of a straight line. Behaviour may even
change suddenly when a certain threshold value is exceeded or with the
introduction of a new factor at a particular time or beyond a certain distance.
For instance, clear quartz sand 0-08 mm in size was observed to slip when
the surface is tipped at an angle very little different from that for larger grains,
when dry, but when exposed to air of normal humidity it tends to stand in
vertical walls.
The many curvilinear relationships where y is increasing with values in x

further increases in x beyond a certain point may actually lead to decreases in
y. Fig. I.6.A illustrates how, in streams incising an alluvial terrace, the grand
mean for valley-side slopes increases with stream order up to the third order,
then diminishes at higher orders; Frame B shows how progressively slower
wind velocities are required to initiate the movement of smaller sand particles,
while for sand particles less than 0- 1 mm diameter, greater velocities are
required. In fig. 1.6.C soil loss increases with the length of time that rain falls
until the inwashing of finer particles produces a protective crust which
reduces the amount of erosion. Similarly, in fig. I.6.D, increasing annual
amounts of effective precipitation in the United States produces greater
erosion until values of 38 cms or more are sufficient to support a complete
vegetation cover which arrests erosion.
Finally, even if a geomorphological factor and a given environmental
variable are unlike these examples by being normally distributed and linearly
related, the complexities of the natural environment will make it impossible to
obtain an excellent correlation.
3. EQUATIONS
Proceeding from the demonstration of a relationship between measurements
of two or more variables, the derivation of a mathematical equation may
provide a useful, concise summary of the finding. However, the precision of
invariable mathematical relationships rarely exists in geomorphology, and
average statistical correlatioqs as generalized descriptions are often as far as
one can proceed without introducing simplifying assumptions that would
reduce the relevance of the results to the natural environment. Mathematical
relationships cannot allow realistically for the possibility of
changes with
time, like climatic changes in the Pleistocene or over
much shorter intervals.
0. O'
o V
1i
1, c 30
X3
-
20
2400
1 Oeêrv shrub
1800 2 Grassland
0 4 3 Forest
I « 1200
0 3
0 2
0 1
20 40 60 80 0 25 50 75 100 125 150

0
Duration ol simulated rainfall «n mnutes Effective precipitation in centimetres
Figine ! 6 Some non monotonic relationships in landform studies
A Variation of mean valley side slope angles with stream order {fiom Carter and
Choile). 1961). for the Lighthouse Hollow drainage network, Poquonock.
Connecticut
B Variation of the threshold drag \elocily in air with particle size {Jrom Bagnold,
1941)
C. Soil loss in runoff water during a 90 minute simulated rainfall application

(Jtom E Epstein and IF J Grant. 1967. Soil Sci. Soc. Amer. Proc Fol SI) The ,
soil IS a Slit loam from Maine
D. Variation m sediment yield with varying amounts of efTective precipitation

{ajtei Langbein and Schuinin. 1958) The data relates to all climatic regions of the
United States,
as this means that certain parameters cannot be assumed to have remained

constant, nor for the ‘kaleidoscope of hydrologic conditions caused by hetero-
geneities of geology’ (LeGrand, 1962). Even where an equation satisfactorily
describes the situation in one case, such as the transportation of sediment in a
river, variations in the stream-depth flow velocity, bed material, bedrock, and
historical legacies make equations very different for each river. Rarely do
predictions from one equation check with the data from another area.
Nevertheless, extensions of the work of R. E. Horton on the geometrical
properties of drainage basins have led to demonstrations of an invariability in
these properties which justifies mathematical treatment. For A. N. Strahler
and his students Horton’s work has been of particular and vital interest. For
others it is an irrelevant appendix. Many are mystified by the invariability of
the mathematical revelations when their own experience of landforms leads

them to anticipate large statistical variations in nature. Since most of the
mathematics applied to geomorphological data involves Hortonian studies, it
is necessary to search for reasons to explain why some find invariable
relations while others believe them to be the exception in nature.

First an air of over-optimism has surrounded the topic ever since Horton
used the word ‘law’ to describe more statistical relationships and geomor-
phologists assumed that a technique designed primarily to improve runoff
predictions would be suitable for landform study. Some workers, while
observing that Horton’s technique of classifying stream order is subjective
and hard to replicate, still refer to the approach as a ‘touchstone’. A great deal
of thework has been spurred on by the initial assumption that mathematics
can be applied to landform study and the deliberate adoption of Horton’s
ideas to demonstrate this.
Secondly, the over-optimism includes a lack of concern about the basic
quality of the data. For the purist the orders of a ranking system would be too
coarse to manipulate arithmetically. A more practical point is the readiness
with which the same stream may differ by an order or more, depending on the
scale of the map, or whether, on the same scale, the map is old compared with
a recently revised planimetric version. Some channels observable in the field
do not appear on maps.

Thirdly, the use of the term ‘law’ perhaps misleads students and makes
them less appreciative of the full significance of what M. Morisawa (1964)
termed ‘aberrations’ from Horton’s laws. A stream may not. in fact, have its
steepest gradient in the first-order segments ‘as is necessary to comply with
Horton’s law’.
Fourthly, patterns develop clearly only on certain types of homogeneous
rocks. Differing lithologies in a drainage basin introduce irregularities and
within a homogeneous rock even weak structural control of stream networks
may influence length ratios. Problems of lakes fed by multiple inlets and
islands bounded by branching channels pose difficulties.
A fifth factor follows from the presence of geological and hydrological
irregularities, in that most workers avoid these difficulties. However, not all
workers are as specific as M. A. Melton (1958) in defining the criteria used to

select study areas. In his study, basins with cliffs, well-developed floodplains
for the main channels, obvious compound slopes, extensively gullied slopes or
trenched channels were avoided. Many Hortonian studies relate to badland
gullies in weakly consolidated sediments, where miniature
patterns are
organized quickly. Because Horton’s scheme makes it necessary to avoid
basins, lakes, and islands, an artifact of the technique automatically excludes

‘ill-organized’ drainage.
A sixth point is that some of the Hortonian relationships and constant

ratios are unrelated to hydrological parameters. They apply not only to
badlands in the United States but also to areas of tropical lowlands in S.E.
Asia submerged by Pleistocene seas, to tidal channels, to chalk dry valleys,
and to the relict pattern of Pleistocene wadis in the Sahara. The point to
emphasize perhaps not the wonder of this mathematical invariability which

is
has confused some theorists about the fundamental nature of geomorphology

but that techniques which cannot discriminate between such contrasted areas
are of limited use to the geomorphologist.
In conclusion it seems that the mathematical regularity demonstrated by
the Hortonian approach refers only to certain, limited areas and that some of
the ‘laws’, like the ‘Law of Stream numbers’ is a statistical probability func-
tion, applicable toany regularly branching system, and is derived not from
the landscape but simply follows as an automatic consequence of the defini-
tion of stream order.
The use of the analogy between a landform and a geometrically defined
shape has a long tradition dating back to the delta and involves fewer con-
fused issues. The degree to which a mathematically defined shape fits a form is
obvious to anyone as is the fact that departures from that shape are real and
not merely a statistically random scatter about that shape. Almost invariably
only an analogy of shape between the curve and the landform is implied, like
Galileo’s suggestion that the long profile of streams approaches a quarter
circle. Workers who have attempted to use empirically fitted curves to recon-
struct extensions of former river long profiles soon realized that the expres-
sions have no geomorphological The inevitability of irregular-
significance.
ities in most profiles due to resistant rock outcrops means that at best a profile
is segmented into a series of curves that would each require separate equa-
tions. The same restriction applies to smooth coastlines between headlands or

to most hillslopes. Even where a smoothed segment is close to mathematical
simplicity a given set of constants usually apply only to one segment and so
have little use for comparative studies.
An agricultural scientist, Troeh (1965) has considered briefly the applica-
tion of three-dimensional formulae to landform and it is not surprising that he
concludes that the complexity of a series ofhills and valleys is so great that it
would be impossible to describe them even approximately with a three-

dimensional equation, that it would be difficult to give any meaning to the
coefficients and that items of interest such as slope gradient would not be
readily obtained from them.
If order in nature extends to the shape of landforms then, like the processes
and the inter-relations between process and form, this approaches mathema-
tical regularity too infrequently to be of more than academic interest. How-
ever, in recent years workers, assuming that the landscape and the processes
operating on should be definable in the terms of the laws of physics, have

it
constructed theoretical mathematical models. These equations should be

sharply distinguished from those which summarize large amounts of actual
measurements because their values are not observations but assumptions and
predictions from the laws of physics. This approach is mainly of interest to
hydrologists and most geomorphologists will probably find the basic assump-
tions unacceptably oversimplified for their purposes. In addition to its inher-
ent complexity, landscape evolution involves as much chemistry as physics
and theoretical ideas about the nature of the landscape nurtured in laboratory-
controlled sciences traditionally underestimate the biological factor. The huge

cost of hydraulic engineering scale models and the trial-and-error procedures
involved are a measure of the degree to which the inter-relationships between
water and landform defy mathematization.
A third type of formula used mainly by technologists occasionally appears
in geomorphological literature and may be based partly on measurements
and partly on theoretical considerations. Its main use is for predicting
approximate conditions where engineering problems are concerned, and,

instead of summarizing measurements, is rather a substitute for them. Highly
skilful work can reduce a comparison between prediction and observation to
about a 100 per cent error in civil engineering which is invaluable to the
engineer but of less use to the geomorphologist. However, when a geomor-
phologist is going to collect measurements it is sometimes very useful to have
available some empirical formulae which might indicate the approximate
range of values to be expected. One device used in deriving empirical formula
for technological purposes should be guarded against in scientific interpreta-
This is the use of the same measurements on both sides of an equation,

tions.
which increases the degree of apparent association between two parameters.

As an hypothetical example a graph may show the relationship between A and
B X W CD, when A might have been defined

EF
in fact earlier as . Stream
D
width is one of many examples that might be cast in the role of D.
5. Laboratory and field work
The first geomorphologists were explorers or the field geologists preparing

maps on reconnaissance surveys. Today the model geomorphologist is for
many still cast in this mould, striding forth equipped with well-dubbined field
boots, a sketch pad, and a well-trained roving ‘eye for the country’. For
larger-scale surveys the Land-Rover geomorphologist has become acceptable,
driving vast distances across little-known terrain. With growing emphasis on
measurement and scientific method the field geomorphologist wonders if
laboratory experiments and remote sensing of the environment
by satellite
will confine him indoors, and the armchair theorist hopes for added justifica-
tion for framing his view of landforms with four square wails.
The fundamental geomorphological experience is that there is an indefin-
able advantage of actually looking at the realities of land form. It is difficult to
explain the logical basis for this belief beyond making the slightly arrogant
but obvious assertion that the linking of the human eye to the human brain
can provide the practised student of landform with an aid of potentially
immense power. Because of the importance of actually observing landform, it
follows that a wide range of experience is preferable both in the search for
generalizations and also to heighten appreciation of the distinctive features
in a given area. This essentially geographical approach was familiar to
Herodotus whose remarkable generalizations on river activity included ex-
amples from modern Greece, Turkey, and Egypt, and innumerable funda-
mental geomorphological facts have been established in this way. For ex-
ample, A. Cailleux, on visiting Alaska, observed the round lakes near Fair-
banks, including one surrounded by half a dozen smaller round ponds 3— 10 m
in diameter, and could record that an identical pattern had been observed
near Tartarin, in the vicinity of Paris.
The mere fact of being in the field increases the chances of the discovery of
crucial evidence either as a distinctive assemblage of forms or an unusually
complete stratigraphical section. As recently as 1960 one of the most com-
plete Pleistocene successions in the western Mediterranean was discovered on
an almost inaccessible section of the south-east coast of Spain (Butzer and
Cuerda, 1 962). There is also the chance that a field investigator might witness
an infrequent event. W. M. Davis advised that it was hardly worth sitting
down and waiting for a rare event like a sheetflood, but on the other hand,
some of the observations that McGee (1897) was able to make, after being
caught in one more than a metre deep, have been of lasting importance.
Since 1 945 the emphasis in some aspects of landform study has tended to
pass from exploration to more systematic research. A stream-dissected hill-
side, as well as being pleasant to view, becomes also a pile of unanswered

questions riddled with problematic features. The problems are not so much
those of interpretation but of how to approach interpretations. The eye for the
country alone is no longer sufficient due to differing preconceptions in the
minds of observers who are looking at the same feature. The decision to make
measurements, however, introduces several difficulties which are scarcely
soluble. To be as representative as possible and to allow him to use prob-
ability theory in stating the accuracy of his measurements and correlations,
the geomorphologist knows that his sample must be random in either space or
time. But this ideal approach is scarcely adopted when experience indicates
the locality where measurements would be most significant, when it is pos-
sible to visit fifty localities near roads within a few hours and use bulky or
1
sensitive apparatus instead of spending a day walking, with little equipment,

to a remote randomly located point on a mountain top, and to have no
comparable measurements for that day. On the other hand, experience both in
the United States and in Russia is that nearly all hydrological instruments are
sited conveniently in the lowlands, with the result that very few measurements
are to hand from headwater areas that are hydrologically and hydrochemi-
cally of greatest importance. It also means that without sediment measure-
ments there is no information for river reaches where gravel is the dominant
bed material and incorporates unstable fragments or minerals. In turn this
makes suggestions about graded streams speculative.
There is the added dilemma that in cooler environments it is far easier to
make very intensive measurements in summer than to collect a single set in
winter, and the wet season in the tropics is a similar limitation. In many cases
the most significant geomorphological changes may occur at times when the
study area is least accessible. For instance, the best time to inspect material
from an eroding coastal cliff is in late winter after a severe storm. This,
theoretically, is also a critical time for collecting samples in the zone of
breaking waves. Even inconvenient hours in the day may be critical. Ostrem,
in excavating ice-cored moraines, found that pit-digging had to be done partly
at night because with sun shining on the pit walls, melted masses slumped
down. Time for surveying hot desert forms by the rapid develop-
is limited
ment of heat haze. The discomfort of extremes of temperature, rain, and wind,
or sand in the eyes, can make accuracy in calibration and measurement a
vexing task. Accidental disturbance of recording equipment is irksome to
those working in populated areas. Several workers find that their enthusiasm
for measurement of geomorphological phenomena is shared by members of
the local fauna who invariably show keen interest and, where possible, inspect
too closely any recording device.
The fieldgeomorphologist sometimes has the additional problem of devis-
ing a method of measurement where no appropriate one exists or is too
expensive to apply on a wide scale. Students of shore processes, with the
ingenuity of castaways, are particularly adept at solving such problems, and
without too much self-consciousness, have traced current movements with
cement blocks, bricks, and seaweed pods, have observed wave motion with
the aid of grapefruit and table-tennis balls and collected sediment samples
with ice-cream cartons.
Traditionally, the geomorphologist, while happy to spend hours indoors
poring over maps, has been averse to compressing the scale of his forms to
create laboratory models. However, C. A. M. King (1959) considered
that
engineering scale models do simulate with reasonable accuracy the changes
on natural beaches, and recently interest in the results obtained by engineers
and geologists from scale models has grown, even if financial and technical
limitations make feasible only the construction of very simple geomorpholo-

gical scale models. One attraction of this approach involves a resort to the
Baconian procedure of studying only one factor at a time, and keeping
constant all others, which is impossible in field conditions where several
factors vary simultaneously. Also attracted are those who are uncertain about
how a landform changes with time and see in scale models a possible means of
accelerating time. However, while the accuracy of scale models is adequate
for engineering purposes, their aid in assisting a scientific understanding of
the processes and inter-relationships involved is limited by too many artifici-
alities to mention. Particle sizes cannot be scaled down because smaller silt-
sized particles behave in a different way from sands. Clays are not used
because flume distances are far too short for clay-sized particles to settle. The
depth of a model has to be exaggerated sufficiently to make turbulent flow
possible. Erosion in a flume or tank of randomly packed sieve fractions may
not be very closely analogous to the natural erosion of an exactingly selected
and packed natural sediment. On the other hand, not many studies used mixed
sizes anyway, which is a serious handicap to an attempt to understand fully
erosion, transportation, and deposition. Added to this is the fact that differ-
ently shaped particles react differently to water flow. Again the fundamental
difference between the aims of the technologist to predict and the natural
scientist to interpret must be stressed. Extension of interpretations of scale
models to the context of natural conditions are difficult because of unanswer-
able doubts on how close the simulation of natural conditions might be. There
is widespread agreement that laboratory studies will provide qualitative con-
clusions only, and that these are usually applicable within narrow limits only.
The conclusion to be drawn from any scale model study is unequivocable;
any interpretation depends on further measurements made in the field.
The observation of chemical, as opposed to physical changes in the labora-
tory has been standard practice in geology for many years. Scale problems
are not involved, but one essential difference between laboratory and natural
chemical reactions is the period of time over which the processes operate.
Materials which show negligible solution under controlled laboratory condi-
tions may, over long geological periods, be eliminated from a weathering
profile. A major difference is the sterility under which most of these experi-
ments are conducted, compared with the natural environment teeming with
life and life processes, where chemical reactions can no longer operate in
isolation. Experiments with physical properties not involving scale reduc-
tions, like soil shear strength, freeze-thaw cycles, or permeability have little
quantitative significance if determined in the laboratory.

All laboratory experiments discussed so far involving the dynamic interac-
tion of parameters should be distinguished clearly from laboratory analyses
where certain properties, like the proportion of clay in a soil or the estimation
of the amount of sediment in a river-water sample involve only sampling

errors in the difference between laboratory and field conditions.
Space research, with promises of environmental observations from satel-
lites, far from threatening to make the field geomorphologist redundant,

assumption by attaching some importance to visual obser-
reiterates his basic
vations from manned spacecraft. With the increasing emphasis on the import-
ance of field measurements and sample collection for laboratory analysis, the
traditionally excellent geomorphological excuse for getting out of doors is
still sound. Measurement and analysis, however, may be tedious and time-
consuming, and sample collection may involve the restriction of systematic
visits to the same point. The adventure and sporting challenge to physique
and perseverance for those who prefer to spend nine-tenths of the day pitch-
ing and breaking camp is today perhaps all the more clearly defined by the
growing awareness of the paucity of geomorphological field measurements

and the possibilities of laboratory analysis of field samples retrieved adding
clarity to explanations. As an engineer recently remarked, ‘data collection in
the mountains is unconventional, expensive, difficult and not without a certain
amount of hazard to personnel and equipment’ (Meier, 1967).
It seems arguable that scientific method in geomorphology does not involve
the telescoping of time and the compression of size of problems in order to
squeeze them inside the walls of a laboratory, but rather to take laboratory
techniques and approach, with suitable modifications out into the field, and to
return with measurements and samples. Increasingly, more time may be spent
on analysis in the laboratory, but it seems that geomorphology remains
fundamentally a field study, with an immense amount of field work yet to be
started.
6. The role of geomorphology
One of the more frequent uses of geomorphological studies in the past has
W. Wooldridge, like W. M. Davis,
been as an aid to geological studies. S.
considered that landforms were the best indicators of earth history in most
parts of the earth’s surface, and he believed that geomorphology should
provide the answers to the last brief chapters of geological time. This was the
aim of the reconstructions of many denudation chronologies in Britain in the
1950s. However, S. W. Wooldridge spoke only a couple of years after the
realization of the potential usefulness of radiocarbon dating,and in the last
two decades techniques of absolute age determination using radioactive decay
rates have relieved the geomorphologist of much of the
responsibility for
chronologies and its perplexing task of attempting correlations of age,
using
as evidence only the form of the land or its altitude. In conjunction with
information from other studies, facts about geomorphological
features and
their interpretations remain useful to geologists in a few specific situations.

The degree of dissection of glacial features assists the Pleistocene geologist in
his attempts to differentiate boulder clays and moraines of different ages. The
same may apply to areas of recent volcanic activity. For instance, in the
Pribilof Islands, Alaska, St George Island is nearly stripped of primary

volcanic features, and is bordered by marine terraces and high sea cliffs. On
another island, St Paul, the volcanic topography is still well-preserved. The
youngest rock dated on St George has a potassium-argon age of 1 6 million
years compared with an age of less than 100 000 years for the youngest
dating on St Paul. Thus confirmed by radiochronometry, geomorphology
shows clearly that volcanism terminated sooner on St George. Another ex-
ample is the intention of W. Penck to use slope-form studies to elucidate tectonic
movements. Although Penck’s hopes are now largely regarded as unrealistic,
W. Q. Kennedy (1962) still believes, with Africa in mind, that geomorpho-

logy might reflect the tectonic conditions that prevailed during the evolution
of the surface relief. Also there are some specific instances where the inter-
pretation of geomorphological features can provide evidence of tectonic

movements. As early as 1886 Le Conte noted that in the Sierra Nevada
volcanic rocks have preserved Tertiary river channels. These ancient allu-
viated channels have steep gradients of about 15-19 in/km compared with
approximately a 6 m/km fall of the actively eroding contemporary rivers,
a contrast strongly suggesting tilting. Also there is a differing ratio between
the two gradients according to compass orientation which could indicate
direction of tilt. To avoid over-emphasizing the broader significance of this
striking example it should be added that, unlike most landforms, these chan-
nels contained gold. Interpretations of landforms are important in some
geological mapping problems where failure to recognize a paleoform, such as
the ancient landslides in the formations along the San Andreas fault, may
cause stratigraphic confusion to puzzle more than one generation. In South
Australia, the Eden fault was recently reinterpreted as a former sea-cliff.
Intensive drilling for oil has shown that buried unconformities are not neces-
sarily peneplane-like surfaces, but on tilted strata may retain the ridge and
vale forms of a scarpland which provide oil traps. The detailed study of raised
beach elevations in which there is a great deal of current interest in Britain
(Walton, 1968) provides another example. From this data the reconstruction
of postglacial isostatic recovery of formerly glaciated areas provides geophy-
sists studying isostasy with invaluable information on the internal strength of
the earth’s crust.
In using interpretations of landforms the geologist, together with the geo-
grapher who queries all aspects of landscape, stand largely alone. Admittedly
studies of soil formation, particularly those in the tropics, would benefit were
the age of a land surface known approximately, as studies in Australia have
shown; might be speculated that the length of weathering time on a land

it
surface could be related to hydrological studies of the permeability of an

aquifer or of a dam site and a potential reservoir area. In general, however,
much of the utility of geomorphological information to a large number of
related studies depends on no more than the descriptive facts, often obtainable
from a contour map. For example, the natural movement of water on the
ground surface and into and out of the ground depends on landforms, as was
demonstrated with measurements by water engineers in the 1930s, and today
‘many of the observational features to be used in solving hydrological prob-
lems lie within the scope of geomorphology’ (LeGrand, 1962). Within a
given area of homogeneous rock-soil types tend to form a predictable
sequence or ‘catena’ downslopes, as was first demonstrated by G. M. Milne in
1936. This relationship, and similarly where vegetation changes are involved,
is often due largely or indirectly to slope form through its influence on
moisture conditions in the soil. Some geographers, including several in the
United States, would argue that the study of man-land relationships may
require only a knowledge of the form of the land, although the importance of
understanding processes in order to appreciate man’s effect on these would be
emphasized by other geographers.
As well as his description and interpretation of landforms the geomor-
phologist’s study of processes may also be useful. However, interest in ero-
sion, transport, and deposition is shared with other disciplines and often
follows their approach and utilizes their techniques. In this context the work
of a geomorphologist may be useful, but it might be confusing to claim that
such work enhances any intrinsic usefulness of landform study.
Applied geomorphology describes the deliberate attempt to concentrate
geomorphological expertise on the solution of practical problems. Although
the term ‘economic physiography’ cropped up in this context more than sixty
years ago, it is only since 1945, with the development of planned economies
in East European countries, that geomorphologists have searched for practi-
cal problems to solve. Despite government policies, the incentive of possible
financial support, the hope and will to do some practical good, or the fear that
changing times are not going to be kind to the scientist who insists on seeking
knowledge for its own sake, the possibilities of applying specifically geomor-
phological expertise to practical problems are limited. One reason is that
problems like landform stability were already in the experienced hands of the
civil engineer and engineering geologist before the word geomorphology was
introduced. However, particularly by collecting
field measurements to supple-
ment theoretical predictions, some geomorphologists working alongside
engineers have made important contributions to practical studies of land-
forms, like the erosion of coastal cliffs. The practical application
of the
detailed geomorphological map is one of the examples most frequently quoted
as an illustration of applied geomorphology. East European workers describe

how, from these maps, other maps can be compiled giving information about
landforms favourable or unfavourable for farming, communication, or hous-
ing. However, with the growing familiarity with process studies and equipped
with a distinctive array of basic inter-disciplinary techniques, the geomor-
phologist, while hesitating to define his work as applied geomorphology, finds
that he can contribute to the study of many of the most pressing practical
problems. One example is the disruptive effect of periglacial processes on
man-made structures in high latitudes or at high altitudes. Another is the
problem of sedimentation in rivers, reservoirs, and irrigation and water-
power plant The supply of this sediment is frequently augmented by
intakes.
artificially accelerated erosion which constitutes a closely linked problem. In
the little-known areas of both dense and sparse population, the geomorpho-
logist can make useful contributions to the work of survey teams describing
and evaluating land capacity.

By far the largest amount of effort and interest in geomorphology comes
from teachers and students. This may justify the suggestion that by far the
most practical role for geomorphoiogy is in education. In its pre-

measurement phase geomorphologists may have continued needless contro-
versies about the relative merits of speculative theories on the grounds that
lively discussions were stimulating. However, increases in knowledge within
recent years, while by no means raising geomorphoiogy to the standard of a
discipline, have meant that landform study has begun to make distinctive
demands on the mind as well as filling the lungs with fresh air. While some
aspects are susceptible to the rigour of scientific method, imagination alone
can provide some of the tentative links which might string fragmented evi-
dence together. Amid form and complexities of
the lawless irregularities of
process, where the words cause and effect do not apply, where scales range
from thousands of miles to a micron, where there is very limited prospect of
mathematical regularity and where even measurement is often very difficult,
one learns to assess possibilities and to be cautious of generalizations. One
learns not only that the environment is dynamic, not static, but also that man
can trigger off catastrophic changes very easily, and further, that the full
significance of such triggered changes is usually only apparent some time

later. To anyone who has studied the inexorable progress of erosion, this
concept of change with time is elementary, whereas the challenge of day-to-
day thinking in four dimensions is not a feature of many disciplines.
In addition to the chronic pressures of food and water shortage in many
areas is the problem in others where general awareness of damage to the
environment has only recently made demands for conservation and control
arouse anxieties about the dangers of uncontrolled exploitation. As geomor-
phologists are increasingly interested in process studies and therefore need to
adopt techniques which, although mostly borrowed from related specialisms,

constitute by their distinctive range an invaluable aid for the study of prac-
tical problems involving soiland water in relation to slope, geology, and man,
students of the subject, after their academic training, are beginning to make
contributions of growing practical significance.

However, those who have pointed to the weaknesses in the work of leaders
like Davis, Penck, and Wooldridge will recall that the first, in devising a
‘Cycle of Erosion’, wanted a geomorphology to use in teaching, for the
second its use was interpreting geological structure and Wooldridge’s aim
was its use in completing geological history. It seems in retrospect that a
deliberate attempt to make geomorphology useful may be mistaken. But by
asserting that the fundamental role of geomorphology is simply the scientific
one of description and interpretation of landforms, one not only keeps the
unique core of the subject in full view, but also the growing use of interdisci-
plinary techniques associated with the increasing appreciation of the impor-
tance of the study of processes, equips students with some expertise that they
could subsequently apply to the collection of data relating to some funda-
mental practical problems.
C. Basic Postulates
‘The action of a single night of extraordinary rain has crumbled it away and made
It bare soil.’ Plato
‘It is to be concluded that what may be natural for one part of the geologic column
may not be for another.’ M. M. Leighton, 1958
1. Catastrophism and uniformitarianism
The view of earth history proposed by the Catastrophists of the early nine-
teenth century was of a succession of abrupt upheavals culminating in a
great Flood. These paroxysms were interpretated as the result of Divine
intervention. In contrast, C. Lyell and J. Hutton favoured slow changes due
to natural processes and considered that interpretations of earth history coufd
be based on present-day evidence. Geology developed from their work, and

A. Geike’s maxim, ‘theTpreserit is the key to the past’, is often quoted, perhaps
partly because the phrase is little longer than the word Uniformitarianism.
However, work in geology and geomorphology is sometimes less con-
sistent with Geike’s maxim, now essentially of historical interest, than its
frequent quotation may suggest. For example, since endogenic processes are
unobservable, for many geologists the present is not so much a key as one
number in a combination lock; geomorphologists have reconstructed many
ancient peneplains without the aid of an uncontroversial present-day
example
to serve as a model. There is growing appreciation of a non-uniform if not

catastrophic element in the rates of both endogenetic and exogenetic changes.
One reason for this is perhaps due to improved communications and with an
extension of increasingly dense inhabited areas there is an accumulation of
eye-witness accounts, added to every year, of some spots on the earth’s crust
changing abruptly in a matter of seconds or minutes. The second reason is
that if the term abrupt replaces the subjective ‘catastrophic’ adjective, abrupt
as opposed to uniform change is seen to be an essentially relative concept,
depending entirely on the time span involved. For instance, within the span of
geological time the Quaternary glaciations were abrupt. Compared with the
10 million years of Pliocene time, the huge Wiirm ice-sheet disappeared in an
abrupt 10 000 years. If it is clear that time-spans much longer than a lifetime
are implied, statements in the Catastrophic idiom become scientifically
acceptable today. For instance, in a recent interpretation of the formation of

the Snake river basin in Oligocene times, D. I. Axelrod (1968) describes ‘the
basin opening up as an ever widening rent’. Therefore reviewing the signifi-

cance of abrupt events is made difficult by the inevitability of using terms
which are relative, depending on a time-span which is often scarcely definable.
Even a catastrophe, by implication a natural force capable of destroying life
and property, is not necessarily moving more rapidly than an average human
being can run. There are some very slow movements which, in the absence of
self-regulating mechanisms, continue to move persistently in one direction.
These may outstrip man’s technical advance, and with cumulative effect may
become a catastrophe. In this sense earth history as we see it. with a rise of
130 m in sea-level between 15 000 and 5000 B.P.. did in fact terminate in a
great ‘flood’, which for unimaginative, uninventive, and technologically back-
ward islanders could have been catastrophic. To have any meaning, a descrip-
tion of the relative abruptness and intensity of a catastrophic event must also
hint at its context in some broadly indicated period of time; similarly for the
time over which an abrupt event recurs sufficiently frequently to be regarded
as commonplace and part of a uniform rate. For periods of a few years or
less, a recurrence interval might be stated accurately, especially in an annual
context. At places where the same catastrophic movement has occurred more
than once, it might be assumed that a recurrence interval can then be
estimated on the assumption that these abrupt movements themselves recur
with a certain uniformity in time. This has been observed to be the case for
rockfalls on the Kiruna-Narvik railway between 1902 and 1960 and is a
basic tenet in the prediction of floods. For longer periods, words appropri-
ately lacking in precision include a lifetime, implying a number of decades,
with a span of a few centuries implicit in the phrase ‘historical time’. Occur-
rence within postglacial times implies a span of a few millennia, but events
catastrophic in the context of a few 10 000s of years might have to be
described as such. At the other extreme geological time implies upwards of a

few million years and Quaternary time implies several 100 000s of years.
In the study of present-day surface forms and sub-aerjal processes the
geomorplfologisrfori6wijTiore~" closely 'IKe notions Included un der _the_broad
heading of ‘Uniformitarianism’ than the geologist conce rned^wkhieconstruc-
tib n He Has to consider how he may extra pola te fromjhe^present
of the pa st.
to the past and How abrupt as opposed to un iforifTrates of ch ange could affect
these extrapolations. Similarly fiie*considers the degree to which the present,
witlTits abrupt and'uniform ch’anges 'compounded, is" typical of the geological
past intpjvhich h e has to delve. In another sense helnay distinguish his
approach as the reverse of that of the geologist by regarding the past record of
sediments and paleoforms as a check on his interpretations of present forms
and proce sses.
Extrapolation from the present to the ^trmXapposition to the reverse
process of prediction, has been call ed p^stdicticm. From_^^bseryatipns,of

present processes, postdiction involves two distinct concepts.. First, a calcula-
tion of present-day erosion is multiplied by some time-span, often 1 000 years,
to gain some idea of the time that an appreciable chan’ge in landform might
take. Clearly the figures are notional, like the counts of modern pollen rain
needed for fossil pollen studies. Their main value is in the order of erosion
indicated and whether this indicates that associated forms are being produced
at present or whether very feeble erosion rates suggest that they are relict
forms. Studies of hurricanes off Florida provide an example of the use of

arithmetic postdiction. In the last fifty years eight storms affecting south
Florida have been of hurricane intensity. Postdictions from these observa-
tions suggest that about 160 hurricanes could occur
1000 years, and that
in
hurricanes were commonplace in the context of Pleistocene time when
1 60 000—320 000 might have occurred.
A second, distinct aspect of postdiction involves the application of the
appropriate knowledge of contemporary process-form inter-relationships to
the interpretation of past events when similar inter-relationships presumably
existed, although perhaps with some difference due to unique local conditions
and possibly on a different scale. For this reason the geomorphologist who
studies landforms in formerly glaciated areas also makes first-hand observa-
tions on the margins of contemporary glaciers.
One assuming present processes to be the same as those
difficulty in
operating over a much
longer time is in deciding on how much significance to
attach to slow but uniform denudation compared with the intensity
of an
abrupt event. For instance, in the upper Rhine valley in Switzerland,
the late-
glacial rockfall near Flims moved material 1540
times farther horizontally
and 140 times farther vertically than calculations of the
annual movement
achieved by present processes. In the 1953 storm surge
3-m-high cliffs
receded 27 m along a 1-5-km sector of the Suffolk coast at Covehithe

(Williams, 1956). In contrast to this abrupt change, field and laboratory
studies suggest that the major portion of sand transport along some beaches
at relatively slow rates throughout the year, rather than by occasional large
storm waves which erode coastal cliffs. Clearly the ratio between the signifi-
cance of uniform and abrupt mechanical processes varies from one environ-
ment to the next, depends on the time-span considered. Beyond indicating the
problem for each individual case-study, general statements are not yet pos-
sible, although the uniformity of chemical processes is conspicuous.
In considering the nature of catastrophic changes at least five aspects are
distinguishable. First, there is usually a protracted period during which a
potentially unstable situation slowly develops. This would be the gradual
accumulation of unconsolidated materials that would offer little resistance to
a subsequent sudden onrush or downpour resulting from extreme atmo-
spheric disturbances. In Louisiana the 1957 storm surge elevated and trans-
ported entire portions of mudflats, including one segment 2 km long, part-way
across the adjoining marsh. Alternatively, where gravity is directly involved,
a long period of weathering finally brings a potentially unstable geological
structure to the brink of catastrophic disequilibrium, or snow accumulates to
form avalanches in spring melts, or rivers and lakes may inch towards levels
of abrupt overspill. The overflow of Pleistocene Lake Bonneville is a classic
example, reconstructed by G. K. Gilbert in 1890. Recent excavations
first
suggest that downstream the river flood was over 1 20 m deep and of the order
of 280 000 mVsec and moved boulders of more than 6 m in diameter. The
discharge compares with the present peak flood runoff in the vicinity
recorded in 1910, 2800 mVsec. A common type of abrupt movement of flows
is where an initial small barrier, by intercepting more material, becomes self-
accentuating so that when the barrier is finally overwhelmed, it is by a surge

of comparatively large force. A carefully recorded example relates to a raft of
logs in the Colorado river, Texas, first By 1824
the raft was 4
noted in 1690.
km long and had reached 66 km by 1925. When released in 1929 a
in length
flood swept the raft into Matagorda Bay where a delta 2930 ha in extent was
built within a dozen years. There is also the possibility of active structural
movements leading imperceptibly to a gravity-driven catastrophe. The diver-
sion of the Brahmaputra as partly the result of gradual tilting of the
Madhupur block has been suggested. In short, some lengthy gradual process
is usually a prerequisite for catastrophic movements.
A second aspect of abrupt events is the triggering mechanism which finally
releases gravity-driven movements of material, snow, or water. Again either
exogenetic or endogenetic processes could be responsible as either the result
of unusual atmospheric disturbances or due to earthquakes in tectonically
unstable areas. In situations of chronic disequilibrium the energy released by
the triggeringmechanism need only be small. In Hokkaido, Japan, falls of

rain of only 50—100 mm are sufficient to release landslides. In many areas
man unintentionally has become the main triggering geomorphological
mechanism.
A clearly defined third aspect of catastrophic change is endogenetic. the
volcanic eruption which combines the features of slow build-up to the point of
eruption and some trigger to effect its release.
A fourth aspect of catastrophic events is the chain reaction which the

results of an initial movement may initiate. This is often due to the constric-
tions of dissected mountainous relief where catastrophic movements are in-

evitably most common. For instance, in1958, a landslide of 30 million m^ of
material, triggeredby an earthquake along the Fairweather fault, Alaska, fell
into Lituya Bay. In turn this slide created a wave which swept up to an
altitude of 530 m. on the opposite shore. The disturbance of ocean water too
may follow earthquakes, landslides, or volcanic eruptions, and huge waves

hundreds of kilometres in length may proceed at velocities up to 800
km/hour.
A fifth factor is that any ratio between the significance of catastrophic and
uniform processes would vary from one environment to the next, even where
these may be in close juxtaposition like the sea cliff and beach, the river cliff
and bed, and would be very difficult to estimate.

Assuming that the present-day forms are adequately described and that an
impression of processes includes a representative range of continuous uni-
form movements and discontinuous rapid ones, postdiction involves the fur-
ther consideration of considering to what extent the. prgsent_obseE.vations are
typmal of past cond itions. This problem applies equally to investigations
where the study area is believed to represent closely past conditions in
another area as to studies interpreting the landforms of the study area itself. It
is important to distinguish between the erosional and the tectonic aspects of

this problem, as the latter will influence rates of erosion and deposition. The
present discussion is confined to the problem of whether in a relatively stable
tectonic area the present is typical of the past, and whether areas at present
involving tectonic movement are typical of similar areas in the past.
Strakhov suggests that the average rate of sediment accumulation in basins
of the geological past and the limits of variation of that rate generally fit very
well within the norm of present-day sedimentation, but due to man’s influence
on land, present erosion rates may be on average twice that of rates prior to
the advent of man (Douglas, 1967). In certain localities, like highlands in the
humid tropics, man’s interference was sufficient to produce chronic disequi-
librium. Man’s interference is not Just restricted to the widespread ploughing
of recent decades, and the difficulties of calculating erosion rates due to water
abstraction, but extends back for centuries when man first started to destroy
the natural vegetation. Farther back in time the Quaternary glaciations repre-
sent geologically abrupt events when phenomenal ice-sheets spread over
nearly 20 million km^ of the earth’s surface, over-riding areas where formerly
mild climates had prevailed. It is possible that the earth had never been so
cold since the beginning of geological time. Scanning back into geological
time, erosion on the unvegetated pre-Devonian relief must have been dis-
tinctive and although vegetation types expanded their range, grasses did not
appear until the Miocene.
It seems that a geographical range of studies of the inter-relationships
between present-day forms and the processes of erosion, transport, and
deposition will still give a good idea of the way in which these mechanisms
operated in the past. But the appearance of the past is not easy to imagine. It
is difficult to conceive that where now there is the landslip scar and irregular
toe of slumped material there was a smooth slope giving little indication that
rain, molecule by molecule, was weakening its internal cohesion. It is difficult
to imagine forests once cloaking grassy or rocky slopes and that in terms of
geomorphological time a cover of ice disappeared overnight. It is hard to
believe that only five hundred generations ago, sea- level was 130 m below
its present stand and that it then started to rise at a rate of about 45 cm per
generation. By the time that the rise halted more than one must have counted
his livestock aboard two by two.
2. Stillstands and the mobility of earth structures
Erosive activity is usually limited by a base-level below which most mechani-

cal processes cease to be effective. In upland areas a resistant rock outcrop
frequently provides a local base-level which might retain a reasonably cdn-
sistent relationship to land surfaces upstream. On a global scale, sea-level
provides a regional base-level at which marine. er^blTmayFscoTe'a'maFk at
the coast and'inland provides a datum at which, or at some distance above,
mechanical erosion on the land surface ceases to work downwards. Rapid
movements of the earth’s crust tend to be somewhat periodic in time,
separated by periods of relative stability. W. M. Davis therefore felt that a
theoretical separation in time of uplift from protracted periods of stability
would facilitate landform interpretation, and in particular the recognition of
evolutionary stages of a progressive flattening of the land surface. From other
quarterscame the suggestion that when coastal movements did in fact occur
they were upward on the continents and downward in the ocean basins,
leading to a world-wide progressive relative elevation of the land surface in
relation to the sea (Suess 1906; Chamberlin 1909). From a combination of
these views it could be predicted that the continents would carry at increas-
ingly higher altitudes, relics of progressively early phases of flattening or
‘cycles of erosion’. Critics of Davis immediately asked how he could assume

that long periods of comparative stability approached the stillstand of base-
level on which his model depended. He was certain that stillstand was insuffi-
ciently long in certain regions, but for other areas pointed out that the
existence of relics of levelled erosion surfaces were the very indication that
stillstands had prevailed. The simplicity of this argument is more striking
than its circularity, and an immense amount of geomorphological effort has
gone into testing this prediction. However, when Davis and his critics dis-
cussed the problem of stillstand neither had available the measurements which
can now reveal the exact age of some land surfaces nor the knowledge that
changes in relative sea-level are rapid enough to be observable within a
lifetime in some areas. Fishermen in localities around Bonavista Bay, New-
foundland, can point to rocks that are above water at low tide today which
they remember as being always below water half a century ago. In the Buena
Vista oilfield, on the west side of the Grand Valley of California, a thrust fault
has shortened the distance between two power-line pylons by 60 cm in thirty
years.
Although the progress movements from eustatic

in separating the crustal
changes will be discussed, the scale of total movements is illustrated first.

There are two types of measurement which express crustal movement. The
first is the result of actual measurements of elevations separated by a known
time, which can therefore be expressed as actual rates, like the levellings along
the Trans-Caucasian railway, where a known increase of 730 cms/1000 years
at one point contrasted with a depression of 640 cms/1000 years in the Kura
depression. Secondly, there are measurable displacements but for which the
time involved can be inferred only approximately. It seems that measured
rates of present-day epeirogenic movements, for both platform and folded
zones are of the order of tens to hundreds of centimetres in a thousand years.
For example the Atlantic coast of France rises at 90-280 cm/ 1000 years,
parts of Britain at 120-130 cm/ 1000 years, and for Africa rates of elevation
are anything nil to 420 cm/ 1000 years. Uplift in the Puerto Rico-
from
Virgin Islands area is about 500 cm/ 1000 years. In the Los Angeles basin
terraces are rising at about 400 cm/ 1000 years. Occasionally, more extreme
values like elevations of 1630 cm/ 1000 years along parts of the Caspian Sea,
or depressions of about 1000 cm in a few hundred years in the Sylhet Basin
are recorded in the most active tectonic regions of the world.
With the aid of radiocarbon dating it is even possible to trace variation in
the rates of movement in late-glacial and recent times. For example, uplift at
the entrance of Chesapeake Bay seems to have been continuous, averaging
850 cm/1000 years between 15 000 and 8000 years B.P. By 6000 years B.P.
Itwas possibly about 6 m above the present mean low water but then crustal
subsidence at 230 cm/ 1000 years followed, allowing extensive flooding of
Chesapeake, until crustal uplift apparently resumed about 2000 years B.P.,
and probably continues to the present day.

Stratigraphical displacements indicate some areas where movements active
today are part of huge disturbances starting in late-Tertiary and within

Quaternary times. A
classic example of this type of area is the Pacific states of
the United States. In the Raton Basin, southern California, several episodes of
later Tertiary vertical uplift, largely along faults, but with some warping,
produced vertical displacements of at least 5 km. In the Carson Range, east of
Lake Tahoe, post-late-Pliocene displacements of lower Pliocene andesites
amount to 1200-1500 m. In the Andes, work in the Cordillera Oriental,
Colombia, suggests that there the main uplift occurred as recently as late
Miocene times.
More tentative arguments, restricted to areas known to be tectonically
active, include references to early Pleistocene or even Pliocene deposits not
consistent with the existence of the present mountains at that time. If signifi-
cant mountain uplift in Turkey took place in very recent times, it might be
possible to attribute the apparent absence of early Pleistocene glaciations to
the mountains being substantially lower at that time. D. A. Axelrod has
recently argued that fossil floras in Pliocene and Pleistocene deposits in the
SierraNevada are anomalously high and that a reconstruction of their
optimum climate might indicate subsequent elevations of some thousands of
metres.
Rates of accumulation in subsiding basins provide an indication of rates of
crustal movement. For instance in the Chilean longitudinal valley, great
thicknesses of clastic fill suggest a net downward movement of about 2000 m
during the Pleistocene.
A separate consideration from rates of uplift is the question of whether
changes in these rates within a relatively short distance cause warping. There
are several indications of axes of warping both approximately parallel and at
right angles to coasts. Indications of the first case, that of marginal down-
flexures of continents, is widespread, the down-warping of the coastal plain
and continental shelf of the eastern United States being one of the areas
documented in greatest detail. This area, particularly in the margin of the
Mississippi delta includes the complication of a hinge line seaward of which
sediments down-warp, whereas inland they have been uplifted. Isostatic
adjustment to sedimentary loading appears to be an important factor. In the
second case features identified with high confidence as former shorelines may
show changes of level along a coast, due to warping in that direction. On a
small scale and within recent time, a marine bench near Night Cliff, in the
vicinity of Darwin, drops from 2-4 m to 60 cm within 180 m but within 2
km rises to 4-5 m and some benches rise as much as 6 m in 1-5 km. On a
larger scale, spanning Pleistocene time and mantled with deposits rich in a
prolific molluscan fauna, are three major erosion surfaces in the coastal
region of north-west Peru. These are the Mancora, Talera and Lobitos
tablazos, which are approximately 270, 150, and 20 m above sea-level at
Cabo Blanco. However, when traced along the coast substantial changes of
level occur. The Mancora tablazo, for example, falls 180 m in a distance of 72
km. In California there are Pleistocene sediments which are not only folded
but may even stand vertically.
Lateral displacements of the crust are of some indirect significance in the
measurement of vertical movements partly because they reinforce the evi-

dence of a dynamic crust and partly because they inevitably include some
vertical displacement. In the north-east Pacific Ocean, two transcurrent faults
involve a lateral displacement of 1425 km, and along the San Andreas fault
there has been a displacement of 320 km since early Miocene times, with
present-day movements measured at 50 m/1000 years. In New Zealand, the
East Ruahine fault has moved at approximately 7 m/ 1 000 years since the last
glaciation.
On stable platforms rates of movement are slow and on a small scale, and
rapidly subsiding trenches like the present-day African grabens are uncom-
mon. However, tectonic history can still be sufficiently varied to give
recorded altitudes on land surfaces or coasts little meaning in absolute terms,
as their regularity of recorded former sea-level heights along the south coast
of Australia suggests. Moreover, the position of shorelines on low-lying
platforms is very unstable because even small vertical fluctuations of sea or
land will cause substantial changes in their configuration. The development of
platforms of moderate elevation appears to lead to the chemical weathering
producing crusts and planation surface but at altitudes unrelated to sea-levels.
One of the best examples of this situation occurs in central and southern
Africa, unfolded since mid-Mesozoic times, where L. C. King and F. Dixey
(1938) were among the first to map extensive surfaces of planation.
Superimposed on the structural movements are elevations of the land in
areas relieved of the load of Pleistocene ice, which amounts to 5 per cent of
the earth’s surface. Isostatic rebound is appropriatelynamed for such rates
exceed in many areas those of tectonic movements. They may be largely
complete before a region is entirely ice-free. These rates are difficult to
compare because rate of recovery diminishes rapidly with time, any given rate
being approximately halved in a span of about 800 years. Also the possibility
exists thatrebound is not necessarily a continuous phenomena and that
maximum rate of uplift in areas more centrally located with respect to the ice
cover may occur later than in the more peripheral areas. This last
possibility
may explain why over the past 3000 years, uplift on the west coast of Baffin
Island averaged 5 m/1000 years compared with 3 m/1000 years on the east
coast.
Some examples of rates of emergence, using the middle third of the emer-
gence curve to obtain an average value, include 29m/1000 years for
Spitzbergen, 24 m/ 1000 years for eastern Greenland and 40 m/ 1000 years
for the west coast of Baffin Island. Examples of maximum rates suggested
include 70 m/1000 years for Boston, which is similar to that for north-east
Canada. For the north of Scotland a rate of 8-5 m/1000 years may have
occurred 12 000 years ago slackening to 0-7 m/1000 years in the last four
millennia. The most spectacular present-day movement is in the Gulf of
Bothnia which continues to rise at 10 m/1000 years, with still one-third of the
expected total postglacial uplift to come. Uplift so far is between 240 and
550 m. In Hudson Bay, a recovery of 270 m may have occurred and a
maximum depression of 600-900 m for the central areas is possible.
Amounts of rebounds vary, being largest where the former ice thickness was
deepest. Thus in Norway uplift was greatest at the heads of deep fiords and
least on the open coast. The highest strand line, marking the marine limit is
220 m in the Oslo and Trondheim fiord areas. At places near the open coast,
an uplift of apparently only 10 m has occurred.
In the Quaternary the incredible spreading of ice over 20 million km^ of
the land surface, inducing on melting catastrophic rates of elevation of areas
depressed during glaciations, should not divert attention from the extra-
ordinary glacio-eustatic fluctuations of sea-level, with amplitudes of at least
146 m. Although present levels are as high as might be expected in an
interglacial, only 16 000 years B.P. the sea was probably about 140 m below
its present level and possibly still 30m below this about 8000 B.P. Compli-
cating the picture of a progressive rise, even on structurally stable coasts,
unalTected by isostatic rebound, is the possibility that the weight of water
added to continental margins has been sufficient to deform isostatically the
coastal areas in proportion to the average depth of water in the vicinity. The
study of the postglacial submergence history of 5 points on the eastern coast
of the United States supports this hypothesis (Bloom. 1967).
Despite the intricate problem of sea-level changes and, inevitably, con-
siderable differences of opinion in detail, two facts on the wane of the last
major glaciation meet with widespread assent. The first is that since about
15 000 years ago sea-level has risen from depths at least 75 m and probably
135 m below the present and that a few millennia ago, all ranges between two
and six thousand having been quoted, the present level was obtained. From
evidence in the stable southern Florida area it appears that sea-level has not
risen appreciably above its present position in the last 4000-5000 years.
Considerable disagreement exists for the period between 35 000 years and
18 000 years ago, during which time the sea-level may have risen con-
tinuously from its low stand of approximately — 140 m. Alternatively, using
evidence from many sources, including former lake levels in the Great Basin,
Broecker concludes that the rising sea-level fell between 25 000 and 18 000
years ago.
Because of glacio-eustatic changes in sea-level and the varied tectonic
complications, sea-level positions before late-Pleistocene can only be recon-
structed where a wealth of evidence exists and then may be of significance
only to areas in the immediate vicinity. Even generally agreed indications of
sea-level positions in earlier Pleistocene times are now realized to be much
more difficult to give than was widely believed a decade ago. It appears that
the glacio-eustatic changes in sea-level corresponding to the major glaciations
were approximately the same order as the last, that the changes have occurred
at decreasing time intervals, but that there is no reason to suppose that sea-
levels would necessarily maintain a constant level during an interglacial.
However, evidence grows to suggest that a few localities may have been
demonstrably stable before and during Pleistocene times, and can give some
indication of relative sea-levels in pre-PIeistocene times before glacio-eustatic
complications began. At such a site in Manatee County, in Florida, estuarine
deposits contain a varied fauna of Hemphillian (Middle Pliocene, 10-4 mil-

lion years B.P.) age. Compared with their Pliocene estuarine environment
these deposits now stand between 1-8 m and 3 m above mean sea-level,
suggesting that sea-level then stood only a few metres above its present
elevation (Webb and Tessman. 1968). This conclusion contrasts with that of
previous workers who, on the basis of less precise evidence and sometimes
retaining the mistaken notion that the capacity of ocean basins was increasing
progressively, considered that sea-Ievel was in general higher at this time.
Impressions of sea-level farther back in time are few, although the interpreta-
tion of bevelled landforms as indicating former base-levels has inspired a wide
range of guesses. P. Kuenen (1950), from evidence not relating to altitude of
land surfaces, considers that sea-level may have fallen about 250 m since late
Cretaceous; R. W. Fairbridge (1961) estimates a possible 200 m.

If it appears an unrealistic and misleading simplification to separate uplift
from erosion, some other concept of the inter-relationship of erosional pro-

cesses and changes of crustal stability of sea-level position may appear
necessary. For huge flows of water to balance uplift with the necessary down-
cutting is possible, as the down-cutting of a rising rock barrier between two of
the Great Lakes shows (fig. 1.7), but in general it appears that equivalence
between rates of erosion and of uplift are rarely achieved.
Although it may be separation of uplift from contem-
difficult to justify the
poraneous erosion, the figures cited show that where there are erogenic
movements of the earth’s crust these substantially exceed rates of erosion as
the latter, when calculated for the land surface as a whole, rarely exceed a few
centimetres per thousand years. In contrast, epeirogenic movements involving
vertical movements of stable platforms may be small, but since subsidence of
only 20 m would involve the flooding of huge areas, and since uplifts of the
same amount lead to extensive marine regression and the drying out of
equally large areas, the enormous lateral changes produced make erosion
rates look negligible. Incidentally, these continental transgressions and re-
gressions are unrelated to the capacity of ocean basins, as was thought at the
beginning of the twentieth century. Similarly, sedimentation rates are com-
monly also of an order of a few centimetres per thousand years, so that for
200 m deep,
areas like the post-Quaternary Baltic, already in places, there is
scarcely a veneer of sediment on its floor.
Years BP
Figure 1.7 Illustration of how downcutting has matched the rate and amount of
emergence of the Sauk rock barrier in St Mary’s River. Ontario (Farrand, 2962).
The rock barrier emerged, due to isostatic rebound, since the Algorna water-level
stage (3200 BP).
In some places, therefore, a separation of tectonic movement or of changes

in sea-level or shoreline position, from erosion or deposition, has been a
virtual reality. In other places a sequence of events would be unrealistic. In
either case the major difficulty appears to be rather that of postulating, at

some arbitrary point in time, some ‘initial’ form which resembles more the
appearance of the present day instead of the unrealistic flatness of an
idealized, hypothetical plain. Apart from suggesting that it is the nature of a
land surface before movement which is the greater problem than deciding on
how erosion and crustal and sea-level changes interact or whether one essen-
tially precedes or follows the other in its subsequent history, only one general-
ization can be offered. The nature of all the factors involved in these move-
ments must be considered and evaluated for each particular case.
3. The Cycle of Erosion
W. M. Davis, at the end of the nineteentli century, proposed a Cycle of

Erosion jn which initially steep-sided -incisions into an abruptly uplifted
surface gradually broadened and flattened until, if continued stillstand of
base-level were a ssumed, an ero^siqnal plain of faint relief resulte^The reason

why he often referred to this sequence a s the Geographical cvcie may riot be
immediately^clear.However, W. M. Davis believed that the races of mankind
were obviously determined by the larger features of the lands and even that
‘religious ideas themselves . . . are seen as if in a mirror held to nature’. It
seems that he termed the Cycle a geographical one because he believed that it
would explain more than the landforms themselves, but it is not clear to what
extent" Davis’s knowledge oT'htiman-historyr combined with his primitive
deterministic view of its progress, influenced his concept of landform evolu-
tion.^However, critics immediately argued that Davis’s theoretical model did
less than explain landforms, and controversy continues to the present day.
Four points might be made in an attempt to put this debate in some perspec-
tive. First, the general erosional reduction of rapidly elevated structural
features has occufred_i:epeat£dlv-in-geological time^faercynian folding, for
instance, was approximately levelled within some iQ million years of com-
parative stability^ A few hundred metres of relief on such structures is trivial
compared witfithe geometrical reconstruction of the hypothetical structural

relief. Although not approaching the levelling of the hypothetical peneplain,
the concept of an erosional reduction of orogenic structures to a relief of
about 300 m, with local instances of 1000-1200 m, provides an acceptable
datum for the study of many problems in stratigraphy. On the scale of the
African continent, L. C. King’s highest ‘pre- African’ or ‘Gondwana’ surface,
attributed to cyclic pedimentation, is usually above 1200 m and is thought to
be traceable in the Abyssinian Highlands at 2400-2700 m. Rough levelling
down is undoubted; it is with attempts to identify and to correlate fragments
of erosion surfaces within very narrow altitudinal ranges that disagreement
has arisen.
^^second point is that Davis’s Cycle was never adopted on the scale that
the amount of parochial controversy in some quarters might sugges^Despite
the translation of his work into German in 1912, and his extended visit to
Germany, his ideas left little impression there, and in the Soviet Union and
much of eastern Europe geomorphology has progressed largely unaided by
the Cycle. In 1 924 S. W. Wooldridge observed that British workers had been
slow in following the lead given them by W. M. Davis. In the United States
many geomorphologists lost interest even in debating about peneplains during
the 1930s and took little notice of H. Baulig’s work on the Cycle which
aroused interest in France and Britain. By comparison, a wide range of
studies of landforms and processes were pursued with little or without any
reference to the Cycle concept.
^third point is that there has always been debate on the reality of the
peneplain. For instance, W. M. Davis saw ‘a complete lack of sympathy
between structure and form’ of a former peneplain in Connecticut,
whereas
R. S. Tarr supported the observations of J. G. Percival, who, in 1842, had

seen that this upland surface corresponds ‘very exactly with that of the
geological formations^ It seems that arguments about the peneplain are
largely insoluble becatise the complexity of landform makes proof or disproof
unobtainable. With varying heights in a disseeted surface it is very difficult to
demonstrate effectively the existence of old denudational surfaces, but equally

difficult to deny that a peneplain, however utterly remote might be the chance,
is theoretically possible.
fourth factor is an increasing awareness that regardless of the possi-
bihfies or disadvantages of the Cycle, sentimental attachment, or dislike, a
growing bulk of geomorphological work proceeds without reference to the
debatejit is possible, without much reference to the tedious seventy-year-old
catalogue of reasons for attacking or dismissing the peneplain on general
grounds, to suggest some reasons why the Cycle appears to be in relative
decline.
One reason is the frustration experienced by exponents of the Cycle when
they find that cyclic interpretations within one area do not agree. P. Birot
(1966) takes the example, where agreement is rare, of an escarpment in a
crystalline massif separating two plane surfaces of different altitudes, and the
difficulty of deciding whether the breaks of slope separate two successive
cycles of erosion or represent a comparatively recent fault dislocating one
former surface. The Appalachian Plateau has become a classic area for
disagreement on the number and eorrelation of erosion cycles. General opin-
ion varied between three and five surfaces, although some workers have
recognized a dozen within a small area. There has never been any substantial
measure of agreement as is shown by hundreds of controversial and specula-
tive pages. In parts of Europe the cyclic but non-Davisian ideas of stepped
erosion surfaces postulated by W. Penck, termed Rumpfflache or Piedmont-
treppe and attributed to the control of interfluve elevation by an accordance
of valley thalweg altitudes, have led to no greater measure of agreement. In
Africa, from the sweeping success of the back -wearing scarp and pediment
one gains the superficial impression of greater unanimity as ‘the dating of
cyclic landscapes in southern Africa is now established beyond challenge’
(L. C. King, 1956).
Another problem is the increasing difficulty of recognizing remnants of a
peneplain now that the eye for the country of some past old masters can no
longer be turned to the problem. The major difficulty is the spanning of wide
open valleys where there is no continuity of erosion surfaces and increasing
doubts as to whether erosional horizons can be treated stratigraphically by
using the analogy with remnants of a definable but dissected lithological
horizon.
A third difficulty is the growing realization that with the effectiveness of
1
erosion that would be necessary to produce a later cycle, remnants of pro-

gresively earlier stages will be increasingly few. Rich (1938) thought
J. L.
that the presentation of two or more cycles would be rare and N. M.

Fen-
neman (1936) that correlations based on altitude alone were liable to error in
proportion to the antiquity of the surfaces concerned. In identifying two
peneplanes in the Ozarks, J. H. Bretz (1962) observed that the evidence for
one had been largely destroyed. In southern Tanganyika, H. Louis
the earlier
(1967) observed that surface erosion by continuing on all present surfaces
meant that peneplanation was persistently recent. As a final example, which is

more specific since it relates to an uncontroversial feature, solifluction debris
obliterates older shoreline sections of the ancient Lake Bonneville at Franklin,
Idaho.
A fourth problem concerns deposits which could provide independent
evidence of the age and origin of erosion surfaces. This evidence is being
examined in more detail but in some instances it only casts doubt on the
interpretation of the erosion surface as a remnant of a formerly more exten-
sive plain. In a classic area, on the dip slope of the South Downs, field
evidence indicates that the basal surface of the Clay-with-flints represents the
approximate northward extension of the Eocene basal plane (fig. 1.8). This
plane, soil scientists claim, coincides in its northward rise with some of the
surfaces of supposed sub-aerial or marine origin described by B. W. Sparks.
They conclude that ‘careful field investigation of every such surface listed by
Sparks showed no trace of marine or other deposits except for Clay-with-
flints, and most of them are ridge crests or summits occurring either at the
intersection of the secondary escarpment slope and the sub-Eocene surface or
between adjacent dip-slope dry valleys’ (Hodgson et ah, 1967).
Fifthly, there is the awesome example of Pleistocene chronology. With
increasing evidence of the complexity and problems in deciphering, correlat-
ing and dating even late-Pleistocene features, the unlikelihood of precise
interpretation of earlier forms is more apparent.
A final factor is the advent of techniques of absolute age determination
which have removed one of the main purposes for attempting to establish the
age of land surfaces from their appearance and altitude alone.
If the Cycle is in decline there are, in addition to the factors contributing to
this, other factors operating positively in other directions. There
is an increas-
ingly wide array of alternative explanations in non-cyclic terms for land
surfaces bevelled across geological structures, although some relate only
to
restricted localities or to restricted instances.
There are several ways in which
geological structures themselves can give an impression of erosional levelling.
On the largest scale, and contained within the stratigraphic column, are
unconformities now
believed to be due only partly to erosion and with
tectonic gravity sliding as the major cause. At the
surface and on a smaller
scale structural instability of rocks like shales underlying more competent

caprock in an escarpment can induce sliding and sagging in the overlying
strata (fig. 1.9). For instance, in limestone-shale scarplands near the Kurkur
Oases in Egypt, local bedrock along the scarpland edges dip east at 1-3
degrees, giving the impression of discordance with the overall westerly dip.
Also extensive fiat land surfaces, often developed on fiat-lying resistant beds
are described as stripped structural surfaces.D. Easterbrook (1969) suggests
that such surfaces are perhaps the non-cyclic fiat areas most commonly
confused with cyclic erosion surfaces. Geological considerations are impor-
tant in an indirect sense too, where resistant strata create a local base-level.
N. M. Fenneman’s view (1936) was that the influence of resistant strata had
probably been under-estimated and the number of regional base-levels over-
estimated, particularly in dissected plateaux and specifically in the
Appalachians where the altitudes might owe more to two Carboniferous
sandstones than to any succession of cycles. More recently Meisler (1962)
describes the example of the Little Swatana Creek where a resistant andesite
lava at the outlet of a basin, underlain by shales, resulted in mean and summit
elevations being set off sharply 12 m above a lower shale basin. Where a
lake upstream is lowered progressively, the importance of local base-level and
its stepwise lowering is often clearly seen. In the Lake Victoria basin, strand-
lines, associated with downcutting of the outlet, stand at 18 m, 12-13 m, and
3-4 m. A radiocarbon date of 3720 years for the lowest may suggest about
20 000 years as the span for the three stillstands.
For erosion surfaces not necessarily related to a local base-level, there is a
range of explanations for specific situations. Pediplanation is the main exam-
ple and surrounds itself in its own system of controversy. For instance, any
suggestion that planed land surfaces are preserved at average elevations,
successively higher with increasing age has, according to E. S. Hills (1961),
no relation to reality in Australia. Altiplanation, a concept formulated by
Eakin in 1916 in Alaska, is sometimes used in an attempt to explain level
surfaces in areas of active or former periglacial areas. There is the suggestion
that so-called Tertiary peneplain remnants in the southern Ardennes are
altiplanation surfaces. It is not usually certain whether these benches are bed-
rock features or include a layer of debris. On the specifically erosional cryo-
planation terrace there may be wash process removing material from the
bench and nivation attacking the riser at its rear. According to J. Tricart
(1967) altiplanation is a theoretical notion seldom realised in nature. An
alternative explanation to peneplanation suggested for escarpments and
plateaux of limestones formerly overlain by an impermeable cover is that
the
concentration of solutional activity at the surface could lead to differential
lowering, the result of progressive exposure accompanying
the gradual
removal of the impermeable cover (fig. 1.10). It seems possible that the
.
ground surface of any rock type more susceptible to any type of weathering
than an overlying cover rock which is being weathered backwards, might
similarly show differential lowering due to progressive exposure.
A further factor contributing to a relative decline in cyclic studies is the
realization of the increasingly wide range of aspects to landform study in

which measurements and observations can be made without the necessity
to interpret the feature beforehand. The expansion of interest in aspects neg-
lected in the Cycle provides the more obvious examples of interests com-
manding increasingly large followings. These include Quaternary geology,
slope studies, and climatic and general process studies. An associated factor is
that the mixed blessing of modern facilities offers opportunities for work with
sophisticated instruments which may tend to invite some researchers away
from topics, like the Cycle concept, to which their application is not readily
apparent.
4. Morphoclimatic zones
Geomorphologists who consider processes then infer the probable effect of

these on landforms, may come to one of two main conclusions. On the one
hand, the physical principles involved in weathering and in resistance to
weathering, although they may vary in intensity from region to region, are
immutable. The resultant forms therefore tend to be similar, and latitudinal
contrasts in climate are consequently of little significance. This physical
approach is that of some leading American geomorphologists, like A. N.
Strahler,who makes relatively few references to climate, and L. B. Leopold.
M. G. Wolman, and J. P. Miller ( 964) who conclude that ‘evidence indicates
1
that the forms of hillslopes may be identical in all climatic regions’. This
approach might be traced back to that of the engineer, R. E. Horton (1945)
who said, ‘The geomorphic processes we observe are, after all, basically the
various forms of shear, or failure of materials which may be classified as fluid,
plastic or elastic substances, responding to stresses which are mostly gravita-

tional but may also be molecular ... the type of failure . . . determines the
geomorphic process and form.’ In addition to the American workers, L. C.
King saw that this physical approach could explain his direct observation that
‘. . all forms of hillslope occur in all geographic and climatic environments’.
In contrast to the physical principles approach, the second conclusion is
that reached by an environmental or climatic approach to geomorphology

whose interest in processes recognizes differences in physical factors and
extends to include the unobservable chemical and biological processes. The
hypothesis is that the well-known zonal contrasts in climate, with their dis-
tinctive effect on vegetation, mean that physico-chemical processes combine
in varying ways and operate at different rates, with inter-related vege-
tational influences emphasizing the distinctiveness of morphogenetic pro-

cesses in contrasted regions. The immutability of physical laws is less im-
portant than the environmental contrasts for landform development, and it is
assumed that the latter should leave some discernible distinctive traces on the
land surface. The emphasis is therefore on studying process and landform in
various environments, with accent on establishing the changes in the relative

effectiveness of biochemical to physical weathering. This approach has, by
describing and comparing the distinctiveness of regions, a strong geo-
graphical emphasis.
A minimization of the importance of latitudinal contrasts in climate is
perhaps an inevitable conclusion which follows from the initial assumption

that physical considerations provide the basis for landform and process study.
The negleet of biological and chemical considerations means that the very
factors which do change with latitude are omitted. So far the physical and the
climatic approaches, apart from J. Tricart’s (1958) and J. C. Frye’s (1959)
rejection of L. C. King’s (1957) notions on the uniform nature of hillslopes,
have co-existed with the fundamental differences being scarcely apparent.
This is probably in part due to the fact that the languages of climatic geomor-
phology are German and French, and those of the physical approach,
English, and the occasional equation.
Even if eventually landforms were demonstrated to have many predomin-
antly azonal characteristics, the study of the inter-relationships between pro-
cess and form in a range of environments much wider than the ‘normal’ mid-
latitudes would remain a fundamental task in geomorphology. One has only
to think of coral islands to realize that some element of zonality exists in
landforms. Therefore some knowledge of the difficulties faced by the climatic
geomorphologist, as he probes further, is an important aspect of geomorpho-
logy.
The distinctiveness of climatic regions and its possible implications for

landform study were first considered by A. Penck in 1883, and he attempted
to classify correlations between climatic areas and surface relief in 1910. In
that year. E. de Martonne was probably the first to introduce the term
‘climatic geomorphology’; and awareness of the distinctive impression of
different climates on landform had already led W. M. Davis to Turkestan and
to the formulation of an arid cycle of erosion. The progress of climatic
geomorphology owes much to German workers after A. Penck, like C. Troll
and J. Biidel, and more recently to J. Tricart and A. Cailleux in France.
Today the phrase climatic geomorphology covers a very broad range of
approaches. The work of A. Cailleux and J. Tricart, searching for the
an area, rather than speculating about hypothetical
essential characteristics of
and indefinable boundaries, is very much in the tradition of French regional
geography. Their attempt is to recognize distinctive morphogenetic processes
and to link these to landfortns. Others are sometimes less specific and merely
produce a physical geography catalogue of climate, vegetation, soils, and
landforms for a recognized climatic zone, with little attempt to link these
into a morphodynamic system. There is the tacit assumption that from the
distinctiveness of the environment it follows that the forms must be distinc-
tive. In this way some morphoclimatic types have been defined on climatic
data alone without initial reference to the forms which are then fitted, or
assumed to fit, into the scheme.
The irregular distribution of moisture where precipitation provides an
inadequate water supply for the weathering and transport of rocks in either
hot or cold climates makes any clear zonation of weathering processes with
latitude unobtainable. A major problem for morphoclimatic geomorphology
is that the fundamental proposition, climate, is not a single definable factor,
but consists of two variables which are partially independent but largely inter-
related in most complicated patterns. The net denudational efficiency of any
one pattern is difficult to predict from climatic data alone. A further com-
plication affects runoff of water as denudation involves evacuation as well as
the weathering of material. At higher temperatures the water supply may be
largely returned to the atmosphere by evaporation and evapotranspiration,
the intensity of this loss being a function of the rise in temperature. Therefore,
although higher temperature may speed up reaction rates, the rate of evacua-
tion of material from a weathering zone may be relatively slow as the higher
temperatures also reduce runoff. A further complication is that higher tem-
peratures speed up reprecipitation processes by as much as they accelerate
weathering, so for this reason temperatures above a certain level become a
self-arresting element in denudational loss processes. In cold areas the
influence of snow and ice cover is similarly difficult to evaluate. Although
temperature may appear to be the critical factor, it is in fact whether preci-
pitation falls in winter or summer that determines if the ground will be
exposed to periglacial weathering.
In addition to the unpredictable influence of temperature changes with
latitude, distinctive geological characteristics are often sufficiently dominant
to make any zonal generalisation difficult. The ineffectiveness of climate alone
to cause a uniform course of rock weathering was shown by the differences
observed in weathering rinds of fourteen igneous rocks, six sedimentary rocks
and two metamorphic rocks weathering under the climate of the north-east of
the United States. In structurally active areas tectonic movements change
erosional processes in a given latitude, and makes a climatic interpretation
difficult. This is the case in the Amazon basin where large-scale uplifts took
place in the Andean headwaters in late-Tertiary time. Even the age of rocks
can make zonal generalizations difficult. In Kenya, Pliocene-Pleistocene
basalts are not weathered to the same extent as basalts in Scotland.
Altitude introduces manifold changes at a given latitude. In central Africa,

Kilimanjaro, 5879 m in height, Mt Ruwenzori and Mt Kenya all carry
present-day glaciers, and well-developed cirques, with floors at about 4750 m,
surround the most easterly of Kilimanjaro’s volcanoes. In the equatorial
Andes the snowline is about 5000 m compared with 4500 m on Ruwenzori.
Along a given latitude climate changes substantially. The disposition of
mountains on the western side of continents and the direction of prevailing
winds combine to create arid areas as discrete patches rather than in belts. In
some areas like Indochina, southern China, and Atlantic Brazil the forest
cover is continuous from equatorial areas to the humid mid-latitude areas,
whereas in Africa, north of the equator, a broad arid area intervenes. In
higher latitudes periglacial activity is practically without zonality, partly
because in the southern hemisphere there is virtually no land in the appro-
priate latitudes, and in the northern hemisphere there are contrasts like those
between southward jutting Greenland and the coasts of north-west Europe
influenced by the north-east extension of the Gulf Stream.
Long rivers can introduce substantial changes in flowing through areas
which contrast with their headwater area. In the northern hemisphere there
are fundamental contrasts between north-flowing and south-flowing major
rivers. In tropical areas several smaller streams create conditions like the Nile
on a local scale. In Equatorial Somalia, the vegetation is generally dry
savanna, through which the river valley is a 2-3-km-wide densely forested
strip simulating the conditions and processes of the humid tropics.
Inmany ways relief and drainage combine to create weathering conditions
which may vary dramatically in geographically adjacent localities. The val-
leys and mountains of the Burma range provide an example, with variations in
precipitation ranging from 500 to 5500 mm annually, and all types of tropi-
cal soilsoccur within this narrow latitudinal span.
The concept of morphoclimatic zones includes the assumption that present-
day forms at least approach an equilibrium with present-day processes. Two
factors make it necessary to treat this assumption with caution. First,
although it is difficult to establish clearly how climate varied during the
glacial periods, the instability of climate rivals that of regional base-levels. In
tropical areas like West Africa abrupt climatic changes produced coarse
gravel terraces, thus simulating the effect of the periglacial episodes of the
temperate and Mediterranean zones. In highlands, like the East African
Mount Elgon, Aberdare Mountains, and the Abyssinian Highlands, glaciers
were present. Weathering crusts,in active formation only in tropical regions
with a pronounced dry season, occur as relict features in forests like those of
the Amazon basin. Conditions at the margins of these latitudinal shifts of
climatic zones are not easy to reconstruct. In central North America there
was a southward shift of isotherms by at least 1200 kms in the ice-covered
areas, but palaeontological evidence suggests that temperate zone flora, as in

central Russia, existed virtually to the ice front. Farther back in time there
were pines and firs in Greenland in the Oligocene, and evidence of tropical
Tertiary flora has been found on the Canadian Shield.
Recently an added problem has become increasingly apparent, as it now
seems possible that marked latitudinal zonality could be a specific characteris-
tic of Pleistocene and Recent times. Badgley (1965) suggests that the North
Pole migrated into the Arctic in Pleistocene times, isolating the Arctic from
other oceans. At the same time the South Pole migrated from the free circula-
tion of southern oceans to the Antarctic continent. Both polar regions became
sources of cold polar air to provide an unusually marked contrast with warm
air from equatorial regions.
Several processes are multizonal. Outside permanently frozen areas, run-
ning water is a fundamental process in all climates. Even in arid areas of today,
relict drainage patterns like the former wadi patterns in the Sahara, reflect this
fact. Distinctive regimes too can be azonal. Intermittent flow characterizes
semi-arid areas due to seasonal drought or periglacial areas, and the heavy
sediment load associated with great fluctuations in discharge favours aggrada-
tion and the development of braided stream courses. On slopes with perma-
frost producing the same effect as hardened clay or solid rock in arid-region
slopes, and with neither type protected by a permanent vegetation, sporadic
or seasonal supplies of surface water produce sheetwash. Wind transportation
is active throughout polar deserts, particularly where no desert pavement has
developed to protect sandy material, and bedrockmay become grooved and
polishedby wind action. Various processes combine to produce changes in
volume. Cracks are functional in the development of some soil patterns in
both cold and warm climates, and expansion leads to bumps as widely
separated as the Icelandic thufurs and the Australian gilgai mounds. Even
chemical weathering processes can have a certain uniformity. In the absence
of water and strong evaporation salt incrustations and salt lakes occur in
north Greenland as well as in hot arid deserts. Some vegetation characteris-
tics which can have an important influence on landforming processes are
multizonal. For instance, the predominance of roots in the tundra commun-
ities (70-80 per cent of the biomass) are like arid communities in this respect
because in both environments conditions are so harsh at the surface that
living matter is best able to maintain its activity in the soil.
As well as process effects, many distinctive landforms are reported from a
broad span of latitude. In some cases it is uncertain whether these forms are
the result of the same azonal processes or whether two contrasted sets of
distinctive zonal processes can produce the same form; or whether they are
essentially paleoforms, relicts from when a given set of processes different
from that of the present-day once prevailed. To avoid havoc in the interpreta-
tion of landforms, the possibility of different processes producing essentially

similar forms cannot be overemphasized. L. C. King (1953) wrote of homo-
logies in landform, H. Wilhelmy (1958) of convergence. The latter word
appears often in the writings of J. Tricart and A. Cailleux. Some examples of
homologies include microforms produced by rock weathering in all types of
climate, as described by Wilhelmy. In unconsolidated material, also, similar
forms appear, like the close similarity between polygons produced by desic-
cation in clay and those in frozen ground. On a larger scale there are the
semi-arid and periglacial ramp-like lower hillslopes, and P. Birot (1966), in
describing the dissection of the Serra do Mar escarpment in Brazil, notes
that the valleys have some similarity in appearance to the U-shaped glacial
valley.
From coastal studies some clarification of the climatic approach is pos-

sible, for it is clear that the general physics of the beach and shore with
constant supply of water and energy has substantial uniformity in operation
as well as in principle. Swales, cusps, spits, tombolas, and similar features are
found in many latitudes. However, beaches in cold climates have additional
features. Pitted beaches occur where buried ice melts, frost cracks some
hundreds of metres long are common and stone circles and polygons may
occur. Ice-pushed ridges also exist. Mechanical erosion by waves may be
restricted by the lengthy periods when coastal fringes remain frozen as spray
from surf, and swash freezes to form a crust of ice on the shore and lower
parts of the cliffs. However, periglacial weathering of the cliffs may produce
solifluction lobes which encroach on to the beaches. Boulder-strewn coastal
flats are restricted to cool-temperate regions, like the St Lawrence estuary,
where drift ice is an important agent of transport and deposition. Any area
that has experienced frost-weathering or glaciation will tend to have relatively
large quantities of coarse material on its shores. By contrast, hot arid lands,
where runoff is limited and allogenic streams absent, have little inorganic
material on the shores amid sand derived from coral fragments. Coral growth
and beachrock development and similar biochemical phenomena are re-
stricted to lower latitudes. As early explorers soon found, intertropical coasts
are fringed with huge cordons of sand with swamps inland, due to the
abundance of fine sediment in the rivers. Headlands are infrequent and coarse
particles rare,due to the deep chemical decomposition of many rocks.
Although contrasts similar to those observed on coasts have yet to be
established for erosional landforms inland, the precedent seems set that differ-
ent latitudes do have some distinctive features. It is important stress, however,
that these may
be relatively minor touches, and that an attempt to construct
homogeneous belts of distinctive processes, let alone the assumed resultant
differences in forms, would be unrealistic and misleading. However, bearing
in mind that it is only by implication that some landform characteristics
might, to a greater or lesser degree, be related to sets of processes implied

under certain broad climatic headings, it is probably justifiable to sketch
briefly the tendencies shown in the areal distribution of certain distinctive
present-day processes.
If one neglects the thickness of ice and its implications, and disregards the
distinction between ice that is or is not frozen on to the underlying bedrock,
and the probability of contrasted reliefs beneath the ice, areas at present
covered by ice are distinctive. Where areas in high latitudes are very dry, with
perhaps 5-7 cms of precipitadon or less, polar deserts may occur. With
greater moisture available, the ice-free areas are dominated by periglacial
environments which include very diverse combinations of processes, although

larger scales in patterned ground are probably restricted to the more severe
climates where permafrost exists. Large areas in Russia can be described as
steppe where there is a sharply expressed moisture deficit, absence of forest
vegetation, and chemical processes accumulate weathering products. For the
humid temperate climates, W. M. Davis’s definition is amply precise. He
described the ‘normal’ climate as not too dry as to lead to intermittent
drainage and warm enough for winter snows to disappear. In tropical regions
with temperatures usually above 20° C. there is a fundamental distinction
between seasonally humid areas and those with a savanna climate charac-
terized by a marked dry season. The specification of a savanna or
Mediterranean-climate environment is made more difficult by man’s distur-
bance of the natural processes. The distinctiveness of hot arid areas is well
known.
The formulation of more precise statements on morphoclimatic geomor-
phology. which are not inferences from climatic data, awaits the measurement
of forms. While the existence of certain subtle differences in landform with
latitude suggest that more traits may be discovered, it is equally well-
established that some earlier claims for morphoclimatic subdivisions in geo-
morphology were over-optimistic, disintegrated on the irregularities of local
conditions and configurations, and became confused because of convergences
in both forms and processes. The realistic morphoclimatic approach, on the
basis of evidence to hand, is almost synonymous with the phrase ‘process
geomorphology’, but also denotes that it is process geomorphology studied in
an environmental context.
5. Structure, process and stage
No phrase has been repeated more often by students of geomorphology than

W. M. Davis’s statement that landforms are a function of structure, process,
and stage. Yet on how to interpret this dictum and which factor to emphasize,
there is such a wide range of opinion that views are almost diametrically
Definitions, nature and basic postulates 6 1
opposed, with in between a spectrum of personal or synthetically grouped

opinions. Difficulties arise when one or two elements of the trilogy are
systematically placed in a subordinate role. By emphasizing the importance of
one element interpretations easily become self-strengthening as the investi-
gator tends as a result to collect evidence relating to the element he has

chosen to consider. It is possible to approach this problem in a detached way
by manipulating Davis’s simple formula. If the dominating influence of struc-
ture is assumed, discovered by investigation, or if a structurally controlled
land surface is deliberately selected for study, one assumes, discovers, or
selects a situation where landform is essentially a function of structure.
Inevitably one therefore assumes, discovers, or selects situations which are

essentially independent of process and stage. W. M. Davis’s Cycle of Erosion
has frequently been criticized on the grounds that it included little mention of
structural complications or that his work included too many assumptions
about processes. To some extent this inevitably follows from Davis’s interest
in stage and in his belief that this was the most important and useful of the
three considerations. Similarly, the current emphasis on the study of ero-
sional processes and their effects introduces the possibility of neglecting the
importance of structure and stage.

There are two further considerations that might be added to structure
process and stage. The first, scale, was not apparent in Davis’s qualitative
accounts, yet the relative importance of structure, process, or stage depends

very much on whether the study area is a portion of a single slope or a sub-
continent.The second is the antecedent conditions which in appearance may
be far more diverse than the idealized flat erosional surface produced as the
hypothetical end-point of an assumed previous cycle or the smooth veneer of
sediments of the abruptly uplifted coastal plain. However, evidence is too
scanty to restore easily past environments and land surfaces, like a pre-
faulting surface, and the advantages of oversimplified initial assumptions very
tempting.
For a given scale and with some unsatisfactory assumptions made about
antecedent conditions, discussion on striking a balance between structure,
process, and stage is made easier for two reasons. The first, a purely technical
point, is that the study of process has already been discussed. Secondly,
whereas adherents to an historical approach advocate stage, and others
emphasize process, promoters of the importance of structure are compara-
tively few. There are several possible reasons why neither the dynamic tec-
tonic aspects of structure nor the static lithological ones are emphasized.
Perhaps one reason is that the importance of structure is obvious for all to
see. Tectonic alignments or lineaments and other tectonically and litho-
logically controlled elements of the earth’s crust can be found in almost every
square mile of the land surface. Yet here and there land form does not accord
with structure, and it is these features, the departures from structural control,
that intrigue the geomorphologist. Compared with the fascination of dynamic
erosional processes or the challenge to abstract thinking posed by the recon-
struction of stages, the interpretation of a structurally controlled landform
might seem dull, self-evident, and dismissible in a couple of words. Areas like
the Canadian Shield, where the surface has been lowered very little and where
the pattern and appearance of bedrock landforms have remained unchanged
at least since Ordovician time, do not attract much geomorphological interest.
Perhaps the geomorphologist’s awareness of the significance of structure in
landform studies is diminished in proportion to his readiness to define areas

like the Canadian Shield as outside the limits of his studies. Another possible
reason for underestimating structure is that ever since C. Lyell (1833)
showed that rivers could erode valleys, any suggestion of the importance of
structures in valley development appears to be a reversion to outdated ideas.
E. S. Hills (1961) on observing the control of stream courses in Australia by
faults and recent warps and folds, realized that, in suggesting that these rivers
were determined by crustal disturbances, he might appear to be taking a

retrograde step. A third factor is that many of the geographers who comprise
the bulk of the followers of landform study, have neither the specialized
expertise nor the facilities to examine the strength or weakness of structures
on a macro-scale nor to examine microscopically the detailed constitution of
rocks. Instead there is a tendency to infer these characteristics from the form
of the land, although there are some exceptions, like the work in P. Birot’s
laboratory in Paris.
When the difference between two opposed views appears difficult to recon-
cile, it is wise to suspect the importance of more than two factors. Perhaps
some of the differences of opinion engendered by the promotion of process or
of stage studies arise because both approaches regard structure as a secon-
dary or even temporary influence to be finally erased by process or by time.
While the idea of two coherent schools of thought would be quite mislead-
ing, there are two approaches in considering the inter-relationships between
landforms, processes, and time.
Some theorists believe that landforms develop in stages and they may
follow an historical approach in the study of the sequence of these events. The
classic example is Davis’s Cycle of Erosion. By considering an end-form that
could take such lengths as those of the Cretaceous or Tertiary eras to achieve,
and by beginning with an idealized flat surface, this view inevitably exagger-
ates the significance of changes in landform through time. Also the recon-
struction of former stages of peneplanation involves a theoretical ambiguity in
believing both in the efficiency of levelling processes and the inter-dependence
of all parts of the advancing Cycle, yet also in the survival of relict erosion
surfaces unadjusted to the new Cycle, The assumption of abrupt change in
base-level is the only way to avoid this ambiguity. The approach by emphasiz-
ing stages in erosional levelling has led to the neglect of vital structural
considerations like the significance of local base-levels where hard rock out-
crops across a valley floor, or to the construction of chronologies with lime-
stones making up the hills on which key levels have been identified. A
conservative estimate of the lowering by surface solution of such hills would
be 30 m in a million years. L. C. King’s pediplanation hypothesis is also
essentially cyclic but is distinctive in the assumed semi-independence of the
component forms of the landscape. It is easier to comprehend his suggestion
of the retreat of scarps towards a high central area where the oldest surfaces
remain theoretically unintegrated with the advancing lower surface.
Workers who study processes as a major interest or those who deduce that
processes will be of primary importance, although by approach less con-

cerned with the possibility of changes in form with time or convinced by the
great difficulties in doing so, usually allow for such changes in their inter-
pretations. Differences of opinion exist among process students on whether
changes in form with time will necessarily lead to greater interdependence of
forms in a landscape. Some leading American workers who have measured
form and processes, like S. A. Schumm and M. A. Melton, believe that this
may be likely. Some European climatic geomorphologists, however, regard
the fundamental inter-relationship to be between climate and form, with the
main changes in form reflecting adjustments to the last major climatic change.
Both J. Tricart and P. Birot have observed situations where hillslope and
channels are evolving with little reference one to the other. P. Birot observes
that the humid tropics are distinguished by decomposition of rocks at rates
more rapid than transportation on slopes, and that transportation on slopes is
more rapid than linear erosion.

In addition to students of stage or of process there is a small school of
thought which has attempted to develop an analogy between landscapes and
physical systems. Although this group is very small and the terminology and
concepts unfamiliar to most geomorphologists, their main conclusion
attempts to force itself to the attention of all concerned with landform inter-
pretation. For present purposes it is difficult to decide not to agree with many
workers who regard this conclusion as sensational rather than sensible, and
dismiss it as irrelevant, but instead to risk continuing needless controversy by
examining the statement in detail possibly disproportionate to its significance.
The conclusion is that landforms do not show appreciable changes during
time, are stage-less or time-independent. In part this suggestion attempts
deliberately to expose weaknesses in aspects of the Davisian approach which
emphasize a sequential change in landform during time. Two separate lines of
thought arrive at the conclusion of time-independent landforms. The first
is
the revival of the concept of dynamic equilibrium introduced
from physics to
landform study by G. K. Gilbert in the late nineteenth century. Although the

recent revival owes much to J. T. Hack (1960) and a desire to seek for an
alternative to the Cycle of Erosion as a fundamental postulate in landform

study, it is confusing to find the concept of dynamic equilibrium, which
supposes that there is an approximate balance between work done and the
imposed load, reflected in the concept of grade in the Davisian Cycle itself.
Also the inflows and outflows mentioned by Davis are not measured by the
contemporary theorists who assume their approximate balance from the
appearance of the forms. In fact where attempts to measure the balance have
been made, the results of workers like A. Cailleux (1948), who studied
rillwash in the Dourdan area, show lack of balance. A. Cailleux found
present-day processes too ineffective to account for the present relief form,
which he concluded must therefore be a relict feature. However, in the ridge
and ravine terrain which may appear to suggest quasi-equilibrium between

stream lowering and the erosion of slope and divide, there is no room for
relict features such as occur on the broad interfluves of areas like the Ozarks
evolving semi-independently of stream channel activity. A further point which
is difficult to follow is that while the assumptions of base-level stillstand in the
Cycle of Erosion may now appear unrealistic, the assumption that certain
landform properties remain essentially constant over long periods of down-
wearing appears to ignore the problem of base-level altogether. In fact
measurements of sediment accumulated offshore from the Appalachian ridge
and ravine area show that recent rates are about one-eighth that of post-
Triassic times (Menard, 1961), demonstrating a substantial change in the
balance between process and form. These measurements which fail to support
the analogy between ridge and ravine topography and physical dynamic
equilibrium could, in fact, be explained by the opposite effect, that of reduc-
tion in relief. The suggestion that landform properties and processes have
been relatively unchanging may, however, be valid. In the 1930s both G. H.
Ashley and N. M. Fenneman described ‘non-cyclic’ erosion causing the
straight, horizontal Appalachian crests to be lowered hundreds of metres yet
looking the same afterwards as before. Their interpretations were based on
the structural control of landform development due to the parallel belts of
rocks with differing resistances to erosion, but were unpopular at a time when
most investigators were searching for constant altitudes as evidence for for-
mer stages of peneplanation. If landform is essentially a function of structure,
it follows from Davis’s equation that landforms will be essentially indepen-
dent of process and stage.
The visual impression of balance of unchanging process and time-
independence that one gains in the Appalachians indeed suggest an analogy
with dynamic equilibrium. This does not alter the fact that the explanation of
these phenomena is probably structural control. It is perhaps significant that
J. T. Hack, in studying areas of domed topography in crystalline rocks in the

southern Appalachians, attributes these distinctive forms to expansion by
sheeting, a structural control.
A second line of investigation has ted to the conclusion that landforms may
be time-independent. This is the study of drainage basin networks in homo-
geneous rocks which has revealed very close correlations between parameters
like stream numbers, lengths, and the frequency of confluences between tribu-
taries. These same correlations have been observed in many areas and have
strengthened the belief that Horton’s Maws’ do describe unvarying properties
of drainage basins, the development of which is therefore independent of
changes with time, including climatic changes. By visual inspection alone of
the distinctive even texture and delicate pattern of drainage networks on

homogeneous rocks it is possible to appreciate intuitively the invariable
geometric pattern which the statistical analysis reveals. However, the invari-
ability of pattern, although a striking and important geomorphological fea-
ture, does not demonstrate that landform evolution is time-independent. Al-
though the homogeneous nature of the rock, which in most Hortonian studies
is non-cohesive or weakly resistant, may appear to make the pattern inde-
pendent of structure too, in fact the very lack of structural or lithological
guidance to drainage in a homogeneous rock introduces a distinctive type of
structural influence. This influence is that which permits surface runoff to
drain downslope in directions determined purely by chance. The stream
parameters and their inter-relationships are therefore determined by statis-
tical laws of averages, chance and probability. It is these which provide the
regularity described by Horton’s Maws’, and it is their strength which shows
through, despite the poor quality of the basic data, not some intrinsic and
general property of landform development. The pattern persists perhaps
because the random lines along which drainage first became established when
a rock is first exposed provided the antecedent condition of reduced resistance
to the transport of material in these directions in subsequent phases. Thus the
regularity of drainage networks, although apparently demonstrating a lack of
structural ‘control’, is none the less a function of structure, due to homo-
geneous materials permitting runoff patterns to develop according to laws of
chance. From their being a function of structure it follows from Davis’s
equation that in such circumstances, process and stage have an unchanging
influence.
It seems that provided the importance of scale and the problem of recon-
structing antecedent conditions, are appreciated, thedictum on structure, pro-
cess. and stage provides the fundamental basis of geomorphology. It also
seems that assumptions about the dominance of any one or even none of the
three have perhaps been made too hastily in the past and that more emphasis
on the careful study of all three elements in contrasted environments, without
preconceived notions about the dominance of any one, might form a useful
diversion from the continuing tradition of speculative discussion.
‘The height of folly is to indulge in wishful thinking and fail to face reality.’ J. B.
Bossuet
6. The necessity for simplification of geomorphological complexity
With the many difficult and complex problems of landform study beyond the
confines of artificial experiments, a degree of simplification is a necessary step
towards some appreciation of the full realities of landforms. The necessity for
simplification raises three problems. First, it means that no generalization can
be dismissed outright, on the grounds that it is oversimplified, without a
careful re-analysis of. and probably also with additions to the evidence.
Because the necessary evidence might be drawn from widespread areas, a
geomorphology is the long persistence of innumerable simplifica-
feature of
tionswhose sole support is subtlety of phrase or plausibility. For instance,
W. H. Hobb’s 1 904 model of four cirques encroaching on an upland area
from four points of the compass is still a textbook favourite, despite the fact,
established in half a dozen scattered areas, that cirques tend to concentrate on

one side of an upland. Secondly, distinction between a useful simplification
and abstract speculation is partly subjective in a scientific inquiry, and the
simplification which, for practical purposes, allows the engineer to build a
technically adequate and financially viable structure is likely to be far too
generalized for purposes of scientific investigations.

Thirdly, there is conflict of aims in a community of scholars between
interpretation and explanation, with simplifications to aid inquiry and the
extension of knowledge on the one hand, and simplifications to make this
knowledge comprehensible to students on The dual role clearly

the other.
does not belong to the simplification, yet geomorphology contains many
examples of confusion on this point, particularly the widespread use of the
Cycle of Erosion concept designed explicitly for students, as a research tool.
There are many ways in which geographical knowledge can lead to a simpli-
fication of landform studies. One is to use information from the classic area
where a given phenomenon is particularly well displayed. For instance in-
formation of global significance on the inter-relationships between tectonic

movement and relief, and their influence on erosion and deposition might be
derived from detailed studies in California, where contemporary movements
are active. Iceland, apart from being one of the few subaerial parts of a Mid-
oceanic ridge, is also the classical land of outwash plains and also for water-
fallsof very varied origins. Possibly the world’s greatest concentration of
pingos lies in the Mackenzie delta area and Spitzbergen may be the classical
land of the soil polygon, but it is in Poland that Quaternary periglacial

features are perhaps best known. This last illustration exemplifies the differ-
ence between the classic area and the type area, the latter taking on some of its
significance from the intensity of existing detailed investigations in an area
often where certain features were first recognized. With changing interpreta-
tions, however, the features of some type areas may become of historic
interest only. In 1947 Western Australia was thought to be a natural eustatic
gauge on account of its structural stability. It is no longer useful as a type area

in this respect because there is now evidence to suggest late Tertiary epeiro-
genic warping and major folding movements in the Pleistocene.
In some situations there is a natural control which eliminates a variable.

The best example of this is a comparison between lake and coastal shores. The
absence of tides in the former, and the smaller size of waves means that
terraces form better and deltas are simpler. The shore zonation has not yet the
complexity of tidal areas. In the Nile, unlike most streams, there are no
complications to downstream progressions because little water and sediment
enters the river during its long traverse of the desert area. Most caves afford
examples of stream environments unaffected by frost.
Complementary to natural controls like these is evidence accumulating

from several areas where the same factor or feature is present or absent. For
instance, meandering currents in the Gulf Stream and on ice are very similar
in appearance to river meanders, the significant factor being that ice or ocean
currents exhibit meanders in the absence of sediment load. The importance of
an abrasive in the development of stream meanders might therefore be min-
imized. As the example of the Gulf Stream meanders show, there is the
likelihood of comparative evidence not just from other areas but also from
other contact surfaces. was suggested a long time ago that the ocean floor
It
would contain forms produced entirely by constructional and tectonic agen-

cies, free from the complications of sub-aerial erosion. For example, beneath
the Arctic ice, the Lomonosov Ridge may be a unique great fold in the earth’s
crust preserved in almost pristine condition. Space scientists would be quick
to point out that forms on the lunar surface have developed in the absence of
water and atmosphere, and under the influence of a small gravitational pull,
and also of their intention of obtaining information from contact surfaces
farther afield. Certain rocks introduce important simplifications into the study
of landforms. Limestones, well known for subterranean forms some of which
are known in other situations like permafrost, peat, or glacier ice. and de-
scribed as pseudo-karst, afford an ideal subject for studies of solutional pro-
cesses due to their soluble, monomineralic composition. In tropical areas,
limestones offer the best opportunities for studying the recession of cliffed
coasts. A conglomerate offers an unusual opportunity for the study of rock-
weathering as ail conditions are constant other than the parent material
provided by the individual fragments. Areas of unconsolidated sediments are

particularly sought after by investigators who wish to apply statistical
methods. Without the necessity for a preliminary weathering phase, water as

a transporting agent has sufficient energy to remove unconsolidated material
and the statistical regularity of integrated erosional topography develops
comparatively quickly.
As climatic geomorphologists would point out. studies within some cli-
matic types introduce a certain control. For instance, arid areas in high and
low latitude might be devoid of vegetation. This certainly was an advantage of
barren dunes in the Sahara desert plains compared with coastal dunes for
R. A. Bagnold’s (1941) study of wind action on sand. The comparing of
characteristics in different areas is difficult due to the inescapable problem
that changes other than those in the object being studied are likely to occur as
well. However, while geographical comparisons may be very difficult to
interpret, the examples quoted above give ample illustration of why the
development of geomorphology has been closely linked with geography, as it
is by geographical search that the natural situation which affords the only
realistic controls on a landform problem will be discovered.
Another problem of simplification of particular concern to regional studies
is that of the antithesis between the uniqueness of the individual case and the
global generalization. The increased use of statistical averages of a large
number of observations of small individual items leads away from the first
extreme. There is also a trend away from the other extreme of over-optimistic
hopes for universal generalizations. As long ago as 1930 W. M. Davis said,

‘On shifting residence from one side of the continent to the other, a geologist
must learn his alphabet over again in an order appropriate to his new sur-
roundings.’ Although Davis may have remained a determinist to the last, it is
increasingly appreciated today that no single theory can explain all features
and that each area is better surveyed and studied individually and the evi-
dence evaluated, at least initially, for that area only.
Amid the accumulation of information is the need to achieve some simpli-

city by structuring diverse facts into a coherent system. Classifications are
seldom realistic owing to the importance of local conditions. Instead theore-
tical systems or conceptual models have been favoured in landform study as
repositories for facts which offer at the same time an interpretation of the
facts. W. M. Davis’s was a major achievement in structuring much of the
thinking and fact-finding of the nineteenth century into his normal Cycle of
Erosion. However, he quickly realized that the scheme lacked universal gen-
erality and devised specific cycles for areas with some ‘non-normal’ individua-
lity. like arid lands. In recent years, while the awareness of the impossibility
of making universal generalizations has grown, the acceleration of the accu-
mulation of facts makes the need for systematizing all the more keenly felt.
This problem applies to knowledge in its entirety, and workers in special
embryo study groups devote increasing attention to the study of information

in a search for general principles about the patterns of knowledge. A. N.
Strahler was the first to apply some of the initial ideas of systems theory to
geomorphology, and interest has grown. For instance, R. J. Chorley has done
much to familiarize geomorphologists in Britain with the approach of
Strahler and other workers in the United States (Chorley and Haggett, 1967).
However, occasionally the use of contemporary systems theory jargon,
instead of its concepts being inconspicuously incorporated into methods of
80 10-1
Yellow
8 -^
o 60
o
>. 6-1
o 40H
20 -\
2d
Red
-I r
2 4 6 8 10
Percent free iron oxide Foreshore slope in degrees
in fine earth
Figures I.] 1 and 1.12 Examples of relationships for which a priori reasoning
suggests a correlation but which measurements do not substantiate. {Left)
Relationship of free iron oxide to Munsell colour code (YR hue x value) in B
horizon of six soil profiles {after J. M. Soileau and R.J. McCracken. 1967. Soil
Sci. Soc.Am. Proc.. Vol.31). {Right) Apparent independence of swash velocity
and foreshore slope {Ingle. 1966). The wide scatter is in part the result of variation
in wave heights and periods.
approach and interpretation, have complicated rather than simplified some

issues. There are other ways in which attempts to simplify geomorphology by
considering systems analysis approaches complicate its study considerably,
because frequently in geomorphology one does not know all the factors
involved nor are the regional boundaries which limit their extents easy to
define. Also, as figs. 1. 1 1 and 1 2 show, correlations which seem reasonable to
assume may not in fact exist or be demonstrable,
if actually measured. Thus
with incomplete knowledge and with ill-defined boundaries it is difficult
to
construct a meaningful model by theoretical reasoning. In addition the models
of systems developed from biological and social sciences data differ
funda-
mentally from technical models in engineering systems. It is difficult to
avoid
confusion in geomorphology where the former appeal to those who still
retain
the analogy between landforms and organisms, while the latter appeal to those
whose approach is based on the principles of physics. However, a specialized
interest termed General Systems Theory represents an attempt to search for
features common to all kinds of systems whether organic or mechanical and
faces the challenge of devising a robot with an electronic mind of its own.
R.J. Choriey (1962) has explored analogies between geomorphology and
General Systems Theory, but since the latter has so far failed to further its
unifying mission, more detailed investigations of the analogy are held up.
Although many may consider that complication, not simplification, has
accompanied systems analysis in its introduction into a few restricted aspects
of landform study, and that it is of interest to only a small group of theorists,
some discussion of this trend must be attempted.
Within a so-called ‘open system’, equilibrium can be achieved if arrival of
one material equals the rate of escape of another. For instance, as described
by W. M. Davis, the outflow of waste material from a hollow on a slope may
balance that of inflow. Where this state of balance or equilibrium can be
demonstrated a ‘steady state’ could be recognized. Over a hillslope as a whole
the hypothetical situation is much more complicated. Rates of departure of
material from a free face or underlying bedrock, or over a slope or from the
base of the slope into a stream may vary and these changes in turn influence
the way in which the other interactions occur. A situation of continually

shifting or ‘dynamic equilibrium’ is then recognized. For many workers these
simplifled situations are too hypothetical in view of the known changes in
erosional processes, climate, and base-level which were particularly frequent
in Quaternary times, with human activities adding further profound complica-
tions in recent times. With these frequent changes where sets of forces are
opposed in geomorphology one set is almost invariably stronger than the
other, so that the result is evolution of forms, at least for a time, not equilib-
rium. Even in the highly dynamic beach environment where easily reworked
unconsolidated materials are involved, inherent instability in wave and sedi-
ment parameters makes only a modified form of steady state conditions
possible. On some beaches, even in the short span of recent years, the
decrease of sand supply in relation to amounts removed down submarine
canyons is appreciable. In fact where budgets have actually been measured in
landform study, similar disequilibria are often found. It seems that without
actual study of the dynamics of a process it is impossible to infer the nature
of a system by reasoning alone. Laboratory measurements of soil-water
dynamics tend to support this conclusion, where a series of moisture contents
were established for soil samples for a range of pressures with a valve either
open to admit an inflow to balance outflow, or closed. For either system the
results were the same at the same pressures (fig. 1.13). It would be impossible
therefore to comprehend from properties alone, in this case soil moisture
Definitions, nature and basic postulates 1 1
values, the nature of the system of inter-relationships involved in the produc-

tion of these properties.
Among some workers who may regard equilibrium concepts as unrealistic,

there is still recognition of some concepts defined in systems analysis. These
are the so-called ‘feedback’ mechanisms which describe modes of change and
include situations where processes are, up to a point, self-enhancing (positive
feedback) or self-regulating (negative feedback). One of the first examples
with which British geomorphologists became familiar was P. Pinchemel’s
concept of ‘auto-assechement’ in chalk dry valleys. An example of a self-
enhancing process is the drying which produces contraction fissures in clays
Figure 1.13 Water content-pressure head drying data {after G. C. Topp et al..
J967. Soil Sci. Soc. Amer, Proc.. Fol. 31). showing the close agreement between
the static equilibrium and steady state cases in the amounts of water retained.
and soils. Once produced a fissure promotes the drying and aeration of deeper
horizons and the fissure deepens. Eventually a complicated set of
self-arresting processes limit the depth of the fissure, and although one might
then expect some quasi-equilibrium to develop between these two sets of
processes, the dry spell might well have ended before this condition was
approached. A good example of a self-arresting process is that which limits
the heights of sand-dunes, because as these become higher the greater the
contrast becomes between wind speeds on the dunes and over the troughs.
Therefore the higher the dune, the greater the tendency for sand to be carried
over the dune crest and deposited in the troughs.
The spans of time taken for erosion of consolidated materials and the
frequency of changes in processes are infinitely greater than in unconsolidated
materials. It may be more realistic to consider how feedback mechanisms are

involved in adjustments to disequilibrium conditions introduced by the latest
changes rather than to generalize about approaches to hypothetical equilib-
rium conditions. The concepts of systems analysis may prove of some use in
simplifying the study of landforms, but so far difficulties have arisen due to
by the investigation and
the application of systems to landforms rather than
measurement of the nature of landforms and process, to establish first which
systems might be of greatest intrinsic importance. In consequence, the geo-
graphical approach of scouring the land surface for naturally controlled situa-
tions is likely to retain its strong appeal in landform study as the most
realistic means of simplifying its infinite complexity.
D. Some Conclusions about Geomorphology
After considering the nature and basic postulates of geomorphology certain

conclusions may be drawn which might to some extent reflect the aptitudes,
training, and preferences of the individual. It is hoped, however, that those

suggested below if not universal truths are a reasonably fair assessment,
as they to some extent form the basis of the subsequent sections of this
book.
The first conclusion is that the geomorphology, as the description and
interpretation of landforms is an important study because it focuses attention
on a factor which has an influence on everyday affairs. Forms like those of the
multiplying branches of delta distributaries or of cols and gaps through ridges
and mountains have had an indefinable but real influence in the courses that
human history has taken.
Secondly, a more immediate and specific point about present-day geomor-
phology is that the closing up of border zones between academic subdivisions
of knowledge has been much to its advantage, and the increasing awareness of
the complexity of landform study has necessarily involved the use of informa-
tion from distant fields. Team projects combining the expertise of more than
one training have appeared, like the work which demonstrated that the
Norfolk Broads were man-dug medieval peat cuttings (Jennings, I960).
Another example is the attempt to summarize and evaluate the recent volcanic
history of Mount Rainier, which included the searching of records of libraries
and local history societies, the stratigraphical study of ash falls, and the tree-
ring dating of forests buried beneath flood debris.
Although in an inter-disciplinary sense geomorphology has made consider-
able advances, in another sense a third feature is the contraction in both time
and space of the problems studied. The size of study areas diminishes so that
measurements, often requiring a great deal of time, can be concentrated in
detail and made frequently if necessary. Also the increased recognition of
.
difficulties in correlation and of the difficulty of identifying homogeneous

regions or systems leads to a contraction of the area to which conclusions are
applied. More conspicuous has been the contraction in the time-spans which
interest the geomorphologist. At a geological extreme workers like L. C. King
remain with their centre of interest focused on past geological eras and for
whom Quaternary landscapes are ‘a multitude of microforms and too detailed
and recent for lengthy discussions where the aim of the work is to demon-
strate the surface history of the globe in geological terms’. The decreasing
influence with time of past actions of the environment upon the land surface
means that the remoter the past, the less may be inferred about past condi-
tions and external variables. It is natural that knowledge of the last glaciation
should be most detailed. Many British geomorphologists who, in the 1950s,
were searching for remnants of Tertiary erosion surfaces, now specialize in
the study of late Quaternary phenomena. The contraction in time-span is
partly a function of the smaller size of areas under review, whereas the broad
lineaments dating back into geological time are evident only in the review of
larger areas. Also the increase in the study of contemporary processes and the
mapping of forms is the description of the present day and a mode of thinking
usually distinct from the geological view of a chronological sequence of
successive landscapes.
A fourth conclusion is that while there has been a shift of emphasis in both
space and time in the objects and problems studied in geomorphology, the
subject probably has changed less than some advocates for change would
claim. One often reads of how nearly all the old masters of landform study
could, in advanced years, walk farther and faster than their students. Yet, with
a possible subtle reflection of the degree of change, it was said that J. P.
Miller, a leading pioneer in the use of quantitative techniques and measure-
ments, ‘.
. could lift the largest rock, and dig a hole with a shovel faster and
deeper than anyone in the party’. For many teachers an aesthetic appreciation
of landform and an enjoyment of outdoor exercise have provided an incentive
for inquiry into geomorphology, and for their students these factors are a
ready source of inspiration. Recent trends in geomorphology have done less
to change this aspect of the study than some commentators imagine. In fact,
the growing significance attached to the systematic measurements in the field
heightens the importance of contributions that can be made by those who
would enjoy walking long distances to make measurements in inaccessible
areas. In the past
decade clubs have been founded in Sweden by interested
laymen to assume responsibility for certain measurements on glaciers. For
those towhom such pilgrimages would offer little pleasure, there are sound
arguments for justifying a laboratory or chair-bound existence. Some of the
urgent tasks in the last category also involve an artistic ability. Geomorpho-
logy in its scientific advance remains dependent on many
qualities like
intuition and imagination, on literary and linguistic skills or on a sense of

time, which might be superficially described as non-scientific.
A fifth conclusion relates to those geomorphologists who utilize mathe-
matics and advanced scientific techniques to solve problems of landform
interpretation, rather than as an aid in a mission to infiltrate geomorphology,
with mathematics, statistics, and scientific techniques, and are free to take a
realistic view of the subject. They point to the limitations to an advance in the
use of quantitative and scientific techniques and therefore from another

source comes the impression that the nature of landform study has changed
less than some believe or are led to believe. For instance, a leading exponent
of quantitative techniques, W. C. Krumbein. asserts that although quantifica-
tion is advancing steadily in some sub-fields of geology, the subject is basic-
ally a qualitative science and that geologists are most concerned with obser-
vational data. A. Jopling who has specialized in sophisticated studies of
sediment transport in flumes describes as the greatest barrier of all. the
fragmentary and incomplete record available to the field geologist. M. A.
Melton (1958) who was one of the first to attempt the correlation of
numerous geomorphic properties of soil, slope, and hydrological parameters
in drainage basins, and to discuss positive and negative feedback mechanisms
in this context, suggests that to argue that all the variability in any natural
environment could ever be entirely explained is absurd. He also stresses the
difference ignored or forgotten by some, between geology and physics or
other exact sciences in the kind of variability encountered as well as the
amount. The fifth conclusion, therefore, is that statements exaggerating the
scientific and quantitative elements in geomorphology are sometimes trace-
able to lack of first-hand experience with the actual nature of problems in
landform study.
A sixth point relates to W. M. Davis and how the position he retains is a
matter of controversy which absorbs a great deal of introspective thinking in
geomorphology. The energy put into this debate is perhaps misplaced. In one
sense W. M. Davis’s hypothesis of the Cycle of Erosion came at a compara-
tively early stage in the development of geomorphology. At that time there
was very little accurate knowledge available about stream discharge data.
Petroleum lay undiscovered and the piston engine had only just been in-
vented. so that aerial views were unknown. Studies of the interiors of the great
deserts awaited the motor transport developments in the 1 930s. It was only in
the 1930s, following the development of innovations like X-ray diffraction
techniques that the general character of clay minerals was worked out. De-
spite the progress which geologists had made in the study of the earth’s
history by the 1890s it still remained a history without years, apart from the
quite misleading predictions based on Lord Kelvin’s physical models, until
1912. when de Geer’s study of varves proved to be a first step towards
establishing a scale of actual times. It was less than a generation after the idea
of sub-aerial denudation had become widely accepted that Davis formulated
his cycle of erosion hypothesis. More than two generations later we can see
that the apparent simplicity of his scheme is more likely a reflection of limited
knowledge at that stage than that Davis’s brilliance could conceive of a

scheme which would anticipate the facts still to be discovered. It seems, on the
basis of an increasingly large body of evidence, no disservice to W. M. Davis

to say that his simplified teaching model is out-of-date, but also too easy to
criticize his work simply because of the time at which he was writing.
A seventh point relates to perhaps one of the most confusing aspects of
present-day geomorphology. This is the parallel that exists between the weak-
nesses of Davis’s approach to geomorphology and that of workers who have
committed themselves to ‘quantifying’ the study, thereby to oust ‘Davisian’
geomorphology. The first parallel is that of oversimplification. If it is, as
J. Dylik (1953) suggested, that one of Davis’s greatest errors was excessive
generalization, it is equally true that any attempt to reduce landform prob-
lems to equations demands extravagant simplifications. A second parallel is
in postulating unchanging conditions. Davis postulated a protracted period

with little change in base-level; J. T. Hack’s concept of ‘dynamic equilibrium’
makes no reference to base-levels. It is said that Davis paid little attention to
changes in climate or to changes in weathering processes in different climatic
zones; these are scarcely mentioned in many studies committed to a quantita-
tive approach. A third and close parallel is in the element of abstract reason-
ing in both approaches. Although W. M. Davis studiously avoided mathe-
matics in the presentation of his ideas, his speculative way of thinking, more
drawn to abstract reasoning than to the establishment of the facts was,
according to J. Tricart, essentially that of the mathematical reasoner. S. A.
Schumm also notes how the quantitative theoretical models of slope develop-
ment are open to the same criticism as those of both Davis and Penck.
Without data upon which the assumptions of a theoretical approach should be
based, theoretical models are little more than speculations, whether elab-
orated by mathematics or by sonorous prose. Tricart advises that mathe-
matical reasoning must not become a convenient means of camouflaging our
Ignorances as this would immediately lead to restarting on the laborious road
of Davisian mistakes,
A
fourth parallel is widespread; it relates to Davis’s
view of youth, maturity,
and old age which can be criticized because the
analogy between organisms understandable in the immedi-
and landforms,
ately post-Darwin age, is seen from greater distance to be misleading. The
stages he defined
may now be largely discarded yet the analogy between
organisms and landforms lives
on. L. B. Leopold, M. G. Wolman and J. P.
Miller(1964) suggest; ‘A river basin is like an organic form, the product of a
continuous evolutionary
line through time.’ Fundamental to Cailleux and
Tricart’s climatic geomorphology is the concept that geomorphology can be

presented as an ecology of forms. The analogy is also to the fore in the ‘open
system’ approach, as this concept from the biological systems point of view
depends on the belief that living systems have to be analysed as systems that
are ‘open’ to matter-energy exchanges with the environment. Davis saw that
the systematic interdependence of land and water forms was so perfect ‘that
one must grudge the monopoly of the term “organism” for plants and
animals, to the exclusion of well-organized forms of land and water’. It seems
that several present-day workers still experience similar feelings, particularly
when confronted with the well-organized electronic circuits simulating organ-
isms, but that geomorphologists might as well accept, perhaps with a lump in
their throats, that landforms are not alive. A fifth parallel is between the
equilibrium that some present-day workers postulate and the near-perfect
systematic interdependence of land and water forms which Davis observed. A
final point of similarity is that the two approaches are similar in the degree to
which their opposed views on time in geomorphology are unrealistic. For
many reasons committed quantifiers have to assume the relative unimpor-
tance of change with time. Historical events are usually highly distinctive and
therefore cannot embody laws, defined as recurrent, repeatable relationships.
A steady state condition is much easier to deal with quantitatively than the
shifting inter-relationships between antecedent and subsequent conditions and
the changes and sequences of forms produced. Thus the committed quantifiers
nose towards the noose of circular argument.
The seventh conclusion is therefore that there are far more similarities
between W. M. Davis’s approach and that of present-day geomorphologists
who are committed to the systematic application of a quantitative approach
than the apparent difference created by the absence of mathematical symbols
in Davis’s writings. It seems that the term ‘Davisian’ can have no precise
meaning in geomorphology since in style some of Davis’s most persistent
critics are the closest followers of his idiom.
An eighth conclusion relates to the unease voiced by some that geomor-
phology’s development has been somewhat slow. If this is true, one of the
main reasons for retardation in the past and quickening pace at present is the
stress ofeconomic demands. In the second half of the eighteenth century the
study of geology was founded, as coal became the principal source of energy,
under the leadership of William Smith, a civil engineer. The great leap
forward for American geology began with the mining boom in the West.
Geomorphological observations were an important aspect of pioneer geolo-
gists’ work because the landforms often provided important clues in delineat-
ing geological boundarieson reconnaissance mapping exercises. A study of
geomorphological processes was also important because explanations for un-
conformities in coal-bearing strata were needed to assist the mining engineer.
In Russia, geomorphology emerged with the founding of Soviet soil
science, and this close interdependence has remained. However, as the hunt
for the treasures of the earth gathered pace, with the discovery of oil in
Oklahoma in 1 9 1 3 as a major turning-point, more sophisticated techniques of

field observation and laboratory analyses became increasingly important. The
search for oil, which has absorbed a major portion of geologists’ efforts since
that time, has depended little on geomorphology and for a generation, with
diminished manpower and financial support invested in directions from which
geomorphological facts might flow, the growth of information depended
largely on those concerned with education. Since 1945 applied work has
expanded to specialisms in which landforms and land-forming processes are
more directly involved. The stimulus of the lead of the engineer R. E. Horton
is one example. Another is that Swedish geomorphologists have found that in
recent years the development of hydro-electric power stations has been
accompanied by greater prospects, from the financial point of view among
others, of studying deltas. Chronic pressure on land and water resources has
led to the intensification of land and water utilization studies in marginal
areas. Directly from such studies, or indirectly from the communication
systems which support them, geomorphologists acquire increasingly large
stores of data on water flow, erosion, and deposition as priorities for tech-
nological endeavour shift towards problems of conservation rather than of
exploitation in the natural environment. Therefore, while it is necessary to
avoid a confusion of emphasis by distinguishing the technologists aims from
that of the scientist, it is
reasonable to suggest that as an eighth conclusion the
rate of progress in geomorphology is closely linked with economic demands.
As a observation, it seems that many discussions about the nature of
final
geomorphology centre around contrasted views where differences may be

exaggerated, and the existence of more than two facets to a problem obscured
in dualistic debates. Conversely, differences in technique or in initial assump-
tions may overshadow a method of approach. The divergence
similarity in
between the views of W.
Penck and W. M. Davis on the evolution of slopes,
conceived as the subject
for a special debate by American geographers in
1940, diminishes in importance
with the growing realization that the fact that
neither measured the actual forms or processes involved is a more fundamen-
tal issue. Few landforms, if any, are understood with certainty. The lively
discussion which therefore results between various theories has traditionally
been highly valued,
particularly for the educational purposes of training open
winds. Particularly
influential has been the stress which Davis laid on the
iwportance of imagination
in geomorphological investigation. There are over
eighty published
theories on the movement of ice from which to choose. On a
smaller scale,
practically every idea concerning beach cusps advanced by one
author is directly contradicted by that of another. Opinions are greatly
divided on the existence of any general lower limit for blockfields. Examples
like the ideas themselves are easy to multiply endlessly. In recent years, partly
due to the assistance of workers in related disciplines, the value of such
controversies has been questioned and formerly esteemed ‘imaginative inter-
pretations’ pale with the cold light of factual evidence to ‘sheer speculation’,
as the number of converts to the discipline of experimental method grows.
For example, in a study in the Southern Uplands of Scotland soil scientists
Ragg and Bibby (1966) note that it is possible to speculate whether erosion
now dominant after an apparently long stable period, was the result of a
modification in climate or of overgrazing and burning. They conclude that
‘only observations of the future course of development and the acquisition of
quantitative data on rates or erosion and deposition can confirm or refute
such speculations’.
A final conclusion therefore is that while ideas are easy to come by in
landform study, it is much more difficult to conceive of ideas that can be put
to practical tests and to gain the conviction that the idea and tests are
sufficiently sound to sustain the investigator through the arduous tasks of
Rapp (1960) urged that ‘we must go out to the
acquiring the relevant data. A.
slopes and make our measurements on them if we wish to evaluate their actual
development in quantitative terms’. This sort of exhortation is commonplace
in geomorphology: the remarkable aspect of Rapp’s statement is that he
actually went to carry out this task. Work like this reflects one of the positive
trends in landform studies at the present time, as the drift away from specula-
tion gathers pace. Speculations are not confined to the nature and develop-
ment of landforms. It is also easy to speculate about the nature of geomorpho-
logy itself and on the course of its development past and future. It is hoped
that the discussion in this first chapter may assist the reader in his evaluation
of further reading in geomorphology, including the remaining chapters of this

book. These are largely devoted to illustrating some of the actual work that
has been done in the past decade to further our understanding of land forms.
II
Landforms and structure
A. Geophysical Considerations
1. Earth movements and mountain building
Most of the major units of the earth’s surface owe a great deal to uplift of land
over very extensive areas, achieved in many different ways during Tertiary
times. The configuration of the broader relief features, like mountain ranges
and plateaux, reflects the differences in folding and faulting, and the spatial
relationships of broad areas of rocks of differing resistances produced by

these recent geological events. Although geotectonics, the study of the broad
structural lineaments of the earth’s surface, falls outside the geomorpholo-
gist’s scale of operations, and to some extent that of many geologists too,
some brief familiarity with the work of specialists like seismologists and
geodesists is important in the investigation of the smaller land-surface sizes
which the geomorphologist studies. Apart from understanding rock disposi-
tions which underlie landforms in structurally complicated areas, or of appre-
ciating that rates of structural movement can exceed by many times those of
erosion, one’sapproach to landform study can be influenced fundamentally
by assumptions about structural behaviour.
Three main points concerning earth structure seem reasonably clear. The
relates to information about the earth’s mantle, which stretches 2900 km
first
from beneath the thin skin-like crust of the earth, half-way to its centre. The
core itself is dense molten iron with nickel and silicon, but the mantle is now
known to be very complex, as yet unobserved directly, and in which many of
the factors influencing thermal convection currents are unknown. Due to the
present inadequacies of knowledge about the mantle even experts’ attempts to
explain mountain building movements in terms of subcrustal processes are
inevitably speculative. Therefore any detailed attempt to summarize these
diverse views within the context of geomorphology would be if not irrelevant,
certainly inappropriate.
A second point is that there is now general agreement on the permanency
of ocean basins which differ in petrology, sedimentation,
and structure from
the continents, at the margins of which the granitic layer thins abruptly to
nothing.
Thirdly, crustal thicknes s of continents is clo sely related to lan d-surface
elevation. As'hasljeenappreciated for more than a century, mo untains are n ot

l^dFsiipported on a rigid substratum. Instead, the height of a moun tain
biodTt^ds to be compensated by a root, made up of the same light er material
whicir^fdjects~b3^sTnuciras'70'irm down into the underlying denser mantle
o 1
— O
'^Sediments Granitic' layer
10
Vi 5 58 km/sec
1-10
o 20 V = 6Km/sec
a>
is 30 v= 7 km/sec [-20
E 40
0
— Intermediate layers £
1 50 V s 8 0 Km/sec
60- Ultra- basic layer

(Crystalline)
70-
Figure!!,! Structural cross'section, southern California (after Gutenberg and

Richter, !954). The horizontal and vertical scales are equal.
(fig. II. 1). On average this lighter crustal layer is 35 km lhick,_^l shrinks, to
as little as 5 km beneath ocean deeps, sharply defined at its base by the M
discontinuity, narh'e d after A. Mobgcovicic who first ide ntified its existence
during studies in the ~~Balkans in^9î With changes in the distribution of
weigHTof this sialic raft, the soliorock material of the mantle may yield to
large stresses somewhat like a viscous fluid, as below approximately 80 km its
strength is only one-hundredth of that near the surface. U nder stable co n-

ditions the large portions of ê crust come to project downward by an
amount in proportio n to tlTe ir surfa ce elevation. The term isos tasy describ es
thfs ideal~state of flotational equilibrmm. The strength of the sup^rting

mantleTs~egarded by some as due to elastic rather than viscous properties,
but an important characteristic of these so-called rheological properties is
that they vary in mode of response according to rates of change of strain as

much as to the total strain which might eventually build up. Recent work
shows that the properties of the upper mantle vary from place to place, and as
its rheological properties are very sensitive to changes of temperature or in

composition, isostatic balance is not wholly related to crustal thickness. The
ideal condition is also probably only rarely attained because adjustments in
weight may involve millions of years. Recovery by buoyancy may not operate
if forces within the crust are sufificient to retain a blo ck belQ.wJtsjequilibrium
1
Landforms and structure 8
position. Conversely, depression by sediment accumulation may be delayed

until loads exceed the strength of crustal rocks. Appreciati on of the natur e of
isostatic balance is fundamental to an understanding of landform evolution in
areas on continental margins, particularly those in higher latitudes repeatedly
depressed~and re-elevated during Quaternary times. There is also the theore-
tical possibility of the erosion of highlands causing isostatic rise. The buoyant
force is such that if erosion spread over millions of years removed 1000 m by
downwearing the net lowering might be only 200 m due to the replacement
from below b y heavy rock as the denu ded block was buoyed up. As yet there
is no information to in dicatejvhether this theoretical possibility is realized or
not.
In the middle of the nineteenth century the conception of the geosyncline

emerged, suîg equendyjto’-be useTm* rhan^di'fferent senseslj n general term s, a
ge osyriclihe is a huge linear or elongate portion of the crust which after
siting below sea-level and bec omin g filled with sediments may_become an
ordgemc*beltjUsuaily a geosyncline is fringed by ^forel^d on the outer side
and a mountainous continental hintCTland'TronrwfTTch the secfimehts arp
supplied. The mechanism involved in subsidence is not clear, but any sugges-
tions as nature must take .into account the crustal thinning or increase
"to its
in density at the base of the crust that would be necessary to compensate for
the lower density of the accumulating sediments. Suggestions about the
motive power that leads subsequently to folding in the geosyncline are several
and encompass a wide divergence of opinion. Since the discovery of radio-
activitymade the concept of a contracting mantle no longer tenable, nine-
teenth-cemmpHâs about compressive forces generated in this way are
possibly a pplicag le on a local scale only. Increasing doubt is also cast on the
classic tectbgSne view that Tertiary uplifts are due simply to isostatic re-
covery of deep sialic roots that had developed in geosynclines. Instead the
energy for lateral compression may come from drifting continents piling up
and picking up folded mountains, like the Andes, on their leading edge.
Where continents pile up against one another and mount to great elevations,
as in the north of India, compression proceeds at a rate of a few centimetres
per year.The viscous drag of hypothetical convection currents might achieve
a similar effect, a suggestion supported by indications from belts like the
African rift valleys and the Red Sea that areas of thick crust may be split
apart and that new crustal material may be created in the intervening gaps
(fig. II.2). A hypothesis which combines these suggestions and other informa-
tion acquired in recent decades is that of Heezen’s (1960) sea-floor spreading.
The suggestion is that new oceanic crust keeps appearing above a hot upward
con vection current in the m antle, and owes much to the discovery of remark-
able parallel zones o f remanent magnetism reflecting past reversals in the
earth’s magnetic field, with progressively earlier zones at increasing distances
82 Introduction to Geo morphology
from the mid-oceanic rise. Thei r width suggests that the ocean floor s might be
replaêd'aurmgTjpah oiJ.Q(h3SiO million years. From this, and from many
mHer lines of evidence, there is increasing_support for a continental drift-
convection current hypothesis of mountain building. In some situations it
seems'possibleTfiiat up\var3~mowme^frts in a geosyncline might be due to other

causes. It is possible in some cases that granitization of the root zone leading
to reduced specific gravity and increased volume causes the whole orogenic
belt to rise as the batholith forms. However, there are many instances of
Figuie 11.2 The explanation of a nft system as an integral part of the theory of
continental drift (Jiom R IV. Ctidlei. 1965. Phil Trans Roy. Soc., Ser A.
Vol 258)
A The formation of a simple nft valley due to a relatively small amount of crustal
extension; there may be considerable transcurrent moxement
B The formation of the Red Sea structure, due to relatively large crustal
extension
C. A possible mechanism for creating the Red Sea structure and the separation of
the crust. Rising mantle convection may induce tensional stress
orogenic movements without a batholith at its core and others where in the
association of orogeny and batholith. the orogeny may be the earlier event. By
whatever mechanism an orogenic belt rises it is becoming increasingly clear
that as it does, gravity movements limit the height to which the central lobe
can rise as a mechanism, termed gravitational spreading, creates a flattening
effect. The evolution of areas like the Alps are now examined in terms of such
glides. Folds or nappes may develop where these gravity glide structures
press against forelands. In this way the buttressing effect of the Vosges and
the Black Forest massifs may have been important in the folding of the Jura
mountains, and in turn this tangential pressure may have contributed to the
elevation of these massifs. Gravity gliding may also be a factor in the tectonic
‘denudation’ of some areas and may also be involved in the initial sinking
phase of the geosyncline.
Landfonns and structure 83
Epeirogenic movements of vertical uplift which do not involve folding of

structures, have produced basins and plateaux in many parts of the world
with the results of more recent movements still clearly defined as structural
relief. The rigid shield areas have experienced epeirogenic movements since
pre-Cambrian time, and there is increasing doubt as to whether there is a
clear separation in time of erogenic and epeirogenic movements. It seems

possible that both types of movement are differing expressions of similar sub-
crustal forces, differentiated according to the stability of the local tectonic
framework. For example, in the Basin and Range province it appears that
compression can take place in one area while tensional forces are dominant
only a few miles away, while in other areas the pattern of late-Tertiary epeiro-
genic movements appears independent of the younger orogenic belts. Further-
more, it is now realized that motion along strike-slip faults driven by non-
parallel sub- continental and sub-oceanic mantle flows, may cause repeated
subsidence and elevation and could produce displaced blocks like the horst
and graben of the Basin and Range province. There, in fact, strike-slip fault-
ing is of late Tertiary and Recent age, and in earlier geological times, the
Great Basin moved south-east in relation to the Sierra Nevada by 130-
190 km.
From the point of view of landform development the contrast between the
two variations of the effects of mountain building is quite striking. Compared
with the clear vestiges if not prominent features of essentially vertical move-
ments, any structurally identifiable units in areas of intense orogeny are
unlikely, partly because of sub-aerial and tectonic denudation during moun-
tain building and partly because many complicated folded structures may be
formed at an early stage beneath the sea when the sediments were still being
deposited.
Earthquakes associated both with mounta in building and with vulcanicity
are probably th e, result of deformation in th e lithosphere* which finally rup-
ture? abruptly, releasing stored elastic energy. The earthquake epicentre is
that point on the earth’s surface vertically above' the poihfi’ o^T'earthq'uake
focus, where energy of a given magnitude is released. Approximately 80 per
cent of earthquake energy develops in the uppermost 60 km of the earth.
When rocks are strained to breaking point and rupture, earthquakes may be
triggered offby some exogenic force such as flood or high tide. Conversely,
earthquakes trigger off catastrophic events in exogenic processes and are for
this reason important in landform studies. One example is the
1870 rock-
slides triggered off on Mount Tacoma, when an area of about 32 ha fell away.
A less disastrous result of the 1 964 Alaskan earthquake was a persistent dust-
cloud 600-1400 m
high which developed and hung about a reactivated fault
to south of Lake Eklunta.
the A
number of spectacular glacier surges
occurred in Alaska after the 1899 earthquake. Tsunamis,
or seismic sea
waves, arrive on coasts, particularly in the Pacific, with great force. The effect
can also be important in unconsolidated materials. There is the possibility
of earthquake trigger mechanisms releasing tensile desiccation stress which
produces fissure patterns on playa floors. Quick clays lose their strength
by shocks and vibrations, assume a liquid state and begin to flow. Although
there may be an order of a million earthquakes a year, and the total energy
released a negligible fraction compared with incoming solar radiation, the
main release of seismic energy is highly concentrated in a few great earth-
quakes which is clearly sufficient to trigger off profoundly significant changes
in land-forming processes. It is difficult to say whether the artificially
generated shocks of underground explosions which may equal major earth-

quakes amount of energy released will accelerate the importance of
in the
movements as the possible effect of these added releases of

triggered exogenic
energy may take some indefinable time before becoming apparent.
2. Continental drift
Continental drift has swung back into favour supported by palaeomagnetic

and other evidence which within a decade has made the theory widely accept-
able, and for some an ‘increasing certainty’. The fit of continental shelves on
either side of the Atlantic, remarkably exact at depths of 900 m (fig. II. 3), and
the symmetry of the mid-Atlantic ridge in relation to the two sides have also
added to the stratigraphic evidence used in the hotly disputed arguments of
F. B. Taylor. A. Wegener (1922) and A. du Toit (1937).
Geological evidence suggests that if large continental movements took
place they have occurred comparatively recently, possibly with initial separa-
tion about 170 million years ago. In the Iceland area current present-day
rates of drift might average about 2 cm/year, involve 1 200 km of new sea
floor since about the end of Cretaceous time and possibly 300 km within the
last 4 million years. The Atlantic might have opened up in 50 million years.
Strike slip faults tend to characterize lateral edges of drifting continents, that
of the San Andreas fault moving 250 km in as many million years. The rates
of movement observed are so great compared with those of other sub-crustal
processes that, as has been seen, mountain building cannot be discussed
without reference to continental drift. The mechanism involved may be diver-
gent flow of convection currents made more effective by the downward
projection of the continents. The rates involved are also orders in excess of
landform change by sub-aerial processes. L. C. King has always maintained
that land surfaces still in existence may antedate the present continents, and
saw the initiation of the African cycle not as a result of uplift but due to parts
of the dismembering Gondwana super-continent (fig. II. 4) drifting away. He
considered that the Gondwana landscape in Brazil sloped west and north-west
Figure 11.3 The fit of the continents bordering the Atlantic at the 900 m contour
(from E. Bullard et ah. Phil. Trans. Roy. Soc., Ser. A, Vol. 258). This computer-
calculated best-fit reconstruction, drawn on a transverse Mercator projection,
represents the possible situation prior to continental drift.
to the Proto-Pacific, and that the Amazon basin below 200 m widens west-
ward right to the foothills of the Andes. In 1927 A. du Toit had pointed out
that when Africa was joined to South America, the rivers must have flowed
towards the Pacific, but the rise of the Andes across the outlets, due to
continental drift, must have compelled the basins on the South American side
to drain back towards the Atlantic. R. L. Sherlock (1933) was struck by the
Amazon basin being for the most part only slightly above sea-level, and that
while the present mouth had no real delta, the great extent of river deposits
inland expanded westward, delta-like in outline. Clearly the geomorphologist
studying larger features like extensive plains or major river valleys may now
have to search for part of his evidence on the other side of the ocean. One
looks at fig. II.3, and may muse about where the St Lawrence flowed, and if
the Central plain of Ireland might have been part of its path.
The on either side of the Atlantic
similarity between the outlines of the land
is so striking that it was at an early age that one formerly learnt that their
drifting apart was considered unlikely. However, for all the efforts to avoid
the obvious conclusion and the heights of controversy that surrounded it.
Figure 11.4 Reconstruction of Gondwanaland at the beginning of Permian time

(from \V. HamiUon and D. Krinsley, Bull. geol. Soc. Am.. Vol. 78). Tillite
(stippled), ice-flow directions on glacialpavements (arrows), and paleolatitudes are
shown.
continental drift is now a topic of sufficient stature to command many pages

of the 1965 Philosophical Transactions of the Royal Society. As salutary as
the recent findings are informative and incisive, are the implications of the
chequered history of the theory itself.
3. Vulcanicity
Because the study of volcanic processes is a highly specialized aspect of

geophysics, the geomorphologist tends to take the form of the erupted vol-
canic material as his starting-point or even to avoid the complication
Landforms and stmcture 87
introduced by sub-crustal processes by studying non-volcanic areas. How-

ever, even if the study of sub-crustal processes properly belongs to other
specialists, volcanic landscapes offer many unique opportunities for tackling
general problems in landform study because the rate of ejection may be

millions of times greater than increments from other depositional processes
or rates of removal by erosional processes. In consequence processes of
erosion may be presented with an initial approximately ideally shaped land-
form on which stages in the history of dissection may become well displayed.
C. K. Wentworth (1927) took advantage of this type of situation in his

studies of the amounts of erosion in Hawaii. Subsequently, the development
of potassium-argon dating methods, being largely restricted to igneous rocks
of geologically more recent ages, has made Tertiary and later extrusions
invaluable horizons on an absolute time scale. Over shorter periods of time
dates of more recent eruptions and extrusions are often adequately docu-
mented The rapidity of weathering of some volcanic
in historical records.
materials, although atypical in rate, is for this reason valuable in a study
where the general slowness of most weathering rates is the source of many
difficult problems. The value of volcanic materials as stratigraphic horizons is
not just restricted to datable lava flows but includes volcanic ash deposits
which may be carried hundreds of kilometres, like the material from the 1 963
eruptions of the Bali volcanoes which travelled to east and central Java. In
other areas volcanic dust may add scarcely noticed contributions to terres-
trial sedimentation but even relatively minor amounts may influence the
chemical composition of accumulating terrestrial clays. It is not yet clear
whether such dust in the atmosphere could have existed in sufficient concen-
trations to induce climatic changes. Finally some brief familiarity with vol-
canic forms and processes is required because in some comparatively re-
stricted parts of the earth’s surface vulcanism dominates the present or the
recent past. For instance. Mount Ararat (5170 m) is one of many huge
Pleistocene volcanoes in north-east Turkey. In the Cascade Range extending
from northern California Canadian border several volcanoes rose
to the
above the plateau surface during Pleistocene and Recent time, including
eleven above 3000 m. In stable areas, apart from along and close to rift zones,
volcanic landscapes are few. In the Soviet Union there is only the Kamchatka
area which is volcanically active today. Vulcanism is also rare in fold moun-
tains of the Alpine-Himalayan type, where superficial layers of sial are
greatly thickened. In contrast, andesitic volcanoes, which tend to be explo-
sive, are closely associated with island arcs and mountains made of
recently
folded geosynclinal sediments; most of the effusive basaltic material
flows out
from points on mid-Oceanic ridges like the Hawaiian and Icelandic
areas.
However, further generalizations are not helpful because some volcanoes
may change their types of activity and contrasts between eruptions may
be observed in adjacent volcanic areas. More than most landforms, the

study of volcanoes tends to be that of the unique form and the unparalleled
history.
The outpouring of lava probably originates from molten regions in the
upper mantle which are quite small in comparison with the total volume of the
mantle (fig. II.5.A). As volcanoes are associated with earthquake belts, it is
possibly the heat released in the shearing of rocks which causes local pockets
of melting. The proximal or immediate reservoirs of most central volcanoes.
Figure 1 1.5 Indications of the range of volcanoes as landforms.
A. An active Hawaiian volcano (from Eaton and Mttrata. I960). Magma from a
source some 60 km deep moves up to a shallow reservoir beneath the caldera.
Occasional discharge of lava from the shallow reservoir through dikes that split to
the surface constitute eruptions. Note the elongation of the volcano along rift
zones and the slight depression of the M discontinuity beneath the volcano.
B. Common tj'pe of Hopi neck (from H. WilUams. 1936., Bull. geol. Am.. Vol. 47)
with diverging columns of lava resting on inward-dipping tuffs. Possible original
form of the now-eroded crater is suggested,
C. Common type of Navajo neck (from Williams. 1936). Dikes branch through
the tuff-breccia shaft. Possible original explosion pit form at the surface, now
eroded, is suggested.
replenished from depth may be relatively close to the surface. That of

Vesuvius may be about 6-5 km below the surface. Of the forces driving
magma upwards, hydrostatic pressure may be the most important because

magma columns are lighter than solid rock.
If the ejected material consists entirely of fragments, known as pyroclastic
material or tephra, coarser sizes will develop around a central vent to form a
cinder cone, characterised by steep 30-35 degree slopes according to the
angle of rest of the fragments. These cones may be up to 150 m in height.
Composite cones, or strata volcanoes, result from alternating phases of explo-

sive activity and outpouring of lava. The strata are partly tephra and partly
igneous rock which may include intrusions as well as extrusive lavas.

Distinct from the size of normal craters which are usually less than 600 m
in diameter, are the huge depressions of calderas, due to the collapse and
subsidence of a volcanic cone. In Iceland a stratovolcano, about the size of the
largest active volcano on the island, disappeared in late Pleistocene times to
leave the Askja caldera with walls nearly 400 m above its floor. It is estimated
that 6500 years ago about 70 km^ of material sank back into the Crater Lake
caldera, Oregon. The walls encircling the present 8-km-broad caldera are
600-1200 m high.
In the absence of explosive activity, basalt flows may build up huge gently
sloping cones, termed volcanic shields. Those in the Hawaiian Islands slope at
2-4 degrees at their submarine base, whereas the summits, which may project
more than 4500 m above the sea, slope at 5 degrees or more (fig. II. 5. A).
Although several of these relatively low cones rise more than 9000 m above
the ocean floor, all rocks are probably less than a million years old. The
outpouring of basaltic lavas on land have formed huge plateau areas, like the
Deccan trap area which covers 650 000 km^ in India. In north-west of the
United States an area only slightly smaller was covered with 150 000-
180 000 km^ of lava in Miocene times. More recently fissure eruptions at
Laki, in Iceland, in 1783 poured out 12 km^ of material. In 1961 lava poured
out of the fissures in the Askja caldera at an estimated maximum rate of
1 000 mVsec., with the lava front advancing at about 1 km/hour. In general,
fluid lava flows very rarely advance at more than 10 km/hour.
In contrast to the fluid basaltic lava, silicic lavas move more slowly, and
extruded in a nearly rigid condition, pile up in domes or solidify in the neck of
the volcano to form an irregular spine on weathering (fig. II.5.B and C).
4. Hydrothermal activity and meteorite craters
One or two areas are world famous sites of present-day hydrothermal acti-
vity. resulting mainly from ground water coming into contact with a source of
heat at depth (fig. II.6). This process must not be forgotten in
landform
studies of any area where this process might have operated in the geological
past because the same basic physical and chemical principles apply to hydro-
thermal reactions as to weathering, and in the first case may occur without
accompanying mineralization. The results may therefore be difficult to tell
apart. However. W. D. Keller (1964) considers as convincing evidence of
Figure 11.6 Model of a high-temperature spring system with deep convective

circulation of meteoric water (after iVhite. 1967). Thermal conduction is the main
source of heat rather than magmatic fluids.
hydrothermal origin, the occurrence beneath relatively unaltered sills, of the

kaolin in Cornwall. In Yellowstone Park, much of the silica deposited around
the hot springs is due to the decomposition of rhyolites as the hot water rises
through them. The initiation of such openings in solid rock prior to sub-aerial
weathering attack cannot be entirely overlooked. A more important con-
sideration is that a fundamental process in all volcanic eruptions might be the
conflict between ground water and magma. Even a cubic foot of water heated
1
to 900°C. at a constant volume yields a pressure of about 7000 kg/cm^ and

there is the possibility of the seismic opening of fissures allowing super-
charged ground water to explode into steam. In 1886 a 14-km fissure was
opened in a couple of hours near Tarawera Mountain by this process.
Unrelated to hydrothermal activity but too infrequent to justify a separate

heading are the some two dozen known ‘authentic’ meteorite craters. One of
the more recent falls was in 1908 in the Tunguska tundra of the Soviet Union.
This felled trees over a 30-km radius. The most clearly defined crater is the
Barrington or Meteor crater in Arizona. It is about 1-2 km in diameter,
170 m
deep and believed to be about 50 000 years old.
B. Geological Considerations
I . Resistance to erosion
Viewed from all angles, landforms almost invariably reflect in some degree
the differing resistances to erosion of the underlying rocks. In cross-section a
relatively resistant rock may occupy low ground simply because it is low in a
horizontal or gently dipping stratigraphical sequence, but there are often

reasons for inferring that higher areas owe their altitude to greater resistance
of the rocks to weathering, particularly in areas of older or unstratified rocks.

For instance, in pre-Triassic rocks of Connecticut quartz-rich rocks stand
relatively high, whereas carbonate rocks, foliated rocks rich in micas and
poor in quartz, and plagioclase-rich rocks generally underlie lower areas (fig.
II. 7). Gaps in the volcanic hills in Central Scotland, near Neilston and
between Johnstone and Dairy, correspond with downfaulted blocks of sedi-
mentary rocks. Along a valley bedrock of varying resistance may account for
much of the differences in form of both valley and channel. In addition to
creating steps in the longitudinal profile, resistant rocks may be related to
valley constrictions with the stream constricted in gorges or in incised
meanders, compared with broader valley reaches in the less resistant rocks
where lateral erosion is unimpeded, like the Brazos river in the High Plains of
central United States. On coasts, viewed in plan, practically any cliffed sec-
tion will reveal more resistant features contrasting with weaker parts cut back
by marine erosion into coves and embayments. North-east of Saigon the coast
is rugged with granite and other igneous rocks forming resistant
points
between short sections of coastal plain. In cross-section wave-cut benches in
igneous and crystalline metamorphic rocks are narrow compared with
broader benches in less resistant sedimentary rocks.
Although resistance of a rock to weathering is obviously a fundamental
issue in the study of landforms, the relative resistance of a rock is difficult to
specify because it comprises several complicated features, each one of which
may respond to either mechanical or to chemical weathering in a different

way. Also, while solutional and mechanical weathering could exploit and
enlarge a line of weakness, it is not always obvious how the initial weakness
originated.
Altitude (ft)
OJ o> CD © o
<0 o o o o o o o
o o o o o o o o
_j
Stockbridge marble
Hartland formation
(mainly schistose & mica rich) >
<
m
~i
o
(Q
1
Mt.Tom hornblende gneiss
(hornblende 50 - 90°Jc)
Diontic gneisses
Younger basics
(chiefly norites & pyroxenites)
Woramoug formation (quartz 50 - 907i>)
Granite 6. granitic gneisses

(quartz is most abundant mineral)
Figure 11.7 Differentia! resistance of rocks to erosion [after Fhni. 1963) The bar
graph shows the mean altitude of seven mapped rock units in the New Preston
Quadrangle, Connecticut.
Theoretically, mineralogical attributes are important in resistance to

chemical weathering. The coherence of crystalline rock is due to the bonds of
the chemical reactions, inter-molecular forces, crystal growth, and mutual in-
clusion of crystals. Within the feldspars resistance varies; in orthoclase atomic

disorder results in weak spots where water can attack, whereas the plagioclase
group and microcline have more ordered structures and have weak spots only
in cracks. Acidic rocks are often erosionally very resistant, whereas chemical
bonds in more basic crystalline rocks with higher calcium or ferrous iron
content are easier to dissociate. Grain size, however, can make a crystalline
rock resistant to chemical weathering, regardless of its mineralogical com-
position. For on basic igneous rocks in south-east Scotland, fine-

instance,
grained basalts give rise to a soil depth of 23-30-5 cm, whereas rocks with
medium-grained equiangular crystals like dolerite lie beneath 50-60 cm of
soil. In coarser sedimentary rocks, with the bulk of the constituent grains
usually of insoluble quartz, cohesion often relates to the nature of the cement,
sandstones disintegrating where their cement contains calcite, iron oxides, or
weathers to clayey material. Porous or poorly cemented rocks are very suscep-
tible to frost-weathering. In sandstones and shales porosities may be about 20
per cent compared with 1 per cent or less for igneous rocks. In finer-grained
sediments the resistance, even in unconsolidated material, is a function of the
mass of individual particles which adhere due to mass attraction, inter-
molecular and electrostatic forces. However, in some unconsolidated materials

the forces, such as the electrolytic effect of salts, which tend to hold the mass
solid, may be more than counter-balanced by the influence of a peptizer which
may render the mass fluid. A variety of organic and inorganic compounds can
liquify flocculated clay. Resistance is equally an inverse function of the size of
pore spaces through which water can move only by laminar flow. Within the
very small interstices, frictional and buoyancy forces are negligible, nor at the
surface is there the frictional drag which makes loose coarser sand-sized
particles susceptible to removal by water or by wind. Pronounced fissility
or schistosity, more than mineralogical attributes, may be a significant
source of weakness in metamorphic rocks. These characteristics cause a rock
to split into thin layers lying parallel to a particular direction, and may be due
to preferred orientation of clay minerals and mica flakes. In shales films of
organic material provide fissile planes (fig. II. 8. A). In consequence rocks like
slate are susceptible to frost-splitting. For instance, north of Oslo the lower
limit of blockfields in areas of fissile rocks at 900 m is 100-300 m lower than
in adjacent areas of igneous rock.
Both for mechanical- and for chemical-weathering the crucial factor is the
ease with which water penetrates into a rock, whether this be essentially along
crystal edges in crystalline rocks, through the pores of sedimentary rocks, or
along the fissile planes of a metamorphic rock. But in this respect all rocks,
regardless both of chemical composition and of origin, may be weakened in a
similar way by jointing and Assuring (fig. II.8.B-E).
The fundamental fracture pattern of a rock mass, it seems, is established by
extension early in its history, as soon as it is susceptible to brittle fracture,
because joints are frequent even in recent sedimentary strata. In igneous rocks
joints can form during or shortly after cooling. However, tension is impos-
sible below a few hundred metres. The fracturing process may be related to
fluid pressures, with systematic joints developing perpendicular to the axis of
least compression. Although no theory on the origin of joint initiation is yet
entirely satisfactory several descriptive facts are established.
In many areas
_
systematic joint patterns may be persistent and extend to appreciable depth,

but for reasons unknown they may be related to the disposition of the more
rigid rocks in the area. These patterns often consist of two directions which
intersect at angles between 10 and 30 degrees. Systematic joints are usually
planar, continuous, and perpendicular to the upper and lower surfaces of rock
A
Preferentially Organic
orientated ' material
particles .
— -Randomly
Fissile orientated
plane (I) (ill) particles
Systematic
joints
surface
D E
Figure 11.8 Lines of weakness intrinsic in many rocks.
A. Fissility in shale (from M. Gipson, 1965. Jour. Sed. Petrol., Vo!. 35): (i) fissile
plane occurring in a zone of preferentially orientated particles, (ii) fissile plane

occurring at the interface between zones of preferentially orientated particles and
organic material, (iii) fissile plane occurring at the interface between zones of
randomly orientated particles and organic material.
B-E. Joint patterns (/ro;« Hodgson, 1961).
B. Plan view showing a typical pattern of non-systematic Joints and their charac-
teristic termination against systematic joints.
C. Block diagram showing relations between cross-joints and prominent bedding

surfaces.
D. Mode in which joints of the same set intersect in section.
E. Mode in which joints of different sets intersect in section.

Landforms and stnicture 95
units. By contrast, non-systematic joints are generally curved fractures which

do not cross other joints and which frequently terminate at bedding surfaces.
Micro-fissures, although ordinarily invisible on a fresh surface may occur as
hairline fractures a millimetre or less apart, and are best developed on more
massive and more quartz-rich rocks. They increase considerably the area of
rock with which water can come into contact. For a capillary of 0-02 mm
radius there is practically no penetration while one of 0-25 mm permits a
penetration of over a metre. Joints may also develop essentially on the surface
due to tension, as in the cooling of igneous extrusions or in the desiccation of
sediments, and occasionally may approach the ideal hexagonal shape.
Joint directions may have a profound influence in guiding the course of
erosional processes which is reflected in alignments in landforms. In the schist
and granite area around Alice Springs several hills show combinations of
straight, joint-determined salients and deep parallel-sided re-entrants formed
on N. Caine (1967) has described Ben Lomond, Tasmania,
cross-joints.
where the polygonal joint pattern of the bedrock has been a major influence
on the form of tors. Above many steep slopes the headwall of scars left by
rockfalls are often the plane surface of a major joint, and avalanche chutes
generally coincide with zones of structural weakness which facilitate disinte-
gration. On valley floors streams follow joint directions as do many valleys

themselves, particularly in jointed igneous rocks and especially in flat-lying
sedimentary rocks where joints are often most markedly developed, and some
linear or aligned lakes may focus attention on this fact. Observations of this
relationship include those of Daubree who, in 1879, noted the striking

influence of joints upon the drainage lines in the upper Mississippi. In south-
west Utah part of a system of deep canyons is strongly controlled by joints
and faults of small displacement. In Yorkshire the Millstone grits. Mag-
nesian Limestone, and Corallian have joint planes in five well-defined
classes with means at 24, 55, 84, 115, and 163 degrees. Harrison and
Thackeray (1940) considered that these directions occur far too frequently in
segments of rivers like the Wharfe, Ure, Derwent, Low Lindrick, and
Newtondale to be merely accidental, and noting that 25, 115. and 163
degrees were especially predominant, concluded that these valleys are con-
trolled by joints. In tropical areas differential erosion is often less obvious,
but in the Serro do Mar escarpment in Brazil, the dissecting valleys are
always orientated by the boundary fractures and include abrupt turns. Also
very close adjustment to structure may develop within the weathering zone, as
the numerous artificial exposures in Hong Kong have revealed. More unusual
in origin is Meteor Crater, Arizona, which is squarish in outline, with the
diagonals of the square corresponding to the direction of the two principal
joint systems.
Several lines of evidence indicate that joint frequency apparently decreases

rapidly with depth and that many joints form near the surface. Fig. 11.9
illustrates the common experience of water engineers who find that most open
fissures in jointed rocks have been encountered by the time they reach depths
of 30 m. It seems that joints must persist downwards but only open up when
unloading occurs. In view of the importance of joints in landscape develop-
Depth below land surface (inrnelres)
Figure 11.9 Cumvilative curve of

depth to water-bearing crevices in
146 Kentucky wells (after E. H.

Walker, 1956. Bull. geol. .Soc. Am..
Voi. 67). showing that 75 per cent of
the wells meet water-bearing crevices
before penetrating more than 90 feet
below the land surface. Depth below (and surface (m feel)
ment and the dependence in turn of joints on unloading to provide lines of

weakness of exploitable size, the phenomena of unloading emerges as one of
the most crucial considerations in geomorphology. Unloading can be of two
main types, due either to linear erosion allowing lateral expansion or to
downwearing removing the superimposed load. To some extent these two
tj'pes of unloading may operate in conjunction with each other, but there is a
tendency for large-scale exfoliation, or sheeting, to characterize areas of
homogeneous crystalline rocks in which primary joints are not apparent. In
this case with the initial relief of load in a vertical sense, the horizontal
opening of joints follows at a distinct subsequent stage. By contrast there are
no distinct phases where linear erosion and relief of load, essentially in a
lateral sense, characterize horizontally disposed sedimentary strata, because
both operate in the same direction and some of the most pronounced joint
patterns result. Sheeting, which is like large-scale exfoliation in appearance, is
often independent of all primary structures in a crystalline rock and may

cross rock boundaries. Sheeting affects massive rocks like granite and also
quartzites, and affects highly foliated rocks like gneiss and schists as well. In
granite quarries in New England freshly exposed sheets on quarry floors have
been known to split suddenly into two or more sheets, and at Barre, Vermont,
expansion of several inches have occurred during quarrying. As G. K. Gilbert
observed in 1904 these partings in a natural setting were always adjusted to
the general shape of the land surface. Sheeting joints trending roughly parallel
to the outline of the slope have since been observed in many parts of the world
and been mistaken for a primary feature in others. Below the surface the
configuration of the sheets is increasingly flat but still conforms roughly to
the broad outlines of the landform above (fig. II. 10. A). In bedded rocks lateral
expansion tends to be more significant (fig. II.IO.B) and joints become more
abundant and open as one approaches cliff faces. In the massive limestone
area of north-west Yorkshire one of the localities where open joints are most
pronounced is above Malham Cove, the highest sheer face in the area. On the
shoulders of cliffs in granitic rocks, the sheeting joints may still tend to curve
round parallel to the topography.
Figure 1 1. JO Inter-relationships between landform and the surface expression of

geological structures.
A. Sheeting arched nearly parallel to the land surface slope (C. A. Chapman and
R. L. Rioii.x. 1958. Am. Jour. Sci., Vol. 256). The rock is a hornblende granite,
observed in Acadia National Park. Maine.
B. Large-scale sheet structures and joints in massive Colorado Plateau sandstones
produced by expansion {Bradley. 1963). Arrows indicate inferred directions of
expansion; X is an exfoliation dome. Y an exfoliation cave, and Z is an overhang-
ing exfoliation plate in a meander scar.
C. A block-glide landslide. Front Range. Colorado {Braddock and Eicher. 1962).

An upper sandstone (stippled), exposed on hogback dipslopes, has slipped along
bedding planes.
Two additional types of joint enlargement, of less general significance,
might be added. On higher latitude coastal areas there is the possibility of ice-
unloading favouring fissure development, and in eastern Canada excavations
on the Quinze river, Quebec, and at other localities, have revealed extensive
near-horizontal fissures a few centimetres wide 6-20 m below the surface for
which glacial unloading is thought to be the primary cause. Secondly, joint
widening by river-water solution on a valley floor of soluble rocks might be as
significant as the lateral expansion of unconfined situations of cliffs on the
valley side. This applies not just to massive limestone areas. H. E. LeGrand
( 962) notes that many joints are enlarged by the action of solution especially
1
in gneisses and schists containing silicates of calcium, and that many of these
enlarged joints are associated with linear sags in the surface relief.
It is not easy to state the precise significance of joints in facilitating erosion

because it seems that their development to exploitable dimensions by unload-
ing is itself a product of erosion, while in turn, as in the east-central Sudan,
the depth of unloading limits the effective depth of water penetration. How-
ever, because the effectiveness of erosion at a given period is influenced by
the efTectiveness of past erosionai conditions in producing unloading, and the
pattern created, perhaps one of the fundamental characteristics to recognize
where joints develop, is that over a period erosion would seem to be self-
perpetuating in pattern, but may work along joint-controlled directions. Also,

because expansion takes place in the direction of least resistance, with sheet-
ing joints related to existing landform. there is probably a tendency for
distinctive forms like domes, canyons, and cliffs to persist.
2. Rock disposition
Faults are fractures in rocks along which opposite sides have moved past each
other. If this movement is relatively minor, it may be hard to distinguish a
fault from a joint, and to some extent the influence of either in landform
development may not be easy to detect. Quite often, however, not one single
fracture is involved, as faults tend to occur in zones where the dislocated
strata may form a belt, perhaps 300 m wide, as on parts of the San Andreas
fault. Such situations facilitate linear erosion, as in Trinity Bay in New-
foundland, an unusually deep elongated basin which is attributed to structural
weakness rather than to glaeiation, as the ice moved at right-angles to the
orientation of the bay. The fiords of the west coast of Norway, according to
some investigators, largely coincide with crush zones, joints, and similar
tectonic or geological contacts, indicating that fluvial erosion and subsequent
glacial erosion were largely tectonically controlled (fig. II.28.A and B). In
other regions, faults stand out boldly, or truncate a structural grain sharply,
but not always as clearly as on the bare rocks of the Canadian shield. In plan
the outline of many inselbergs in several parts of Australia and Africa are
coincident with fractures. Vertical movements on faults may exert a control
on the orientation of drainage, like the fault bounding the north-east flank of
the Barind, in India, where the Karataja river and its tributaries flow south
west into the Jamuna river, probably due to south-west tilting, and examples
of lateral offsetting of streams occur in the areas of active transcurrent
faulting.
Faults may be an important influence on landform development, but since

it is possible to describe faults according to a variety of geometric, geological,
and genetic criteria, an attempt to present a simple classification would be
misleading. A fault displacing strata essentially in a vertical sense may be
rotational. On a pivotal fault the vertical slip and the height of an associated
escarpment gradually decrease along its strike, then gradually start to in-
crease beyond a given point to produce a scarp facing in the opposite direction.
In a hinge fault there is also a rotational movement, but one which produces
increasing displacement along a fault-line in one direction only. All faults
have a certain rotational movement, but where this is insignificant a transla-
tional fault is recognized. The fault may be ‘normal’ where the dominant
movement is downward on one side, with the dip of the fault plane usually
exceeding 45 degrees, or the dominant movement may be the thrusting
upward of one side, often along a fault-plane inclined at less than 45 degrees,
to produce a reverse fault. By contrast in strike-slip or transcurrent faults the
principal stress and resultant movement is essentially horizontal, although
vertical displacements, both non-rotational and rotational may be produced.
In a number of particular tectonic settings, possibly associated with rising
convection currents in the mantle, faults in approximate alignment may pro-
duce not just steps in the landscape and diversions or alignments in the rivers
but the distinctive fault-bounded blocks and depressions ofhorstand graben re-
lief. However, the movements which leave horsts upstanding either by
vertical
actual upward movement or due more to the sinking of grabens, may be an in-
tegral part of dominantly large-scale strike-slip movement, as already discussed.
Faults are sometimes described by the trace of the fault-line in plan, seen in
relation to the disposition of the bedrock. In sedimentary strata these are
strike, dip, oblique, or even bedding faults. Peripheral and radial faults occur
on domes like those of some igneous intrusions. In relation to regional
structure as a whole large faults may either be longitudinal striking parallel to
the long axis of the major structure, or transverse, if striking across it.
Three points about the relation of relief to faulting might be added. First,
there are many references in the literature to the distinction between fault
scarps where any difference in altitude was due to vertical displacement, and a
fault-line scarp which is essentially an erosional feature due to the removal of
less resistant material from one side of a fault. A second point, related to the
fault-line scarp, is that altitudes on either side of a fault may not correspond
to the direction of initial vertical movement if less resistant material is on the
side of relative uplift. Thirdly, a large number of faults exist without trace in
the land form.
100 Introduction to Geotnorphology
In areas of more gently folded rocks the structural geologist differentiates

between symmetrical and asymmetrical folds, the limbs of the latter differing
in Where
dip. folding has been more severe three degrees of increasing
asymmetry are recognized, the limbs in an isoclinal fold being essentially
parallel, in an overturned fold the steeper limb has passed through the vertical
and may be essentially horizontal in a recumbent fold. In an area of essen-
tially low dips, a monocline is a fold which produces a local steepening of dip.
In addition to these hypothetical, idealized cross-sections of folds, in plan and
more recognizable in less disturbed areas are anticlines and synclines, charac-
terized by the width of the fold being narrow in relation to its axial length.
Where numerous smaller folds are compounded into broad structural arches
or basins, an anticlinorium or synclinorium might be recognized. In areas
where folds are less systematically developed domal structures, which are
broad in cross-section in comparison with their short axial length, may occur.
Structural salients and embayments on the margins of folded areas may
complicate a systematic pattern of folding, major transverse faults may dis-
rupt it, and strike-slip faulting tends to drag along a series of folds en echelon.
Moreover, with the exception of one or two classic examples like the half-
cylinder of the Zagros Mountains in south-west Iran, erosional processes

appear to be sufficiently effective during slower rates of folding to restrict
topographic elevation to less than its hypothetical structural elevation and to

reduce it comparatively rapidly to a fraction of this height after folding
ceases. In addition there is the complication of an inversion between
structural relief and land surface form, with the axial lines of some anticlines
deeply trenched whereas synclinal folds in thick resistant strata resist erosion
and within a short period of geological time after folding commences, become
the highest points in the area. Because of the complications in pattern of
folded rocks and the significance of erosion during folding, attention to rock
disposition in landform study of these areas often focuses on the scarpland
relief of individual limbs of folds. 1
The inter-relations of three aspects of the disposition of strata in the limbs

of folded structures, the inclination, lithological sequence, and relative thick-
ness, are particularly significant in landform study. These distinctive aspects
are also apparent in structures that are essentially tilted blocks. Although
none can be considered realistically in isolation one important effect of
inclination, or dip, of the strata is that the relative erodibility of a rock is
expressed most accurately in land elevation when dips are vertical and least
accurately where dips are low, where if a resistant horizon is exhumed, it
shields the underlying strata. Structural benches then become significant

landforms, like those in the Old Red Sandstone of Caithness in northern
Scotland, or more extensive structural plateaux may develop. Moderate dips
accentuate the importance of gravity in erosional processes, particularly
1
Landforms and structure 1 0
where the lower end of a tilted block or limb of a fold is eroded away. This
may induce landslides along the bedding planes in deeply dissected relief (fig.
II.23.C), in quarries or along coasts where the dip is seaward, like the slips of
Chalk and Upper Greensand, overlying the impermeable Gault Clay, seen
along the south coast of England. In addition to the susceptibility of mass
movement at the unsupported lower end of an inclined block, there is the
accentuation of erosional attack at its upper end, concentrated laterally by
gravity against the exposed edges of less resistant strata. This may apply to
ice movement, tongued and grooved into the strike ridge and vale relief as
much as to stream erosion, as strike valleys in glaciated areas in the Seal Lake
area of eastern Canada or the Cairnsmore of Fleet area in southern Scotland
show.
The sequence of varying lithologies of rock is a dominant factor of land-
forms developed in regions of folding. A stratum can play one of three roles.
First, a more resistant stratum may produce a cap-rock efTect, protecting less
resistant material buried beneath from erosion and may also provide resistant
fragments of debris which move down the scarp slope from its outcrop, which
reduces erosion of the less resistant material forming the lower part of the
down-dip valley side. It follows that the presence of an inclined resistant
stratum between two less resistant lithologies tends both to protect the under-
lying stratum from erosion and to accentuate the erosion of the overlying one.
A third effect is that of an unstable horizon. This may be as a slip-surface
which appears to be present as a distinctive horizon in cases where mass all
movement takes place in the direction of dip, or which by its local absence,
may make certain points along a valley-side stable. A useful example is the
much studied slide, dislodged by the 1964 Alaskan earthquake on to a
tributary of the Sherman glacier. Generation of the slide of massive coarse
sediments was favoured by the presence of weak fractured beds of phyllites
and slates dipping steeply and almost parallel to the original surface of the
slope. A period of chemical decomposition of the material at the slip-surface
may be necessary before a basal plane becomes a potential slip-surface. The
schist and gneiss with foliation dipping steeply downslope which on failure
led to the 1959 Madison canyon rock-slide avalanche in south-west Montana,
was highly sheared and deeply weathered. Broader outcrops of incompetent

material by weathering or by shearing under the weight of the cap-rock, cause
the latter to camber over, with the consequent opening of joint, favouring
reduction in the prominence of the cap-rock edge.
In folded rocks along coastal margins the attitude in plan of the strike of
the structures in relation to the direction marine erosion is
analogous to that,
seen in cross-section, of the inclination of strata to sub-aerial and
sub-glacial
erosion. Like the elevations and depressions seen in land in
cross-section and
related to near vertical dips, structures striking perpendicularly
to the coast
produce in plan promontories and coves in close relation to the erodibility of

the rocks. Like the horizontal strata of a plateau, resistant ridges striking
parallel to a coast will protect less resistant rocks inland or provide resistant
fragments to add to the resistance in coves where the sea may have broken
through to less resistant rock.
A third factor is the ratio of the stratigraphic thickness between strata. For
a given dip, it seems theoretically sound to infer that the thicker a cap-rock is
in relation to the underlying less resistant material the slower will be the
erosion of the cliff or scarp, although there are not the measurements avail-
able to demonstrate this relationship. With changes in the angle of dip,

however, the outcrop width of a given stratum as opposed to its strati-
graphical thickness will vary. Thus a prominent scarp former like the Chalk
forms broad relatively unbroken escarpments where the dip is low. as in
eastern England, but where tilted into steep hog’s-back ridges, as on the Isle of
Wight and adjacent areas in southern England, the narrow Chalk outcrop is
broken through at several points by valleys which run transverse to the strike.
In addition to landform inter-relationships with structural movements at
the surface, there are some limited areas, particularly in unconsolidated
sediments where surface forms are related to the upward propagation of
structures that exist or move at depth. For instance, the alluvial plain of the
Beni basin in north-east Bolivia is an area with pronounced surface linea-
ments and orientated lakes. The latter have neither inlets nor outlets, but have
distinctive flat floors and relatively abrupt sides and may be up to 19-8 km in
length. Some lineaments can be traced from the Brazilian shield outcrop into
the basin. G. Plafker’s suggestion is therefore that the orientated surface
features in the Beni basin form through slight movements in the basement
blocks although faulting may be a more likely explanation for some orientated
features in other areas (fig. II.Il.C and D). Increasingly there has been
recognition of the possibilities of intrusion of plastic salt into a restricted
opening causing deformations in overlying strata and forming some form of
surface disturbance or perhaps some surface definition as a dome or linea-
ment. Pressures of only 150-200 kg/cm’. applied at room temperatures, are
sufficient to induce flowage deformation of salt rocks, and this is an important
consideration in the landforms on Triassic and other strata including
evaporites throughout the world. There is also the possibility that the shape of
some static structure at depth will influence landform development on the
surface. Fig. II. 1 3 illustrates an instance where a valley running transverse to
the general alignment of surface ridges follows the pattern of sub-surface con-
tours of an underlying rock.
Surface features due to the presence or movement of sub-surface structures

may have been neglected in landform studies. Rather more widespread but
perhaps also somewhat neglected are features related to the action of
water on
O t/J
s < i-<
C :So
^
“S: 3
.5? CQ
c a
§c Su;
g
S i2 Cu
‘
O
ou x:
X
O
t/5
1-a
.5 •
T3 C •a cd
C3
cs
<u
</3
p
c
Q, lU
w) .i2 '
c JS
naetres 3 O 'in H*
j= .s
H
Q
cj
^
tN
.
<u c \o
m P B w o\ :
GD 2d
heights
-H £.
c o
O
CJ o
Contour
o =
2;
o
O tJl
^ o
« «
o
•S'
XJ U a> TO
o o o c
p jO rt E
o O D.
IS ^
S u § £
m .t«
w c ^
« ‘-T.
o «
5 h ^ E^
tj rr ti T-
c c3
•—
a> 3 1/3 V3
5 o ino c3
j::
° I
c p
-O c5 c
p «=
C <u
*0
:S .=
_
2-1
S- n V, ^
E ui ^
S £ 3
W vj rt O
w^ p
»- 0\
« -s o
o ^
p <u
.3 O
x> -5
= .
C3 CD
'Z C
-3
C
<u ?
ô
eo
*-1 c 2c ^
N >»
•£P V
^ s CQ >
strata at depth, termed interstratal solution, because it affects carbonates as
well as evaporites. In centralCanada evaporites reach a thickness of over

240 m and may be covered by 300-1500 m of younger rocks. Over a belt of
1900 km structural lows are depressed by as much as 115 m and may be up
to 160 km long, approximately equal to the amount of salt removed by water
circulating downward through fractures (fig. II. 11. A and B). Southward in
Kansas and Oklahoma, along the eastern edge of the High Plains, a similar
process has produced more obvious surface basins, formerly attributed by
G. K. Gilbert (1895) and others to deflation. Beneath the North Crop of the
South Wales coalfield, extensive interstratal solution of Carboniferous Lime-
stone produce pronounced collapse features on the ground surface of
the overlying Millstone Grit, the enclosed depressions being up to 45 m in
1 1
I 1
0 1 50 300 450 m
Figure 11.14 The collapse of Basal Grit on the North crop of the South Wales
coalfield due to interstratal solution of limestone {T.M. Thomas, 1963, Trans.
Inst. Brit. Geogrs, IVo. 33). The tongue of collapsed grit lies immediately to the
north of Carreg Cadno.
diameter and as frequent as 8-1 2 per ha. Similar features occur on the ground
surface of many similar so-called ‘covered karsts’, and may be the product of
interstratal solution over millions of years, and it is possible that some

features in present-day bare-limestone karsts might have developed beneath a
now-vanished cover. In fact, in existing landform studies there are probably
far more references on present-day relief of former covering
to the influence
structures that have now disappeared, than on invisible sub-surface structures
and lithologies, and there are many examples of the surface characteristics of
a given stratum superimposing some imprint of their pattern on the underly-
ing strata and the structures.
The importance of rock disposition at the earth’s surface in influencing, or
even in essentially controlling, landform is at times too obvious to offer much
challenge in interpretation. On the other hand, sub-surface features, because
they cannot be directly observed in most geomorphological investigations, or
1 06 Introduction to Geomorphology
now-vanished former covers, because of uncertainties in their reconstruction,
illustrate how geomorphological evidence is intrinsically incomplete.
3. Distinctive lithologies
It is well known that the pioneer of British geology, William Smith, recog-
nized unhesitatingly the Chalk outcrop in east Yorkshire by the appearance of
the Yorkshire Wolds when he viewed these for the first time from York
Minster tower, two dozen miles away. From a review of geological considera-
tions in landform study it becomes do the relatively few
clear that not only
rock types possess distinctive lithological characteristics and a certain inde-
finable distinctiveness in their landform expression, but that they add further
distinctiveness to landscapes by tending to occur in certain areas only, where
distinctive structural settings are conducive to their formation. Therefore,
although the importance of rock types has been assumed rather than studied
in geomorphology, their more clearly established characteristics should be
summarized briefly.
In sandstones the particles have a reduced chemical range due to their

detrital origin and, if well-cemented, are relatively insoluble. Their dominant
characteristic is their well-defined joint surfaces and bedding planes which
makes them susceptible to wedging by tree roots and clay fills. Their suscepti-
bility to frost wedging is less clear due to free drainage through their porous
texture and down the joints themselves.
Shales in all climates are frost-susceptible and changes in water content
alone are sufficient to exploit their fissility. Weathered shale transformed to
clay moves readily downslope. Clays, by their impermeability are susceptible
to surface erosion and due to their ability to absorb large quantities of water
are prone to mass-movement. Earthflows tend to be closely related to clay
minerals which expand on wetting, thereby decreasing the strength of the
layer. Kaolinites have two properties which do not encourage slipping. One is
the small coefficient of shrinkage which means little Assuring on drying and
little expansion stress on wetting; the second is its high liquid limit, which
means that high water contents are necessary to induce flow. In contrast, illite
clays have much lower liquid limits and montmorillonite clays have a higher
coefficient of shrinkage.
Chalk, being essentially calcium carbonate in composition, is readily

soluble in acidulated soil water, and if saturated its high porosity makes it so
susceptible to frost shattering that a layer of pulverized debris some centi-
metres thick can accumulate in a few years on an initially bare surface. The
well-known characteristic of massive limestones is also solubility. However,
these rocks are often less micro-fissured than crystalline rocks, porosity is
absent and joints are often widely spaced, all three factors reducing suscepti-
Landfonns and structure 1 07
bility to frost action both on small and large scales. One well-marked
characteristic, due to the near-perfect cleavage of their principal mineral,
calcite, is low resistance to impact. As limestone valleys are typically stream-

less or have flows regularized by underground flow, the apparent neglect of
this characteristic may be justified. However, in examining the origin of
gorges which typify extensive limestone terrains in mid-latitudes, the low
resistance of massive limestones to impacts from gravel carried by spring-
melt floods in cooler phases of the Pleistocene, might be important. Dolomite,
although again soluble, is made up of crystals not firmly bound together by an
interstitial cement, and if saturated, will undergo intense granular disintegra-
tion like chalk.
Granite stocks or batholiths simply due to their mode of emplacement tend

to produce domed forms. On smaller scales sheeting tends to produce domal
forms too, with joints developing after sheeting to outline sharp angular
relationships between plane surfaces. Rarely are cliffs produced which have
the sheerness of vertical joint-controlled sandstone, limestone, or basalt cliffs.
In consequence of the distinctive properties of granite, R.J. Russell (1967)

can cite examples of granite coasts south-east of Gaule in Sweden, near Rio de
Janeiro in Brazil, or near Albany on the southern coast of western Australia
which show striking similarities. On the other hand, with more intense micro-
fissuring, granites may form gentler boulder-covered or even sand-covered
slopes, and in porphyritic granites the large felspathic crystals are vulnerable
to chemical weathering and they crack easily under impact due to their
excellent cleavage. Other acidic crystalline rocks like diorite and porphyry
share some of the distinctive properties of granite but due to smaller crystal
size tend to be more resistant.
The main characteristic of extrusive basalts is the highly distinctive hori-
zontal surface of recent flows. Where interbedded, like intruded dolerites
also, they are often distinctive as well-jointed cap-rocks. With soft mineral
constituents like chlorite and olivine, chemical decomposition is often
rapid.
Of the metamorphic rocks, gneiss and schist tend to split along mica-rich
planes and weather into readily transported fragments. In consequence, in
dissected relief, these metamorphic rocks and fissile slates tend to produce
long relatively even slopes of moderate steepness close to the angle of repose.
Quartzite, cemented by crystalline quartz, can approach indestructibility and
ISthe rock to emerge most clearly from the deep weathering mantle of the
humid tropics which effectively reduces the influence of most distinctive rock
properties in landform development. The highly metamorphosed rocks of the
ancient shields are very resistant and preserve a surface form that may date
back to pre-Cambrian times.
4. Variations within an individual rock
An inevitable feature of rocks, of fundamental significance in the study of

landforms, is that whatever may be their properties at a point these will
change in degree if not in kind laterally. Inevitably, at some other point a
stratum or rock body when traced laterally will disappear. This characteristic
is perhaps most readily appreciated with reference to igneous rocks fed from
a central pipe; a phacolith is by definition a lense-shaped intrusion at a fold
axis, a laccolith a dome-shaped intrusion. The more elongate igneous bodies,
like the vertically intruded dike or the horizontally intruded sill or the extru-
sive lava flow also inevitably die out laterally. The effect on associated
landform is often clear. Near the northern boundary of the Central Lowlands
of Scotland, J. B. Sissons (1967) describes how the volcanic rocks reach their
greatest altitude in Ben Cleuch in the Ochils (718 m), where the lava beds are
at their thickest. These beds then fall in altitude north-eastwards towards
Montrose as the lavas become thinner and increasingly interbedded with
sedimentary rocks.
Reflecting changes in channel positions in their depositional environment,
many sedimentary rocks show marked lateral variations in grain size, com-
paction, and cementation, as well as changes in lense or stratum thickness.
Within the Lower Greensand series in the Weald area of south-east England,
the outcrop of the Hythe Beds, extending east and west across the Dorking
and Reigate district, is seldom over half a mile wide; in the Leith Hill area it
widens to 5-5 km as chert seams present in the sandy formation in this area
provide a cap-rock effect. Compared with the prominent salient of Leith Hill,
itssummit at 294 m being the highest point in the area, the land around
Dorking, where the Hythe Beds consist almost entirely of sand, is low. Even
in plateau areas, lateral changes are significant in influencing the details of
landform. In the Arizona plateau where many sedimentary rocks are per-
sistent units across the area, others are lenticular on a broad scale, or show
marked lithological changes laterally, indicating the effects of local basins and
swells in the sedimentational history. Along the eastern escarpment of the
Chuska Mountains in north-west New Mexico the front is cuspate in plan at
several places. The concavities are relatively thin units and the points of the
cusps are developed where the units are locally thickened. Even in uncon-
solidated material like morainic debris local concentrations of boulders and
blocks will form small promontories on a shore like the north coast of
Galway Bay, close to the town, or along parts of the margin of Lake Vattern
in southern Sweden. Fig. II.15.C illustrates a situation on a Cotswold dip-
slope where incision of a stream shifting downdip on insoluble clay appears to
deepen into underlying limestone where the clay dies out. While sediments are
deeply buried, lateral changes may be induced by folding, particularly of
incompetent strata, producing thinning on limbs and thickening in axial areas

by flowage. Jointing is of fundamental importance in weathering and since
systematic joints tend to be more widely spaced in thicker more massive
units, the significance of lateral changes in joint frequency is redoubled. In the
Appalachians there is a general decrease in the size of joint surfaces in shales
and coal from east to west, the range of spacings from centimetres to several
metres occurring on a local scale also. In the extensive strip mines of the
Houtzdale-Snowshoe area, narrow zones of closely spaced joints are
separated from one another by broad areas of few or no joints (Nickelsen and
Hough, 1967). Observations like these are particularly valuable because
variations in joint spacing are frequently inferred in the interpretation of
erosional features without being able to demonstrate that the apparently
weaker zone was in fact crossed by closely spaced joints before it was eroded
away. Similar zoning of joint frequency may occur in igneous rocks. In the
southern Beartooth mountains of Wyoming, erosion appears to have cut
elongate ridges parallel to zones of maximum microjoint development.
D.U. Wise (1964) suggests that the microjoint stresses coincided with the
closing phases of block uplift of the Laramide Rocky Mountains. When lifted
free of the confines of adjacent basins, the local fault pattern permitted
expansions, but in taking place in different directions, local dead spots in the
expansion pattern left some granite masses without microjoints. Therefore,
although difficult to demonstrate due to the removal of the required evidence,
there is a logical basis for suspecting that some inselbergs could be the more
resistant portions of a crystalline mass that possessed few if any joints. There
are many effects of joint zones other than guidance of linear erosion. In the
Skjomen area, northern Norway, postglacial weathering has been insignificant
on the less-jointed solid rock and there may be no talus at the foot of
cliffs.At other probably well-fissured points talus may accumulate to heights
of 20 m. As well as changes along a cliff or slope, differing densities of
jointing are an important consideration downslope too, particularly in crystal-
linemasses (fig. II.15.D). In north-western Eyre Peninsula, Australia,
some granite slopes are determined by huge curvilinear sheets of massive
granite, white those on closely jointed granite are boulder-strewn. Differential
weathering where joint frequencies change over short distances produces sharp
breaks of slope. In the Vosges, irregularities appear to be related to joint
density. The more densely jointed areas are susceptible to frost-weathering
while more resistant prominences gradually emerge from the surrounding
more fissile material. Fig. 11.15 illustrates some situations in which landforms
appear to be related to lateral variations in resistance of rocks.
In the study of smaller features vertical changes in properties of rock have
to be considered. These are most pronounced in sedimentary rocks where the
term basal conglomerate itself is perhaps sufficient illustration of vertical
1
changes. In unconsolidated sediments the pressure needed to shear one hori-

zon might be twice that required to shear another a few metres above or
below it. In consolidated sedimentary rocks unevenness in cementation leads
to corresponding irregularities of form including curious pedastal rocks
formerly attributed to wind action, like those near Montpellier-le-Vieux

in Languedoc.
Figure 11.15 Lateral variations in rock resistance and associated erosional and
weathering forms.
A and B. Vector diagram of valleys and fjords {left) and strike frequency diagram
of fractures {right) in South Lyngen, near Troms in northern Norway (from
Randall. 1961).
C, Vertical incision of a strike stream shifting down dip related to the lateral
disappearance southward of a clay beneath the lower limestone {after L. Richard-
son et al., 1946. Geology of the country around Witney); the river is the
Windrush. south of Leafield. and the disappearing stratum, the Jurassic Estuarine
Clay, underlying theTaynton Stone.
D. Steps in the longitudinal profile of a glaciated valley with over-deepening in
well-jointed parts {after F. E. Matthes. 1930. U.S. Geol. Surv. Prof. Paper 1
60) as
suggested by studies in the Yosemite valley.
E and F. Density of jointing controlling the distribution of penitent rocks at the
sub-surface weathering stage (£. Ackennann. Zeit. f. Geomorph., Vol. 6).
Physical, chemicaland biological
basis of geomorphological
processes
A. Regional-Glimate
1 / Temperature
Solar radiation received by the earth is the transfer of thermal energy by

means of electro-magnetic waves, no material medium playing an essential
role in its transmission. This is the energy which drives the chemical, bio-
logical and mechanical forces orsuS^fial weathering processes. This amounts
to 13x10*’ calories oT radiant energy each year, only 47 per cent of which
reaches the earth’s surface due to absorption, scattering, and reflection ofjjie
incoming waves by'the earth's atmosphere. Furthermore, snow'and ice may
reflect nearly half of this fraction, grassland nearly a quarter, bare ground 7-
20 per cent, and woodland may reflect 10 per cent or less. As all surfaces
reflectmore and absorb less energy when the angle of incidence of the light
waves is low. the amount of energy received equatorward of 38° latitude leads
to a net heating, compared with net cooling towards the poles. Oceanic and
atmospheric circulation counteract this lack of balance and prevent tempera-

tures at the poles and equator from becoming progressively more extreme.
Part of the total input of energy thus becomes an azonal influence on geo-
morphological processes because the water evaporated is distributed the world
over, maintaining ice-sheets, glaciers, streamflow. and winds are created.
The other portion of the incoming energy, involved in the actual heating of
the ground surface shows a pronounced latitudinal trend. The annual amount
of energy eventually filtering through to the ground surface is 60 000-
70 000 calories/cm’ in the humid tropics compared with a mere 2000-
5000 calories/cm’ in arctic deserts and tundra. The ground surface of mid-
latitude forest, grassland, and steppe receive between 10 000 and 40 000
calories/cm^. In cold areas, although the energy input is min imaL. there is in
less extreme climates enough heat to melt ice. This makes refreezing possible.
Physical, chemical and biological basis of processes 113
This process involves the reorganization of water molecules into a rigid

crySalline network with an associated 9 per cent increase in volume. Refreez-
ing thus generates stresses in confined spaces in soil and rock. With higher
temperatures although water molecules are increasingly in the random posi-
tions of a liquid and have a greater energy than the molecules of the solid, ice,
turbulent flow dissipates much of this energy as frictional heat. However, a t
higher temperatures, chemical, biochemical, and biological weathering
becomes increasingly important, rates doubling for every 10° C rise in 'tem-
perature. It is impossible to predict how effective the heat reaching the ground
surface is in land-forming processes. With the equatorward increase in clay
decomposition products and the increasing relative importance of solutional
weathering, thermodynamics have little to do with the transport as well as the
erosion of the bulk of the weathered material. Poleward, snow is a very poor
conductor of heat, particularly if in a dry state. The thermal conductivity of
snow may be almost one-tenth that of the mineral part of the soil, markedly
limiting frost penetration in winter. In Siberia the greatest differences between
air and soil temperatures occur everywhere in January as soon as the snow
cover begins to persist. Snow cover may even keep the ground-surface tem-
perature at or near melting point if there is heat transfer from the sub-soil. As
a result, in areas like the greater part of the high mountains in north Sweden,
permafrost is unlikely to be widespread although it is not possible to establish
a clear relationship of soil freezing with air temperature and depth of snow
cover. Even may be deep enough to keep bottom deposits unfrozen and
lakes
the ground beneath may remain unfrozen at all depths creating punctures
through frozen ground even in areas of extensive permafrost. In the Taylor
valley. South Victoria Lands in Antarctica, where average annual tempera-
tures are —20° C some of the deeper ice-covered lakes like Lake Bonney
remain liquid at depth. In summer in cold regions much of the small amount
of solar energy is absorbed merely in melting snow and ice, and increases in
soil temperature to promote biological and chemical activity are minimal. In
European U.S.S.R., the total inflow of heat between April and August is 1 000
calories/cm^ in the north-west, rising to 3000 calories/cm^ in the south-east;
the expenditure of this energy in thawing accounts for all that supplied in the
north-west and half of that supplied in the south-east. In humid and cool
temperate latitudes extensive peat deposits exist because summer tempera-
turesdo not provide adequate supplies of energy to evaporate the colossal
volumes of water held in these highly porous organic veneers. In warmer
humid latitudes photosynthesis and respiration of plants route an appreciable
proportion of the solar radiation input into the biological cycle. In the humid
tropics the task of pumping out water from dozens of metres of deeply
weathered mantle absorbs much energy. In addition there is not just an

acceleration of weathering reactions but equally pronounced accelerations in
processes involving the reprecipitation of dissolved solids, sometimes referred

to as back-weathering, which also makes it very difficult to assess without
measurements how much energy is converted into the work of net removal of
material from the land surface.
Several speculations about the influence of temperature on land-forming
processes are possible. The greater solubility of carbon dioxide at lower
temperatures has been used as an argument for expecting greater solution in
higher latitudes, whereas the increased reaction rate at higher temperatures
appears in qualitative terms as an apparently sound reason for expecting the
reverse. A decrease in temperature, by lessening the settling velocity of
suspended sediments, might suggest that sediment transport is more efficient
in higher latitudes. On the other hand, with viscosity decreasing with higher
temperatures, so that of tropical rivers is about half that of temperate rivers,
the greater turbulence could .suspend more easily the smaller-sized sedimen-
tary particles of those regions, thus offsetting the more rapid settling velocity.
Clearly the mere statement of these temperature-controlled chemical and

physical relationships can, without measurements from comparable natural
situations, lead to contradictory statements. A very large number of measure-
ments will have to be made before it will be possible to understand how the
immense energy of solar radiation is converted into the work of sculpturing
landforms.
B. Amounts and Motions of Water and Ice
1. Hydrological considerations
Water, the only compoun d that occurs naturally at the eart h’s surface in
gaseous, liquid, and solid states, is indispensable for weathering, erosion, and
for the removal of weathering products. In many situationsrwhatever may be
the detailed mechanisms involved in rock breakdown, it is ultimately the bulk
of water in runoff which is a major factor in influencing rates of denudation
(fig. III. 1 ). Climate determines the availability of this agent and its volume
and period of flow in streams, through soils and down slopes. Three hy dro-
logical characteristics of water available for geomorphological woj k are more
significant thanjhejotal precipitation.
First, the effective precipitation may, after the demands of evapotranspira-

tion have been met, be an increasingly small fraction of the total precipitation
where the supply of energy to vaporize water increases. At higher tempera-

tures, air can hold a very large quantity of water without its relative humidity
increasing. Evapotranspiration in the tropics is therefore potentially very
intense, and a month with only 20 mm is a dry month. The coefficient of flow
describes that percentage of precipitation which enters river channels, the
that at
volume of running water increasing at a slightly greater rate than
which the temperature decreases. Examples include the forested River
Njong
basin in the Cameroons where only 16 per cent of the 1800 mm precipitation
enters the river and on the Ivory coast coefficients are similar.
About 25 per
cent of the 1500 mm of precipitation falling in the Congo basin enters the
river. In low flat areas like the Yucatan platform, coefficients may be 10 per
cent or less. In the areas of more accidented relief the coefficients
may exceed
of Andean
50 per cent, and in the Amazon basin, due to the altitude its
Run - o»( tmm) Pfecip»lQtion during fneosurement mtervol (inches)
Figure III. 1 Some hydrological and temperature factors leading to greater

removal by fluvial transportation. (Z-c/r) Relationship of suspended sediment load
to runoff for catchments in eastern Australia {/’ro/n Douglas. 1967).
(Right) Relation of stream bank recession to precipitation and frost (J'roin M. G.
Wolman. 1959. Am. Jour. Sci., Vol.257). The numbers are percentages of the
observation interval during which mean daily temperature was below freezing.
The observations were made on the Watts Branch. Maryland.
headstreams. 40 per cent of the 2300 mm precipitation flows off. In general,

however, evapotranspiration in equatorial regions could account for about
1200 mm of precipitation, and runoff varies between 250 and 1000 mm
according to precipitation and slope. Great actual losses in semi-arid or arid
areas are experienced along wooded margins of allogenic streams or even
from the margins of ephemeral streams. For instance, in the Tombstone area
of Arizona losses from such situations are the equivalent of 300 of mm
precipitation per month. In the United States 70 per cent of the average
precipitation of 750 mm returns to the atmosphere due to evapotranspiration.
Comparable with this deficit is that of French river basins, which range from
475 to 520 mm. In the steppe zone of Russia precipitation decreases from
400 to 200 mm in southward and eastward directions, exceeded by potential
evapotranspiration figures of 700-800 mm. As an instance of mid-latitude
conditions in the Far East, the Han river basin in Korea has a deficit of
540 mm compared with a total precipitation of 1200 mm. In cooler environ-
ments. like northern Sweden where precipitation may be of an order between
1 1 6 Introduction to Geomorphology
600 and 2000 mm, evapotranspi ration is about 150 mm/year. In the tundra
of European U.S.S.R., precipitation of250-450 mm/year compares with an
evapotranspiration loss of about 100-150 mm/year. The Lena leads off about
175-1 90 mm of the 370-385 mm of precipitation and the loss in the
Yenessei basin is similar. To the south in the taiga part of the forest zone
precipitation of 400-600 mm/year similarly exceeds evapotranspiration of
150-300 mm. In the milder mixed broad-leaved forest zone, precipitation
increases to 500-700 mm, but because evapotranspiration increases also, the
moistening of the ground is no greater.
Intensity of precipitation is a second important factor in landform study
because not only is the importance of mechanical impact increased on bare
soil but also a high percentage of a downpour will run off with little lost to
evapotranspiration if it exceeds the infiltration capacity of the soil. Tropical

areas are prone to repeated downpours. In the high Guinea area of the Congo,
ten downpours of more than 50 mm falling at a rate of 5 mm/min have been
recorded. At Ibadan near the Nigerian coast, more than 70 per cent of the
annual total of 1200 mm falls in downpours of over 12-5 mm. At Bulawayo,
over a period of 13 years, falls sustained for at least 15 minutes included
one with a rate of 200 mm/hour, eight at 100 mm/hour, and fifty-two at 50
mm/hour. Even within a year downpours may be frequent. Over much of
West Africa falls in the 150-200 mm/hour range may be sustained for very short
spells. Monsoon Asia is another area experiencing huge downpours. In
southern China precipitation in excess of 400 mm may fall within 24 hours.
Occasionally extremely heavy downpours occur in other environments.
For instance, in 1966 heavy rain in Italy included falls of more than
500 mm in 2 days in the Alpine parts of the Tagliamento basin. During the
Exmoor storm in August 1952, which produced intensive erosion in the Lyn
valley, precipitation in excess of 185 mm was recorded. In the Crimea
midsummer torrential rains of between 50 and 100 mm are sufficient to
release mudflows. While intense erosion is a feature of the occasional once-a-
generation downpour in higher and mid-latitudes, in equatorial and mon-
soon areas intense downpours are experienced several times a year or even
annually.
A third faetor adds to erosional capacity of surface water in cool environ-
ments, the melting of snow and ice. For example, the channels leading to the
Kvikkjokk delta in Swedish Lapldnd are generally frozen from the end of
October to the second week in May. Annual total discharge is 870 million m^
to 1080 million m^, with daily discharge during ice-free periods of 25-
40mVsec rising to 200mVsec in floods. As melting reaches its peak, rivers
may increase in volume by 300-400 per cent in 24 hours. On the River
Colville in Alaska,42 was discharged
per cent of the total annual runoff
during 4 weeks of ice break-up. During spring flooding on the Mackenzie
discharges in excess of 500 000 cusecs (14 000 mVsec) have been recorded,
more than twice the maximum observed for the Colville. These figures com-
pare with the Yenissei where the annual flow of 17 400 mVsec is the flow of
only part of the year. Despite the huge width of these valleys the Mackenzie
m
may rise more than 6 during the spring melt and the Yenissei by 10 m. Ice
jams may cause overtopping of adjoining levels.
Although the chances of abrupt and substantial downpours or snow-melt
surges of flow are not high in temperate latitudes, floods are still an important
consideration in landform studies. Interest centres round the recurrence in-
terval. This is the average interval of time within which a flood of a given
magnitude will be equalled or exceeded once. It also is a prediction of the
chances of a flood of a given magnitude occurring in any one year. For

instance, a flood with a recurrence interval of 50 years has a 2 per cent
chance of occurring in any one year. The ratio of discharge for some recur-
rence interval to the mean annual flood gives a useful indication of possible
geomorphological effects of the extreme event in relation to average con-
ditions. In the United States, bankfull stage occurs once in two years out of
three. Similarly a recurrence interval of 1 •5—2 years for the bankfull stage has
been observed in north Sweden, and likewise the inundation of the Ob flood-
plain does not take place every year. In England and Wales discharges greater
than bankfull may occur on two days each year on average.
2. Hydraulic considerations
The quantity of water leaching through weathering profiles or draining down

slopes is the most important single factor in controlling the amount of
breakdown of parent materials and in their removal. Large volumes of water,
on average, even if dilute in sediment and solutes, remove by their sheer bulk
far more material than small flows with a dense sediment load and high solute
concentrations. However, in the study of certain distinctive localities, usually
those where unconsolidated sediments are being reworked, the lack of cohe-
makes some understanding of the motion of the water
sion in the alluvium
important an understanding of the features developed. Also the geomor-
in
phologist who might be interested in the hydraulicist’s studies of alluvial

channels requires some familiarity with the terms describing the ways in
which water moves.
In a thin boundary layer just above the static floor of a stream channel
water particles move in a parallel way with no cross-currents. This motion is
termed laminar flow. This is usually a mere boundary film with stream
or
slopewash velocities nearly always rapid enough for the
cross-currents of
turbulent flow to develop. Due to the rough surface of
channels in moun-
tainous areas there may be almost as much backward as forward motion in
8
the cross-currents and eddies. Turbulent motion consists oF two types, the
division depending on the Froude number
F = —!—
\JgR
where v is the mean velocity, g the acceleration due to gravity, and R is the
hydraulic radius of the channel. The last term is equal to the depth of water
for wide shallow channels. Streaming turbulent motion occurs when the
velocity is relatively low and is less than the square root of the hydraulic
radius times the acceleration due to gravity. Velocities with a Froude number
greater than 1 are those of shooting turbulent flow. In deeper channels with
smooth perimeters in cross-section and sinuosities in plan, cross-currents are
arranged in a circulatory motion because in a bend the faster-moving water

particles nearer the surface are subject to greater centrifugal force than cross-
currents near the bed. These motions combined with the overall downstream
flow cause a spiral motion termed helicoidal flow. This motion is observed in
meanders and at channel forks but is easily obscured by localized turbulence
caused by channel irregularities. Where a confined stream of water enters
a broad area of tranquil water it is useful to consider the analogy with
submerged jet flow, an hypothesis developed by hydraulic engineers, in con-
sidering the geomorphological implications of the sharply defined discon-
tinuity where the incoming flow is lined with intensely turbulent eddies and
flanked by zones of complementary reverse flow. However, the way in
which a sediment-laden stream decelerates and diffuses laterally is different
from that of a piped supply of sediment-free water discharging below a water
surface.
Changes in stream velocity are a difficult characteristic to discuss because
of the empirical constraints in the calculation of mean velocity from a series
of points in the cross-section. One constant which varies over a considerable
range is von Karman’s constant of turbulent exchange which compensates for
the damping effect of turbulence by dense concentrations of suspended sedi-
ment. In fact, formulae for velocity distribution in an open channel are not
valid at high concentrations of suspended sediment. Another variable factor is
channel depth. For a given volume of water velocities will be less in a broad
shallow channel than in a relatively deep narrow channel where channel
resistance is much less because the wetted perimeter is shorter. Increasingly
large volumes of water tend to move more quickly because, until bankfull
stage is reached, the relative proportion of flow involved in overcoming

frictional resistance of the channel becomes less. For instance, water flowing
into the Laitaure delta in Sweden with a volume of 253 m-’/sec, nearing
bankfull discharge, had a mean velocity of 1 m/sec whereas a summer low
flow of 39 mVsec. flowed at 0-4 m/sec. With flood discharges on the Ob,
Physical, chemical and biological basis of processes 1 19
velocities are between 1 and 2-0 m/sec. The average velocity of the Colville
- 5
river, Alaska, is 1-5 m/sec; for several streams in the United States mean
velocities vary between 0-9— 1-2 m/sec. Increases in the depths of channels
downstream improve the efficiency of the channels to transmit faster flows

and larger flows moving over diminishing gradients in lowland areas. The
study of these inter-relationships is of crucial importance to hydraulic
engineers concerned with flood control.
3. Waves and offshore currents
The description of waves advancing on a shore involves two basic measure-

ments of scale. The wave length, L, is defined as the horizontal distance
between corresponding points on two successive waves. The wave height, H,
is the vertical distance between the wave crest and the preceding
trough.
Waves consist of orbital movements of water which are not quite complete,
with water particles moving forward slightly as each wave passes producing a
slight displacement of water. It is the transmission of energy through water
waves that is the important consideration rather than the relatively slight
displacement of the actual water particles. The diameter of the orbital motion
diminishes rapidly downward in a geometric progression related to wave
length. The orbital diameter is halved for each increase in depth of |L. Wave
velocity in deep water is primarily dependent on wave length, but high waves
of a given length run somewhat faster than low waves. On coasts the wave
height gives a good indication of the total energy which is usually propor-
tional to the square of the mean wave height in most theoretical equations and
is largely a result of wind velocity, the fetch or the distance of open water
over which the wind blows, and the water surface gradient created by the tidal
range. Thus waves greater than 2 m in height are not frequent off the surf
zone in the Mediterranean where the mean tidal range is only about 25 cm.
Tidal range in the Great Lakes is as little as 7-5 cm. In contrast the classic
case is the Bay of Fundy funnelling tides of 3 m at its entrance up to a range
of 1 5 m at its head.
Waves where the depth to the bottom is less than half the wave length, L/2,
are defined as shallow-water waves. L/2, referred to as the wave base, is often
about 9 m below the surface. In entering shallow water the horizontal particle
velocity at the wave
becomes increasingly greater than that retarded
crest
near the bottom. At a depth equal to 2H the wave profile becomes very
peaked and asymmetrical. Breaking of the wave finally occurs when the depth
is about 1 -SH. If the bottom gradient
is gentle the crest of the wave spills over
the advancing front of thewave without completely destroying the wave form.
In contrast to the spilling breaker, a plunging wave forms
above steeper
bottom gradients when heavy swells pitch the wave crest into the
preceding
trough. After a wave of oscillation breaks, a wave of translation bores across

the surf zone until it meets the increased slope of the foreshore or swash zone,
where it exhausts its energy either by running up the foreshore slope as
swash, or by collision with the returning backwash which tends to form a
layer on the bottom in the swash zone. Where the bottom contours in shallow
water are not parallel, waves slow down

where the plan of bottom
first
contours projects seawards and the crests of waves swing round and tend to
approach an alignment tending to parallel the bottom contours, thus concen-
trating energy on the headlands.
Currents differ from waves in that there is a progressive movement in one
direction. Due to the hydraulic head of water piled against a shore by waves,
compensating seaward currents may form narrow lanes which exist right to
the water surface and cut through the breakers. These rip currents, although
first described only in 1941, are now recognized as a major element of near-
shore and surf-zone dynamics. They may maintain velocities of over 1 m/sec
for periods of minutes. Waves approaching a shore obliquely cause a uni-
directional movement alongshore. The zig-zag pattern of these longshore
currents may, however, be interrupted by seaward-flowing rip currents. The
greatest longshore velocities, which may be a few dozen metres per hour,
occur midway between the two high-energy zones, the breaker and swash
zones, which cause drag at the seaward and shoreward edges.
Tidal currents, produced by the ebb and flow of tides rarely exceed
3-2 km/hour, but where the flow is channelled through some constriction in
coastline form, currents may at times be as much as 16 km/hour. Currents
seaward of the breaker zone do not usually exceed 30 cm/sec, although seas
around the southern half of the British Isles are notable for the strengths of
their currentswhich are 2-3-5 km/hr over large areas, and 4 km/hr in
smaller areas. Off the New South Wales coast, a variable southerly current
moves at an average of 5-7 km/hr 8—15 km offshore with a more indeter-
minate and variable current running north closer to the coast. Ebb tidal
currents move through Hell Gate in the East River at New York at 9 kmAir.
Near-bottom currents in the Bering Straits average 5-4 km/hour. Other
similar currents are not restricted to the surface. In the Indonesian Strait,
strong current movement exists to a depth of 3000 m and a deeper flow
through the Straits of Gibraltar is in a direction opposite to that at the
surface, as dense saline water, a hundred times greater in volume than the
'Mississippi’s discharge. The Gulf Stream moves 5-12 km/hour,
at rates of
with the eastward-flowing equatorial counter-currents moving at three times
these speeds. Smaller-scale circulations due largely to the strength and direc-
tion of the wind may move a few kilometres a day like those moving anti-
clockwise in the North Sea. In the Mediterranean where tides are insignificant
there is nonetheless a south to north current in the Levantine basin which is
about 0-9 km/hour, probably influenced by the Nile’s discharge. Although

these currents may all play some part in redistributing river-borne sediment
entering the sea, they do not exert sufficient drag on the bottom, below 10m
down, sufficient to resuspend deposited material.
4. Ice and snow
Glaciers cover about 1

1
per cent of the land area of the earth, 30-50 per cent
of the land area is covered with snow and about 23 per cent of the ocean area
is covered by sea ice. The Arctic ice. although it has a profound influence on
heat exchange in the oceans is only 3 m thick. The Greenland ice-sheet covers
1 700 000 km^ with a depth of 3 km
at one point, and the Antarctic covers
12 950 000 km^ and in thickness exceeds 4 km at places. These two masses
account for about 99 per cent of the volume of the ice on the globe and their
marginal tongues of ice may be larger than valley glaciers, like the Beardmore
glacier in Antarctica, which is 200 km long and about 40 km wide. There are
also some very much smaller ice-caps on uplands in higher latitudes. At
70° N the Barnes ice-cap in Baffin Island covers 5900 km'^ and is about
165 km long and 22-62 km wide. In the Icefield Ranges the most extensive
ice-cap in the North American continent is 240-670 m thick. In contrast to
the dome shape of summit ice-caps, glaciers occupy valleys and a few con-
tinue beyond the confines of the valley to broaden out as piedmont glaciers.
The Malaspina glacier near Yakutat emerging from several narrow valleys in
the St Elias Mountains spreads out in a broad lobe about 50 km across and
about 40 km long. Valley glaciers may reach lengths of 30-50 km and 3-
4 km in width, whereas in other areas the ice-covered area consists of a large
number of small glaciers. In Transcaucasia there are 487 glaciers with a total
area of only 635 km^.
The process of the conversion of snow to ice starts with the recrystalliza-
tion and partial consolidation of porous fresh snow, transforming its low
density of about 0- 1 gm/cm-* to the densities of 0-4-0 -8 gm/cm-* of granular
snow termed firn or neve, and finally to ice with a density of about 0-9].
Transformation of neve to ice takes place at depths of about 30-40 m but
may be nearer 1 00 m for ice-sheets. In temperate zones higher temperatures
speed up the transformation process which takes about 50-100 years,
whereas in polar ice-caps recrystallization, taking place without passage
through a liquid stage, takes 200-300 years.
An
important concept in discussing the distribution of snow and ice is the
‘snowline’, recognized in 1887 by Bruckner and today commonly defined as
the lower limit of perennial snow. The definition of the climatic snowline
disregards local influences like wind, insolation, and aspect which make any
close correlation between temperature— precipitation averages and the
snow-
line unrealizable, whereas the orographic snowline links the points at which
the preceding winter’s snow will just disappear. In Spitzbergen, at 78“ N. the
snowline is approximately 500 m. In Iceland at 65° N the snowline rises
towards the south-east from 600 to 1 000 m, is about 1600 m in southern
Norway at 60° N, about 2700 m in the Alps, and at 4250 m on Mount
Kenya. A very important fact is that east-west changes in snowline are
usually more abrupt than the general equatorward rise. The snowline on the
dry eastern slope of the St Elias Mountains, Alaska is 1500 m higher than on
the wet west slope.
Close to the observable firn line is the theoretical equilibrium line, dividing
the zone of ice accumulation near a glacier source from that of net loss in the
zone of ablation in its down-valley part. Studies of glacier regimen, consider-

ing the balancing processes of accumulation and ablation, are simplified in
comparison with water-balance studies by the absence of significant evapora-

tion losses. Such changes, together with the condensation of ice from water
vapour known as sublimation, are less than 1 per cent. This is a result of the
very large supplies of energy needed for vaporization, being 540 calories per
gram, and because absorption of moisture by cold air is very limited, as
amounts of water vapour range from 4 per cent by volume in hot tropical
climates to only 0-01 per cent in glacial climates. Near the snouts of many
Scandinavian. Icelandic, and Alaskan glaciers the ablation loss of ice may
amount to 10 m or more in one season. In a more southerly location, in
British Columbia, the Salmon glacier at 56- N. with an accumulation zone at

1650 m, loses about 5 cm/day. One of the more striking characteristics of
glacial budgets, and one which emphasizes how sensitive some interactions
are to changes, is the degree to which ice-caps generate the local climatic
conditions to maintain a supply of moisture and snow. The Barnes ice-cap, for
instance, is just a low dome on an even upland surface, generating its own
nourishment in the form of orographic precipitation. Later workers have
enlarged on Leverett’s (1916) view that the North American ice-sheets grew
westward because of storms coming from the south-west. Warm moist
Tropical-Gulf maritime air masses would have been cooled both by coming
into contact with the ice-front and by being forced upward over the glacier
and its associated wedge of cold air, creating a major zone of precipitation
along the ice-front. Similarly, self-enhancing conditions in the development of
the Greenland ice-cap were suggested by Wegmann in 1941. He surmised
that as the climate cooled at the end of the Pliocene, the relief of Greenland
was embayment, surrounded by bordering chains on which glaciers
that of an
developed which spread into the interior lowlands as piedmont lobes. It seems
possible that if such sheets coalesced and the ice surface rose to the perman-
ent snowline neve supply could have then added directly to the surface.
Like the effect of channel floor on modes of stream flow, the irregularities
of the sub-glacial floor make the movement of ice a very complicated pheno-
menon, with the variables influencing the mass budget and neve formation
changing with altitude, aspect, latitude and in a downvalley direction. From
early attempts to describe ice-flow, using laws of viscosity, impressions were
gained of a spreading body unable to conserve its form indefinitely. Recent
measurements exclude the possibility of viscosity existing in glacier ice.
Instead J. F. Nye (1952) suggests that ice yields to compression stress and
strain rate imposed by an external force like a plastic solid which retains the
form it took under the effect of deformation. It seems that due to pressure
within a glacier increasing with depth, melting and refreezing facilitate inter-
granular movement, crystal growth, and transfer of material. On a smaller
scale grains tend to be drawn out in response to the forces acting on them,
while on a larger scale, intergranular movement takes place. Melting at grain
boundaries reduces friction between the grains and facilitates movement of
the grains relative to each other. In some temperate glaciers ice aggregates
may be surrounded by a weakly saline membrane which, by tending to remain
liquid when refreezing occurs, may facilitate inter-granular movements. This
mode of mechanicalmovement is termed plastic deformation and depends on
a certain depth of ice, about 20 m for temperate glaciers and on some down-
valley gradient. Under a laboratory stress of about kg/cm^ ice flow becomes
1
appreciable, but in field conditions with loads imposed for much longer
periods of time, loads of less than 100 gm/cm^ are sufficient. Probably much
of the movement of thick temperate-latitude glaciers on gentle slopes occurs
by plastic deformation, although the rate of movement due to plastic deforma-
tion decreases rapidly near the base of a glacier. However, a second process,
that of sliding on the bed, is an important, if spasmodic, addition to the total
movement of the glacier. It is accompanied by tensions which, if shearing
occurs, create fissures at the surface. The French geophysicist,
large
L. Lliboutry, Weertman. have done a great deal to clarify concepts of
and J.
basal sliding. This mechanism may account for as much as 90 per cent of the
total movement, particularly where there is a steep gradient, a high absolute
velocity and if the ice is relatively thin. Clearly, temperatures at the base of
the glacier must be near melting point for sliding to occur and in very cold
areas all but the most thin glaciers are frozen solid to the rock at their base.
B. Kamb and E. La Chapelle ( 964) made the first direct field observations of
1
mechanisms involved in the sliding of a glacier over its bed. The observations
of N. A. Ostenso (1965) have added to evidence which suggests that melt-
water lubrication is an important factor. As water has a greater density than
ice, water-filled spaces in and below the ice cannot be compressed without
being previously emptied. At the base of the Casement Glacier, the winter slip
rate in 1966-7 was 2-3 cm/day compared with summer rates of 2-9 cm/day
in 1966 and 2-6 cm/day in 1967. Over a 24-hour period there is a direct
1 24 Introduction to Geoinorplwlogy
relation between slip rate and water available at the base for lubrication. The
breaking up of the glacier into blocks makes it possible for the ice to flow by
still another mechanism, moving like a powder on a giant scale, particularly if
water fluidizes the ice. Meltwater circulating in the ice in tunnels, running
under pressure in siphons and enlarging fissures provides nearly all the water
at the base of a glacier. Even in Iceland where the geothermal heat supply is
two to three times the normal supply, minimal source of heat accounts for
this
only 5 per cent of the runoff in mid-winter. However, in the case of some
glaciers, ground water may be an important part of the discharge at the winter
minimum, according to observations in Switzerland and northern Sweden,
and may make significant contributions to the lubricating effect.
In high latitude glaciers and ice caps where the ice is welded on to the
bedrock, the slow movements take place along a zone of shearing in the
lowest 00 m, and in formerly ice-covered areas of deep narrow valleys
1
some experts find it conceivable that ice sheared across the subjacent ice
filling these valleys. Near the termini of glaciers shear planes similar to
thrust faults are often observable, but this process appears to be of only very
local significance. It is important to distinguish ice-carapaces from glaciers as
the area on which the first may occur, as seen in Alaska at the present day.
could easily lead to an overestimation of the thickness of a former glacier. For
instance in central parts of the Trinity Alps. California, there are extensive
glaciated rock surfaces on sloping valley walls far above the level of the
former ice streams. However, here striae, scoured and plucked rock surfaces
and schrund lines show that the valley walls in these areas had a mantle or
carapace of ice possibly no more than 60-90 m thick, that flowed towards the
valley axis.
The velocity of displacement of ice tends to be proportional to the cube or
fourth power of stress. Velocities usually range from a few millimetres to a
few metres per year. The very cold glaciers may move only a metre or two in
a year despite their immense thickness, although the Ferrar glacier near the
Ross Sea moves at 5 cm/day and the Beardmore glacier at 0-8 m/day. On the
west coast of Greenland the velocity of several glaciers entering Disko and
U manak baj's is about I • 5 km/year. In the Alps speeds of ice surface flow are
of the order of 40 m/year, small glaciers in China move 10-30 m/year, in
Patagonia the Grey glacier moves at 450 m/year, and extreme cases in Alaska
move at over 60 m/day. In addition to the factors like ice thickness, slope,
and temperature, mean velocities may change for a variety of circumstances.
Velocities, for instance, are not constant down-valley; the Tasman glacier in
New Zealand flows at a rate of 0-5 m/day 19 km from the snout and at
0-35 m/day 10 km from the snout. Mean velocitiesmay change from one
decade to the next. In Swedish Lapland, the Mikka glacier which now termin-
ates just below 1000 m altitude had a mean velocity of 6-2 cm/day in 1895-
7; in 1899-1901 it was 7-3 cm/day but was half this figure in 1958-9 and
only 2-9 cm/day in 1961-2, In fact, during the last century most glaciers
have been shrinking. Distinct from flow velocities are rates of change in the
position of the ice-front. Studies of varved clays suggest that the ice-cap front
of the last glaciation retreated at 100-150 m/year from Scania, the southern-
most province in Sweden, while in vacating areas beyond Stockholm, rates
increased to 200-300 m/year. Retreats of ice-fronts, however, are essentially

due to glacial thinning rather than to processes operating at the ice-front
itself.
Movements of frozen snow and ice, in addition to glacier flow, merit some
eonsideration. Snow-creep is a slow continuous glacier-like movement of a
snowpack. Stress is too small to produce shear failure as in avalanches but
appreciable slip may take place along the interface between the ground sur-
face and base of the snowpack. Investigations at 1 100 m on Mount Seymour
in British Columbia found movements of 60-105 cm within a winter with
maximum rates of movement of 12-5 cm/week. The localities where ava-
lanches occur are usually those where precipitation, increasing with altitude,
leads to great thicknesses of snow cover which shear under their own weight
at altitudes where atmospheric circulation multiplies temperature oscillations
through freezing point. In consequence mountains in temperate oceanic areas
are the domain of avalanches and may occur there within an altitudinal range
of less than 500 m. Wet snow may slide on slopes as low as 15 degrees.
C. Mechanical and Frictional Forces
The mechanical and frictional forces involved in the physics of denudation

are simply those of the shear stress exerted by the moving agent and of the
resistance of the land-surface material to that stress. Most of the essential
features of flow in fluids and deforming solids can be represented as in fig.
III. 2 by a sheet of infinite width and length flowing down an inclined plane. If
^ ax/s
Figure III.2 Stresses and strains for laminar flow on an inclined

plane. H is the
thickness of the flow, Z the height above the inclined, A is the
slope angle, K the
average velocity, Fs the velocity at the surface and Fb the velocity
at the base.
the flow is assumed to be essentially laminar the shear stress acting on a plane
perpendicular to the z axis is equal to
Density Acceleration /Thickness of Height above \

X slope
of flow due to gravity \ flow, H .
^
,
inclined plane, z/
angle,
,
A .
The maximum shear stress occurs at the base of the flow, where z equals 0,
although offshore in depths greater than 9 m this case occurs above the basal
plane and no shear stress is exerted on the sea-floor. Shear stresses within
actual flows are usually too complex, even in theory, for most geomorpholo-
gists to consider that the study of the greater complications within actual
flows in the natural environment fall within their sphere of competence.
The force of resistance to movement of
land-surface material involves two
quite contrasted aspects. The
breakdown of solid rock by weathering
first, the
processes may take millions of years and is considered in some detail in
Chapter IV.A. The second concerns the initial dislodgement of a weathered
particle or the reworking of unconsolidated deposits. Here much of the
resistance is simply internal friction. As the shear stress exerted by some flow
on its basal plane increases, there comes an instant when a few of the loose
fragments on this plane are entrained. Fig. III.3 illustrates how the force of a
flow turns an individual particle about its leeward points of contact with
subjacent particles. The force of the grain’s weight, or its immersed weight in
a fluid, acts through the centre of gravity of the particle to pull it backwards
and downwards. When the drag force of the flow just exceeds this, movement
takes place. On a scree, steeply piled sand, or similar accumulation, the value
of the angle A is very close to that of the angle of repose.
Figure I1I.3 Physical forces acting on a particle at the threshold of movement

{after Bagnold, 1941). W is immersed weight, p the drag force due to
the grain’s
the fluid acting parallel to the general surface, and A is the angle at which the grain
is poised on adjacent grains, close to the angle of repose.
D. Sediments and Mechanical Characteristics of Soils
•On beaches . . . with the motion of the waves, oval pebbles are driven to the same
place as oval and round to round . . Democritus
There are four reasons why the geomorphologist may find some knowledge of
the physical properties of soils and sediments useful. First, sediments may, as
already discussed, develop properties characterizing their depositional en-
vironment, and it is often a simpler, if not the only approach, to measure
sedimentary parameters rather than the depositional processes themselves.
Secondly, thereis a close interdependence between soil formation and land-
form development, and, since Barton 1916 described accelerated weather-

in
ing below old soil lines in Egypt, several ways have been recognized in which
soilsand sediments can either accelerate or retard further weathering of
bedrock. These inter-relationships will be discussed in Chapter IV. D. 9.
Thirdly, fossil soils are a record of phases of erosion and stability over
thousands of years and of environmental conditions during periods of stabi-
lity when soil formation took place. Fourthly, because the spans of time
during which soils change tend to be shorter than times involved in landform
change, notions on rates of the latter might be grasped from studies of soil
adjustments to changing conditions, which could guide thinking on the degree
to which landforms might adjust from one change in environmental con-
ditions to the next.
Soils and sediments become an integral part of landform study, not simply
because they may afford useful aids but mainly because any advantage to be
gained depends on a full awareness of their complexities or of limitations on
interpretations which might expose over-optimistic approaches or superficial
generalizations.
1 . Particle sizes
Since the publication in 1914 of J. A. Udden’s attempt to relate particle-size

modes to different kinds of particle movement, the significance of particle size
has been much studied. Sizes determined by sieving and settling velocities
indicate approximately the diameter of a sphere of equal volume to that of the
particles. Particles smaller than 2 mm,
usually taken as the upper limit for
sand-sized particles, are often described in microns. micron is one thou- A
sandth of a millimetre, often symbolized by the Greek letter mu t. References
4
to particle sizes may appear in three contexts. First, there are
boundaries
between particle-size classes which are arbitrary but necessary in establishing
convenient subdivisions. There are several schemes of subdivision, but most
use a reversed log scale with each equal subdivision spanning
a progressively
narrower size range, so that a small number of large particles does
not
Particle size (phi units)
weight)
(by
liner
percentage
Cumulative
Estimated porticle size (millimetres)
Figure 111.4 Cumulative frequency curves showing particle-size distributions of

sediments.
A. Percentage points used in the calculation of indices to describe particle size

distribution.
B. Illustrations of the distinctive sorting achieved in wind-transported sediments:

the fine silt is Saharan dust, wind-blown to Britain (A. F.
Pitty. J968, Nature, Fol.
220) and the coarse from the Derbyshire limestone plateau is attributed to
silt,
wind-transport during former periglacial conditions by C, D. Pigott (/’/fO’' 1966).
C. Particle-size distributions drawn on Rosin’s Law of crushing paper {from J. T.
Andrews. 1963. Geog. Bull. No. 20). an alternative to the popular semi-
logarithmic paper, illustrating the similarity between crushed gneiss and tills from
cross- valley moraines.
—
dominate the distribution. Secondly, there are specific diameters to describe a

cumulative frequency distribution of a sediment sample, of which the useful-
ness of the Pjo, the median indicating the central-size value of the transformed
distribution (fig. III.4), is most readily apparent. This value may indicate the
average velocity conditions of the depositing agent. Thirdly, there are critical
sizes that appear to have some functional significance. For instance, frost-
shattering appears to be incapable of splitting grains finer than 1 0 microns;
the higher fraction of particles smaller than the 60 microns range in river
sands compared with beach sands appears to be the main distinction between
these two sediments. However, in general it is inadvisable to make direct and
simple inferences from diameter measurements alone because so many

variables are involved. Equally diagnostic are sorting indices which describe
the range of sizes involved in the rearrangement or adjustment of particles to
specific dynamic parameters, like current velocity. For the purposes of sedi-
mentary petrology, the Greek letter phi is often used to describe the logarith-
mic transformation in terms of units of equal arithmetic width,
0 = — logj d or cl = (t)
where d is the particle size expressed in millimetres, as suggested by

W. C. Krumbein in 1934. However, partly because tables of negative logar-
ithms to the base 2 are not readily available, a certain amount of difficulty in
the use of the phi transformation has arisen. Fig. III.5 is an attempt to provide
sufficient detail on the micron and phi scales to permit interpolations of
accuracy adequate for general purposes. For single expressions of central
tendency, a description in millimetres or microns is probably more explicit.
For sorting indices values of phi are obtained from cumulative curves for
certain cumulative percentages (fig. III.4). The most commonly used sorting in-
dex is Krumbein’s (1938) phi deviation, — whereas A. Cailleux sub-
tracted 050 from 075 and 025 from 05o, and used the smaller value, repre-
senting the steeper part of the curve as an index. He. In more recent work
more efficient indices are based on a wider spread of percentiles, R. L. Folk
and W. C. Ward suggesting a modification to Inman’s (1952) statistic
^ 084 -<l>lf. 095 -05

^ ^
4 6-6
Fig. III.4 illustrates the steepness of curves reflecting the efficiency of wind
transportation as a sorting process. A further measure is skewness, an expres-
sion of whether material is predominantly in either the coarser or the finer
side of the median diameter or symmetrically distributed on either side. In fig.
III.4 the median is coarser than the mean and the particle-size distribution
is
therefore positively skewed. This skewness index can have some value where
the finer tail of a distribution might be removed, like the fraction smaller than
62 microns winnowed from beach sands or gravel lags which, due to the
truncation of sizes, tend to have negative or near zero skewness values.

However, negatively skewed beach sands become positively skewed when
either the coarse tail of a symmetrical log-normal curve has been removed or
(-3 3 - 6 6 0 range) 70 40 60 60 100

(0 - 3 3 0 rar^ge) 2 4 6 e 10
0
-2 -1
e I
-5
5
t -4
6
’0
(0-330 range) 02 04 06 08
(3 3 G 6 0 ra-'ge) O 02 0 04 0 06 0 08
(6 6 9 9 0 range) O 002 0 004 0 006 0 006
mfihmetre «au *za«er»i to phi values on vert'cai scales
Figure ttl.5 A metric-p/ij unit conversion chart. The rows are metric units,
preceded by the phi range which they correspond;
(in brackets) to the latter are
read off the appropriate column on the ordinates.
a fine tail added by finer material filling up a porous frame of coarse material.
In contrast to beach materials, transport by unidirectional flow tends to
produce positive values of skewness as is often the case with riv'er and dune
deposits. Folk and Ward suggest a phi skewness inde.x
hH - 20.,
Sk ^
\ '/'.s4-</’l6 ''’vj-'T /
In addition to a reversed log scale sediment size distributions are occasion-

ally plotted on Rosin's ‘law of crushing’ paper, as both mechanical and
chemical disintegration of rocks has a distribution following this ‘law’ for
artificially crushed material which apparently results from random breakages

of material (fig. III.4).
Great care and caution are necessary in interpreting textural parameters.

In addition to reflecting environmental conditions they are influenced by the
availability of materials of a given size. In some environments several pro-
cesses may operate simultaneously or occur within a short interval. More-

over, even at one point more than one population may be present. It may
include grains deposited from the bedload, material settling later from the
suspended load and contributions from an underlying veneer deposited under
different conditions. There is also the possibility of post-depositional resort-
ing by by organisms. These problems are magnified by the difficulty
frost or
of recognizing polygenetic soils and sediments in the field. In consequence
there is no guarantee that textural parameters will necessarily establish clear
differences between dune and beach, glacial and proglacial, blockfields and
boulder clay, or between alluvial and colluvial deposits.
2. Particle shape
There have been many attempts to use particle shape in reconstructions of

erosional and depositional environments. Indices are basically of two types.
First, in coarser materials, indices are based on measurements of the three
axes length, breadth, and depth, which might be abbreviated to L, B, and D
L+B
respectively. A. Cailleux introduced a flatness index which has some
discriminatory power, as fig. III.6 shows. E. D. Sneed and R. F. Folk (1958)

suggest a detailed classification of a sample into several shape categories. The
second type of index concerns the smoothness of particle outline. Powers
approach depends on the use of a set of reference images with which each
particle is compared and classified. For larger particles part of the circumfer-
ence of pebbles is measured. Where corners are rounded the radius, /•. of an
inscribed circle is estimated by superimposing the corner on a set of concen-
tric circles. A. Cailleux and J. Tricart have made extensive measurements of
an index of wear, —
L
Although criteria distinguishing the effects of distinctive environments on
shaping pebbles are necessarily somewhat tenuous, one or two tendencies
have been observed, or significant distinctions established (fig. III. 7). It seems
that most well-rounded boulders of glacial transport may retain traces of

original concave surfaces to a far greater extent than is generally observed
on
stream or beach cobbles. The most important factor governing pebble shape
IS usually, like the initial triangular form of some glacial boulders, litho-
logical composition. Also a doubling of the mean size of particles may
L- B
L-D
Figure III.6 Some morphometric properties of sedimentary particles.
A. The main axes of L = length ) B = breadth ) D = depth. B

a pebble. These arc
and D need not intersect at the same point in the horizontal plane, P; also shown is the
radius of the circle inscribed in the least rounded corner in the horizontal plane, P.
B. Triangular graph of pebble shape (Jrom Sneed and Folk. 1958). Shapes of
pebbles falling at various points on the triangle are illustrated by a series of blocks
with axes of the correct ratio. All blocks have the same volume.
C. Various pebble shapes and their Cailleux indices of wear, —x 1000 [after
Kuenen. 1956. Jour, Geol., Vol. 64).
D. Examples of the appearance of sand grains under the microscope [after F.

Ottman. 1965. Introduction a la geologic marine et littorale); i = angular and
unworn, ii = sub-rounded, iii = rounded and polished, iv = pitted.
E. Examples of the appearance of sand grain surfaces on election micrographs
[from Krinsley and Donahue. 1968)-. (i) mechanical V, showing central depressed
area [above) with rough surface in cross-section [below) and (ii) chemically etched
V [above), showing a flat raised centre in cross-section [below).
Physical, chemical and biological basis ofprocesses 133
Glaciomarme drift 'o Upper
Tills Raised beaches A Intermediate
A B ^
1
• Lower
400
09-
0 7-
ratio
0 5- A
o 300 o A
Void
o
o
0 3- /a'
0 1 - o
"T 1 1 1 1 1
T — T
\ \
1
1
1 5 1 7 1 9 21 23 25 27 140 150 160 170

Bulk density Index of flatness
Figure III. 7 Differentiation of depositionai environments by sedimentary properties.
A. Contrasts in compaction between sediments {after Easterbrook, 1964), the tills
being the Vashon and Sumas tills in the Puget Lowland, Washington.
B. Contrasts in the Cailleux indices of wear and flatness for different levels of
raised beach {from Godard, 1965). The measurements were made in the vicinity of
Helmsdale, Sutherland, where the lower raised beach is about 1-5 m above sea-
level and the upper raised beach averages 40 m.
affectform and sphericity of pebbles to a greater extent than several hundred

kilometres of river transport in extra-tropical areas. It seems that in general
larger cobbles are less susceptible to breakage and therefore may become
more rounded. It is often very difficult to estimate whether progressive
change of shape might take place with transport, whether a balance of round-
ing and fragmentation is established, or whether there is a continuing supply
of fresh material, perhaps frost-shattered or solifluction gravel, downvalley or

along the shore. A
major factor appears to be the presence of sand because,
like ventifacts polished and shaped by wind-blown sand to the most distinc-
tive of seems that gravels in water acquire their shapes when
pebble shapes, it
stationary by wet sand-blasting. Global generalizations about pebble shape are

probably unobtainable partly due to changes in parent material, varying com-
binations of suites of rocks in a sample and latitudinal changes in weather-
ing conditions. These may range from periglacial areas where the supply
of coarse angular material on slopes or entering stream channels and beaches
is abundant to tropical areas where rapid chemical weathering makes coarse
material rare on coasts and inland recementation makes it impossible to
separate much of the highly weathered material into discrete particles.
3. Surface appearance of particles
The surface appearance of sand grains may show degrees of dullness or

shininess due to distinctive modes of etching of the surface (Cailleux and
Tricart, 1959). Apart from classifications by surface appearance, these sur-
face textures were first studied in detail with an electron microscope by R. L.

Folk and C. E. Weaver (1952). Recently a range of distinctive patterns has
been established, like small V-shaped indentations, characteristic of beach
environments and possibly due to grain collisions which have a maximum
density of about two notches per square micron. Combinations of four or
A B
Figure 111.8 Particle disposition and orientation.
A. Rose and dip diagrams showing pebble orientation in an East Anglian till
{from R. G. West and J. J. Donner, 1956. Quart. Jour. Geol. Soc., Vol. 112), In
the rose diagram (above) each circle represents half a stone in a given 10-degree
class. Half of each frequency is placed on opposite sides of the diagram. In the dip
diagram (below) the radii indicate long axis direction and the concentric circles the
angle of dip. from 0 at the circumference to 80 on the innermost circle.
B. The Schmidt net (lower hemisphere), a three-dimensional fabric diagram illus-
trating data from a laboratory delta (front Johansson. 1963). (Left) A point diagram
for the T-axes of a foresei bed. (The crosses indicate pebbles 0-10 cm from the
delta lip. dots those settled farther away.) A segment of co-ordinate counting
network is included and two pairs of circular counters (1/100 and 1/180) are
also illustrated. (Centre) Data plotted on the point diagram frequency contoured
with a 1 per cent circular counter, centred at square grid intersections on a co-
ordinate counting network, (/t/g/t/) Frequency, as points per unit counter area.
more distinctive markings serve to separate glacial, beach, and dune environ-
ments. It has also proved possible to distinguish between high, medium, and
low energy beaches. These microfeatures are identifiable in ancient sediments
(Waugh. 1965), are reproducible experimentally, and in Quaternary studies
in East Anglia have provided interpretations generally in close agreement
with stratigraphical fossil evidence.
4. Disposition of particles
Textural parameters alone may not provide the information required for
certain reconstructions, and the geomorphologist may at times have to ven-
ture even farther into sedimentary petrology and soil studies and consider
bedding structures and soil fabrics with his textural studies. Disposition of
large fragments is increasingly studied since K. Richter ( 1 932) made the first
detailed studies of stone orientation in boulder clays. It is now established that
the orientation of stones in tills is parallel to the direction of transport,
although the mechanisms involved are little understood. It is also appreciated,
however, that the orientation may change within a few metres vertically or
over a few dozen metres horizontally, and that a later ice advance may
complicate a pattern by inducing some reorientation in an underlying till
deposited during an earlier glacial phase. The selection of a plane in the till for
orientation analysis poses a further problem. In solifluction deposits the pre-
ferred orientation of the long axes also tends to parallel the direction of
movement. In streams the preferred orientation tends to be on average per-
pendicular to the direction of flow. However, some flood deposits orienta-
in
tion may be markedly longitudinal. It seems that two forces are involved and
that where gravity is important, orientation tends to be longitudinal, while the
dominant influence of hydraulic tractive forces tends to generate perpen-
dicular orientations. For smaller particle sizes in the sand range, however,
parallel orientation is widely observed, whether the current be backwash,
stream, or air flow. A rose diagram is useful for representing orientation data
(fig. III.8). Classes 10 degrees in width are commonly used and the number of
stones in a given orientation class are counted. Half of these are placed on
the diametrically opposite class on the diagram because, unlike a wind rose,
there is no movement in either direction along the axis of orientation. The
departure of each peak from the centre of the rose diagram indicates half the
total frequency of pebbles lying along a given orientation.
Any process which produces some preferred orientation in a horizontal
plane in sediments must also affect its disposition of inclination in a vertical
sense. Near Wadena Minnesota, an up-glacier inclination of the long axes
in
of stones was observable in some parts of drumlins. whereas in other parts
where the orientation was still well-marked there was no preferred inclination.
In East Anglian tills, the latter situation prevails with no preferred inclination
in the orientated stones. A systematic upstream inclination is widely recog-
nized in river gravels, whereas on beaches changes occur over short distances,
due to the abrupt and fundamental changes in the characteristics of water
movement. This is an important and widely recognized feature of sedimentary
deposits. S. B. McCann (1962) was able to suggest that two large gravel
deposits at Corran near the entrance to Loch Etive in west Scotland were
unlikely to be marine gravels, but fluvioglacial fans because the pebbles
dipped systematically in the opposite direction to the bedding sets and had
apparently escaped the reworking which marine action would have caused. In
screes, debris, if sufficiently elongate, tends to pack with an inclination less
than that of the angle of repose, but instances where the converse holds may
indicate important contrasts in the mode of scree material movement. In the
arrangement of rock fragments in superficial soil layers on slopes it might be
assumed that preferred orientation or inclination would not be well-marked,
but few measurements have been made.
Diagrams show both the inclination and the orientation of a pebble
to
sample may be based on a series of concentric circles. Radiating lines, as in a
rose diagram, represent the orientation, with the concentric circles represent-
ing the angle of dip, horizontal at the outermost circle and diminishing to 90
degrees at the centre point (fig. III.8). A dot on this diagram represents each
stone. Polar co-ordinate paper makes a similar but more sophisticated display
of pebble fabric data possible. Here the radius of the concentric circles varies
so that each sector bounded by these and the radiating lines has equal area.
This is an equal-area azimuthal projection, more simply described as a hemi-

sphere. The radii of each n circles r,„ is on a square root scale so that
/•= R\/n), where R is the radius of the whole projection and n the selected
angle of dip. In drawing contours round the frequencies, the number of points
counted in the four cells adjacent to any intersection is used.
5. Soil fabrics
Recently fabric studies have been extended to features indicative of move-

ment. reorganization and concentration in soil materials. The primary units of
study, analogous to the coarse fraction in a pebble fabric, is the ped, a cluster
of soil aggregates separated from neighbouring peds by recognizable surfaces
of weakness or voids. Soil fabric analysis, therefore, deals with the size, shape,
and arrangement of soil peds and voids, and as this approach, initiated
by R. W. Brewer (1964) begins to specify more accurately the processes
involved, it may become of increasing relevance to landform studies.
6. Soil hydrology
The mode and rate of movement of soil water is of fundamental importance in

influencing the proportion of surface runoff in an area. It is well known that
heavy downpours are particularly efficient geomorphic agents when ante-
cedent precipitation or the thawing of ice. concentrated in the surface soil by
freezing, has already saturated the soil layer. The amount of water in a soil,
expressed as a percentage of dried soil in a sample, is closely controlled by the

porosity of the soil. Porosity in turn is influenced by size distribution, but
varies with different degrees of compaction and will vary with the moisture
content itself if colloidal particles are present. Also, there appears to be a
distinction of size orders of pores between inter-aggregate porosity, as occurs
between peds, and intra-aggregate porosity. This distinction was observed in

studies of chernozem soils with 42 per cent of their particles in the 50-10-
micron range which revealed a marked deficiency of pores in the 60-5-
micron range. Within the pores, water is held by both the mutual attraction
between soil and water, termed adhesion, by cohesion, due to the mutual
attraction that exists in water molecules, and by absorption on to the electro-
negative surfaces of colloids. Capillary water can exist in pores as fine as 0-03
microns.
Permeability in turn tends to be closely related to porosity, but rates
change during the course of one shower, as fines are washed into open pores,
and rates at the surface are in part influenced by the nature of the sub-soil
which might be quite different. In cool temperate conditions there is often an
abrupt reduction in permeability on descending into the illuvial horizon.
Fissuring due to desiccation or to root pressure or burrowing by animals
accelerates permeability. In many free-draining situations water may move
down through a metre of sandy soil in about 3 minutes. If there is an
admixture of silt and clay in a sandy soil permeability rates may be up to a
dozen hours, while soil water may take longer than 4 months to move through
heavy clays. Permeability rates for intermediate textures fall between the two
extremes. In the loess of Voronezh Province, U.S.S.R., permeability is 0- 15-
l-30m/day. Similar field measurements in the forest steppe of the Trans-
Volga region indicated rates of water penetration into a ploughed silty clay of
O' 1-0-9 m/day. Here thawing of the sub-soil caused lateral flow of soil water
in spring at a rate of 7 1 m/day and other measurements suggest that 3 m/hr is
a reasonable indication of rates of lateral flow in soil water when vertical

filtration is arrested in the sub-soil.
Although there are many reasons why particle size alone is an insufficient
indication of the mode of water movement, it js a fundamental factor in all
aspects of soil water movement and the importance of particle size analysis in
geomorphology is seen to have redoubled value.
7. Mechanical properties of clays
Clays can absorb huge quantities of water partly because the total surface
area of tiny clay-sized particles is vast and water content is the main con-
sideration in many dynamic physical properties of soils. This is partly because
clays increase in volume as they absorb water. Montmorillonite can take up to
ten to fifteen times its initial dry weight and increase in volume by more than
a third. This swelling introduces abruptly increased pressures and
swelling
clay minerals appear to be an inherent factor in landslips.
Engineers studying
soil mechanics make frequent reference to the Atterberg limits, which de-
scribe water contents at which soils tend to change their state.
A soil is said to
138 Inti eduction to Geomoi phology
be in a plastic state when the water content is sufficient to permit changes in
shape without the production of surface cracks, the latter characterizing
volume changes in the semi solid state At moisture contents less than the
shrinkage limit no further reductions in volume accompany diminishing
water content This limit is usually between 10 and 12 per cent for a wide
range of soils When the difference between the plastic limit and the liquid
limit, the Plasticity index, is small, soils are highly susceptible to erosion by
rainwash and running water For clayey soils in a semi solid or solid state the
shear strength of the materia! is the most important mechanical property
influencing its susceptibilit> to erosion In unconsolidated material the shear
strength is directly dependent on the soils’ compaction achieved mainly by
wetting and drying processes acting on the clay portion to produce cohesion
and on a range of particle sizes to increase internal friction Some examples of
values of compression shear strength include 7800 kg/m’ for postglacial
clays susceptible to earthflow near Ottawa. 1700 kg/m’ for laminated silty
clays at Rutherglen, Glasgow, with extremes for such materials ranging be
tween 1000-4000 kg/m- Values for weaker consolidated rocks are much
higher At Mam Tor m Derbyshire, the site of one of the larger landslips in
Britain, the compression shear strength of shales, with ilhte and kaohnite the
mam clay minerals, was 48-67 kg/cm- and in the Ashop valley to the north
svhere postglacial landslipping is again spectacular, values are 67-89 kg/cm^
Another soils mechanics concept, defined by K Terzaghi (1965), is that of
sensitivity which is the ratio of the shear strength of a soil in an undisturbed
condition to that of the same material remoulded at the same water content
Sensitive clays have values between 4 and 8 and in the extreme case of quick
clays, values are over 16 The regain of strength during remoulding is termed
thixotrop) and is favoured by the very small particle size and high water
absorbing capacity of montmorillonite clays
So far attempts to apply methods of soil mechanics in landform study have
not been noticeably successful However, it would be a mistake to assume that
the physical analyses by civil engineers for technical purposes and measured
by empirical tests, are involved in the production of particular landforms
Most of the measurements are synthetic, refer to a situation where a soil is
burdened with an artificial load, and are not realized in nature Even liquid
limit values determined by laboratory tests may define volumes of water
greater than the porosity available in the field Therefore, while landform
study may make increasing reference to the mechanical properties of soils and
weak rocks, these will be made in full awareness that in the natural situation
the crucial factors for variations in the properties of a given material are
water content and gradient and the other factors like permeability which
influence it. as water content and gravity are the fundamental influences on
most of the dynamic mechanical properties in soils in natural settings
8. Frozen soils
One group of dynamic mechanisms in soils which have been part of geomor-
phological investigations for at least as long as they have required the engin-
eer’s attention is that concerning frozen soils. Soil materials have a specific
heat ranging between 0-3 and 0-8, thus gaining or losing heat on average
twice as quickly as the same volume of water. For this reason temperature
effects in a soil depend on the amount of soil water rather than on differences
in the soil material itself. Depths below 0-75 m in soil profiles on Reading
Beds are often structureless, impermeable, and unaffected by former frost-
heaving activity.The absence of water-filled fissures, unlike the chalk areas, is
possible the main factor limiting this depth. In most soils, for temperature
above freezing, a wave of temperature change advances 2-3 cm/hr, whereas
the rate of frost penetration is approximately 5 cm/day. This rate may be
nearer to 7 cm/day in more porous, drier sandy material or slowed down to
3 cm/day in damper constricted pores of a humus loam.
Segregations of ice form where there are large supplies of water. In samples
taken from frozen ground in the Mackenzie delta area, the average ice content
may easily reach 500-1000 per cent in relation to the dry soil material.
Water may be held in a frozen layer that would normally have drained.
Freezing of the soil may have a desiccating effect on adjacent unfrozen layers
due to the emptying of the capillary-sized pore spaces (fig. III. 9). Osmotic
phenomena are important in the freezing process, ice attracting water because
of its higher electrical potential. In Victoria Land, Antarctica, rates of growth
of wedges range from 0-3 mm to more than 5 mm per year and average
2 mm. Segregations grow particularly rapidly in silts. This is probably
because water moves appreciable distances in a short time through the pores
of silt-sized materials, which, particularly in the 20-50-micron range, provide
the most efficient passageways for capillary movement. Capillarity, the
phenomena of forced ascension of water in fine tubes, operates equally along
lateral moisture gradients in soils. Sands are too coarse for capillary-sized
channels to be extensively developed and in clays the pores are too small for
rapid movement, whereas water might move 2-2-5 m/month by capillarity in
silts. Sands and coarse materials do not freeze as a block because ice occupies
only part of the pore space and without capillary circulation, no ice segrega-
tions develop. Where they do develop the sandwiching of the seasonally
thawed layer, or mollisol, between closing layers or lenses of frozen ground,
leads to the development of pressure structures such as involutions and
injections. Their development requires a minimum seasonal frost penetra-
tion of perhaps a metre, which appears to be related to nearness of mean
annual temperature to 0°C rather than to the intensity of winter cold. A
2,
by 1,
work
numbers
on
Var’egan
based
the
and
profile,
ground,
this
respectively.
In
frozen
17).
of
No.
permafrost
layer
Bull.
lower
relict
Geog.
and
extensive
1962,
an
Bone,
b discontinuous,
M.
permafrost,
R.
{from
of continuous,
Siberia
body
indicate
upper
Western
an
3
and
in
indicates
permafrost
of
Zemtsov,
Profile
A.
C. A.
distinctive feature is the persistence of frozen soil as a relic from a past

colder phase.
E. Geochemical Considerations
When rocks, being formed at higher pressures and temperatures than exist at
the surface, are initially exposed to the sub-aerial environment, reactions
occur to produce new compounds of greater volume and lower density. This
mobilization and redistribution of elements during weathering is a complex

and variable process open to few generalizations. Although the geologists and
soil scientists who study geochemical processes involved in chemical weather-
ing report on their complexity, it is an advantage to the geomorphologist
following their studies that the number of rock types, mineral groups, and
chemical elements involved is relatively few. There are only half a dozen main
rock types, and eight main elements, oxygen (47 per cent by weight), silicon
(28 per cent), aluminium, iron, magnesium, calcium, sodium, and potassium.
Moreover, no other liquid can compare with water, either in abundance or in
the number of substances it can dissolve, or in the amounts that it can hold in
solution.
1. Physicochemical factors
One important factor influencing the pattern of selective loss of minerals is
ionic potential which is in part dependent on the dipole nature of water

molecules, with cations attracting the negative ends of water molecules and
anions the positive end.The bigger the ion and the less intense the charge on
its number of water molecules that gather around it.
surface, the greater the
The factor Z/r, where Z is the charge and r the ion radius, is known as the
ionic potential. Where the concentration of positive charge on the surface of a
cation is sufficiently high to repel the protons in water molecules, precipita-

tion of an insoluble hydroxide may result. In contrast, elements like sodium,
calcium, and magnesium, with low ionic potentials of 1-0, 2-0, and 3-0
respectively, remain in solution during processes of weathering and trans-
portation. Elements with intermediate Z/r values are precipitated by hydro-
lysis, although elements with still higher ionic potentials form anions con-
taining oxygen which are usually again soluble in this form. Ionic potentials of
other elements commonly involved in rock weathering include aluminium,
5-9, and silica, 9-5. In general, although monovalent ions are leached out more
quickly than divalent ones, and the latter
more quickly than trivalent ones, the
behaviour of some elements in weathering and transportation processes
is very complex and often
at variance with work on theoretical solubilities in
distilled water. A comparatively well-understood example is iron, which in the
ferrous state is stable in solution (Z/r for Fe"^* = 2- 7) so that oxidation to the
ferric state (Z/r for = 4-7) must precede precipitation of iron.

Some of the most significant physiochemical processes in geomorphology
are due to materials, clays in particular, and also humified organic matter,
which exist in the fine subdivision of the colloidal state. Colloidal systems
consist of distributed particles approaching molecular dimensions and a sol-
vent. the terms being the ‘disperse phase’ for the former and the ‘continuous
phase' for the solvent or dispersion medium. A colloidal solid-liquid system
incorporating a small amount of colloidal matter in a much larger quantity of
water is termed a sol. Like liquids a sol is not rigid and flows readily. Where a
relatively small quantity of water is enmeshed in the interstices and crevices
between and within colloidal particles, the viscous mass shows some rigidity
and is called a gel. Pastes are systems in which the concentration of discrete
particles is such that they form the bulk of a plastic mass. Coagulation of
colloids is governed by the presence of free ions. If the number of free ions is
high they coagulate the colloidal particles into ‘clots', crumbs of greater
volume which reduces the contact forces responsible for cohesion of the
colloidal substance which thus loses its eonsistency when flocculated. The
distinctive effects of sodium are seen in the precipitation of river-borne clays
on entering the ocean and in the instability of recently uplifted marine clays.
Colloidal systems in the weathering profile are normally in the gel condition
because of the limited amount of water usually present, although clay min-
erals are usually suspended in a colloidal solution when they first leave a
weathered mineral. On the other hand changes of state which accompany rela-
tively short-lived additions of abundant water are some of the most significant
events in the evolution of many land surfaces. There are also two highly
significant properties of a solid-liquid colloidal system in addition to changes
of state. First, a characteristic of the surface of contact of a colloidal particle
is its interface actions, due to its electro negative charge. As a result colloids
can fix and retain by adsorption gaseous, solid, or liquid particles of very
small size. Secondly, as the average figure for the size of colloidal particles
is less than 1 micron, their numbers are in hundreds of billions per gram, with
a surface area of thousands of hectares in a cubic metre of clayey material on
which interface actions, like the adsorption of water, may take place.
2. Weathering reactions
Base-exchange involves the mutual transfer of cations like Ca”, Mg". NaL
and K” between a thin film of water rich in one cation and a mineral rich in
another. Exchange reactions are reversible, and different ions may replace
one another. The rate of exchange depends on the acidity, organic matter,
temperature, and other properties of the solution as well as on the abundance
.
of the various cations and their chemical activity. Generally the rate of
exchange is rapid, requiring only a few minutes. Increases in cation-exchange
capacity (CEC) is one of the more dramatic results of clay formation, and
there is a strong direct correlation between changes of concentration of
certain elements and variations in the proportion of the clay fraction.
Many clay minerals are hydrated, involving the addition of the entire water
molecule to the mineral structure. The disruptive effect of minerals expanding
due to water incorporated by hydration is again essentially mechanical and

one factor causing granite decomposition. In a 1 0-year experiment of actual
weathering, a 9 per cent decrease in the amount of material in granite frag-
ments larger than 200 microns was attributed to hydration. An example of
hydration is the conversion of hematite to limonite:
2Fe203 + 3HjO - 2Fe20j
but such reactions are rather easily reversed by heating and there then has
been no fundamental chemical change.
In hydrolysis, by contrast, the hydrogen ion, becomes part of the
atomic structure of a clay mineral. This is the most important weathering
reaction for silicate minerals, even in deserts. The physical effect is an
exchange of H* ion from the water for a cation of the mineral, leading to
expansion and decomposition of the silicate structure. A chemical effect is to
increase the pH The reorganization of the silicate structure
of the water.
makes it possible to accommodate even more absorbed water in the crystal
lattice. The way in which hydrogen ions and water decompose potassium
felspar illustrates the effectiveness of the H* ions:
2 KAIS 13 O 8 4- 2H* 4 HjO 2K' + AljSijOslOH)^ + 4Si03

OrthocJasc + Hydrogen +Water •* Potassium + Kaolmite Silica
felspar ions ions
Here the H^ ions force their way into the potassium felspar structure and
displace potassium ions which then leave the crystal lattice. Water thus not
just dissolves and alters minerals but also acts as a source of H* ions,
although carbonic acid is a much better supplier of H"^ ions for hydrolysis
than pure water would be. Carbonic acid ionises to form hydrogen ions and
bicarbonate ions:
HjO CO, H 3 CO 3
H" + (HCO 3 )
Water + Carbon -» Carbonic— Hydrogen-*- Bicarbonate
dioxide acid ion ion
Commonly in the weathering of felspar, hydrolysis and carbonation operate

together:
2K;AlSi308 + 2H,0 + CO,-» AI,Si,0,(0H)4 + K 3 CO 3 + 4SiO,

Orthoclase Water + Carbon Kaolin + Potassium -t- Silica”
felspar dioxide carbonate
Here hydrolysis produces the clay residue, anhydrous aluminum silicate
while potassium carbonate, the result of carbonation, is carried away in
solution.
Carbonation is the reaction between carbonic acid and minerals, water
acquiring its acidity largely from the carbon dioxide generated by humifica-
tion processes in the soil. Fresh rainwater is also slightly acid because it
dissolves a trace of carbon dioxide from the atmosphere. The effects of

carbonation are best known in the solution of calcareous rocks
CaCOj -H HjCOj Ca" + 2(HCO,)'

Calcium Carbonic Calcium and bicarbonate
carbonate acid ions
The solubility of iron is depressed in carbonated water but that of silica
enhanced. These responses favour a theory of iron oxide accumulation

by water rich in carbon dioxide dissolving silica and concentrating iron
oxide.
Oxidation is common in freely draining parts of the weathering zone, as it
is the oxygen dissolved in infiltrating rainwater which oxidizes metallic iron

and changes ferrous iron to the more oxidized ferric state. Although oxidation
commonly involves combinations with oxygen as part of the weathering
process, this is not always the case. By definition a substance is oxidized when
it loses electrons and reduced when it takes on electrons. The combination of

iron or other elements with oxygen weakens the original mineral structure,
freeing the remaining minerals for participation in other chemical reactions,
although hydrolysis usually precedes oxidation. Iron is a very common
chemical element and its distinctive rusty colour on weathering is widely
observed. For instance, in weathering at Marble Point, Antarctica, the only
measurable chemical change is the progressive loss by oxidation of ferrous
oxide in iron-bearing minerals. During chemical weathering ferrous silicates,
such as the rock-forming minerals pyroxene, amphibole, olivine, and biotite

form hematite, and hydrous iron oxides like limonite and goethite, by oxida-
tion.
3. Physicochemical characteristics of weathering environments
Some important physicochemical factors relate to environment rather than to

the elements themselves. The degree of acidity, for instance, is of great
significance in chemical reactions in the natural environment, reflected in a

weathering profile by the hydrogen ion concentration of its aqueous suspen-
sions. The concentration of dissociated hydrogen ions in this aqueous suspen-
sion is in equilibrium with those actually adsorbed on the weathering pro-
ducts. Thus the hydrogen ion concentration of an aqueous suspension from a
weathering profile is an index of the intensity of acidity in the weathering
environment. Its expression is in terms of pH units which is the negative
logarithm of the hydrogen ion concentration. Thus if a suspension has a pH of
6, this value indicates that its hydrogen ion concentration is 10'*’ or loofooo of
a gram per litre. In a neutral solution like pure water pH is 7; for an acid
solution the number is less than 7, reflecting the greater hydrogen ion concen-
tration. The pH of an aqueous suspension is particularly significant in con-
trolling the precipitation of hydroxides from solution, the solubility of iron

being 10^ times greater at pH 6 than at pH 8-5. The solubility of ferric iron is
so low, 0-01 ppm in pH range 5-8, that significant solution and transport of
iron probably involves reduction to the ferrous state. At pH of less than 4
alumina is readily soluble, which may lead to the preferential removal of
alumina in extremely acid environments, like those of podzolization, whereas
from pH 5 to pH 9 the solubility of silica increases considerably but alumina
is practically insoluble. However, due to the very high binding cation-
exchange energy that clay has for alumina, the mobility of alumina due to
solution in an acid environment may be cancelled out by immobility due to
cation exchange. An increase in pH aids precipitation of carbonate.
The oxidation-reduction potential varies with varying concentrations of
the reacting substances. This is in part dependent on pH as most reactions
involve hydrogen or hydroxyl ions, with oxidation generally proceeding more
readily the more alkaline the solution. In reducing environments, compounds
including hydrogen ions are common and organic matter tends to accumulate.
For instance, air trapped in waterlogged interstices of an anaerobic environ-
ment may develop substantial concentrations of methane, CH^. compared
with the combination of carbon with oxygen to form carbon dioxide under
aerobic conditions.
4. Mineral solubilities
From the discussion of the three preceding geochemical aspects of weather-

ing, it will be clear that it is not easy to generalize about the solubility of
various minerals. Even though it is useful to distinguish between the several
different reactions involved, in practice, and as an unavoidable consequence
of the composition and properties of water, they operate simultaneously. On
the other hand a particular geochemical environment may dictate that a
distinctive pattern of weathering reactions be followed. Thus, although perfect
mineral grains fall into a sequence according to their susceptibility to decom-
position. regardless of the rock type in which they occur, an agreed absolute
scale is still wanting, and there is probably a need for different scales for
different environments. In general monovalent ions are leached out more
quickly than divalent ones, and the latter more quickly than trivalent ones. The
lattice site of an element is in many cases an important factor in determining
its relative mobility during weathering. Calcium generally exhibits much

greater mobility, especially during the early stages of weathering, than any of
the other elements, like magnesium, being taken up into solution at rates tens
to hundreds of times faster than for other elements. It has been observed in
the weathering of volcanic ash on St Vincent that it was the pyroxenes,
olivines, and plagioclase of more calcic composition that altered more
quickly. As the first phase of felspar decomposition is rapid disintegration by
hydration, no loss of material is involved initially. In fact in 1947 P. Birot
(1947) found that a 5-7-month treatment of samples of acid crystalline rocks

under alternating 12-hour spells of wet then dry conditions at 70° C resulted
in a 0- 1 gm 20-gm sample due to hydration. Ordinary
increase in weight of a
room temperatures Deep chemical weathering is a
did not produce this effect.
phase subsequent to hydration. The first sign of plagioclase weathering is
surface pitting and widening of fractures by solution. Alteration to clay
minerals results in an expansion of the plagioclase lattices, particularly with
alternate wetting and drying and rock-shattering results. Despite the presence
of highly mobile cations, the alkali felspars are relatively stable because the
lattice restricts access of water. The bases, followed by silica, are leached out
of a surface zone of weathering, leaving sesquioxides of iron, aluminium, and
manganese as a residual accumulation. A residuum of iron sequioxides de-
pends on high temperatures and does not take place under continuously dry
conditions. Mica weathering is essentially a process of potassium depletion.
There is hydrolytic interplay between water and the interface of the mica
lattice. In an acid environment this cation is displaced and the bonding effect
between mica sheets loosens with the dissolution of magnesium and the
oxidation of iron, producing further loosening and expansion of the crystal
lattice. The weathering to clays of biotite and, to a much lesser extent,
plagioclase, is a vital factor in thebreakdown of granite. Like plagioclase,
pyroxene and olivine minerals may show etching, indicating that dissolution
is part of the weathering process. The solubility of silica depends greatly on
whether it exists in the form of quartz or as amorphous silica, the latter
including hydrated and dehydrated silica gels and the skeletal remains of
silica-secreting organisms. At 5°C the theoretical solubility of quartz is about
6 ppm compared with 60 ppm for amorphous silica. The undersaturation of
terrestrial waters may be attributed, in addition to the slowness with which
silica or silicates dissolve, to the activity of organisms that use silica.
Specific examples of observed differences in mineral resistance to weather-

ing include studies by W. W. Smith (1962) who found a mineral sequence in
the weathering of basic igneous rocks in Scotland, in order of increasing

stability, of olivine, labradorite, augite, magnetite, ilmenite, and hematite.
Studies in Maryland provide an example of a scale of resistance to weathering
of heavy minerals. The scale, related to a garnet value of 1, includes zircon at
kyanite 7, horn-
100, tourmaline about 80, sillimanite and monazite at 40,
blende 5, staurolite 3, and hypersthene less than 1.
that of
An important geochemical concept in the study of weathering is
weathering ratios, calculated either between the proportion or amount of a

given element in a soil compared with similar quantities in the parent
material, or between concentrations of different elements in a soil or clay.

Examples of the latter include the molecular ratio between silica and the
sesquioxides:
SiO^
AljOj + FeÔj
A related expression is the SiOj : FcjOj ratio. Indicative of the general con-
centration of silica and alumina in Cuban weathering crusts, for example, are
values of this ratio at 3—5 and frequently higher. For latosols in the humid
tropics, values of the ratio SiOj : AI 2 O 3 are generally between 1 and 2.
F. Biological Activity
Ever since 1882 when Charles Darwin’s measurements showed that earth-
worms could move 10 tons/acre/year, attention has been drawn, from time to
time, to the tendency for studies of sub-aerial processes to emphasize physical
and chemical factors, perhaps at the expense of biological considerations. But
in many investigations bacteria and other forms of life, an irksome contamin-
ant in controlled laboratory experiments and often a source of variation
irreducible to the confines of an equation, remain excluded. Today awareness
of the risks in neglecting the biological factor are keen, as leading authorities
in a range of disciplines independently come to emphasize its importance. For
instance, it is organisms that keep the oceans so low in dissolved silica;
experts like W. D. Keller regard bacteria as one of the most important and
essential adjuncts to the processes which result in argillation; the lichen
Caloplaca was found at 85 degrees south; it was found impossible to weather
biotite in laboratory conditions until tree seedlings were incorporated into a
recent experiment; in Florida Bay, south of the Everglades, sediment size and
physiochemical parameters were not strongly related, and even the physio-
chemical parameters were not strongly inter-related, but that there was a
close correlation between the distribution of sediment size and turtle-grass
{Tlialassia restudimim Konig).
Biological factors provide an aid to landform studies, and, as an expression
of climatic controls and local environmental factors, are an important inter-
mediary agent in land-forming processes. Plants and, to a lesser extent,
animals are able to contribute substantially to rock weathering because
the
sun transmits to them the necessary energy, whereas inorganic reactions
must
be driven by built-in chemical energy with fewer direct contributions from

solar radiation. These processes require very careful consideration and obser-
vation because it is difficult to predict whether the intermediary agency of
plants will accelerate or retard erosion. Mechanically plant roots can bind or
break and, simultaneously might enhance solution or alternatively induce
precipitation and storage of chemical elements. In addition, it is necessary to
consider the organic factor from three separate viewpoints, that of its bulk or
the biomass at the surface, the e.vpression of different environments seen in
different ecological tj^pes of plant, and thirdly, the variations in physiological
processes in contrasted environments.
1. Acceleration of mechanical processes
It is not certain how much mechanical work the e.vpansion of growing tree
roots might perform. P. Birot ( 1 966) states that a living root 1 0 cm broad and
1 m in length is capable of moving a block weighing 40 tons. Where soils are
stony and in the upper horizons of bedrock where root space is reduced, roots
are forced to develop around and between boulders and to penetrate into the
openings in bedrock. In valley-side cliffs in Magnesian Limestone a dozen
miles south-east of Sheffield, the rock-disrupting roots of yew trees (Taxiis
baccata). after seeding some distance back from the cliff may grow to be
15 cm in diameter at depths of 6 m (fig. III. 10). Penetration depths of more
than 6 m have been observed in sandstones elsewhere, and in fractured granite
in theColorado Rocky Mountains, Ponderosa pine roots penetrate to depths
of 10-12 m. Extreme depths of 30 m have been suggested, but in spruce and
pine woods near Moscow. 75 per cent or more of the tree roots lie within
40 cm of the surface. In the taiga spruce forest roots are confined to the upper
30 cm. and in deciduous forest still only to 50 cm. A significant fact may be
that in deserts where vegetation is apparently scant there is an exceptional
preponderance of the biomass in the root fraction, on average SO per cent,
which may not spread in the upper horizon which dries out rapidly. In the
shrub and arctic tundra where conditions are also severe at the ground
surface, up to 90 per cent of the biomass may be below ground. In all zones of
deciduous trees and coniferous forests roots are 15-24 per cent of the
biomass.
Apart from the prying action of growing roots and the movement of
superficial material when trees are under the stress of high winds, the main
contribution of trees to mechanical weathering is probably due to the tremen-
dous leverage exerted by the trunks of falling trees. H. J. Lutz (1960) studied
wind-thrown trees, 30-60 cm in diameter in New England forests and found
several examples where rock masses more than a cubic metre and weighing
over 0-25 tons were mox'ed vertically or horizontally 0-5-3 -5 m often with
Figure III. 10 Inter-relationship between tree roots and cliff breakdown (from G.
Jackson and J. Sheldon, 1949, Jour. Ecol., Vol. 37), as suggested by a study of
yew trees (Taxus baccaia) on Magnesian Limestone cliffs at Markland Grips,
south-east of Sheffield.
fresh fracture surfaces in the hollows. Mounds produced by wind-throw are

common in the deciduous forests throughout the eastern United States. In
Guyane, A. Caillieux counted ten wind-thrown trees per hectare. The
Peruvian Andes is another locality where wind-throw is reportedly common,
facilitated by shallow rooting, 0-2-0-3 m in depth, due to the supply of
nutrients being limited to litter fall on the ground surface. Similarly, but
usually because permafrost creates a hard impervious substratum, trees in
arctic conditions are easily overturned. The possibility of some latitudinal
effects due to root action has been suggested (fig. III. 1 1 ).
Figure I II. 1The role of trees in the sub-surface development of tors {from I.
Gams, 1966. Geografski Vestnik. Vol. 38). A, initial stage. B. disintegration of
more closely bedded and jointed blocks by tree root activity and lowering of the
soil surface, C. Further disintegration and lowering of the soil surface: (i)cool
climate, (ii) warm climate.
Three important facts about the distribution of wind-throws are the ten-
dency for them to occur on exposed sites like ridges and escarpments; their
importance particularly in the tropics, due to tree colonization of bare rocks
where nutrients are most readily available; and that the ultimate fate of most
trees if not broken by wind would seem to be to topple over. It might be most
illuminating to establish the degree to which the contrast between the slopes
which typify tropical coasts and the sea-cliffs and jagged stacks of extra-
tropical latitudes reflects the descent of dense forest to sea-level in the tropics
compared with the treeless coastal strip on spray-bathed extra-tropical sea-
cliffs. In fact the degree to which the landform in any forested environment
differs from that of a treeless domain might be worth considering in terms of
root action and wind-throw.
The amount of energy expended and the mechanical reworking achieved by
soil fauna in the soil is substantial. Their action in moving material from the
upper layer of soils on to the surface is probably the most important method
of truncating soil horizons, especially in the tropics. In eight experimental
plots atRothamsted total consumption of soil by earthworms ranged from 10
to 90 m. tons/ha/year, involving up to 8-7 per cent of the total weight of the
top 10 cm of soil. Thus all the material in this depth would pass through the
1
alimentary tracts of the earthworm population in 1 1|-80 years. Although the

temperature for optimum earthworm activity is as low as 10° C, in tropical
latitudes the termite takes over a comparable role, consuming similar volumes
of earth if its preferred diet of vegetable debris is deficient. In the savanna,
termite mounds may be 2-4 m high with a volume of 2-4 m-* (fig. III.12.B).
Sizes in the cerrados of Brazil are similar, with a spacing measured north-east
of Andarai, averaging 15 m. Near Elizabethville the mounds cover 7-8 per
cent of the ground surface. One estimate of the net work by termites is a
complete reworking of the top metre of the soil every 1000 years.
Ant mounds may be 1-15 cm in height and from 7 cm to 4 0 m in dia-
meter. The number of mounds per hectare ranges from 62 000 for small sizes
in humid temperate forests to 25-50 for large ones in steppe areas. Com-
monly they occupy 1-4 per cent of the total area and the ants may move the
top 6 cm to the surface once every 100—300 years. In Silwood Park in
Berkshire, mounds are up to 40 cm in diameter and 15 cm high and may
occupy up to 10-11 per cent of the ground with a density of 5300/ha in
parts. In the Wisconsin prairies ant mounds are 30 cm high, 60-90 cm in
diameter and number 1 00-125/ha. Therefore* each point on the surface could
have been occupied by a mound at least once in 600 years. The combined
activities of earthworms, rodents, and ants could turn over the upper 60 cm of
grassland once a century.
Mole hills studied near Moscow were about 0 cm high and about 30 cm
1
in diameter, ejected at a rate of ll.l mVha/year which represents 0-3 per

cent of the weight of the soil in the top 10-40 cm. In the recycling of
unconsolidated material even the contribution of elephants, where appro-
priate, might be taken seriously. In the Wankie National Park, the volume of
solids removed by an elephant at each wallow amounts to a cubic metre.
The large scale of reworking of coastal sediments by benthic animals has
only been appreciated for a couple of decades. However, it has now been
established that Arenicola will, for each square metre of sediment surface,
rework 150 000 cm-* of sediment in a year, or rework the entire thickness to
depths of 20-30 cm in 20 months. Another worm Clymenella can rework to
a depth of 20 cm in 2-4 years. In subaquatic environments like the Hudson
estuary, man-made debris occurs to a depth of at least 2 m into the bottom
sediment, believed to represent the scale of overturning of debris by aquatic
worms.
A significant implication of these immense expenditures of energy is the
amount of incoming solar radiation diverted into the reworking of uncon-
solidated material in the soil layer rather than directed to the
breakdown of
rocks. However, on slopes the heaping up of loose material would inevitably
lead to some downslope displacement by gravity alone. On the other
hand a
pronounced effect of many of the reworking activities is to increase porosity
2
Figure I II. 1 Reworking of superficial material by organisms.

A. Ant mounds in the Wisconsin prairie {after F. P. Baxter and F. D Hole, 1967,
Soil Sci. Soc. Am. Proc., Vol. 31). Figures under the mound on the left are the
numbers of ant channels observed on the floor of the pit at the depths indicated.
B. Termite mound in Nigeria {from P. H. Nye, 1955, Jour. Soil Sci., Vol. 6).
C. Effect of mole tunnels on soil moisture content {from A. B. D. Abaturov and

L. 0. Karpachevskiy. 1966, Soviet Soil Science) observed in the Krasno-
pakhorskiy Forest near Moscow.
D. Sediment sorting by Amphritrite ornata. Barnstable Harbour, Massachusetts

(from D. C. Rhoads, 1967. Jour. Geol., Vol. 75). Grooved ciliated tentacles of A.
ornata collect and transport sediment 1 mm
and less in size to the mouth. Non-
ingested manipulated remains as a conical deposit around the anterior opening of
the U-shaped mud tube and egested sediment surrounds the anal end.
E. Reworking of intertidal sediments by the burrowing of a crab (from Evans,

1965, after IV. Schaefer).
which therefore reduces surface runoff. The activity of moles can reduce the
bulk density of a surface soil from 0-8 to 0-5 g/cm^ and increase porosity by
about 10 per cent. Measurements made on an experimental farm near
Moscow revealed a dense network of passageways 5-7 cm in diameter, aver-
aging 255 cm/m^ in the top 10 cm of soil. This represented 15-3 per cent of
the area and 7-2 per cent of the volume. The effect on soil-water conditions is
evident from fig. III.12.C. The sum of the volume of earthworm burrows in a
sandy soil in eastern England was the equivalent of a 4-4-cm pipe beneath
each square metre. On the shore, burrowing activities may lead to a 1 0 per
cent increase in porosity, a figure closely comparable with the results of
moles’ activity in the soil.
2. Retardation of mechanical processes
Increases in soil porosity are just one indication of a remarkable variety of

ways in which organic influences work to retard some erosional mechanisms.
The involvement of vegetation in retarding erosional processes starts with the
interception of precipitation by foliage. Proportions vary, but a threshold
value of precipitation of at least 2 mm before any moisture penetrates to the
ground beneath Canadian hardwoods illustrates one aspect of interception.
On the Ivory Coast 50-95 per cent of the total precipitation may be inter-
cepted. Even vegetation as lowly as mosses, being vigorously hygroscopic,
absorb moisture not just intercepted from precipitation but from atmospheric
vapour too. In addition to interception vegetation indirectly influences the
infiltration rate into a soil because of water losses due to evapotranspiration.
If a plant uses up large proportions of soil moisture before and during a rainy
season, there is more space available for the intake of the subsequent rains.
Most draw up water from depths of 6 m or more. Studies in the Carmel
trees
mountains, Israel, showed that the maquis shrub can utilize moisture from
depths down to 7-5 m. The average annual moisture withdrawal was 450 mm
from ground covered by maquis. 410 mm by pines, and 330 mm from be-
neath pasture grasses. These figures compare with an annual precipitation of
700 mm
and a loss of 265 mm
from bare ground. Moisture depletion in the
upper metre was similar under the various cover types.
The many experiments comparing controlled plots on agricultural research
stations give the clearest impression of the scale on which differences in
vegetation cover, by influencing surface runoff, regulate the erosional poten-
tialof precipitation. Table III. 1 illustrates a typical example. These figures
suggest that in drier climatic areas where grassland prevails most of the flow
may be on the surface with little soil-water flow compared with small volumes
of surface flow under forests where infiltration capacity is very high. Under
some oak forests surface runoff has never been recorded even in years of
1

Table III.l Surface runoff coefficients for spring precipitation in the mixed forest
zone of European U.S.S.R. (M. I. L’Vovich, Soviet Soil Science, J966).
GRASSLAND PLOUGHED LAND FOREST
Sand 0-20 0-01

Sandy loam 0-33 0-23 0-03
Clay loam 0-53 0-39 0-19
maximum runoff. Where the surface is bare, peak flows may be up to twenty
times the volume of peak flows from comparable forested areas. Associated
increases in erosion rates are usually substantial if not catastrophic. A typical
example was the seven-fold increase of soil slippage on grass-covered areas in
the San Gabriel Mountains, California, in areas formerly covered with a
natural chaparral vegetation.
As elTective in reducing the erosive potential of precipitation as the living
foliage above the ground is the organic matter decaying at and below the
ground surface.
It is important to differentiate between thoroughly decomposed residues,
termed humus, and organic matter in less complete stages of decomposition,
the humus having colloidal properties. Humus in the soil, because of its
electro-negative properties, not only retains large proportions of absorbed

water within the soil after precipitation but is a major bonding element in the
delicate cohesiveness of soil aggregates or crumbs, the porosity of which
increases infiltration while the cohesion of the crumbs themselves makes them
resistant to erosion. Table III.2 illustrates the decline in water-stable aggre-
gates and the diminution in pore space which accompanies the removal of
vegetation.
Table 111.2 Measurements of porosity and water-stable aggregates in the 0-
7-5 cm soil horizon in three types of tropical forest clearings, Ghana {from
Cunningham, 1963).
WATER-STABLE AGGREGATES
POROSITY PER CENT AIR-DRY SOIL
3 MM
Capillary Non-capiiiary Total Total True crumb

pore space pore space pore space structure structure
Shade 37-0 14-7 51.7 55-3 39-8

Half exposure 35-0 16.4 51-4 50-2 28-9
Full exposure 32-6 10- 42-7 48-7 29-0
Perhaps more significant than humus itself is its rate of production. Humi-
fication of organic material, an oxidizing process, is more rapid at higher
temperatures, partly because it is a temperature-controlled reaction and

partly because high temperatures minimize the quantity of water in the soil
that would impede air circulation. In consequence there is little accumulation
of organic residues in the humid tropics, despite the very high biological pro-
ductivity, and thick accumulations in cool temperate climates despite low

productivity. The result is a definite tendency for the thickness of the organic-
residue blanket at and beneath the ground surface to increase towards cooler
latitudes, one of the few geomorphological parameters to show some sem-
blance of a systematic latitudinal trend.
Due to rapid rates of humification the surface of soils in the humid tropics
are significantly low in litter accumulation despite the prolific leaf fall.
Amounts range between 1000 and 5000 kg/ha but reserves as low as 400
have been measured in south-east China. On average the amounts are half or
even a third those in deciduous forests of the temperate zone and the chances
of erosion of the forest floor correspondingly increased. In savanna areas the
accumulation of organic matter is smaller, amounting to 50-1 50 kg/ha in
many areas. This is less than in steppe areas where organic material is
conspicuous in soils because rainfall is too low to remove it from the soil
horizons. For instance, beneath the relics of a climax vegetation isolated on

two steep-sided plateaux in North Dakota, where precipitation is about
375 mm/year, organic material in the soil amounts to 17 000 kg/ha. In the
Russian steppe decomposition rates similarly lag behind supply of plant
residues and a characteristic steppe matting accumulates to about 4000-
5000 kg/ha in meadow steppe and 1500 kg/ha in arid steppe. The role of an
organic mat in retarding erosion in steppe conditions is conspicuously deli-
cate in balance, for two reasons. First a very large proportion of the total
biomass cycles annually, being 50-55 per cent in meadow steppe and down
to 40 per cent in arid steppe. Secondly, there may be a sharp decline in the
ground mass in dry years, which in extreme conditions might be only one-
tenth that of very wet years. In arid areas the amount of organic matter at the
surface is many times less than in steppe areas. However, it is not yet possible
to suggest the degree to which landforms in desert areas are distinctive and
the degree to which any distinctive traits are related to the absence of
an
effective organic blanket between the bedrock and the sky.
In humid temperate areas temperatures are such that the organic residues
are effective in reducing potential erosional activity.
On the one hand oxidiza-
tion rates are rapid enough
produce ample quantities of colloidal humus
to
which increases water-holding capacity and water retentiveness within the soil
and creates water-stable aggregates. On the other hand more resistant
consti-
tuents of forest litter remain undecomposed for several years and, by filtering
out silty particles from runoff waters, assist in keeping pore spaces and
passageways open in the underlying soil, thus favouring high infiltration rates.
In cool temperate environments distinctive conditions prevail as the accu-
mulation of organic matter begins to exceed decomposition rates. This is the
result of the low temperatures being ineffective not merely in oxidizing
organic material rapidly enough to balance supply but in evaporating suffi-
cient quantities of ground-surface moisture to create an oxidizing environ-

ment. However, poleward of the damp cool-temperate areas temperatures
soon become too low for biological productivity to make any appreciable
contributions to the litter fall. In eastern Canada, peat thicknesses diminish in
thickness from about 9 m near the St Lawrence valley to about 2 m north-
wards at 55° N. Similarly the net primary productivity of tundra willow and
dwarf birch thickets is only one-seventh to one-fifteenth that of birch woods in
the central taiga. However, the organic residues remain little altered for so
long that a mass reminiscent of peat layers still accumulates. Organic mats
are often widespread covering, for instance. 10-15 per cent of Labrador-
Ungava. They retain water, three times its own weight in the case of peat, due
to absorption by decomposition products, the swelling of plant residues and
the filling of the porous space between the macrostructure. One geomorpholo-
gical consequence is that this prevents substantial proportions of precipitation
from coming into contact with the underlying mineral soil and bedrock, and
also reduces runoff on gentler slopes. On the northern edge of the taiga peat
covers all slopes less than 10 degrees, but may disappear with more rapid
water flow on steeper slopes. It is perhaps only a coincidence that slope-angles
of 1 1 degrees are highly characteristic of the lower parts of many rectilinear
slopes in Britain, but it does seem possible that the importance of the inter-
relationships between a peaty or organic-residue blanket and water draining

off slopes has been underestimated. Perhaps there is wider significance in the
fact that in the Alaskan tundra it is the protection of shores by drifting mats
of vegetation debris that appears to determine lake basin morphology.
In cold climates where saturated soils and runoff are the rule due to
reduced evapotranspiration losses and to frozen sub-soils, perhaps the major
effect of diminished vegetation cover, which becomes discontinuous when
annual means fall below 6° C, concerns temperatures. An organic cover
prevents thawing temperatures, which would produce solifluction movements
on slopes, from reaching underlying frozen soils, and it has been observed
that the removal of the insulation of a moss cover does greatly hasten the
thawing of frozen ground. A water-saturated organic cover layer conducts
freezing temperatures downward when frozen, but all the energy of tempera-
tures above freezing are absorbed in evaporating a portion of the melting
water, warmer temperatures do not penetrate to the underlying soil and the
cold reserve in the ground thus persists. However, in some permafrost areas
where the organic mat is thick enough, frost-heaving can be confined to this
superficial peaty layer. In slightly milder conditions of cool temperate en-
vironments the insulating role of a humus or peaty layer is reversed, as sub-zero

temperatures, in freezing only the topmost layer of water saturated organic
material, do not penetrate to depth. In South Wales at an altitude of 600 m
peat was observed to be frozen to no more than 5 cm depth. Living plants and
trees also affect temperature conditions in the soil. By decreasing air current
v'elocities they impede heat radiation from the soil to the cold air. Their cover
checks ground cooling by nocturnal radiation particularlywhen it retains a
snow cover. J. Tricart (1967) quotes measurements in Germany where frost
reached a depth of 47 cm in a field, 45 cm beneath Norwegian Pine, 38 cm
under beech, and 34 cm under Scots Pine. In Paris basin depths of frost
penetration recorded were 27 cm in well-drained sand, 18 cm in poorly
drained sand, 13-1 cm beneath grass mat, and cm under forest.
1 1
In addition to the variety of blanket effects of roots, litter, and humus

layers, vegetation has a highly significant braking effect on the velocity of
water and air movement by dissipating their energy. This may induce deposi-
tion of material in transport from areas where higher velocities prevail, and
may have a filtering influence on the lateral distribution of sediment sizes.
One of the reasons why this effect requires careful consideration is because
equations for calculating stream-flow velocity, being inapplicable to condi-
tions where vegetation grows in natural channels, have necessarily led to the
systematic exclusion of information from vegetated channels in measurements
made by water engineers.
In the deeper part of the outlet section in Lake Laitaure, sparse vegetation
grows to 0-3 m above the bottom at the time of high-water discharges in
summer, and as fig. III. 13 shows, the braking effect is such that the flow
velocity may be proportional to the log of the water height above the vegeta-
tion but not to the log of the water height above the channel bottom. In south-
west United States, the gradients on some valley terraces are steeper than
modern arroyo gradients, particularly in head-stream areas. This contrast
exists despite the existence of large cobbles and boulders on the arroyo floors,
which is the reverse of expectations based on hydrophysical principles which
indicate that a steeper gradient is an adjustment to a coarser load.
M. A. Melton considers that this apparent contradiction is due to the dense
grass 0-6-0-9 m high, which occupied the former river valley floors, unlike
the canal-like cleanness and straightness of the modern arroyos.
Vegetation in channels is limited by basin altitude, gradient, stream velo-
city, bottom material, and stream depth. Although the broader geomorpholo-
gical implications of its role have not been investigated, examples of a few
criticalmeasurements are suggestive. In European U.S.S.R. stream channels
are only overgrown at elevations up to 300-400 m above sea-level.
Critical
longitudinal gradients are 0- 1-0-2 per cent, but occasionally are as
steep as 1
per cent. Velocities which prevent vegetative growth are 0-56
m/sec in canals
in India and 0-6 1-0- 76 m/sec in canals
in Spain. However, in sandy chan-
nels. the abrasive action added to the velocity of flow makes critical velocities
158 Introduction to Geoinorphology
of the latter 0-45-0-65 m/sec in the plains rivers of European U.S.S.R.,

whereas in bouldery or gravelly channels vegetation grows if velocities
remain less than 0- 65-0- 95 m/sec. Yet again, in this context of the effect of
sand in streams, one catches a glimpse of a more subtle aspect of the self-
Flow velocity m cm/sec
Figure III. 13 The damming effect of the summer growth of vegetation on water-
flow through the outlet of Lake Laitaure, northern Sweden {from Axelsson, 1967).
The observations, made on 26/8/1957, show that the velocity was proportional to
the logarithm of the approximate height above the O-S-m-high vegetation and not
to that of the height above the channel bottom.
accentuating nature of accelerated erosion unleashed by man’s disturbance of

the environment. Vegetation growth in downstream reaches depends on low
water not exceeding 1-0-1 -5 m and is in consequence usually absent when
catchments are greater than 8000—10 000 km^. In general it seems that chan-
nel vegetation will tend to occupy middle reaches of many streams, limited in
headstream areas by the low temperatures at the higher altitudes and by
steeper gradients, and by water depth in the lower reaches of larger streams.
For ephemeral streams, limiting conditions will be those which affect the
amount of growth during the intervals between floods. The occasional flood in
the Gebel Akdhar hills near the North African coast may encounter trees
several decades old in its path.
In coastal areas plants resistant to high concentrations of salt may have a
braking effect on water movement. If Fucus become established on wave-cut
platforms, their presence reduces the force of the waves. In estuaries, plants
like Salicornia and Spartina reduce water movement to the extent that
accre-
tion may take place at a rate of lOcm/year. Sand trapped by eel grass
{Zostera sp) forms irregular mounds, as on the Wash flats, and carpets of
algae like Enteromorpha sp can trap sand also.
On hillslopes completely covered with tough grass the braking effect may
be sufficient to keep sheetflow below erosive velocities, and even-contoured
slopes remain undissected. The streams of such an area in the Ngong Hills,
Kenya, are sediment free, and on forested slopes a solid bed of dead leaves is a
substantial obstacle to rillwash.
In sub-aerial situations, as is well-known, vegetation decreases substan-
tially the velocities of winds close to the ground surface. The planting of an
open stand of dune grass may increase the zone of virtually calm air close to
the sand surface more than thirty-fold. On a larger scale it seems that many
seif dunes originate where vegetation forms sand-traps rising above the gen-
eral surface of a sand sea, as illustrated more clearly on a smaller scale by the
nebka (fig. III. 14).
Figure HI. 14 Deflection of air currents by a nebka (from Coqite, 1962).
The braking effect of vegetation on rates of water and air movement is
clearly substantial and produces distinctive geomorphological effects,

although its importance, particularly in slowing down water flow has perhaps
been underestimated. One reason why the braking effect of vegetation
requires careful consideration is because below a certain velocity it is a self-
enhancing process, since overgrowing of a channel bottom, by increasing
channel roughness leads to slower velocities and hence creates more favour-
able conditions for further growth.
Like the biomass above the ground surface slowing down movement of
water and air, roots may hinder soil movement. The roots which may sud-
denly wrench up a ton of material on wind-throw may throughout the tree’s
life act as a binding and anchoring agent environments ranging from deep
in
tropical soils to the mollisols of periglacial areas. Their significance is par-
ticularly marked in tundra conditions where up to 90 per cent of the biomass
may be below the ground surface. Vegetation may make scree angles some-
what steeper. In northern Sweden a scree 65 m high, 37 degrees in the upper
part and 38 degrees lower down has scattered Oxyria digyna rooted in finer
stable material 20-30 cm beneath the scree surface. Roots resist lateral ero-
sion as well as downslope movement. On the shores of Lake Vattern in
southern Sweden the Rosenlund cliffs retreated at 0-21 m/year in the east and
0-10 m/year in the west between 1795 and 1908, whereas there was no
retreat in the forested section. On tropical coasts the mangrove, adapted to
very saline conditions and to oxygen deficiency in its root zones, is a pioneer
community, particularly in many Australian and African estuaries and adja-
cent open coasts. With roots amounting to half the biomass, the mangrove
plays an important part in fixing banks of unconsolidated sediments. The
effect of colonizing pioneer communities in stabilizing dunes is particularly
well marked on tropical coasts where vegetative growth may be rapid and
may extend close to the shoreline. It is unknown whether the mechanical
resistance afforded by roots, which account for up 60 per cent of the
to
biomass in savanna grasslands and 80 per cent in desert areas is geomorpho-
logically significant.
3. Biochemical weathering
The involvement of organisms is a fundamental characteristic of many

aspects of chemical weathering. In as much as the scale of chemical denuda-
tion is essentially a function of the volume of river water discharged into the
sea, plants are involved indirectly to the degree in which they restrict the
coefficient of flow by returning moisture to the atmosphere by evapotran-

spiration. In this sense they are involved in a fundamental mechanism which
limits the amount of dissolved materials as well as suspended sediment that is
removed from the landmass and is of increasing significance towards lower

latitudes. It is difficult to generalize about the more direct influences of
organisms on chemical processes. Although every 1 0° C increase in tempera-
tures introduces an approximate two-fold increase in rates of biological and
biochemical reactions, it is only in certain situations that these reactions
contribute to uptake of solutes. In many situations, particularly in lower
latitudes, there is an equal acceleration of biochemical processes leading to

the precipitation of dissolved material.
4. Acceleration of chemical weathering
The role of organic acids in weathering is the subject of occasional specula-

tion but scarcely is investigated systematically. It seems that some more com-
plex organic compounds are powerful chelators which are capable of forming
soluble co-ordinate covalent compounds with various metallic cations. One of
the most significant characteristics of lichens is that they generate a range of
organic compounds not found elsewhere which are a powerful agent

in nature
in extracting essential from rock and can produce small

trace elements
weathering pits on the upper surfaces of rocks like granite. Otherwise most
humic acids produced by anaerobic fermentation are too weak to form stable
!!" to
complexes and despite their acid reaction supply little solutions.
Organic acids are more effective in transportation processes, particularly that
of transporting iron as complexed ions. The complexes are sufficiently stable
to be transported by percolating water but, if brought into contact with ferric

oxide, they are sorbed and the ferrous iron is then readily re-oxidized. None-
theless, more than half the total quantity of iron, and also manganese and
copper, dissolved in the water of the Dneiper travels in metallic-organic

compounds. The amount of dissolved iron in rivers flowing through swamp
lowlands may be as much as 1
5 ppm instead of the more usual values about
0-5 ppm.
The aspect of decomposition of organic matter of greatest geomorpho-
logical significance is its eventual fate under aerobic conditions of complete
oxidation to carbon dioxide, water, and other simple end products. As a result
the carbon dioxide concentration. 0- 03 per cent in the atmosphere is often in

the range 0-2-3 -0 per cent in soil air. If biological activity is depressed by
averages of either temperature below 10° C or moisture contents less than 10
per cent, carbon dioxide concentrations are unlikely to be above 0-2 per cent.
In aerobic conditions, but with ample soil moisture available, carbon dioxide
concentration describes a smooth summer-maximum curve, following tem-
perature closely (fig. III. 15.A). However, mid-summer desiccation of a soil
below 1
0 per cent moisture content may depress the carbon dioxide peak, but
under favourable conditions bacteria and other micro-organisms involved in
the oxidation of organic matter may generate 1 3 times their own weight of
•
carbon dioxide in 24 hours. Microflora are able to do this work because they
have the necessary energy, transmitted to them by the sun. In habitats with
the same temperature and moisture conditions, carbon dioxide is highest
under oak and birch, intermediate under larch and lowest under spruce and
pine, although both carbon dioxide and vegetation type are probably both
variables dependent on soil fertility. Carbon dioxide output is essentially the
product of microbial activity, root respiration itself accounting for perhaps
only about 20 per cent of the total soil air. However, microbial activity is
probably closely linked with root activity and its exudations on which the
organisms flourish. Seasonal changes of the amount of calcium carbonate in
solution in seepage water in Poole’s Cavern, Buxton (fig. III. 15.
B), and at
several points in other nearby caves, show a well-defined early autumn peak
believed to reflect the carbon dioxide conditions in the soil, allowing
for a
time-lag of some weeks or a few months for water flow-through
time. In fig.
million)
per
contenttparts
copbonote
Cafcfum
Figure 111.15 Relationship between temperature and biochemical processes in

soil and in water.
A. Soil carbon dioxide output at 20 cm depth. Wicken Fen, Cambridgeshire {from

V. \t. Co/iivav. 1936. New Phytologist, Vol. 35).
B. Calcium carbonate in seepage water, Poole’s Cavern, Buxton {from Pitty.
1966).
C. Calcium carbonate in a freshwater lake. Malham Tarn, "iforkshire (from A. F.
Pitty. 1967. Brit. Assoc. Lecture, Section E).
III. 15. B an unusual feature, attributed to the degree and abruptness of the
thaw following the exceptionally severe 1962 winter, is high values of dis-
solved calcium carbonate that might reflect the ‘spring burst’ of organic
activity that follows such a thaw. Although the significance of carbon dioxide,
when dissolved in water to form carbonic acid, is seen particularly clearly in
the solution of limestone, carbon dioxide has a profound and general signifi-
cance in rock-weathering. The role of carbon dioxide in increasing the solubi-

lity of iron, for instance, has also been stressed, and indeed traditionally the
theoretical significance of atmospheric carbon dioxide in rock-weathering has
frequently been indicated. This emphasis, however, has been misplaced where
the comparatively colossal output of biologically produced carbon dioxide in

the soil has been passed over.
There are many forms on and small scales indicating accelerated
large
weathering that are associated with organic activity. Fig. III. 16. A shows how
Glacial slriations Living moss Roughly weathered

rock
B
Pebbles While Cyanophyta 1 Polydofa Fucus Large borers
zone I
and and
\ Cyanophyta Lithothamnium
Approximate 2 -7m
width of zone
Mam process Biological Biological Biological

or agent of (Cyanophyta) fCyanophyta (molluscs)
erosion and Potydora)
Secondary Biological
(Cyanophyta
and Potydora)
Others Occasional abrasionai Potydora and

channeling seacurrenls
Figure 111.16 Biochemical weathering on rocky shores.

A. Pseudo-pillow structures in greenstone on Sakami Lake shore, Quebec {from
J.E. Anderson and J.P. Mills. 1968. Bull. geoi. Soc. Am., Vol. 79)'. (i) outcrop
glacially striated at N N 65° E, (ii) mosses growing along striations
30^ E and
where outcrop is wet, and
mosses outlining pseudo-pillows where outcrop is
(iii)
wet, weathered rock outlining pseudo-pillows where outcrop no longer stays wet,
and little-weathered striations on outcrop well above lake level,

B. Erosional zones on a shore platform in chalk at Eletot. Pays de Caux (after IV.
D. Nesterojf and F. Melieres, 1967, Bull. Soc. geol. de France, Ser.
7, Vol. 9).
colonization of lines of weakness in a rock surface by vegetation can produce

low elliptical domes in the intervening area. In favourable conditions in the
shore zone, biochemical processes can consume rock at least as quickly as
physical and purely chemical erosion.
5. Retardation of chemical weathering
In still or slow-moving water the carbon dioxide supply in the water is
diminished by the photosynthetic processes of aquatic plants. This may lead

to the precipitation of part of the dissolved load and some plants, notably the
stonewort Chara, absorb calcium bicarbonate in order to extract carbon
dioxide, leading to further precipitation. Like the increased intensity of solu-
tional processes, theamount of precipitation in such situations increases with
the more vigorous biological activity at higher temperatures. Water draining
out of Malham Tarn in the Carboniferous Limestone area of north-west York-
shire shows the results of this process, and the contrast between figs. III. 15.C
and III. 15. B epitomizes the antagonistic roles of organisms in chemical
weathering. Due to a different environmental change ferric hydroxides may be
precipitated by the action of specific bacteria which utilize all the oxygen
in the water for their respiration. In soils, particularly in lower lati-
tudes and especially under savanna conditions, desiccation leads to the

precipitation of iron solutions particularly close to tree roots due to the high
osmotic pressures. Thus elements like aluminium and calcium, as well as iron
and organic compounds taken into solution in the wet season are immobilized
almost entirely in the sub-soil horizons, which accordingly becomes ferru-
ginized. Similarly on coasts in a warmer environment, algae and other organ-

isms probably play an important part in creating the conditions for intensive
chemical precipitation in the formation of concretionary deposits like beach-

rock, not to mention coral islands.
Plants may appear to retard chemical weathering by incorporating a range
of mineral elements into the biomass. Even in the tundra, where the decom-
position of plant remains follows the same pattern as in many of the coni-
ferous forest communities of the temperate zone, and is accompanied by
relative accumulation of silicon, alumina, and iron, considerably less accu-
mulation of calcium and magnesium and losses of the mobile substances like
potassium, chlorine, and phosphorous. The retarding effect in these cool

environments is that minerals remain locked up in plant residues for long
periods. There may be more than 4000 kg/ha of minerals immobilized. In

areas of active peat accumulation rates of addition are of the order of 20-
35 cm/ 1000 years. In the B horizon of a soil at Garpenberg, central Sweden,
the residence time of soil organic matter is 370 years ±100. Even in temper-
ate latitudes it may take two centuries to transform a fallen tree-trunk into
humus and the organic mass on the forest floor is two to five times that of
litter fall. By contrast, in tropical rainforest the very rapidity of biological
processes is involved in the retardation of chemical weathering. The accu-

mulation of organic matter, although litter fall is four times as great, is rarely
more than a third or a half that of deciduous forests of the temperate zone and
contains only between 80 and 300 kg/ha of minerals. This is a very small
reserve in comparison with the annual re-uptake of about 2000 kg/ha, of
which about 800 kg/ha may be silicon, with calcium and potassium about
200 kg/ha each. There would therefore appear to be little chance of minerals
released from humified organic matter entering drainage waters. In fact some
tropical plants accumulate significant percentages of certain minerals, par-
ticularly silica, the cycling of which distinguishes tropical forests from those
of other zones. Five per cent of certain savanna grasses is silica by ash weight.
As long ago as 1 896 A. Grob found clots of amorphous silica in the internal
cavities of bamboo. In tropical forests of China silicon contents may be more

than 7 per cent. Although the degree to which minerals are irreversibly lost
from the biological cycle could be of profound geomorphological significance

if it were substantial, the problem has received little attention. Lovering
(1959), however, has suggested that a forest of silica accumulator plants
averaging 2-5 per cent silica and 16 tons dry weight annual new growth
would extract 2000 tons/acre/5000 years of silica, and this calculation merits
closer consideration. To the contrary, in fact, it seems that the balance

between uptake and return of mineral nutrients approaches the ideal steady
state. Measurements of uptake in arctic tundra and forest tundra were 38, 111,
and 166 kg/ha respectively, compared with returns of 37, 108, and 157 kg/ha.
In deciduous forest communities with quantities of 200-400 kg/ha of minerals
involved, there is a very small imbalance with accumulation exceeding losses
In the dry savanna of India measurements include an uptake of
slightly.
1000 kg/ha and a return of 800. These and similar figures show two things.
First, that for a given reserve of nutrients, the biological cycle is closed with
most minerals, on release fall, being immediately reutilized by growing plants.
There is therefore in a natural situation, with roots tapping solutions at depths
of several metres and the restricted or non-existent surface runoff in the litter
layer, little chance of erosional loss of mineral elements released from litter
fall. In the high oak forests of Voronezh province, U.S.S.R., measurements
have shown that of the elements which experience some irreversible loss, this
proportion of the total of these elements retained with litter fall is less that I
per cent. Secondly, even if all such elements were removed due to human
interference, they would constitute for a few years only a scarcely noticeable
addition to the total denudational losses. It seems therefore that the role of
plants in cycling nutrients is essentially to retard denudational losses
and that
the basis of Lovering’s calculations is an overestimate of
the degree to which
166 Introduction to Geoinorphology
the litter fall is swept off the forest floor, and an underestimate of the degree to
which the biological cycle is closed.
A final consideration is the possibility that organic acids, far from being a
significant agent in weathering as is perhaps too readily assumed, might act in
a protective role. From laboratory experiments which showed that amor-
phous silica solubility in several humic acid solutions was much less than that
shown in distilled water, R. Siever (1962) concluded that the colloidal
organic compounds were absorbed on the free silica gel surfaces and thus
prevented the solution of that surface.
6. Organisms as aids in landform study

\
The stratigraphical use of organisms and other ways in which biological

evidence may help in establishing chronologies will be illustrated in a later
section concerned with methods of dating stages in landform evolution.
Vegetation is dynamic depositional geomor-
an equally useful indicator in a
phological environment tending towards some equilibrium condition as it is

of indicating stages of sequential development. In the latter case plants, being
exceedingly sensitive to variations in soil quality and development, closely
reflect the uni-directional change in their environment. However, in a contin-
ually changing environment, like a delta, the one factor which in itself gives
the closest approximation to the continued average effect of the other pro-
cesses may be the vegetation. No plant community is in a static constant
condition and the stands continue to evolve according to channel activity and
may not approach a climax. On the landward fringe of small deltas in tem-
perate environments woods, meadows, and sedge fens do not normally receive
sediment, and in willow thickets there may be a very thin veneer at the most.
Conversely if a delta is growing rapidly, due perhaps to greater sediment
transport or to shallow basin depth, pioneer communities will occupy a large
part of the land areas. Reeds cover 29 per cent of the 9000 km'^ Hi river delta,
U.S.S.R., and their average height declines from 4-2 m in semi-flooded areas
to either 3-0 m in wet valley conditions and 2- 1 m in dry valley conditions.

Stems on the left bank are 0- 1-0-2 m higher than on the right bank. In a
dynamic geomorphological environment, the longer a plant lives, the greater
are its cumulative chances of being undercut and swept away. B. L. Everitt
(1968) in the study of part of a valley occupied by the actively meandering
stream, found the valley area of any age to be decreasing exponentially,
69 years being the time involved for an area of any age to decrease by half
(fig. 111.17).
Secondly, again in addition to its value in establishing times of change is
the use of biological evidence in calculating amounts of change. The position

of the base of trunks of older trees in relation to the present land surface
Figure 111.17 The process of flood-training and the use of its results in dating
flood-plain features (after Everitt, J968).
A. Supple saplings less than 10 years old are bent over and buried with only
shoots, now vertical, rising above the sand.
B. Age map of part of the Little Missouri valley floor compiled from age data
obtained by coring cottonwoods.
(fig. III. 18.B) may be useful in calculating amounts of pronounced erosion

or deposition. One of the more spectacular examples is the bristlecone pine
(Piniis aiiistata) in the dolomite areas of the White Mountains. California.
Here, on living trees more than 4500 years old, the stubs of ancient side roots
may project into the air 0-9-1 -2 m above the present ground surface. The
depth of movement of a solifluction feature is often indicated by a con-
Figure HI. 18 Botanical evidence of amounts of land surface change. (A) Ac-
cumulation, as recorded by nodes on a buried marram grass stem {from J. S.
Olson. 1958. Jour. Geol„ Vot. 66). Wide spacing of internodes occurs in years
following rapid deposition of sand in winter. Sudden decreases in annual grass
elongation indicate stabilization. (B) Degradation, as shown by progressive root
exposure {from V. C. La Marche. 1968. U.S. Geol. Surv. Prof. Paper 352-1).
spicuous layer of organic matter in the soil profile resulting from overrun
vegetation.
G. Human Activity
‘A change amounting to but little less than a revolution in the long established
processes by which the features of the earth’s surface are modified and developed,
accompanied the advancement of man from a state of barbarism to one of civilisa-
tion.’ I. C. Russell. 1904.
About 2 million years ago the ancestors of man emerged, with Early man
{Homo erectus) standing up straight about half a million years ago. Human
society and the deliberate organized work of Modern man began no more
than 10 000-15 000 years ago, and has since disturbed biological and ero-
sional balances over two-thirds of the land surface of the earth. The first
major study of this phenomenon, C. P. Marsh’s Man and Nature; or Physical

Geography as modified by human action was published in 1864 and by this
time large scale plantations were started in areas like the Aigoual area in
France, where intensive erosion by floods in the Tarn basin were understood
to reflect a link between deforestation and erosion. Despite the precedents,
Economics man has during the past century pressed a vigorous claim to being
one of the main converters of energy into erosional work.
One source of profound change has been man’s development of water
resources and drainage networks. Even on the smallest scales, like drainage
ditches, the increase in surface drainage density may be substantial; in
southern Scotland natural drainage densities of about 2 km/km^ have in-
creased twenty-fold. On a larger scale, water levels in the interior basins of

south-west United States have steadily declined over the past 50 years. For
example drops of more than 1 0 m in water levels beneath playa surfaces are
very common. Farther south the ground level itself in the Mexico City basin
first began to subside as water was removed from pore spaces by drainage
channels installed in Pre-Columbian and Colonial times and accelerated after
1879 when the basin was systematically drained. In the last half-century
intensified pumping lowered water levels even farther, and with the added
factor of the sheer load of a city of 4 million inhabitants, some parts of the
land surface have subsided at rates of up to 15 cm/year. Reservoirs not only
alter flow regimes but the impounding slopes may become unstable with
saturation at their base. Landslides occurred with ‘unusually great and unex-
pected frequency’ in the bordering Pleistocene deposits as the Grand Coulee
dam reservoir filled. The trapping of sediment in reservoirs may accelerate
erosion downstream in the river channel or on a coastline deprived of sedi-
ment. The total weight of stream-borne sediment reaching the Gulf of
California would have been about 370 million m tons/year. Since the inter-
ception of the Colorado river by the Hoover Dam its contribution to Gulf
sediment, formerly half the total, has been reduced to only some 8 million
m. tons/year.
Man’s removal of the natural vegetation, apart from his ploughing of the
cleared areas, has had such a profound effect on erosional processes because
of the variety of changes which all combine to increase surface runoff and at
the same time reduce the soil’s resistance to erosion. Apart from reduced
interception and evapotranspiration losses, the porous litter and root layer
vanishes and the soil, depleted in humus, has less power to absorb water and
to create a fine porous crumb structure. Fine soil particles no longer filtered
out in the litter layer are washed into lower horizons in the soil resulting in
reduced permeability. Table III.3 shows how the hydrological regime in the
Chernozem regions has changed in the past millennium. The simultaneous
effect of increased amounts and intensities of surface runoff acting on soils
with reduced cohesion is to increase erosion rates. In the Midwest of the
United States, sediment yields became as much as seventy-five times greater
than those under non-cultivated conditions. In India, deforestation in the
Godavari drainage basin has exposed vast areas which have subsequently
been eroded on a massive scale. In humid tropical areas the deeply weathered
mantles were vulnerable to even the smallest changes. The lavakas of
Madagascar are steep gullies 1 50-250 m
100-120 m broad, and 30- long,
40 m deep cut in the plateau edge, probably over the last 500-1000 years. In
Hong Kong gullies 150-200 m long and 2-30 m deep are due to progressive
deforestation over the last 2000 years or more, and in many more upland
areas of tropical forests, like Cuba, severe erosion followed deforestation.
Table I1L3 Transformation of the water balance due to human interference over the
past millennium in The Central Chernozem regions. U S S.R {from H I L'Vonch.
Soviet Soil Science. 1966)
Elements of the vater balance (in mm) are P precipitation. R = full ruer flow.
S = ground surface runoff. U = underground flow to ruers. E evapotranspiration
and \V soil water
ELEMEVTS OF THE l\ ATER BALANCE
p R s u E w U/Rô
Natural (some time before ninth-

tenth centuries) 480 50 18 32 430 462 64
Elementally transformed nineteenth-
twentieth centuries 480 100 72 28 380 408 28
Partly transformed b) socialist agri
cultural measures 1925-50 480 95 65 30 385 415 31
1960s 480 85 55 30 395 425 36
In other environments accelerated erosion has been produced in different
ways. In the savanna areas of Africa, the widespread firing of vegetation may
have removed half the biomass in a single sweep, prevented the addition of
humus to the soil and made the surface less permeable by baking irreversible
changes into the clay colloids. More recently engineers have been surprised
by the delicacy of balances in areas of frozen ground. Within a few weeks of
the bulldozing of 30-60 ems of gravel and sand off an arctic beach, a
hummocky and pitted surface developed because the underlying ice lenses
came within reach of the zone of seasonal thaw. The removal of natural
vegetation in such areas, scant though it may be, produces the same effect by
allowing deeper penetration of thawing temperatures. Around Anadyr in
Siberia, the removal of the tundra vegetation led to gullying, which, by

improving drainage, facilitated deeper penetration of the thaw, in turn favour-
ing intensified gullying. In drier climates removal of vegetation or lowering of
a watertable reduces the resistance of soil towind erosion. Farmers in most
sandy areas, like those in the sandveld of South Africa, have found that
ploughing greatly increased wind transport of sand, a problem recently acce-
lerated in eastern England by the removal of hedge boundaries. Agricultural
engineers suggest that in the Kara-Kum desert the uncontrollable grazing of
livestock is sufficient to initiate wind erosion and that as a result severely
wind-eroded soils develop at villages and at nomad camp sites. Again, as in
the case of gullying in the humid tropics, the immense power of Technological
man is not a prerequisite in creating the relatively small disturbance required
to trigger off accelerated erosion in the more delicately balanced environ-
ments.
Human activities in modifying land form have in many ways assisted in an
understanding of their development, particularly by the uncovering of fossil
evidence and by the exposure of artificial sections revealing the contacts
between biosphere, soil, landform, and subjacent rock. In fact it might be
suggested that one of the handicaps in the study of tropical weathering
processes is the paucity of artificial exposures in an area where natural
exposures are almost non-existent.
IV
Inter-relationships between
processes and landforms
The complexity of the factors involved in determining how the physical,

chemical, and biological forces decompose or disintegrate rock, and carry
away or deposit detritus, makes generalization about landform development
much more difficult than is sometimes appreciated. Rock breakdown trans-
portation and deposition are inextricably interlinked, but in order to study

their wide variety of inter-relationships in the development, perpetuation or
modification of landforms. an initial arbitrary separation to focus attention
on the main characteristics of each of these three phases is required, together
with discussion of aspects of landform that are largely the product of one only
of these phases.
A. Rock Breakdown
There are three aspects of rock breakdown. First, denudation is a general
term describing the breakdown of rock and the removal of part or all of the
weathering and erosionir^daucls~rfom~tHe site of their detachment. As
denudation ma y involve merely the transfer of weathering 'and erosional
products from one part of the land surface to another, the term net denudation
orlietlremoval describes the amount of material lost from the land surface to
the sea. Secondly, in its most restricted sense, rock breakdown may, due
either to chemical weathering or to mechanical pressures building up in situ,
involve little movement. A tI^d_aspect*of rock breakdown, erosiom is also

dueô^mechanical pressure, that of the abrasive action of weathered frag-
ments during transportation and other stresses due to the motion of running
water and moving ice.
In considering rock breakdown in situ it is vital to differentiate between

situations in which removal of material is also involved and those in which
parent material is merely reorganized rather than removed. The latter
Inter-relationships between processes and landforms 173
becomes increasingly the case in hotter climates where the ratio widens
between water entering the soil from above and water leaving the weathering
zone as runoff. Materials may be added to a weathering zone from an ex-
ternal, allogenic source, although detecting this may be possible only by
detailed mineralogical analysis.
As all weathering reactions involve water either as a reactant or in the

transportation of the reaction products, more than one chemical process is
often involved during decomposition. For instance, the weathering of feldspar
to clay involves both hydration and carbonation. Frequently the dominant

process changes within short distances and even within the span of a soil
profile. For instance, weathering of a ealcareous boulder clay will involve
rapid changes due to oxidation and hydration near the surface, the carbona-
tion of limestone fragments at greater depth, and thirdly, near the base of the
weathering zone, the comparatively slow dissociation of silicates and their
formation into clays.
Although the course of weathering is related to the physical and chemical
properties of water, the amount of denudation is equally related to the
amounts of water leaving the weathering zone. This is obvious from the
appearance of limestone surfaces in arid areas where few of the solution
features of karsts in humid climates develop. Conversely one of the factors
associated with areas of more intense bedrock corrosion in the gritstones of
the southern Pennines is an abundant supply of slow-moving seepage mois-
ture giving continuous wetting of the bedrock. Undoubtedly the most impor-
tant single factor controlling denudation is the quantity of water leaching
through the weathering profiles. Another factor which influences the amount
of chemical denudation is the time during which water is in contact with the
rock. For instance, the mineralization of soil water and superficial ground
watermay be two to three times that of slope surface water. Fig. IV.26.B
shows how water in the limestone areas in the Pennines, after acquiring
140 ppm calcium carbonate relatively instantly, gradually increases its con-
tent of calcium carbonate with progressively longer spells underground.
Chemical and mechanical weathering processes are often closely inter-
related. For instance, as chemical reactions occur on the surface of materials,
areas containing loose, mechanically fragmented rocks and minerals will tend
to yield far more and solids than areas devoid or stripped of com-
solutes
minuted fragments. Another factor which makes it difficult to distinguish
between chemical and mechanical weathering is the increase in volume which
accompanies the weathering reaction, hydrolysis, which has a disruptive force
greatly in excess of that exerted by most physical mechanisms.
1. Erosion
Erosion is a term_p^haps best reserved for rock breakdown by the dynamic

action of an agent like moving water or ice. Erosion may include ^rasion if
there are collisions between debrisjransported by water, ice, wind, or^ravity

fall and the adjacent bedrock. It is of vital importance to distinguish between
erosion, implying the breakdown of coherent rock, and the mere reworking of
unconsolidated material produced by a previous phase of weathering or
erosion. An immense volume of literature, particularly in fluvial geomorpho-
logy, is devoted to fluvial ‘erosion’ where discussion relates only to reworking
of unconsolidated material and almost entirely ignores the problems of
describing how running water might erode solid rock and of deciding on the
relative significance of this process in denudation as a whole. Discussion of
the significance of glacial erosion is also scant with many examples of neat
circular arguments inferring the scale of glacial erosion solely from the size of
features believed to be the product of its action. In both instances there are
several known circumstances which could theoretically contribute to erosion
but whether they play a major, minor or negligible role in rock breakdown is
as yet unknown.
Bedrock erosion in stream channels requires the operation of much larger
forces than those which cause soil erosion on slopes or the reworking of
channel alluvium. Abrasion of channel bedrock by the impact of bedload
moving at flood velocities is believed to have some erosional significance in
streams in middle and higher latitudes. Despite the probable significance of
reduced abrasion in pebble-deficient tropica! areas, the absence of ice could
also be significant. Even by its weight alone ice could aid lateral pressure-
release jointing on the lips of waterfalls in cool temperate and arctic environ-
ments. Abrasion by the saltating load of sand is also possible and the smooth-
ness of some bedrock surfaces in channels may be due to wet sand-blasting. In
humid tropical rivers where the process may have increased significance in
relation to other bedrock channel processes, J. Tricart has observed that some
bedrock surfaces are in consequence smoothed and like roches moutonnees in
appearance. A distinctive product of abrasion associated particularly with the
cutting through of rock barriers are potholes where initial hollows are drilled
deeper by eddies, swirling pebbles, and sand in spiral paths. On steeper slopes
the loosening of blocks by the force of impact of falling material is a distinc-
tive type of abrasion, and in a periglacial environment grooves up to 20 m
long have been ploughed by gliding boulders.
Water erosion may involve two purely hydraulic forces. The first,
hydraulic lift, is due to suction forces of eddies in turbulent flow which, on a

small scale, keep sand grains in suspension and can in larger flows lift
boulder-sized blocks. If lines of weakness in bedrock have developed due to

pressure release, or by chemical or mechanical weathering, it is conceivable
that hydraulic lift could detach such blocks. In fact as many observations of
flood conditions reveal, like recent flash floods in Vermont, erosion is
essentially in a horizontal direction, and degradation down through bedrock

is the result of long-term weathering. The frequent discovery of solutional
widening of limestone joints to 30-60 m below some river courses would
support this view. Dam engineers have found that rock resistances in valley
floors like the granite base of the Dours gorge may be fifty to a hundred times
less than that of laboratory samples.
A second hydraulic force, observed in turbulent flow confined in artificial
conduits and moving at speeds faster than 7-5 m/sec, is cavitation. Here
water molecules are pulled apart by momentary shocks to form airless
bubbles. The collapse of such bubbles involves pressures exceeding 30 000

atmospheres and the rapid hollowing of conduit walls, even if made of metal,
testifies to the erosional potential of this explosive hydraulic action. Although
cavitation is not to be expected in the unconfined condition of a surface
stream, it may well be involved in producing cavities along phreatic cave
passages and sub-glacial streams.
One most widely recognized focal points for stream erosion appears
of the
to be immediately below rock ledges or cliffs, particularly where the underly-
ing stratum is relatively weak. Undermining proceeds due to the action of
deepening and widening in the plunge pool, but the problem of exactly which
factors exert the dominant forces has not been examined in detail. Nonethe-
less, huge gorges have been developed at rapid rates by waterfall recession,
particularly along the lines adopted by large streams after glacial diversions
of drainage. The classic example is the postglacial Niagara gorge where
measurements over the past century show an average rate of retreat of
1-2 m/year for the Canadian Falls. Indications of rates of downcutting
through bedrock by streams can only be approximate due to the problem of
defining an initial reference point. Also, as the present example of the Niagara
gorge shows, apparent downcutting might be due essentially to headward
retreat of a waterfall position rather than to slow downwearing along an
extended line. Several examples suggest that an average rate of bedrock
channel lowering in middle latitudes might be roughly of the order of
15 cm/1000 years. In the Front Range, Colorado, downcutting of 60-120 m
appears to have occurred within Quaternary time and probably started in pre-
Wisconsin times or Beartooth Mountains, Montana, streams
earlier. In the
are now 90-180 m
below gravel-covered benches which may be early
Pleistocene or even younger. In the humid tropics even the larger rivers, in
crossing a resistant outcrop, scarcely cut a gorge through it. There may be
waterfall or rapids recession, but rarely does this extend more than a few
1 00 m upstream from the outcrop.
If a river course, instead of flowing over a resistant stratum, flows parallel

with its edge, the underlying less resistant stratum might be excavated by
lateral erosion. If lateral erosion in inclined strata is dip-controlled it may
also involve a certain vertical component in the erosion of the channel
bedrock. The contribution of a stream to lateral erosion is not easily
separated from the activity of weathering and rillwash on the undercut slope.
R. P. Sharp (1940) concluded that about 40 per cent of the mountain front
retreat in the Ruby-East Humboldt Range, Nevada, was due to lateral ero-
sion by streams.
The same hydraulic and abrasional processes that operate on the bedrock
of a stream channel are active on abrasional platforms and cliffs along coasts.
There are several Isolated records comparable with that of a 60-kg rock
hurled 30 m
above sea-level on the Oregon coast. It is impossible to say if
abrasion more important than hydrolysis, salt crystallization, biological
is
activity, or mass movement, and whether their relative significance varies
with latitude. However, in the absence of abrasive materials the most power-
ful storm waves are relatively impotent against walls of massive well-
indurated rocks. R. J. Russell (1963) observes that there are instances where
sharp and complex changes in hydraulic pressure, frequent alteration between
wetting and drying, and the activities of water-level organisms have produced
inconsequential changes in the past 4000 years. For instance, granitic cliffed
coasts near Rio de Janeiro are somewhat stained but display few abrasional
features along the strandline.
Where it occurs cliff recession might be largely the result of landslipping
rather than erosion, and depends on some seaward gradient on the erosional
platform for the evacuation of debris from the cliff foot. Therefore, marine
erosion tends to be self-arresting after a wave-cut platform some tens to
hundreds of metres in width has been developed. Abrasional platforms on the
Israel shore are continuous for long stretches and may be 30 m or more wide.
On the north coast of Jamaica, west of Rio Bueno, the reef flat bench is
0-4 km wide. Probably due to the importance of biochemical weathering,

platforms on coral limestone are often wider and flatter than average. One
example is the Kenya to Tanga coast where the inclination of a coral platform
45—750 m wide is less than 1:200. In New Zealand a gradient of about
1:100 is often observed, closely approaching the suggested theoretical equi-
librium value. However, slopes of erosional platforms along the south coast of
England range from 1:14 at Eastbourne to 1:90 at Studlands. The influence
of fetch is well illustrated on the shores of small islands. The slope off the
south-east shore of Visingso island in Lake Vattern, Sweden, is 1 : 1 9 to 1:11
(3-5 degrees) whereas the north-east shore slope is 1:70 (0-8 degrees).
Although some sea-cliffs may have remained in essentially the same posi-
tion for 100 000 years, postglacial rise in sea-level or local circumstances
have led to measurable rates of retreat of others. On Gotland, built ex-

clusively of Silurian limestones, marls, and sandstones, wave-cut platforms
are 100-200 m wide and cliffs retreat at 0-4-0-6 cm/year. On Barbados, a
boulder from the Krakatoa eruption indicates 30 cm of marine erosion in 60

years, suggesting an average rate of 0-5 cm/year. Rates of marine erosion are
often considerable in unconsolidated materials. Many stretches along the west
Louisiana coast were retreating at rates of up to 1-8 m/year prior to the

arrival of Hurricane Audrey in 1957. Cape Cod cliffs retreat by approxi-
mately 0-9 m/year. Boulder clays and cliffs on the Yorkshire coast from
Flamborough Head to Spurn Point have retreated 3 km since Roman times.
On the southern coast of the Lleyn peninsula, Caernarvonshire, cliffs 1-5-

9 m high are cut back at 0-1 5-0- 60 m/year.
These figures suggest that retreats of boulder-clay cliffs of a few kilometres
will have occurred along many coasts since the last glacial retreat, and that
during the repeated sea-level rises in Pleistocene times it is conceivable that
cliff retreat in even moderately resistant rocks might have also amounted to a
few kilometres during the span of Pleistocene time. However, for cliffs like
those of the English Channel in south-east England and Normandy lining sea
breaches through narrow necks of land, initial rates of erosion comparable

with average rates of erosion in boulder clay seem possible. The degree to
which coastal erosion in coherent strata is a self-regulating mechanism has
not been established. As a wave-cut platform develops, increasingly the
energy of wave advance is expended in crossing the platform and dissipated in
reworking sediment veneers before it reaches the base of the cliff. In such
circumstances cliff erosion depends on the possibility of little understood
processes lowering the platform. Perhaps the intense organic activity in the
intertidal zone should not be overlooked in studies attempting to assess
whether appreciable lowering of wave-cut platforms takes place.
To an observer a glacier or ice-sheet may convey an overpowering impres-
sion, yet to any obstacle it may transmit a maximum pressure of 2 kg/cm-^
only. The physicists concerned with the mechanisms of movement have
ice
not given the same consideration to the problem of specifying the actual way
in which ice might, with or without the aid of abrasion by bed material, erode
rock. In consequence there is a range of opinion on the efficiency of ice as an
agent of erosion. Large boulders, regardless of the force by which they are
transported may by their own weight be sufficient to crush much smaller
obstacles and to scratch smooth surfaces. For instance, striations on quartz-
ites in the Fort Churchill area, Manitoba, occur once every 12 cm with
lengths varying from 3 to 280 cm with breadths between 0-2 and 4-5 cm.
Depths usually a fraction of a millimetre rarely exceed 2 mm. Striae and
larger sinuous grooves and similar scouring forms often occur on the lee-side of
resistant protuberances, indicating that the erosional agent was not associated
178 Introduction to Geoniorphology
with ice deformations due to pressure but rather to a separate scouring sub-
stance, J. Gjessing (1967) suggests that scouring forms might be asso-
ciated with soaked ground moraine or a compound of water, ice particles, and
rock material, flowing in these lee-side positions between the bedrock surface
and the underside of the ice. On a larger scale, testimonies to the efficiency of
erosion by glaciers rests on the scale of forms assumed to be the product of
glacial erosion. Glacial action appears to remove irregularities in valley sides.
For instance, sharp ridges, pyramidal peaks and many ravines characterize
the land surface above 1850 m along the south-west wall of the Shakwak
valley in the St Elias Range. Yukon. The 1850 m level is an upper limit above
the lower valley sides, smoothed apparently by glaciation. As present-day
large-scale landslides in mountainous areas show, many steep slopes have
been weathered to an extent that little internal cohesion remains. These
features suggest that ice might therefore truncate obstacles like valley-side
spurs, thus straightening valleys of the preglacial topography, and also that
ice failed to dislodge parts of slopes where lack of internal cohesion was
approaching critical limits. Conversely there are e.xamples in glaciated areas
of valley constrictions corresponding to resistant rock outcrops like those in

the Aberdeenshire River Dee upvalley from Banchory. Neither is a glaciated
valley invariably U-shaped. Like Glen Nevis they may vary from a U-shape,
narrow to a V-shape, and then open out into a broad U-shape. Other instances
that complicate the specification of landform characteristics uniquely asso-
ciated with glacial erosion include the truncated spurs of recently faulted
relief and the valley cross-sections of the U-shape. that may be pronounced
along the sides of fluvial valleys, like the lower Driesam in the Black Forest.
Just as there is an inter-relationship between pressure-release jointing and the
dome shape of hills in areas of crystalline rocks a similar association for the
inverted dome shape of the U-shaped valley is easily envisaged. In more open
lowland situations the resistance of certain rocks is important, seen particu-
larly well in areas where small igneous intrusions resist erosion and in the lee
of which less resistant rocks escape erosion. The many such ‘crag and tail’
features in the Central Lowlands have been discussed and illustrated for
nearly a century. In other lowland situations, straightened sides of an adjacent

upland mass may give the impression of ice activity. In moving north, the
Keewatin ice-sheet is supposed to have straightened and smoothed off the
eastern scarp of the mountain overlooking the Mackenzie delta.
Although it is usually supposed that glaciers have straightened out valley
irregularities in plan, the argument is reversed for irregularities in longitu-
dinal section. Valleys at present or formerly occupied by glaciers often have a
longitudinal profile like a staircase (fig. IL15D). Each step or tread is rela-
tively flat or even overdeepened with a steep riser at its upvalley end and a
riegel, a knob of resistant rock, at its downvalley end which constitutes the
1

Sognefjord
level
sea
below
Hardangerfjord
Metros
Figure IV. Longitudinal profiles of Norwegian Qords {aj'm Holtedahl. 1967).
riser for the upvalley end of the adjacent, lower tread. Again, however, this
feature of a staircase longitudinal profile is not uniquely associated with areas
of glaciation. In valleys attributed largely to fluvial erosion analogous series
of levels have in the past often been linked with former, higher sea-levels and
the steeper sections regarded as knickpoints introduced by rejuvenation as
successively lower levels encroach on those above. In Corsica the longi-
tudinal profile of the upper Vecchio river valley in the Monte d’Oro massif is
a series of steps only a few hundred metres long, separated by a series of
subvertical cliffs about 20-40 degrees in inclination. Observations in areas
where there is no transported load to fill up hollows show that rivers, like the
Gdta Alv in Sweden, may erode hollows as deep as 20 m below the level of a
downstream rock barrier.
In fiords, long narrow arms of the sea with parallel, steep-sided walls, often
rectilinear in plan and common on higher latitude coasts, the longitudinal

section is distinctively basin-shaped overdeepened inland and with raised rock
thresholds near the outlets. The overdeepening of these basins in relation to
the thresholds, often cited as the clearest indication of the erosive power of
ice, may be substantial. For the Skelton inlet, an arm of the Ross Sea,
Antarctica, a maximum figure is 1933 m. This compares with depths like
1288 m for Messier Channel, Chile, 998 m for Scoresby Sound, East
Greenland. 764 m for Finlayson Channel, British Columbia, or 527 m for
Breaksea Sound. New Zealand. Along the Norwegian fiord coast, Sognefiord
is overdeepened by 1308 m, although neighbouring fiords are not usually
more than half this depth (fig. IV. I). Not all fiords show a systematic relation-
ship between fracture pattern and their orientation, as seen in the Glomfiord-
Melfiord area farther north. For the formation of fiords. R. F. Flint suggests
2
that the importance of various factors must vary substantially from one fiord
to the next. A. P. Crary (1966) considers the significance of possible inter-
relationships between fiord ice and the ocean. He suggests that once floating
ice is formed downvalley, erosion is limited to an area near the junction with
the grounded ice (fig. IV.2). As the depth of such ice thickens inland, the
erosion of bedrock could therefore lead to overdeepening inland, particularly
if land margins were rising slowly through the order of 1000 m that occurs
during isostatic rebound. In contrast to ice floated out of an inlet he considers
grounded ice as most efficient in clearing away obstacles to gravity flow above
sea-level. The spreading of an ice-sheet on leaving the constriction of valley
walls also involves a thinning of ice and this effect therefore has some
similarities to the thinning produced by the floating olT at coastal margin.
Figiiie IV. Glacier ice floating on to sea-water in Skelton Inlet, Antarctica (after
Crary. 1966).
There is also the view that ice erodes comparatively little. In 1890, A. C.
Lawson considered that there was no evidence to suggest that the surface of
the Canadian Shield had undergone any material reduction in level due to
glaciation. Around Flin Flon. 650 km south-west of Hudson Bay, an acci-
dented pre-Ordovician topography, with a relief of 20 m where it was covered
by the Quaternary ice-sheet, is essentially the same as where it remained
fossilized by an Ordovician cover. Similar features have been recorded in
Lower Cambrian strata in Finland. In the Fall Zone of Connecticut, R. F.
Flint considers that glacial erosion, and preglacial erosion too, appear to have
altered structural alignments very little. A. Godard (1965) suggests that much
of the landscape of north-west Scotland appears to be largely the product of
preglacial weathering. In areas of valley glaciation like Alaska, it is observed
that glaciers can advance and recede without greatly modifying the terrain. In
Iceland, lava flows indicate that a large portion of the landscape is of pre-
Pleistocene or very early Pleistocene age, and in British Columbia show that
the late-Tertiary relief of the Interior Plateau has been little modified by
Pleistocene glaciation. In the Alps, some estimates, including the possible
effect of long periods of interglacial sub-aerial erosion, suggest 300-400 m of
overdeepening, while other investigators suggest that almost no overdeepen-
ing by ice took place. Evidence on a smaller scale of inefficiency in ice erosion
includes the weathering pits that remained unerased by the Farmdale glacia-
tion in the Rock River area of northern Illinois. On a much larger scale are
Inter-relationships between processes and landforms 1 8
huge overdeepened troughs lying at right-angles to the direction of ice move-

ment and as deep as those lying in the direction of its general movement.
J.K. Gilbert (1904) remarked on this unexpected pattern in Alaska. Earlier J.
Geike (1878) discussed a similar pattern in north-west Scotland, where basins

along the whole of the east side of the Outer Hebrides reach about 200 m
below sea-level near Barra Head. However, he did not stress that the direction
of ice-sheet movement was almost perpendicular to the alignment of the
basins. J. H. Winslow (1966) discusses the problem posed by deepwater
continuations of some fiords, like Storfiord in Norway, extending beyond the
known limits of glacial advance.
Although the necessity for some breakdown of rock prior to glaciation is
increasingly recognized as a prerequisite for removal of rock debris by ice,
there are several examples of loosened materials that have remained un-
affected by the passage of ice. In the lowland area from the Orkneys south to
Central Scotland deeply weathered rock remains beneath boulder clay
deposits. In Swedish Norrland around Junsele, soft peaty lacustrine deposits
are similarly buried, and a thick bed of kaolin at Ivo in Skane survived
vigorous ice movement. In northern Baffin Island a body of ice, admittedly
less than 30 m thick, receded recently to reveal undisturbed patterned ground
features and vegetation.
Cirques are another overdeepened feature, sunk into the higher parts of
areas where valley glaciers formerly existed or persist today. Their distinctive
hollow form, remarkably regular for an erosional feature, appears to show
little modification with differing structures. In the Lake District, cirques
unmodified in form cut across complex structures like the faulted contact
between the Skiddaw slates and the Borrowdale volcanic series. In higher
latitudes cirques become progressively few as the area occupied by major
continental ice-sheets increases. Their development appears to be greatest
near the limit of permanent snows and the importance of oscillations through
freezing point is a significant consideration in many hypotheses. Originally,
following the work of Willard Johnson who in 1904 was lowered down the
gap between a cirque headwall and the ice of a cirque glacier, freeze-thaw at
the base of this gap, or bergschrund, was believed to be significant. Other
hypotheses have depended on the assumed erosional activity of ice. In the last
of a series of hypotheses proposed by W. V. Lewis, rotational slip was
suggested. However, this hypothesis appears to depend on the pre-existence of
a hollow shape and a further objection is that measured ice velocities are not
greatest at the base of a cirque glacier near its outer threshold end as the
phrase ‘rotational slip’ implies. Many areas may be like the Lake District with
cirques located in structurally weak zones where pre-glacial hollows might
have readily developed. Once again distinctive landforms
are not easily
causally linked with areally correlated glacial phenomenon.
FigwelVS Forms typical of middle and high latitude mountains formerly

occupied by ice
A The U-shaped valley cross-profile, as illustrated by the middle part of the

Sognefiord (J- Gjessing. 1965-1966. Norsk Geog Tidsskr . Vol 20)
B. Cross-section of a cirque. Mount Olympus, centralTasmania (£ Dei byslw e in
Fairbridge. 1968) This section across the Lake Enone cirque typically displays;
( 1 ) a steep headwall. (2) a concave floor, in this case a rock basin modified by talus
accumulation at the foot of the headwall, and (3) a threshold of bedrock covered
by a thin veneer of glacial drift.
C. Plan views of comes (cirques) in the Cairngorms (Z?. E. Siigdeii. 1969. Scot,
geog. Mag., Vol 85): (i) Come an t Sabhail, and
an-t Sneachda. (ii) Come
(ill) Come
Ruadh, emphasizing the symmetry of these forms by showing their
approximation to the arc of a semicircle. The headwall of Come an t Sneachda
includes a secondary come
It is not easy to draw conclusions about the importance of glacial erosion.
Even without considering the mechanisms involved it is difficult to observe
points at which ice might erode rock and unsafe to consider that the amount
of debris in melt water represents material actually eroded by ice. For land
form evolution in many areas once occupied by ice, contemporary opinion is
that tectonic and preglacial erosion may well have been at least as important if
not dominant factors. For instance. H. Holtedahl (1967) considers that the
presence of well-developed, typical fiords along the west coast of Norway is a
natural consequence of the oblique Tertiary uplift of the Norwegian land
mass, leading to increased fluvial erosion which was especially active in
cutting deep preglacial valleys on the seaward slope. It is suggested that a
glacier is, above all, an agent of transport with ice too plastic and too weak to
Inter-relationships between processes and landfornis 183
there are other areas For

attach itself firmly to a coherent rock. In contrast
forms them-
which interpretations are based largely on the appearance of the
the conclu-
selves and some investigators record that they are unable to avoid
has taken place on a vast scale. J. B. Sissons
sion that deep excavation by ice
erosion has been extremely
(1967) suggests that ‘there is no doubt that glacial
effective in many parts of Scotland.’
Although many workers have increasing doubts about the efficiency of ice
in the erosion of unweathered or unloosened rock, their

views should not be
confused with the ‘Protectionists’, a school of anti-glacialists who at the
beginning of the twentieth century went beyond the belief that ice had little
erosive power and claimed that an ice cover protected the underlying
rock
from erosion, notably from frost-shattering. Many present-day workers,

whose beliefs in the efficiency of ice erosion may vary widely, agree that in
certain localities there is substantial erosion beneath ice due to the pres-
surized flow of meltwater confined by the ice cover. The heavy loads of fine
material in meltwater from valley glaciers indicate the importance of com-
minution of transported material in sub-glacial meltwaters if not to actual
degradation of the channel floor. Giant potholes appear to be associated with
former sub-glacial environments (fig. IV.4). In the Flam valley, a tributary of
Sognefjord. organic material in the bottom of one of the large potholes close
to the summit of Furuberget was dated as 9350 ± 300 years B.P. suggesting
that active erosion took place in pre-Boreaiand earlier times, presumably in
Figure IV.4 Sections through giant potholes {from Holtedahl. 1967). The
potholes were surveyed by I. Klovning at Furuberget, in the Flam valley, Norway.
Erosion by sub-glacial streams under hydrostatic pressure is suggested.
a sub-glacial environment For comparison, potholes at Taylor’s Falls,

Minnesota are up to 4 m in diameter. In addition there are smoothly sculp-
tured micro-forms on rock surfaces including channels, troughs, and hollows,
termed p-forms by Scandinavian geomorpHologists and ascribed to the flow of
sub-glacial melt water. Hjulstrdm (1935) was the first to suggest that cavita-
tion might be involved in their formation. A large number of localities
containing /j-forms are situated on mountain plateaux in broad interfluve

zones. In many parts of the British Isles formerly covered by ice, channels cut
across interfluves with a characteristic ‘up-and-down’ longitudinal profile
distinguishing them from present-day stream occupied valleys, are now
widely attributed to sub-glacial melt-water erosion. In some parts channels
descending at the gradient of the local slope have, following the work of C. M.
Mannerfelt (1945), been described as sub-glacial chutes and attributed to
accelerated drainage in this direction following a supposed release of water
held under hydrostatic pressure in the lowest parts beneath an ice-sheet.
In addition to meltwaters flowing under hydrostatic pressure beneath ice,
there is the possibility in stagnant ice of melt waters flowing freely at the base
of crevasses reaching through the ice to the ground beneath. M. M. Leighton
and J. A. Brophy (1961) have termed straight and shallow channels believed
to have been cut in this manner as crevasse traces and have described many
examples showing striking alignment in western Illinois. Many are 5-10 km
in length and one is over 50 km long. Melt- water channels in areas formerly
covered by stagnant ice are among the more spectacular glacial features of the
Plains region of the United States.
2. Frost action
The degree to which frost action is involved in general weathering processes,
as opposed to the production of a specific form, is not clear. An initial
difficulty is that for freezing of water in microfissures to be effective the

crystals must possess microfissures already enlarged sufficiently for the freez-
ing of contained water. There are reports that particle size in arctic soils is
largely controlled by the grain size of the parent rock and that in the
Antarctic, soil-forming processes are barely discernible. Generally, in arctic
and sub-arctic areas, diurnal freeze-thaw can cause only differential movement
within a surface layer of a few centimetres and may not make a signi-
ficant contribution to the amount of weathered material. However, frost break-
down of mineral particles may be significant on glacial outwash plains where
saturated conditions, seasonal changes in freezing conditions and fluvial
reworking of the sands appears to reduce mineral particles to sizes pre-
dominantly in the 1 0-1 00-micron range, indicated by C. Troll (1944) as the
size-range typical of frost-rived particles, and subsequently observed in high
1
latitudesboth in North America and in Russia. Many attempts to evaluate, by

controlled experiment, the importance of freeze— thaw action in the initial
disintegration of rocks have been attempted since A. G. Hogbom made some
preliminary experiments in 1899. In one experiment a simulation of 16 years
1—
of freeze—thaw on rock fragments increased the proportion of the 0-
1-0 mm range by 10-20 per cent. S. Wiman (1963) used 36 ‘Icelandic-type’
-1° to 6°C cycles and -30° to 15°C cycles, but produced
9 ‘Siberian-type’
debris which was only a small percentage of the original weights. Samples in
dry plastic containers showed no weathering, but it is difficult to decide
whether this shows that water is a prerequisite for frost action or whether it
illustrates hydration producing a similar disruptive effect. Table IV. 1 pro-

vides a further illustration of materials subjected to laboratory freeze—thaw
cycles. Blumche, cited by Tricart and Cailleux (1967), showed that the
number of cycles needed to develop a fissure in a sandstone of 25 per
cent porosity was 3 whereas in a sandstone of 5 per cent porosity 43 cycles
were necessary. In a chalk sample with 30 per cent porosity 1 cycle was
sufficient and several experiments have observed how porous chalk dis-
integrates completely with a few freeze-thaw cycles.
Table IV. Artificially frost-weathered products as a percentage of the original

weights of the rock samples (/rom Wiman, 1963).
ROCK TYPE ICELANDIC SIBERIAN
Slate 116 016

Mica-schist 0-25 0-04
Granite 015 0-07
Porphoritic granite 0-29 019
Quartzite 0-02-019 0-007
Gneiss 0-65 0-01
Any conclusions on the role of freezing and thawing in the disintegration

of coherent rock material must emphasize the difficulty of separating this
from the role of other processes and point to the likelihood of its effective
operation being restricted to a certain number of rocks
or porous
like schists
sandstones where pre-existing openings are large enough to permit ice-
formation. R. J. St. Arnaud and E. P. Whiteside’s (1963) experiment
shows
the range of results between duplicate samples (Table IV.2).
This reduces the
significance of the contrast between their reduction of a
granite sample by
0-48 per cent compared with the 0-15 per cent reduction of 36
‘Icelandic’
cycles by S. Wiman (1963). Table IV.2 also shows a decline in
debris
produced in the second 100-cycle treatment, perhaps because
the initial
stages in artificial weathering exploits pre-existing
weaknesses. A more funda-
mental reason for believing that the results of laboratory experiments cannot
be applied to natural situations is that where frost-shattering is demonstrably
effective the process, if the weathered material is not removed, becomes self-
arresting by producing a layer of material through which freezing

temperatures fail to penetrate to bedrock.
Table /K.2 Effect of repeated artificial freeze-thaw cycles on the breakdown of

mineral grains {/lom Ainaiid and Whiteside. 1963)
FIRST 1 00 CYCLES SECOND 1 00 CYCLES
Initial Size range, First Second First Second

material millimetres sample sample sample sample
A 0 05- 2 0 1 30 1 93 0 64 0 44
Horizon 2 0 -10 0 0 61 I 58 0 44 0 32
B 0 5-20 3 49 2 84 0 32 0 35
Horizon 2 0 -10 0 1 68 0 66 1 40 0 60
Crushed 10-20 0 16 — — —
Quartz 2 0 -10 0 0 13 0 18 — —
Granite cube 60 0 48 — — —
3. Other mechanical processes of rock breakdown
As early as 1 925 E. Blanck and S. Passarge attributed the crumbling of rocks

in the Egyptian desert to the crystallization of salts. In recent years workers
have reported increasingly on the apparent importance of salt crystal growth
in rock breakdown. One of the most important weathering agents in the semi-
arid zone may be the salty dust which lodges in the openings in the rock and
swells when light rain falls. Basalts near Laguna Pueblo. New Mexico, have
been weathered in this manner. Similarly this mode of mechanical disintegra-
tion is possibly significant on tropical coasts where evaporation is intense and
the salt in sea-water spray crystallizes in fine cracks in the rocks. In dry areas
in Antarctica salt-weathering may be the dominant process of rock break-
down and appears responsible, at least in part, for cavernous weathering
features up to 2 m in diameter, and for the reduction of quartz diorite in
McMurdo Sound to a poorly sorted sand with diameters as small as 3

microns. Some workers believe that an essential characteristic of salt-
weathering is its more rapid action on the lower rather than the upper side of
rock surfaces. A somewhat simitar mechanical-chemical weathering process
may be the crystallizing of caliche. It is conceivable that over long periods of
time this might be a sufficiently disruptive force to disintegrate both super-
ficial deposits and bedrock in semi-arid regions. Temperature changes can
Inter-relationships between processes and landfonns 187
theoretically induce some expansion and shrinkage of rocks. During the past
30 years many investigators have concluded that this process is ineffective.
The role of wind in the disintegration of solid bedrock, as with that of
stream action, is easily exaggerated if the scale of transportation of debris
usually ascribed to ‘wind erosion’ is not carefully evaluated. Again by far the
larger part of material removed and transported by wind represents the
reworking of previously disintegrated, loose material with little contribution
from the breakdown of solid rock.

Only very soft rocks will disintegrate under the blast of wind-blown sand at
some distance above the ground. However, as ventifacts show, in arid
climates the comminution of debris and the cutting of resistant rocks is
possible in a shallow layer close to the ground surface. Near Marble Point,
Antarctica, the amounts removed are usually less than 3 mm of rock.
P. Kuenen (1960) suggests that such modification might take a dozen or even
a hundred years. In an experimental plot set up in a windy pass in the Western
Coachella valley, California, R. P. Sharp (1964) found maximum wear at the
discrete height of 23 cm above the ground, which is above the height of 1 3 cm

below which half the saltating material travelled. In this area the critical wind
speed of 18 km/hour is exceeded 16-5 per cent of the time. The height of
maximum wear probably represents a level at which grain size, number, and
velocity combine to give greatest impact energy. Using bricks of a 2-5-3
hardness on Moh’s scale in a natural setting, he first observed pitting on the
228th day of the experiment, and after 1
1
years found a maximum cutting of
5-5 cm. Two other features were noted. The firstwas the development of a
residual sill at ground level on windward sides; the second, fluting on the top
surfaces of bricks produced by descending grains. In desert areas the effect of
wind abrasion is now generally thought to be restricted to the slight under-
cutting of low-standing outcrops. The drying effect of wind, in assisting
evaporation and thus accelerating salt crystallization in arid areas, is prob-
ably a more significant aspect of the role of wind in erosional processes than
abrasion by wind-driven particles.
However difficult it may be to judge the relative importance of chemical
weathering from the appearance of disintegrated rock, and regardless of the

degree to which a separation of the interlinked processes of chemical and
mechanical breakdown may be arbitrary, one fact is clear; the scale of the
invisible operation of chemical denudatipn has probably been underestimated
in a large number of landform interpretations. Some theoretical schemes of
landform development are remarkable in the
degree to which inferences about
processes are restricted to mechanical weathering and erosion.
Measurements
of the actual amounts lost show increasingly how widespread and substantial
is the scale of chemical weathering. The measurements made by A. Rapp
(1960) during his nine seasons’ study of mechanical weathering and mass
movement in a mica-schist area of arctic Sweden have, in themselves, aroused
great interest. The fact that he demonstrated also that this contribution to net
denudation was no greater than chemical denudation of 26 m. tons/kmVyear,
even at a latitude of 68-j°N, receives less attention. In the adjacent Skojem
R. Dahl (1967), from the weathered appearance of granite surfaces
district,
concluded that chemical weathering was possibly an important factor in the

formation of cavernous recesses. In most arctic areas the precise nature of
physical and chemical weathering has never been well-established although
for more than 30 years there has been a suspicion, first expressed by Glinka
(1914), Taber (1943), and others, that the importance of mechanical dis-
integration could be over-emphasized. In well-drained sites it appears that
hydration is quite active, although the presence of feldspar in the clay-size
fraction suggests that hydrolysis is not strongly developed.
Although little is known about the changes in the chemical character of
water as it passes from the surface through soil and rock to a zone of
saturation below, some measurements are available. In general, concentra-
tions tend to drop where precipitation is heavy and to increase where it is
slight. Figures may be very much a reflection of other local conditions,

difficult to place within broader zonal generalizations. For example, concen-
trations of calcium carbonate in limestone areas in Britain commonly fall
between 150 and 250 ppm, yet in the distinctive environment of a shallow,
seasonally dry lake in the McMurdo Sound area, Antarctica, concentrations
Table IV.3 Amounts of calcium and magnesium in solution in contrasted weathering

environments ranged across European U.S.S.R. (from B. G. Skakalskiy, Soviet
Hydrology, 1966).
Values are in mg/1, and represent the range of several measurements made in each
environment.
CALCIUM MAGNESIUM
Slope Ground Slope Ground

water water water water
Tundra I- 5 5- 63 1- 3 2- 23
Taiga 2-17 14-165 1- 4 4- 21
Mixed forest 5-17 23- 82 1- 4 8- 36
Northern steppe 26-54 85-156 I-ll 15- 81
Southern steppe 4-81 112-303 3-53 71-161
Forest steppe 17-33 65-433 1-12 8- 59
of 3630 ppm have been observed. Table IV.3 lists the range of values
observed for calcium and magnesium concentrations from north to south
across the Russian plains. Due to rapid reprecipitation, values from tropical
areas are difficult to interpret. At Koullon in the Congo a range of 90-270
ppm calcium carbonate has been recorded. It would take several years if not
centuries for terrestrial waters to approach equilibrium in silica, even in
humid tropical environments. Amounts of silica dissolved in superficial
groundwater as well as in river water are low. Calculations suggest a
weighted mean concentration of silica in all rivers of the world to be 13 ppm.
Even though removal of silica is an intrinsic part of humid tropical weather-
ing, rivers in the tropics have only a slightly greater load of colloidal silica
than those of the ancient massifs of Europe. Concentrations in waters drain-

ing from tropical forests are 10-40 ppm compared with the 14 ppm average
for North America. In fact silica is the least variable of the major constituents
in ground water. Values for the River Po range from 4 to 8 ppm. In
Honolulu, the range in superficial groundwater is 8-16 ppm silica with a
mean of 12 ppm, values for large springs average 20 ppm, and for deep wells,
38 ppm. In springs draining from Pre-Cambrian ferruginous cherts in Cerro
Bolivar the silica concentration is 10-5 ppm and 15 ppm in drill holes. In
weathered epidiorite in British Guiana, the figure for springs is 12-5 ppm and
33 ppm for wells 50 m deep. For the Mahadanz area in India, the amount is
26 ppm. In the 35 years after the 1883 Krakatoa eruption the silica content
of the ash decreased from 67 to 61 per cent. At present rates of removal in
forested areas of the Ivory coast, at 0- 7-2-5 mg/cmVyear all the silica in the
silicates would be removed in 44 000 years. On Oahu Island, Hawaii, silica
removal is 3-6 mg/cm Vyear. 0-
Some controlled experiments on 1-weathering show substantial losses. In
1930 the French workers Demolon and Bastisse broke down 800 kg of
granite into sizes in the 2-4-mm size range. Natural processes achieved the
further break down of sizes between 1 935 and 945 as shown in Table IV.4.
1
Over the period 1931-45 elements in the drainage water were removed in
substantial amounts (Table IV. 5).
Table IV.4 Percentage change in various sizes of granite fragments after 10 years’
natural weathering (from A. Demolon and E. Bastisse, 1946, C.r. Acad. Sci.. Vol.
223).
FRAGMENT SIZES IN MICRONS
<0-75 0-75-2 2-20 20-200 >200
Percentage in 1935 0-32 49 2-96 25-20 71-0

Percentage in 1945 0-45 00 3-85 29-50 64-75
General weathering rates are relevant to landform studies but not neces-
sarily easy to interpret. The nearly uniform and complete weathering shown
by the first 4-5 m of the soil on Kaui in the Hawaiis would require about
70 000 The weathering of dacitic ash to a depth of 0-9 m on the El
years.
Salvador would take at least 5000 years. Volcanic rocks erupted in recent
times in Indonesia and the West Indies show that the establishment of a
complete weathered zone could take 1 million years, and in Venezuela data
from springs suggest that the present iron-rich formations could be the result
of weathering processes operating over a span of 20 million years. In a
Table IV.5 Amount of dissolved solids removed from granite fragments after 10
years’ natural weathering {after A. Demolon and E. Bastisse, 1946, C.r. Acad. Sci..
Voi 223).
CaO MgO KjO SiOj
Initial weights (kg) 29-680 7-52 1-28 1-20 1-16

Amount removed (kg) 0-276 0-043 0-059 0-003 0-038
Annual average removal
in solution (gm) 19-8 3-0 4-2 0-2 2-7
forested area in the Ivory Coast, the calcium could be removed in 3000
all
years and all 000 years. In loess deposited in the Slims

the potassium in 1 1 7
river valley, south-west Yukon, 9780 years ago decalcification is complete in
the upper 22 cm and partial in the subjacent 20 cm, whereas Neoglacial loess
is unweathered. Soil profiles on moraines left by the receding Mendenhall
Glacier, near Juneau. Alaska, show that podzol profile formation is slight
after 250 years in this environment and that the establishment of an equili-
brium condition between environment and soil profile would take 500—1000
years. In deposits of the ancient early Quaternary or late Tertiary River
Teays weathering has destroyed all but the most resistant original materials.
Little remains other than quartz sand and siliceous gravel and boulders.
In appearance, changes of colour often indicate the progress of weathering.
Weathering by oxidation is typically indicated by a red or yellow surface
layer on weathered rock. Because of its abundance and easy oxidation, iron
shows the general progress of this type of weathering. Most frequently described
is the deep-weathering of acid crystalline rocks, on which the signs of
weathering are, first, a whitening of the rock apparently due to the development of
fine fractures in the feldspars. When the plagioclase is partly decomposed and as
the attack on orthoclase begins, the rock breaks down to platey fragments of
decomposed granite called grus. As corestones develop, the most prominent
fractures are parallel with their boundaries, and zones of weathering in the grus
are concentric around the corestones (fig. IV.5). Corestones rarely remain
coherent at sizes less than 1 m in diameter. Of the visual im-
pressions of rock breakdown the depth of weathering often attracts notice. In
Table IV.6 the Assuring of granite and gneiss in the Kola peninsula, to depths
comparable with many areas of deep-weathering is particularly noteworthy.
Figure IV.5 Development of granitic boulders by sub-soil weathering (fiom

W. M. Davis, 1938, Bull. geol. Soc. Am., Vol 49)
In certain favoured localities it is possible to observe the effects of denuda-

tion by the amount of ground surface lowering in relation to some reference
datum. These tend to be greatest in areas of relatively soluble rocks like those
rich in calcium where precipitation is moderate to heavy and where evapo-
transpiration losses are relatively slight. In a granite area near Narvik in
northern Norway, R. Dahl (1967) calculates a lowering of 1-1 mm/1000
years at 85-95 m above sea-level and 1-3 mm/1000 years at 120 m above
sea-level. This suggests an order of a negligible metre of lowering if present
climatic conditions prevailed for a million years. For the humid tropical
areas, P. Birot (1965) states the amounts of dissolved material removed from
crystalline rocks into approximate equivalent of 1-3 cm/ 1000 years surface
lowering. For the Hawaiian basalts, the theoretical lowering would be 13
cm/ 1000 years. In total, the Koolau extinct volcano could have been lowered
by 650 m in a period starting 1-3—5 million years ago. The rates of de-
nudation of the Hydrographers volcano, north-east Papua, increase inland
from 8 cm/1000 years near the coast to 52 cm/ 1 000 years at an elevation of
533 m. Bermuda’s precipitation, now about 1400 mm a year, has been suffi-
cient to lower the land surface by 125 cm since Sangamon time. At 2230 m
altitude in the Maren mountains, lowering of limestone surface takes place at

T5 cm/ 1000 years. In Kentucky, lowering rates are about 15 cm/ 1000
years. In chalk areas in southern England, open fissures and voids only partly
filled with clay indicate the solution active at the chalk surface. Differences of
level in localities
where the chalk surface is locally protected by prehistoric
earthworks suggest a lowering of the chalk surface at a rate of 10-12-5
cm/ 1 000 years. For most areas it is possible to calculate a rate of weathering
indirectly from amounts of material transported from a drainage basin. Illus-

trations of these rates and difficulties in their interpretation will be discussed
in the next section.
Table IV.6 Some examples of type and depth of weathering.
DESCRIPTION OF NATURE
OF WEATHERING AND
LOG ALITY LATITUDE ROCK TYPE DEPTH
Ballangen, 38 km west 69° N Mica schist and Grus 2-3 m

of Narvik phyllite
Kola peninsula 68°N Granite and gneiss Fissured to 10-30 m

Moraineless Buchan 57j-°N Weathered to 0-5-)0 m
Aberdeen 57°N Granite Weathered to 9 m and even
to 12 m
Scotland 56°N Granite, syenite. Chemically weathered to at
norite, felsite, least 12 m

schist, quartzite. Old
Red Sandstone, gneiss
Shap Fell, West- 54-[°N Granite Average depth of weathering
morland 0-6m
Banbury S2°N Ironstone Partial decalcification and
oxidation, averaging 3-
6 m
Boulder, Colorado 40°N Granodiorite Maximum depth 23 m
Sierra Nevada 38°N Granite Disintegration to 15-30 m
Teays river area 39°N Sandstone Weathering to more than
12 m
Japan 36°N Granite In places, decomposition of
20-30
Wiluna-Meekatharra 26- Granite and gneiss Weathering to more than
area, Western Australia 27°S 15 m
Sao Paulo, Brazil 24° S Weathering of open joints
100-130m
Hong Kong 22°N Granite Weathered to more than
30 m
Hawaii 20° N Basalts In places, partly weathered
to 30 m
Minais Gerais 19°S Average weathering about
100 m
Salisbury, Rhodesia I8°N Granite Decomposition to more than
10 m
Malacca 2j°N Granite Average depth of weathering
4 m
Singapore 2°N Gabbro and grano- Regolith depth 10-20 m
diorite
5. Weathering forms
A variety of rock surface features have been described and attributed to some
form of weathering. However, a great deal of uncertainty surrounds some of
the explanations linking certain forms with an areally associated weathering
process. The main difficulty is that the association of a weathering form with
a given process does not necessarily imply that the latter can exert the forces
required to produce the given form.
Many features are too small to be classed as landforms but assume an
importance in landform study out of proportion to their size because they
illustrate, at manageable scales of time and size, the interaction between land
surface and weathering process. Widely observed are localities where small-
scale recesses develop by ‘cavernous weathering’. These may give a pitted
appearance to certain vertical rock or boulder surfaces, or form weathering
pits on horizontal surfaces. The dry Victoria valley, Antarctica, provides an
example of the first mode of weathering where the hollows on boulders (fig.
1.4. A) on the desert pavement surface and on moraines tend to coalesce to

leave mushroom-like forms rather than honeycomb surfaces. The hollows
have no apparent preferred orientation or position in the boulders. For these
features and for similar forms developed on coasts where higher temperatures
involve periodic drying of sea spray, salt-weathering appears to provide a
satisfactory explanation. These pits are frequently noted in tropical and sub-
tropical regions, particularly on the fiat surfaces at high levels on inselbergs of
acid igneous rocks. They even occur on hematite outcrops in Minas Gerais,
Brazil. Their diameter is about a metre and their depth may be 10-20 cm.
Those on the huge inselbergs in South Australia may be up to 10 m in
diameter and a metre deep. In middle latitudes they occur on the upper
surfaces of isolated tor rocks, blocks, and on other flat surfaces at relatively
high levels and on shore platforms. Diameters of about 0-3 m are often
equalled by their depths. In Tasmania, pits are about 75 cm across and less
than 15 cm deep. They are well known on the Tertiary plateau basalt at
Gobbins Island, County Antrim, and have been observed as far north as
block-field zones in north-east Baffin Island. Weathering pits may develop in
sandstones and schists as well as in granites, lavas, and limestones. Although
H. Wilhelmy (1958) has pointed out that weathering pits form sub-aerially
without covering layers of soil their interpretation is uncertain. Possibly
different processes can produce similar self-accelerating mechanisms that lead
to the concentration of weathering activity at sites where initially
some
chance irregularity existed. The presence of a pool of water if periodically dry
seems to be essential. In fact, drying out is a feature of at least some pits;
those on tors on Monte Montarone near Lake Maggiore in Italy, dry out a few
hours after heavy summer rainstorms. Pits in limestones, karst wells, are
often several times deeper than their diameters, suggesting the importance of
chemical weathering and the importance of ease of removal of weathered
products for deepening to proceed. In addition to the removal of soluble
elements by biochemical and chemical weathering, it seems that hydration
will exert a mechanical effect in disrupting the grains or mineral crystals in
damp spots and that on shore platforms in lower latitudes, salt-weathering

may be involved. The surface of massive limestone monoliths of a prehistoric
monument in Derbyshire, receiving perhaps 1000-1125 mm of precipitation
annually, suggest rates of karst-well deepening of about 15 cm/ 1000 years.
On adjacent gritstone escarpments weathering pits are often filled with algae
and the pH of the water may be as as acid as 3-3-5. One self-accelerating
factor might be that the initial depression favours not only water accumulation
but also provides a slightly less harsh environment in which algae colonies
might survive or provide natural flower pots in which plants become estab-
lished. The bottom of pits in the Narvik area is lined with detritus consisting
of a mixture of gravel and mud-like silt suggesting the decomposition of
organic material. In many situations high winds perhaps provide the most
effective force in removing resistant mineral particles from the base of the pits
and this may explain their siting on exposed tor summits and account for why
periodic drying out is a characteristic of the hollows.
Some depressions are too large in scale or too broad in relation to their
shallow depth to be termed weathering pits. Some differ by occupying the
floors of ill-drained basins rather than level summits or isolated peaks. In
tropical rainforest deep weathering can introduce reversed gradients in the
underlying rock surface by the more rapid and deeper decomposition of less
resistant material. Again the process is most clearly displayed in limestone
areas, where steep-sided depressions develop into the distinctive cockpit karst
landscape well seen in West Indian islands like Jamaica. In more open lime-
stone depressions, the distinctive feature is the seasonally flooded level floor
bevelled across the limestone structures between steep cliffs in the humid
tropics, moderate slopes in less hot environments and with sides as gentle as 4
degrees in the basins or turloughs in western Ireland. Enclosed depressions
possibly due to biochemical weathering may even occur on quartzite, as
observed in the Blue Ridge, west of Morganton, North Carolina. The depres-
sions range from a few metres to as much as 60 m in diameter with a deepest
point of a metre. The dip is nearly horizontal which is probably an important
and aquatic moss which floats on the surface
factor and a thin layer of algae
of the ponds and covers the bottom materials when the ponds are dry may be
involved in the weathering process.
In South Africa basins known as pans are widespread. Sub-circular, oval
or irregular in shape, they range in diameter from a few tens of metres to
several kilometres, and in depth from a metre to as much as 60 m. Although
many appear to be wholly or partly the result of chemical weathering, the

deepening of many results from the removal of weathered debris by wind
transport. These pans are very much like the basins which occur by the
thousand on the Great Plains of the United States, ranging south from
Montana New Mexico
to in a belt where precipitation is between 250 and 500
mm/year. The piping of silt into underlying coarse-grained strata is another
possible explanation for these basins.
In cooler environments enclosed depressions have long been recognized.
However, as the semicircular walls around a cirque and the depression itself
have always been linked with erosion by ice, it would be unwise to discuss
these forms in the context of weathering however much they may in appear-
ance resemble the small-scale weathering pits. The case of nivation hollows
(fig. I.4.D) is somewhat different. In form they may not involve a reverse
gradient. The floors of those near Resolute Bay, North West Territories, have
a downhill slope of 3-|— 7 degrees and range in width from 10 to 60 m and in
length from 1 0 to 65 m. In occurrence they are associated with snowdrifts,
but the way in which the melt water from the snow acts in weathering a
nivation hollow is obscure. In addition to the problem of shallow depth of
freeze-thaw penetration there is the complication that snow cover if anything
blankets the underlying ground from sub-zero temperatures and from oscilla-
tions through freezing point. Abundant water supply and unfrozen soils might
conceivably favour localized acceleration in rates of chemical decomposition.
On moderately to steeply inclined bare rock surfaces, particularly in trop-
ical latitudes and on some limestone surfaces in most humid areas, channels
or flutings develop, following the slope of the rock surface. They often occur
on the steep slopes of sugar-loaf-shaped domes. One of the earlier descrip-
tions of their development on igneous rocks was that of H. S. Palmer (1927)
in discussion of the flutings on joint faces of basalts in Oahu, Hawaii. In
southern Malaya they are usually found on surfaces steeper than 60 degrees,
very few on inclinations less than this and never when below 23 degrees. A
shallow groove may be only 1 cm in depth, the deeper ones about 0-5 m but
most are less than 30 cm deep. On acid crystalline rocks in Liberia, the
channels may be 1 m in depth. On the surfaces of inselbergs in South
Australia, precipitous and even slightly overhanging faces are scored by
narrow grooves 20-30 cm deep and occasionally as deep as 60 cm. These do
not reach the foot of the wall but either fade out gradually or end abruptly in a
small hollow set into the rockface. This last feature suggests that these
channels cannot be regarded simply as originating as drainage channels, even
if subsequently this is
the role they perform.
Just as weathering pits on vertical surfaces have counterparts in exposed
horizontal areas, so also do flutings develop on horizontal as well as on
steeply inclined surfaces. These are usually associated withjoint widening and
Figure IV.6 Weathering forms in limestone.
A. Steepening and undercutting at the base of a cliff in a tropical environment
{J. N. Jennings and M. M. Sweeting. 1963. Bonner Geog. Abh., Voi. 32). a cliff-
foot cave near Barnet Spring. West Kimberley. Western Australia,

B. Small enclosed depressions on Cow Close Fell. Littondale, in north-west
Yorkshire {from K.M. Ciaylon, 1966. Field Studies. Vol. 2). The depressions
follow contacts in the alternating limestone-shale sequence of the Carboniferous
Yoredale beds and are formed largely in a mantle of boulder clay due to piping
down open joints in the underlying limestone.
C. A polje, the largest limestone enclosed depression form in middle and low
the rectilinear pattern of channels reflects the appearance of the joint direc-
tions. In addition to the enlargement of joints on limestone pavements, the
same process operates on joints in gneisses and schists and other jointed
rocks containing silicates of calcium. However, in situations where joint
expansion could be due largely to lateral expansion due to pressure release or
to slight failure in an incompetent underlying substratum, the amount of
channel widening along joints to be attributed to solution must be evaluated
with caution.
some processes combine to promote deepening of hollows in certain
Just as
situations, others combine to produce upstanding residual eminences. Tors
are residuals of bedrock isolated on all sides by cliffs a few metres high.
Those in Tasmania consist of a number of blocks, each one 2-4 m high but
with most tors totalling less than 6 m in height. Such isolated rock pedestals
and pinnacles are found in all massive coherent rocks like limestones, sand-
stones, granites, and basalts. They tend to occur in sites high up in relation to
their immediate surroundings. They may represent the remnants of the nuclei
of more massive portions of rock formerly surrounded by less resistant
material which was more thoroughly weathered and more readily removed.
There is a possibility that in massive rocks some little-understood combina-
tion of self-enhancing processes leads to the progressive reduction of weather-
ing on tor summits in comparison with surrounding areas, whereas at a
different scale other processes combine to produce depressions by focusing
weathering on a few specific points. On some flat summits the absence of any
appreciable downslope gradient may be a factor contributing to their preser-
vation because an important consideration is the removal of the weathered
material assumed to have previously surrounded the residuals of sounder
rock. Tors are usually attributed to a phase of differential weathering fol-
lowed by a phase dominated by the removal of the loosened material by mass

movements. D. L. Linton (1955) emphasizes the importance of how a fall in
base-level could produce this change. For arctic environments the differential
weathering was attributed to frost action by Hogbom (1912), Eakin (1916),
and J. Palmer (1956). In tropical areas workers have developed on the early
latitudes {after Jugoslav 000 map

official 1 '.100 series). This map of the Livno
polje also shows the seasonal flooding regime on the flat, enclosed floor of the
polje.
D. Weathering forms on bare limestone pavements
(from Jones. 1965', D iii from
Sweeting. 1966)'. (i) the effect of gentle dip on solutional channelling,
(li) on a horizontal pavement (dint) block, including (top centre) an
channelling
instance of centripetal channelling,
(iii) the dominance of joint control on a less
weathered dint, Chapel-le-Dale, and (iv) the dominance
of solutional weathering
on steeply inclined surfaces; the German term karren is sometimes used to
describe such runnels.
Figure IV. 7 Isolated rock domes in tropical environments.
A. Morphological map of the Oyo bornhardt, western Nigeria {after M. F.

Thomas, 1967).
B. Contour map of the Oyo bornhardt {after Thomas, 1967)-, contour interval 10
feet.
ideas of J.D. Falconer (1911) with the initial phase being one of chemical
action prior to mass movement in a removal phase. With chemical denudation
known to equal that of physical denudation in some arctic localities, the mild
temperatures of interglacials, higher temperatures in Tertiary times, and the
shallowness of freeze-thaw penetration in solid rock, the initial phase of rock
breakdown probably involved some chemical weathering even in arctic en-
vironments. The corners of loosened and residual blocks, however, remained
essentially angular. In mountainous areas, particularly near cliff edges in
jointed sedimentary strata, open joints are sometimes observed and may reach
depths of about 30 m and be as wide as 1-5 m at the top. Any freeze-thaw
activity at such depths would have an amplitude of only a degree or two and
free drainage in open joints would limit the amount of moisture present. This
point increases the possiblity that pressure-release jointing or expansion due

to slight failure in an underlying incompetent stratum keeps near-surface
rocks in a continually fractured state, open equally to mechanical, chemical,
and biological agents of rock breakdown.
Of other processes possibly involved in the development of tors, wind
abrasion, despite the appearance of pedestal rocks, is not now believed to have
sufficient force to undercut steep sides of tors. The possible significance of salt
weathering is now considered in some areas. In the arid Central Otago area in
New Zealand, there is widespread cavernous weathering near the base of
schist tors with admixtures of decomposed rock and salts within the hollows.
Fig.III.il is a reminder of the possible significance of tree roots in this

context.
B. Transportation
Some land surfaces inclined across bevelled structural features are now
referred to as transportational surfaces rather than as surfaces of erosion or
as erosional slopes. This usage reflects an increasing awareness of the intimate
link between transportation processes and many landforms. In studying the
transportation of debris, therefore, the geomorphologist gains not just an
insight into the mechanisms involved in the removal of weathered material
and in influencing depositional characteristics but also, in many instances,
into influences which directly or indirectly are an integral part of the
C. Postulated development of a bornhardt by the removal of the deep chemically

weathered profile around the dome and etching at its base {Thomas in Fairbriclge.
1968).
D. Kegelkarst (cone-karst) in Cuba (from H. Lehmann. 1960. Ann. de
Univ.
Lyon. Spec. No. 1 along a north-south section through the Sierra de los
1 )
Organos. with flat polje floors terminating abruptly at the foot of the residual
masses.
Figure IV. Illustrations oC difficulties in interpreting hillslopes as erosional
forms. (l)The role of differential resistance of rocks in the development of
benched slopes in past and present periglacial and glacial environments.
A. Profile of a stepped slope on Blomesletta in Vestspitsbergen {from R. S.

iVaters, 1962, Biuletyn Perygiacjalny, Vol. 11).
B. Profile of a stepped slope on Dartmoor. Devonshire {from R. S. Waters, idem.).

C. Profile of a stepped slope in the Bohemian-Moravian Highlands with perigla-
cial processes inferred (from J. Demek, 1968, Przeglad Geograficzny, Vol. 40).
D. Profiles of a stepped limestone pavement in Glencolumbkille South. County

Clare for which glacial scouring is often inferred {after Williams, 1966): ‘B’
indicates a bench. ‘S’ a scar.
processes shaping the landforms themselves. However, it is essential to note
at the outset, as hydraulic engineers and sedimentologists stress, that the
heterogeneity of sizes in natural sediments are not readily susceptible to

theoretical treatment, and in the absence of basic measurements, even the
expert’s understanding of the mechanisms involved in the entrainment and
transport of sediments is incomplete.
1. Transportation of particles in a fluid
The transportation of particles in a fluid involves three forces. These are the
velocity of the fluid motion, the force of gravity acting downwards and the
fluid resistance acting in a direction opposite to that of the motion of the
particle. With other factors equal, the force of gravity is proportional to the
cube of the grain diameter and therefore, while the settling velocity due to
gravity becomes increasingly high for particles of larger size, it is reduced to
negligible proportions in comparison with fluid forces for very small par-
ticles. Although it seems obvious that particles of diminishing size from
gravel or sand-size scales are progressively easier to move, F. Hjulstrom in
1939 recognized a critical factor in the transport of sediment in his studies of

fluvial processes by noting that the smallest size of material which can be
transported as bedload is 180 microns. This effect is due to the settling
velocity of material with a specific gravity of 2-65 being higher than the
frictional drag or threshold velocity above this size and lower for sizes less
than 180 microns (fig. The comparable turning-point on R. A. Bag-
IV. 10).
nold’s graph for wind transport is about 80 microns, close to the upper
median grain sizes recognized in loess material. As smaller sizes become

progressively more difficult to move, the velocities required to dislodge par-
ticles are still greater than those needed to transport them. It is as a result of
this phenomenon that dunes are not normally found in loess deposits because
the fine-grained nature of the material presents
a smooth surface texture, and
once particles finer than 80 microns have settled they cannot be swept up
again individually because they sink into a viscid surface layer and are out of
reach of turbulence. In water too the laminar sub-layer shields the finer grains
in interstices of coarser material. Also there may be much greater cohesion
among smaller particles due to inter-particle forces, an effect dominant at
about 100 microns. Fig. IV. 10 shows this very important feature of transporta-
tion.This necessity for a greater mean velocity to entrain material than that
needed to carry the load is sometimes termed the Hjulstrom effect. Thus in
streams where velocities fall to a rate at which particles begin to settle out,
such velocities are inadequate to dislodge particles of a similar size on the
channel floor. On beaches cusp growth depends mainly on the fact that
velocities of flow must be higher to entrain than to carry load. In a gross
A
1
11
( ( )
meires
Verticat exagger-aiion X 2
Mass wasting
Woxing s/ope
Downwarpmg w»lh ci-acKing
of crosl
and ti&suring
Percolation through >

Springsapping and Free face or
washing out scarp
Slumping under gravity of blocks

Accession ot^~
weathering rocK
Into talus l5^»0vpt»NjPhystcal dismlegratton
Surface mantle '^^-•wJlJ^TSwChemical attrition
Hard oUnlhile Weathering ta'iujs r-ivfi'jsi

Altered and weathered
sandstone
Ferruginous sandstone Non-mottled (erraililic

with mudstone bands sandy clay on waning
and lentides slope or pediment
statistical sense the diminution of particle sizes coarser than fine silts away
from the bed follows a logarithmic law of decrease in both air and in streams.
In breakers, as far as sampling difficulties permit, it seems that there is a
gradation upwards from coarse sizes and larger volumes near the bottom. It is
not easy to observe other effects. Fines cannot settle out within the length of
laboratory flumes or tunnels and the observation of the movement of coarse
material in natural situations is difficult and hazardous. There is also the
problem of currents moving at different velocities and operating in different

directions, like hot desert winds. On a beach, transport up a beach in turbu-
lent swash may by suspension, whereas that down the

tend to be essentially
beach is by sliding and Thus material with a high settling velocity
rolling.
tends to move downbeach while that of low settling velocities moves up a

beach. A further complication is that particles of the same mean-size category
but of different shape may differ by a factor of 2 or more in their settling
velocities. However, hypotheses are as yet inadequate to explain the transport
of debris involving a wide range of particle sizes under the wide variety of
conditions found in nature.
Bedload is usually coarser than 200 microns. Typical of the bedload of
major streams is the Ob at Barnaul of which 70 per cent falls in the 200-500-
micron range. Although there is some sliding in surface creep of bedload
material the movement is largely that of rolling particularly for particles

larger than 500 microns because the increase of current velocity upward from
the bed. by exerting greater force on the upper half of a pebble, tends to cause
rotation. Pebbles with one long dimension will tend to roll into a position with
Figure IV.9 Illustrations of difficulties in interpreting hillslopes as erosional

forms. (2) The influence of a distinctive lithology, the problem of homologies,
indications of the operation of more than one process, and the chemical recom-
bination of weathered material.
A. The influence of the lithological characteristics of a massive sandstone on a

slope in the Javorniby Mountains in the Moravian Carpathians (J. Demek, J966.
Geographia Polonica, Fol. 10).
B. A slope south-south-west of Tahat. typical of many cases in the Atakor moun-

tains in the central Sahara (from Rognon, 1967). The distribution of blocks and
the stepped profile are similar to that observed in areas of arctic mollisol, but in
this case may be due to block-glide on silty soils during very wet phases of
climate.
C. A slope-profile in Glen Etive, Argyllshire, showing the genera! shape attributed

to glacial overdeepening, but with its midslope third showing a constant declivity
suggesting debris control at the scree-repose angle (from Pitty, 1969).
D. (from R. P. Moss, 1965, Jour. Soil Sci., Vol.

16). The diagram shows how
slope processes in south-west Nigeria become closely interrelated with the
presence of a ferricrete (plinthite) crust 0-7-2.0 m thick
and its mode of retreat.
Relative concentration
0 9 05 01 0-01 O 001
1 I \ I [ ! i n
60 60 40 20 00 -20 -40 phi units
Figiiie IV. 10 Approximate relations between flow velocity, grain size and state of
sedittientmovement (afiei Siindboig, 1967) The sediment is uniform with a
density of 2 65 gm/cm^; the velocities, critical for erosion or for cessation of
bedload movement, refer to a level 1 mabove the flume bottom; the relative
concentration of suspended load is the ratio between the concentrations at half the
water depth and at a level close to the bottom
this axis perpendicular to the current. Currents may raise discoidal pebbles
on edge and roll them along like wheels, although when packed into a beach
deposit discoidal fragments may shuffle along as a downbeach member of a
pack is displaced. Usually bedload material travelling at about half the velo-
city of the floodwater moves in a series of discrete steps, involving distances
of the order of 100-grain diameters for particles of average sphericity. In
traction, spherical and rod-shaped particles tend to move faster than the
discoidal. However, due to its relatively slow settling velocity, discoidal
shingle is transported farther and piles up at the base of cliffs near the summit
of beaches whereas spheroidal ones remain seaward, as confirmed by
numerous field observations. The ease with which currents roll a pebble may
depend on whether the relative sizes of adjacent particles support the pebble
or leave it exposed to fluid-dynamic forces. With a wide range in particle size
the larger particles may roll readily over a surface made up of finer particles.
Once a coarse particle stops rollingon a bed of fines, up-current scour starts
immediately and the particle settles in the scour pocket. The deeper the depth
of burial of larger particles the more immovable they become, whereas the
largest particles will remain in movable positions for longer periods of time.
Therefore there are certain situations where the transport of the larger par-
ticles remaining as bedload, perhaps surprisingly, is favoured. For example,
the selective removal of coarser grains from beaches might be attributed to

A. Disc migration on a gravel beach (from Stuck, 1967). By displacing a few of

the discs at the seaward end of the column, and with the others conforming to this
new dip, the gravel discs move past the stationary points a and b.
B. Pebble tilting on its upstream side to adjust to the groove made by undercutting
{from S. Sengupta, 1966, Jour. Sed. Petrol.. Vol. 36).
C. Ridges formed by ice-push {from Ktipsch, 1962).
large diameters projecting upward from the bedload into higher seaward
velocities. While surfaces of fines may removal of exposed large
facilitate the
particles, dominantly pebbly surfaces tend to discourage sand accumulations
because of the increased trajectories of rebounding grains. With wind trans-
port. grains too large to be lifted may be impelled forward
by smaller
saltating grains which may move stationary particles
up to six times their
diameter or two hundred times their weight by high-velocity impact. In
water,
the impact momentum of a descending grain is sufficient to raise a surface
grain only by a very small fraction of a diameter. The degree to which the
threshold drag velocity is exceeded has not been studied,
but for the Blue
Creek, California, at least, the value appears to be equalled
or exceeded per 5
cent of the time for the coarse bed material in
this mountain stream. A further
point about bedload is the maximum size that might be moved by the
occasional extreme flood. Wolman and Eiler record a 3-m-deep flood travelling
at a mean velocity of 2 7 m/sec, with a peak possibly of 4 5-5 5 m/sec, moving
• • •
a boulder 2-7 x 1-5 x 1-2 m. A mean current velocity of 7-21 m/sec bed-
load velocity has been known to move a 3-m-diameter boulder. From calcula-
tions for the Truckee River near the Caiifornia-Nevada border. Pleistocene
flood velocities, apparently sufficient to move boulders up to 12 x 6 x 3 m in
size, might have been 9 m/sec on a 0-007 grade. Bedload carried by excep-
tional melt-water floods released from the amount of the Nisqually glacier in
1955 included boulders greater than l-8m in diameter. Floods in the

Chiapas river in southern Mexico can carry boulders at least 2-4 m in
diameter. In Colombia, sudden violent floods entrain much coarse material

including boulders as much as l-2m in diameter, and in general, tropical
mountain torrents can carry a boulder a metre in diameter.

McGee introduced the term saltation in 1908 to describe the manner in
which particles appear to be lifted from the bed at a steep angle, are
accelerated forwards by the current and then settle back on a descending
gradient which is much less steep. In water, the apparent lift may be a few
particle diameters and most effective for sizesbetween 500 and 62 microns
(1—4 o). whereas in air it may be hundreds if not thousands of grain
diameters. Particles somewhat larger than 125 microns may attain heights
of 20 m during saltation, and grains up to 2 mm in diameter have been found
wedged into cracks in wooden power line poles in semi-desert areas at heights up
to 6 m. In the saltation of particles in water it is difficult to dissociate apparent
particle lift from turbulent movements of the water itself. In plan, however,
the paths of saltating grains are straight lines parallel to the general current
direction, and differ significantly from those suspended by turbulent eddies.
In airborne grains, it is much clearer how the energy of forward motion of

wind is converted into an upward one by impact with the floor. This is also
seen on steep slopes where saltation of rockfall fragments may be important.
In this case a snow or ice cover facilitates motion and the occurrence of de-
pressions caused by this process suggests its relative frequency. On a 200-m-
high scree in Insigsuin fiord. Baffin Island, beneath a rockwall several
hundreds of metres high, there is evidence of free-falling blocks arriving with
considerable impact approximately two-thirds of the way up the scree. G. K.
Gilbert (1914) noted that the particles in saltation in streams occupy a
definite sheet like zone above the bed. Inter-particle collisions maintain the
density of the traction carpet as few rebounding or lifted grains escape
collision to attain a height in proportion to their initial rebounding energy.
This traction carpet may have a transitional status between the bedload,
always in contact with the bottom, and the particles carried along well above
the bed in a state of suspension. In wind-blown material 50 per cent of that
observed by R. P. Sharp (1964) travelled at heights from 7-5-15 cm above
the ground. A
similar smooth-topped traction carpet accompanies backwash
on beaches. However, the depth of this layer may only equal a few grain
diameters outside the high energy zones. On the Santa Monica beach,
California, a measurement of the traction carpet depth is 2-5 mm, compared
with 0-5 mm on La Jolla beach. Also in contrast to the saltating motion of
unidirectional currents, the oscillatory motion of waves tends alternately to
lift and to deposit material as an intermittently suspended load.

Finer particles up to a certain size limit, which is related to the amount of
turbulent energy available to support the sediments, may be maintained in
more or less permanent suspension by the turbulent exchange of particle-rich
fluid from near the bottom with fluid from higher levels less dense in particles.
Suspension in water affects sizes less than 140 microns (2-80) and particu-
larly sizes less than 62 microns. R. A. Bagnold (1937) observed that the
velocity of upward air movements near the ground were on average approxi-
mately one-fifth that of the mean forward velocity of a wind. In rivers the
material in suspension is normally almost completely independent of that in a
traction carpet and turbulent exchanges within the suspended load are
balanced. In some circumstances it is possible, however, that water moving
upward may contain a greater silt content than that moving downward. Finer
sizes which can be held indefinitely in colloidal suspension with the very fine-
grained particles kept dispersed by ionic-electrostatic forces are sometimes

termed the washload. Unlike the suspended load the washload is relatively
unrelated to the hydraulic conditions and their changes. Occasionally, but
from a range of environments, come reports of material being transported by
flotation. In 1926 J.S. Behre estimated that in an hour the River Ohio
transported 130 000 kg of material by flotation.
There is no realistic definition for critical velocities just sufficient to
suspend the sediment. Instead there is continual transition as progressively
more particles of a given size are entrained. The fluid velocity theoretically
required to dislodge grains of a given diameter is sometimes termed the
threshold drag velocity. The velocity for moderately efficient suspension of
sand-sized particles might be 2-2-5 times the threshold drag velocity. On
A. Sundborg’s (1967) modification of Hjulstrom’s well-known illustration of the
orders of size and velocities involved (fig. IV. 10), four different areas may be
noted. The area of high velocities and small particle sizes describes conditions
of net scour. The bottom left part of the diagram represents conditions of
transport in suspension, but of net deposition of bedload. The grey shaded
area indicates the condition of entrainment of coarser sizes from
the bottom
and the transport of bedload while, with lower velocities, the unshaded area
indicates the absence of
prolonged transport and the deposition of bedload
and possibly suspended load too. Recently it has been shown
that the critical
factor is not the average velocity of flow but the bottom shear velocity. In
consequence, the height of the column of water above the bedload is critical
and in deeper water at normal flows a larger average velocity is needed to
transport particles of a given grain size than in shallow water. For this reason
water flow in a beach swash zone is more competent. That the usefulness of
fig. IV. 10 is notional should be stressed also because it applies to uniform-
sized material only. Where a mixture of particle sizes or discoidal shape are
involved there is as yet little precise theory or observation available. Tentatively

A. Jopling (1966) suggests that velocities might have to be increased by
perhaps 20 per cent for poorly sorted sediments. Another complication is that
for a given flow past a fixed point neither the fluid velocity nor the shear stress
which varies approximately as the square of the velocity are constant, but
fluctuate considerably due to turbulence and momentary increases in velocity.
Inevitably contrasts are far more pronounced on shores where there is a high,
momentary velocity under the crests of waves which are about to break, and
under breakers in association with the development of vortices, which can lift
particles far above the bottom. Another related factor which again applies
indirectly to a beach environment, is the concept of the abruptness of flood.
Experiments in Russia with surges released down laboratory streams revealed
intense dragging and suspending of sediment associated with the passage of a
surge. It seems that in an accelerating current, velocity at the bottom
approaches velocity at the surface. In a natural situation in small Scottish
streams in high flood gravel disturbance appears to be associated with smaller

sharp increases of velocity than with the slower rises of larger floods.
Where surface water flows over a ground surface instead of concentrating
in channels some additional factors have to be considered. Often the entrain-
ment of particles is the result of rain splash which in badland gullies may raise
fine sands and silts by at least 5 cm. In Turkmenia, comparable heights of 30
cm have been observed involving lateral displacements of 40-50 cm. but

where vegetation covers more than 60 per cent of the area, there is practically
no remov'al of solid material. Experiments have shown that raindrop splash
may carry a higher percentage of large particles and aggregates than surface
flow. Particle sizes in the 50-400-micron range are most susceptible to
movement. Coarser sizes are too large to be dislodged and smaller sizes seal
the surface so that a water film on the surface dissipates the energy of
raindrop impact. Sheetflow or sheetwash involve shifting rills a couple of
centimetres to half a metre in depth, dying out with distance from the hill-
front. On fans, a maximum water depth of 15 cm has been observed, with

channels continually filling up with debris before shifting to another line. The
process is essentially a transportational one. moving loose material of sizes up
to coarse sand. In all situations experiencing sheetwash there are extensive
areas unprotected by a permanent vegetation cover and where there is a
seasonal or sporadic supply of water. Rainwash, however, may be important
on the scantily covered forest floor in tropical forests as rainwater pouring
down tree-trunks is sufficiently concentrated to expose roots downslope.
Piping is a transportational process which produces tubular sub-surface

by the grain-by-grain removal in
drainage channels in insoluble clastic rocks
suspension by moving ground water which opens into free drainage. Broad
subsidence may follow. This essentially mechanical process termed piping is
particularly well-developed beneath weathering crusts where cavities several

metres or even 00 m in length may develop. The process occurs in granite
1
areas in Hong Kong. In dry environments the coarser material that fills
desiccation cracks favours the development of sub-surface seepage lines. In
cool environments the transportation of fine material in blockfields may take
place below the ground. On beaches fine materials may move through a
framework of coarser material.
The critical general factor in the movement of particles by wind is a
limitation on the amount of effective precipitation rather than wind velocities
although the limits of 17-5 km/hr needed to suspend dry sand and the
inability of wind to turn particles larger than 5-7 cm are clearly defined.
Otherwise, in middle and low latitudes vegetation may provide insufficient
cover on sands in continental interiors with precipitation as high as 650
mm/year. Areas of active longitudinal dunes in south-west United States and
in central Australia are limited approximately by the 250-mm isohyet. In
arctic deserts the necessary surface dryness for sand to be movable may be
realized only when precipitation is less than 75 mm/year.
2. Amounts of sediment transport
Apart from the mechanisms of transport, the amounts moved are another
vitally significant aspect of landform studies. In streams there are the concen-
tration of the suspended load, the percentage contribution of the bedload and
the net amounts removed to consider. Sediment concentration or turbidity is
important because it gives a clear indication of the rates of supply of detritus

to thechannel system. Exceptionally high concentrations arise momentarily,
due to mass movements or to an intense phase of slopewash at certain
vulnerable points in a catchment. High concentrations occur frequently in
semi-arid areas where supply may include a large proportion of fine-grained
wind-transported material which fails to cross a perennial stream or the
depression of an ephemeral stream channel. In these circumstances amounts
may exceptionally reach 50 per cent or 500 000 mg/1. In contrast in winter in
high latitudes, measured concentrations may be less than 1 mg/1. As examples
of intermediate values
some results from northern Sweden might be noted.
Upstream from the Laitaure delta the highest concentration is 1260 mg/L,
and near the water surface at bankfuil stage is about 950 mg/I. Nearby the
highest concentrations are of the order of 3 10 mg/1, in the Kanajokk and 275
mg/l. in the Tarraadno rivers. The Rhine, upstream from Lake Constance,
with an average flow of 224 m^/sec has an average sediment concentration of
approximately 825 mg/l. In the relatively flat European U.S.S.R., the turbidity
of the Polomet river, with a catchment area of 63 1 km^ at approximately 250
mg/l is higher than most rivers. Here values 7000 mg/l in high may rise to
water due essentially to bank erosion and channel reworking. Of the larger
streams, the mean turbidity of the Ob, at approximately 500 mg/l, is some-
what low. In the Negev where discharges range from 2 to 1000 mVsec
turbidities observed range from 60 to 680 mg/l.
A relatively unknown factor in the calculation of amounts transported is

the contribution of bedload movement which is usually estimated from for-
mulae. Russian scientists have estimated that of the sediment discharged from
continents, 9 per cent is bedload. From mountainous areas the proportion is
1
1
per cent, from plains 8 per cent. Individual examples may fall within or
outside this framework, usually for reasonably clear reasons. In the lower
Mississippi and the Amazon bedload is only 5 per cent. An estimate for the
Congo is 6 per cent. For the Rhone near Villeneuve. the proportion is 12 per
cent, for the Linth near Walensee, 22 per cent. In a drainage basin including
the most ice-covered high mountain area in Sweden, the proportion is 14 per
cent. Fifty-three kilometres downstream in Rapaalven, bedload amounts to
about 16 per cent of the suspended-load discharge. One feature of coarse bed-
load is the relative slowness of travel. J. Tricart observed that during the 1957
flood on the Guil, unprecedented in postglacial times, 10-cm pebbles rarely
travelled more than 1-2 km. Within a rocky gorge in Herault province subject
to violent floods he estimates an average progression of 0-2 km/year. In wind
transport about 25 per cent of the load moves by surface creep, perhaps re-
flecting the fact that if sizes equivalent to the washload of streams were initially
present they could soon be swept a few kilometres above the ground.
G. K. Gilbert defined ‘capacity’ as the maximum load that a stream can
carry. However, a natural stream cannot be saturated with sediment as it
might be with a salt. Again it is the wide range of particle sizes that may be
carried which introduces a complication as does the range of transport
mechanisms. Thus a stream unable to move a boulder a few kilograms in
weight may be competent to shift thousands of tons of finer silt, which in the
case of semi-arid floods, may exceed one-quarter of the total weight of the
moving suspension. Another problem is that concentrations, if exceeding 5
per cent, begin to exert profound changes on the hydraulic laws that apply for
pure liquids. In fact the increased density of the fluid mass means that the
relative density of particles diminishes, thus facilitating transportation. Also,
shearing forces on the bed increase. The submerged weight of a rock in a

mudflow reduced by perhaps more than 60 per cent compared with its
is
submerged weight in water. Thus, although there must be a theoretical limit to

the amount of material a stream can transport in suspension, the practical
limit is nearly always by a deficiency in the supply of suitable material
set
available for transportation. By far the most geomorphologically significant

statement on the amount of stream load is its actual measurement. Nonethe-
less. the significance of such a measurement in landform interpretation may
be limited, as it is impossible to infer the relative importance of processes

which dislodged or eroded the material at the sediment source area within the
catchment, the degree to which the sediment is being reworked or evacuated
from the river network and the degree to which human activity has accel-
erated the previous rate. As these complications are added to the diverse
resultsproduced by local factors, only the broadest outline is given in Table
IV.7. Examples of more specific local observations might include the trans-
portation of loessial areas in China, equivalent to an overall surface lowering
of 7-9 mm/year. In the Italian provinces of Emilia, Romania, and Marci,
comparable figures are 0-2— 1-4 mm/year. Reflecting in part the fact that the
erosiveness of the same amount of runoff might be one or two dozen times
that of humid subtropics, in the arid sub-tropics the theoretical lowering in
Central Asian mountain drainage basins range from 1 to 6 mm/year in the
south to 0-003 mm/year in the north.

On beaches, even the slightest current will cause some horizontal displace-
ment during the gravitational fall of suspended sediment. Thus in the breaker
zone where particles move to and fro under the waves, they are laterally
displaced from the net direction of this current by even the slowest longshore
current. Within the inner breaker zone the oblique uprush and more perpen-
dicular downwash will lead to particles including pebbles following the well-
known zig-zag pattern of longshore drift. The velocities of these movements
on the California coast average about 30 cm/sec with a maximum recorded of
1-3 m/sec. Individual grains move at 2 -4-3 -6 m/min on Californian beaches,
at 1- 1-2-0 Wmin on average on Florida beaches, and at 0-60-0-75 m/min
as a maximum rate on Sandy Hook foreshore. New Jersey. There are reports
of grain movements at 0-15 m/min on Baltic coast beaches, rising to 1-65
m/min under wind forces of 2-6. On the shore of Ikroavik Lake, Alaska,
littoral drift may reach 0-30 m/sec during 30 km/hr winds,
moving pea-sized
gravels at a rate of 1-28 m/min. Velocities of 0-8- 1-0 m/sec are strong
enough to divertmost sands alongshore and even moderate longshore velo-
cities are sufficient to deter offshore movement. Thus
volumes of alongshore
transport are substantial. At Vridi, south of Abidjan. Ivory
Coast, the average
annual displacement from artificial obstructions is 800 000 Along the ml
coast of British Guiana longshore drift transports
150-200 million m. tons
annually as suspended sediment and in vast migrating
mudbanks. In the Baltic
Table IV 7 Transported load and turbiditj of selected rivers ijiam h I oiiinici, 1969,
Bull Inst Ass Sci Hydrol , Vol 14)
For each order of drainage basin si/e the lowest L the median M and the highest H
values have been selected trom Fournier s data The number ot rivers in each basin size is
given m brackets The next to the lowest value in each group is on average 2 7 times the
lowest for load and 2 8 times the lowest lor turbidity The highest value m each group
IS on average 1 7 times the next to the highest for load and 2 7 times lor turbidity
ORDER OF SIZE LOAD

OF DRAINAGE AREA tons/
BASIV KM RI\ ER COUNTRY LOCAl.IT\ KM KM
10-100 L Mangami New Zealand Tanki Road 82 8
(1!) M Wanganui New Zealand headwaters 80 50

H Manaatepopo New Zealand Katetahi 15 5 727
100-1000 L Thames Canada Ingersoll 518 15
(37) M Neveri Venezuela Batalon 976 301

H Marccchia Italy Pietracuia 357 4 570
1000-10 000 L Pembina Canada Windy gates 7 800 4
(50) M tiler W Germany Krugzcll 1 1 18 292

H Waipioa New Zealand Kanakanaia 1 580 6 983
10 000-100 000 L Meuse Belgium Hedel 29 000 24
(22) M Tone Japan Matsudo 12 000 273
and Toride
H Yesilirmak Turkey Ay vacik 36 000 1 228
100 000-1 m L Oder Poland Gozdow ice 109 400 1
(18) M Waal Belgium Huihuizen 160 000 13
H Chao Phva Thailand Nakorn Sauau 103 470 106
Turbidity
mg/1
10-100 L Melito Italy Olivclia 41 23
(9) M Mam W Germany Marktbreit 27 74
H Eleuterio Italy Risalaimi 79 971

100-1000 L Thames Canada Ingersoll 518 23
(35) M Tronto Italy Tolicnano di 911 1 350

Marino
H Hii Japan Nadabun 924 28 619
1000-10 000 L Yodo Japan Hirakata 7 120 23
(51) M Me Wang Thailand Mew Lorn 2 708 559

H Maticora Venezuela Don Pancho 2 490 23 500
10 000-100 000 L Meuse Belgium Hedel 29 000 68
(20) M Me Ping Thailand Wang Kra 26 386 283
Chao
H Yesilirmak Turkey Ay vacik 36 000 4 893
100 000-1 m L Oder Poland Gozdow ice 109 400 9
(18) M Saskatchewan Canada Saskatoon 140 000 149
H Euphrates Svria Tabqa 120650 1 709

an annual transport of 1-9 million m. tons/year moved from south to north

past Klaypeda. On the Californian coast figures range from 76 500 mVyear
to 215 000 mVyear at Santa Barbara.

In contrast to the very turbulent swash there is very little turbulence in the
downwash, with the exception of rip currents which transport a lot of sand
beyond the breaker zone by their stream-like flows. Similarly there is not
usually much movement seaward of the breaker zone. Off the East Anglian
coast tracer movement suggests a drift of perhaps 6-4 km/year with a peak
km/hr where a sand stream moves from the north end of the Norfolk
rate of 3
sand banks 160 km north as far as Flamborough Head and spreads across
100 km. At the southern corner of the North Sea, sediment from the English
Channel enters at a rate of 600 mVyear. In contrast the current through the
Bering Strait is perhaps unusual in being the main source of sediment in the
area. Such slow offshore currents are of direct geomorphological significance
where the sea-floor material drifts shoreward, as from depths of 10 m or less
itcan be moved on to the beach. Sand moving on to Mediterranean beaches in
France comes from depths down to 9 m. Where foreshore bars develop,
deposition on the landward face may favour an onshore movement. On the
Georgia coast, the rate of this movement is generally about 10-30 cm/day.
In coastal areas, dune sands more than 100 m above sea-level illustrate
qualitatively the transporting some environments. Measure-

power of wind in
ments include those of H. J. L. Beadnel! (1910) who observed a year’s dune

advance at the Kharga Oasis, the highest dunes (20 m) moving at 10-9
m/year, with the smaller ones (4- 10 m) moving at 18-4-1 8-8 m/year. For
coastal sand dunes of Guerrero Negro, Baja California, mean travel velocity
of dunes is 4-9 cm/day or 18 m/year. Nearby, inland from Bahia Sebastian
km across a lagoon in
Vizcaino, a dune field has travelled approximately 12
the last 1800 Between 1956 and 1963 the Salton dunes in Imperial
years.
Valley moved 14-40 m/year, compared with 7-25 m/year in the preceding
15 years. However, in observations over 12 years of the Kelso transverse
dunes, R. P. Sharp while observing a maximum rate of movement of 15
cm/day found that, despite great activity, this involved so much shifting back
and forth that net displacements were small indicating 30-45 cm/year over
the 12-year period. In southern Peru, there is some reversal throughout the
year in the mornings and during the four winter months this reversal pre-
dominates Jn south-west France, the 100-m high Pyla dune advances from
an
eroding shore at a rate of 1 m/year. The volumes involved include
measure-
ments of 41 kg/hr/m for the Kharga oasis. In January in southern
Peru,
barchan advance involved transport of 79 kg/hr/m, and for the
Kelso sand
ripples R. P. Sharp estimated 89 kg/hr/m under
a 50 km/hr wind.
The degree to which very strong winds on isolated rocky
summits are
effective in reworking material has not been widely investigated,
although a
lowering of 2—3 m on the summit of Mount Nussbaum, Taylor Valley,

Antarctica, has been attributed to deflation. It is also difficult to see how the
possibility of the action of wind in moving sediments downhill could be
evaluated, although the importance of this factor is obviously related to the
high transportation rates of semi-arid streams.
The significance of wind transportation of finer silts can usually only be
inferred from deposits. From the amounts of fine dust added to the land
surface on desert or high dry plateau margins the significance of the process
appears appreciable. In 1863, 4 million m^ of dust fell on the Canary Islands,
westward of the Sahara. In the vast expanses of Turkmenia, wind accounts for
60 per cent of sediment transport, compared with 20 per cent by water and
20 per cent for the combined action of wind and water. In the periglacial
environment wind, as a transportational agent, is often more significant than
rillwash, particularly along the margins of arctic outwash plains. Loess
deposits in Alaska following the last (Kluane) glaciation are 35-100 cm
thick.
If it has so far proved impracticable to measure actual amounts of bedload
transport by streams, it follows that similar measurements for a glacier would
be even more difficult, G. Ostrem (1965) measured the amount of englacial
material along a 40-m stretch of the Isfalls glacier snout, but the delivery of
1-3 m. tons/year appears negligible compared with 100 000 m. tons in the
same 40 m of the adjacent end moraine. Transport on a glacier surface may

occur where lateral moraines of valley glaciers lead on to the ice surface as
medial moraines below glacier junctions; examples on the Siward glacier, on
the east coast of Baffin Island are about 1 2 m high. However, in general, it
seems that dominant bottom transport should be assumed for ice, and that the
competence of glaciers is undoubtedly due to the compactness of ice and.
compared with water, its relative resistance to compression. The huge size of
some ice-moved boulders testifies to the efficiency of ice as a transportational
process. One of the largest erratics in Scotland, on the Arran coast near
Corrie Burn, is 3 m^ at its base. 3-7 m
high and weighs approximately 400
tons. Ice may up reversed gradients even in some
also transport material
terminal zones. Erratics were carried obliquely up the southern slopes of the
Kilsyth hills to 210 m above their outcrop within a distance of 3 km. As far
south as the Illinois Ozarks, the Illinoian glacier, after radiating from the
Labrador dispersion area, surmounted obstacles 90 m high. Another testi-
mony to the efficiency of ice transport is the considerable distance over which
some large erratic blocks have been carried. The mechanisms in continental
ice-caps are even more obscure than valley and piedmont glaciers. However,
from the results of recent investigations in Antarctica, it appears that some
basal transport occurs, but that the rates could be unimportant. The ice
bottom 5 km down at Byrd Station contains a lot of debris, including beds of

siltand sand and larger fragments up to 5 cm across. Erratics suggest a
distance of transport of 360 km.

In valley glaciers where sub-glacial water exists it is difficult to apportion
the transportation achieved between the ice and the sub-glacial stream. The
volume of material in outwash plains compared with the size of moraines
suggests the dominance of fluvioglacial transport in some areas. They owe
part of their efficiency to flow under hydrostatic pressure. Subglacial streams
entering the Slims river fountain 2-4 m
into the air with maximum reported
heights of 6 m. Around the openings are large accumulations of well-rounded

boulders 0-6-0-9 m in diameter, suggesting the efficient removal of all sizes
finer than this.
In coastal areas drift ice can transport pebbles of considerable size. An ice-
front on disintegration becomes a powerful transportation agent as beach

material frozen to the underside is rafted away. Drifting ice seems to be the
main agent of transport of the large quantity of coarse sediments found on the
muddy flats along the St Lawrence.
Ice can also be an agent in deforming land surfaces over which it passes,
causing movements en masse but without entraining the material. As these

deformations have some resemblance to those of gravity mass movements and
because original sedimentary structures remain essentially intact, these ice-
drag or ice-push features are perhaps appropriately discussed together with

other aspects of glacier transport rather than in the context of glacial erosion.
Because of the preservation of sedimentary structures, ice-pushed ridges
appear to depend on the existence of a thick permafrost tract in front of the
advancing ice-sheet. After postglacial thawing the pushed-up blocks remain
largely intact due to the internal friction of their constituent materials. Similar
ice-pushed ridges are a feature of frozen shores. They are common along the
north Alaskan coast, but the ridges, 0-6-4 -5 m high, probably amount to
only 1-2 per cent of the sediment above sea-level. In some places ice-pushed
boulders are so regularly arranged along the shore that the impression is that
of artificial features modified by erosion. A wall of turf and stones pressed up
by ice is a common feature around many Swedish lakes. In ice-drag features
movement may take place on gliding surfaces often parallel to the bedding.
An example studied in some detail in the north side of Beaver Creek Valley,
near Elkton. Ohio, is a rift 10-5—18 m
m wide and 9 deep cutting across a
bulbous spur. This appears to have been pulled about 15
rift apart by the m
passage of ice, filled with glacial debris, then capped with sandstone and shale
displaced 45 m by a second ice-drag mass movement.
Another transportation mechanism in which ice is probably involved is the
rock glacier. The tendency for rock glaciers to be composed
of equiangular
debris which provides for the greatest interstitial accumulation
of ice, without
itself possessing any tendency towards
mobility may add to the reasons for
supposing that rock glaciers owe their mobility to interstitial ice. The present-
day activity of some rock glaciers is evident where they override trees or
saplings.
Apart from ice-covers, ice-drag, and
interstitial ice, underlying ice, where it
exists, may have some on transportation. In Antarctica, ice-cored

influence
talus aprons have been observed below cliffs on which the movement and the
sorting of the debris has been attributed to vibrations caused by the cracking
of the underlying ice, induced by rapid temperature changes accompanying
the passage of shadows across the aprons.
The transportation of material by moving snow may be significant in
colder areas if snow accumulations are appreciable, particularly if the move-
ment is that of a slush avalanche. The site for these movements is the upper
parts of slopes where a collecting basin exists. In winter ice may block the
gully and hold back wetsnow and water which on release in the spring thaw
may carry much debris in the mass of water-saturated snow and melt water.
In northern Sweden a boulder 5 x 3 x 2 m in size and 75 tons in weight was
moved 120 m down a slope of only 5 degrees in this manner. Total volumes
moved may be as much as 2000-4000 m^ slush avalanches moving ten times
the total load of other avalanches. However, avalanches and snowblock falls
from hanging cliffs are important in clearing away smaller sizes of rock
waste.
3. Mass movement
There is a range of mechanisms involved in transporting material down slope
gradients in which the amounts of water involved become progressively less.
At one extreme is the mudflow, its liquidity and its channel floor position
making it transitional between muddy stream flow and fluid mass movements.
The water content involved in soilflow tends to be less than the theoretical
liquid limit with debris yielding to continuous plastic deformation above a
shear plane. Soil creep may operate in unconsolidated materials on slopes less
than approximately 35 degrees. Materials again tend to be drier than the
theoretical liquid limit but, unlike soilflow. the irregular motion of loosened
individual particles or aggregates, characteristic of soil creep, does not
involve a basal shear plane. Landslips describe mass movements of solid
material over a sliding plane with perhaps some changes in structures but
without continuous deformation. Finally falls, at the opposite end of the scale
to the mudflow, involve the free fall of dislodged solids alone. The distinc-
tiveness of mass movements is not, however, invariably closely related to water

contents as gravity is often the overriding factor. It is not easy to categorize
types of movements and it may be misleading to specify distinctions too
rigorously. The susceptibility of a slope surface to mass movement and the

landforms which result are influenced by a wide range of factors continually
changing in space and time and even within the same slip area. Geological
factors which influence the shape of shear or tension fractures may play little
part in the form assumed downslope by the transported material. The momen-
tum and effects downslope of mass movement depend not merely on the
geological conditions favouring its release but also on the infiltration capacity
of the overwhelmed ground. Surface weathering conditions range from the

frost and snow melt of cooler environments to the deep chemical weathering
in humid tropical areas. The depth of decomposition, the weathered clays
adding to their weight with their capacity to absorb the large quantities of
water and expanding in the process, the abrupt contact between weathered
soiland subjacent rock providing an ideal sliding plane, all make the humid
tropics the environment in which mass movements are most characteristic.
Mudflows are fluid masses moving in surges down stream channels. They
may develop from steep scantily vegetated slopes made up of unconsolidated
material containing enough clay to make the mass slippery if wetted by
abundant water received in a short period of time. The pressure of water in
pore spaces also reduces intergranular friction. The weight of water is a
critical factor also, because flow starts when the weight component parallel to
the slope exceeds the internal cohesion of the mud and the friction at its base.
At some point a between a heavily loaded stream and a mudflow
distinction
becomes The water content may be 20-60 per cent by weight or by
arbitrary.
volume, and densities may be as high as 2-0-2-4. The high density of mud
exerts forces on stream bedload substantially different from those of running
water and quantities of channel alluvium are incorporated into the fluid mass
as it passes. An example of a larger area of mudflows covers an area of more
than 40 km^ in Chile, at 27° S on the southern slope of the Carro Cadillal.
Mudflows also occur near the south Crimean coast. In 1949 when 178 mm of
precipitation fell in 12 hours, a mud- and rockflow involving at least 1-5
million m^ of material moved down the Uchau-Su river valley into the sea
within a day. Some mudflows are so viscous that they come to a halt in the
stream channel.
On slopes, soilflow or earthflow depend
on the water content of the
less
mineral material and more on gravity as the major factor, and are common on
declivities between 5 and 30 degrees. In the tropics, much steeper slopes are
often involved. The movement is essentially one of plastic deformation, but

initially, either at the beginning of seasonal melting or following an abrupt
failure, flow may be more affected by liquidity. The shortness of the lobes
which develop downslope are due to the loss of water by drainage and
infiltration which eliminates any element of liquidity in the
flow mechanism
and tends to extinguish rapid movements. Where sufficient volumes
of debris
are involved, boulders several decimetres in diameter
can be moved.
Many workers use the term solifluction non-genetically to describe the

viscous flow of soil under saturated conditions, but as the term was first used
in the context of an arctic environment, others imply soilflow only where
freeze-thaw conditions prevail or even only where a permanently frozen sub-
soil exists.
The similarity between soilflow in the humid tropics and in periglacial
environments arises because the moisture changes in the top 10-15 cm of the
clayey tropical soils produce the effects of alternating expansion and contrac-
tion which also accompanies freeze-thaw activity. In damper periglacial areas
solifluction is the main agent of transport favoured where permafrost is

present as this provides an impermeable layer below the seasonally thawed
soil which easily becomes waterlogged. Where frost penetrates to depths
sufficient to induce large-scale solifluction mean annual air temperatures are

not usually above PC. Material in the lobes is often only 0-25-1 -5 m thick
on steeper 20-25 -degree slopes. Measurements of solifluction lobe movement
include rates of 2-6 cm/year in Alaska, 4-5 cm/year in Spitzbergen. and on a
gentle 5-degree slope in Sweden. 0-9-3 -5 cm/year. On steeper slopes rates
increase as shown by S. Rudberg’s measurement of a 12-cm/year rate in
northern Sweden in 1958 on a 20-degree declivity. In an adjacent locality A.
Rapp (1960) found that the centre of lobes moved at 25-30 cm/year but that
the sides moved only at 4-7 cm/year. Also, at times of initial release, move-
ments may be more rapid. In an area of active retreat of supersaturated frozen
ground in the Mackenzie delta area, the enlargement of active scarps at 1-5-
4-5 m/year may give some indication of such rates. In many areas recog-
nizable solifluction features are no longer active. Measurements on Mount
Northcote in the Snowy Mountains of Australia, made 60-220 m below the
summit, while indicating a rate of 0-4 cm/year until 1540 ± 160 years B.P.,
suggest little subsequent movement.
Soil creep is another slope process encountered in most latitudes occurring
on slopes as gentle as 2-4 degrees with again more than one mechanism
producing similar effects. In cooler damp conditions, freezing temperatures,
although they do not penetrate sufficiently to produce a solid mass of ice,
convert interstitial water in superficial soil layers into segregated ice crystals
overnight. These crystals of needle ice, called pipkrakes, develop below
individual particles or aggregates of soil which are better conductors than the
surrounding soil. On a slope the subsequent melting of the segregated ice
crystals and the vertical drop of the raised soil particles involves a slight
downward displacement. On a bare 55 per cent slope near Wenatchee, where
annual precipitation of 525 mm falls mainly as snow, 25 per cent of particles
labelled with a radioactive tracer moved 0-7-5 cm, 36-41 per cent moved
7-5—15 cm. and 20 per cent 15—30 cm during a winter spell. Frost creep was
the main process involved. In the Tatra Mountains. Poland. T. Gerlach
Inter-relationships between processes and land/arms 219
(1959) has recorded particle displacements of 2-6 cm in one freeze-thaw

cycle. A critical factor, the height of pipkrakes, varies with local conditions,
diminishing as diurnal changes through freezing point at the soil surface
become less frequent. In mountainous areas in central Germany, pipkrakes
may be 10-15 cm long, Schmidt (1955) calculated that weights of
and J.
material lifted in the Black Forest and the Taunus areas were 650-
7200gm/ml The weight of individual particles lifted is also appreciable.
Pipkrakes 3-4 cm long have been observed on the Chambarrau plateau.
Records show pebbles 0-1-2 kg in weight lifted 1-5-3 cm and that one
9-5 kg fragment was raised 1 cm.

Displacement of particles in accumulations of steeply inclined coarse
material is also pronounced. In the English Lakeland hills, marked stones
moved 12-15 cm during a winter period. In Spitzbergen stones in areas of
stone stripes moved 3-4 cm/year on 6-18-degree slopes, and on 3-6-degree
slopes by 1-2 cm/year. In Sweden there are records of coarse material mov-
ing at 0-2-1 -6 cm/year.
As environments become warmer, and where moisture supplies are
adequate, organisms become increasingly important in soil creep processes.
Examples of the scale of disturbance in superficial soil layers by organisms
were given in the preceding chapter. Rates may range between 0- 1 and 10-0
mm/year.
Movements discussed so far relate to those at the surface or involving
superficial layers. In landslides entire sections of slopes are removed to depths
of several metres. The most common type of landsliding is the slump in which
a long slice of cap-rock or surface material separates from a scarp face and,
without being greatly deformed, moves downslope on a curved slip surface
(fig. IV. 12. A). The strike of the slump blocks is controlled by the trend of the
The backward tilt of the rotating block’s ground surface creates a

cliff.
depression at the headward end of the tilted block which becomes marshy or
the site of a pond. As in solifluction a crucial factor may be the intake of large
volumes of water, which by displacing air from the pore spaces and building
up pore-water pressure, imparts buoyancy to constituent materials if these are
sufficiently loosened or weathered. In 1952 when there was the largest and
most spectacular of landslides in the upper Columbia river valley, springs in
the area dried up 10 days before the failure occurred. Perhaps initially,
therefore, some minor displacement, by damming back these underground
springs, triggered off the much larger movement. Where a massive cap-rock
overlies less competent strata rotational movements may also occur. How-
ever. some blocks glide along intact, tipped outward away from their source,
particularly where cambering precedes mass movements. These blocks then
load contiguous segments of the slide paths. In some instances the detached
blocks are displaced vertically without tilting,accompanied by lateral
4. Transportation of dissolved material
In higher latitudes there is little dissolved material transported in streams

partly because much of the small quantities initially taken into solution is
captured by organisms. In the tropics the synthesis of clay minerals means

that in the transportation of weathered products the suspended load may be
three to four times that of the dissolved load, even though the initial weather-
ing processes were largely solutional. In mountainous areas mechanical trans-
port removes up to seven times the quantities carried in solution. By contrast
in many areas the amount of material transported in solution approaches that
of mechanical transport, as in much of Europe, and in some lowland areas
transport in solution may account for nearly twice the amount removed by
mechanical transport.
The scantiness of information on how the dissolved load is transported,
compared with the detailed attention to mechanical transportational pro-
cesses. is clearly out of all proportion to its relative significance. In fact this
imbalance may have placed one of the greatest limitations on progress in the
understanding of landform development. Its significance relates not simply to
the appreciable if not preponderant quantity of materials removed in solution
but to the fact that transport by solution is as rapid as the rate of waterflow
and is effective on and beneath all land surfaces where water moves, however
low the gradient on the surface or the degree to which gradients are reversed
beneath land and water surfaces.
The main element transported in solution is calcium which, in bicarbonate
form, is chemically stable over times involved in river transport although it
may be reduced by the activity of organisms in slow-moving reaches. Next

may come either silica or iron, their proportions varying according to how
climatic conditions influence weathering processes and the other substances
present in the solution. As an example of the latter circumstance, the solu-
bility of silica decreases appreciably when alumina is also in solution, posing
an obstacle for large-scale transport of kaolinite in solution. K. B. Krauskopf
(1956) found that little or no silica was removed when kaolinite, montmoril-
lonite, calcite. or FcjOj were added to silica solutions made up in distilled
water. The mode of transportation of iron is also problematic. As ground

water enters a stream, oxygen from the atmosphere oxides iron bicarbonate to
ferric oxide. Because of low solubility of the oxide, a greater part precipitates
from solution as agel, but the remainder is protected by organic colloids or in
the form of iron-organic compounds. For the most part these gels join the
suspended material and are evacuated from the drainage basin. Because of
either hj'drolysis. adsorption, or both by alumina ions, the transportation of
significant amounts of this element even through a weathering zone may be

possible only in the form of complex ions or as protected sols. Stream
sediments may also alYect the chemical composition of the dissolved load.
Some minerals will dissolve, some may cause precipitation of certain dis-
solved ions, and because of their exchange capacity, others may stabilize the
chemical composition of the dissolved load. In the eastern United States,

where the washtoad clays less than 4 microns in size include kaolinite, illite.
and vermiculite. their exchange capacity is less than 20 m.e.g./lOO gm. In
western states, montmorillonite increases and in consequence so does the
exchange capacity. For washload clays in the Crooked river, near Post,
Oregon, the figure is 63 m.e.g./lOO gm. It is significant that during periods of
very high concentrations of suspended sediment, the ratio of cations adsorbed
on colloidal sediments to those in solution probably reaches a maximum. Of
the maximum flows observed in the United States, the smallest ratio of
adsorbed cations to cations in solution. 01, was on the Juniata river at
Newport, Pennsylvania and the highest on the Rio Grande, in New Mexico,
where there were three times as many cations adsorbed on the suspended
sediments as there were in solution. Intermediate ratios at maximum flow
include 0-2 for the Crooked river at Post in Oregon, and for two rivers well-
known in geomorphological literature, the Brandywine Creek, Delaware,
ratio was also 0-2, and for the Green river at Munfordville, Kentucky. 0-3.
These and other figures suggest that during periods of very high sediment
concentration, cations adsorbed on suspended sediment may approach or
even exceed the cations carried in solution. This conclusion is of profound
significance for geomorphological studies as it shows that measurements of
transported solid and dissolved loads in such circumstances could give the
impression that the importance of mechanical weathering in relation to solu-
tional weathering was three times its actual significance. There is also the risk
of further exaggeration because there is also the possibility that organic
Table IV.8 Average figures for mechanical and chemical denudation (from N. M.
Strakhov. 196 7. Principles of lithogenesis) based on calculations by G. V. Lopatin.
DISCHARGE OFDISCHARGE OF INTENSITY OF

DEViVJ DAT ION. toms/km-
AREA.
CONTINENT 10"KM- 10" tons/year 10" tons/year Mechanical Chemical
Europe 9-7 420 305 43 32

Asia 44-9 7445 1916 166 42
Africa 29-8 1395 757 47 25
N. and Central
America 20-4 1503 809 73 40
South America 18-0 1676 993 93 55
Australia 8-0 257 88 32 1 1
matter may be responsible for an appreciable proportion of the exchange

capacity of stream sediments. Under average to low flow conditions,
suspended sediment concentrations are usually very low, and with the ratio of
adsorbed cations to cations in solution being negligibly low and probably less
than 0-001 there is no chance of confusion in equating solid/dissolved load
ratios with the relative importance of mechanical and solutional weathering.

some general overall calculations of the amount of material
In addition to
removed shown in Table IV. 8. some specific examples are worth
in solution
mentioning. The loss in the main rivers of the northern half of European
U.S.S.R involves 21 million m. tons of calcium and 15 million m. tons of
magnesium.
C. Deposition
Modes of deposition and origins of depositional landforms are often contro-

versial issues. One of the problems, apart from the practical problem of
observing the process in operation, is that usually more than one factor is
involved, either simultaneously in one environment or in effecting similar

results in different environments. For instance, some boulder clays laid down
on land are sometimes difficult to distinguish from glaciomarine drift
deposited from floating ice. Trough-shaped sets of cross-strata are produced

in different media in different environments, the products of nearshore
marine, beach, river, and eolian processes wherever the depositing currents
flow primarily in one direction.
The most widely applicable generalization is that deposition follows a
decline in the competency of a transporting agency. In a stream this is when

the velocity which dislodged a particle of a given size slackens by about one-
third. There is a time-lag between the moment at which a decelerating
current can no longer suspend a particle and the moment at which this
particle reaches the bottom. This settling lag is an important factor in inter-
preting the accumulation of coarser material on beaches. Deposition may

initially be self-enhancing because the deposition of the coarser particles
increases bottom roughness. This, in slowing the stream down, leads to the
deposition of less coarse material. This mechanism may. however, become
self-arresting with the deposition of sizes which are sufficiently small to
increase the bottom smoothess. However, over a large size range, from 70
microns to 30 mm at least, the balance of forces acting on a particle does not
change in such a way as to cause a major change in the mechanism of particle
selection during deposition. Once deposited, a particle may remain in a given
position for a time that could range from a few seconds to tens of millions of
years.
Decrease in competence may not be constant throughout a transporting
medium. Deposition takes place in areas that are beneath, or to the side of.
some thread of maximum velocity and its flanking threads of maximum

turbulence. Enough material may accumulate at a given place to give rise to
a distinctive form, starting with two particles coming to rest so that one lean-
ing on the upstream side of the second is itself in part blocked and in part
shields the downstream particle from the full force of the current. As this
is a self-accentuating process, a traction clog may develop which could

eventually become a central bar or even an island in a braided stream (flg.
IV.15.A).
It is possible to recognize three types of depositional landform. A form
resulting essentially from deposition in the ‘past’, where the definition of the
past reflects a subjective preference on the part of an observer, is only one of

these. The description of the characteristics of this static form, if the deposi-
tional material remains little unaltered, is relatively straightforward. A second
type occurs where a given volume of material moving more or less contin-
ually. the definition of time-spans again involving a subjective decision, may

assume a distinctive form during its some deposi-
transportation. Thirdly
tional forms depend essentially on a quasi-equilibrium between amounts of
material removed and continued renewals in supply. In this case the amount
of material passing through the system may be of much greater volume than
that temporarily incorporated in the form.
The purpose of the following discussion is to describe static depositional

landforms and the unchanging characteristics of dynamic depositional land-
forms.
1. Depositional landforms on slopes
The most distinctive depositional forms on slopes are those assumed by

material released and transported downslope by abrupt mass movements. The
chaotic arrangement of several irregular ridges, mounds, and depressions is
characteristic of many landslide areas. The more liquid the moving mass the
more regular and lobe-like the form assumed by the deposited material. On
slopes in cool environments where higher vegetation is absent, material mov-
ing slowly downslope may assume distinctive small terrace shapes, fronted by
a bank of turf, or of coarse debris (fig. IV. 13). In general, the form assumed
by solifluction deposits is that of a smooth sheet, usually 0-5-1 m deep on

slopes and thinning on summits but often thickening to several metres
in hollows and valley bottoms. The term ‘head’ describes such crudely strati-
fied debris produced by solifluction. The extent of these deposits is much
greater in many areas than was formerly believed. J. M. Ragg and J. S. Bibby
(1966) have observed the stony mantle in the higher parts of the Southern
Uplands. The rubble layer, where examined, is well sorted and has a uni-
Figure IV. 13 Slope deposits.
A and B. (A) Earth-flow deposit {from Zaruba and Mench, 1969), one of the
largest in Czechoslovakia, produced in 1960 near Handlova. The movement
involved debris of volcanic rocks and clayey and silty sediments of Sarmation
(Neogene) age which had already been entrained downslope by previous move-
ments. Factors involved were probably changes in clay consistency by periglacial
climatic effects and also possibly the squeezing out of a plastic substratum by the
load of the overlying volcanic sheets. (A) Longitudinal section. (B) Cross-section
near the base of the tongue of debris.
Figure 1V.14 Alluvium deposited on a talus slope (from Nichols, 1960). Alluvium
on a slope is often distinguished by the term colluvium if it has been transported
by slope processes.
modal peak. The mode of the surface rubble layer is generally less than
32 mm whereas the coarse modal peak in the deeper layer is greater than 32
mm. The subjacent layers are often bimodal and contain considerably more
fines in the coarse silt to fine sand range (31—250 microns). There is a
coating of sand and on the upper surfaces of stones whereas these lower
silt
surfaces are clean and often rest on coarser material and nests of small stones.
The superimposition of soil layers on hillslopes commonly results in stone
fabric differences between surface and buried layers. The arrangement of rock
fragments in the overlying layer may be chaotic whereas undisturbed bedding
planes of the parent rock may show through the buried layer. In central
Poland, rhythmically bedded slope deposits, due to the rillwash, dominant
during phases when ground ice melted, were formed during the climax of the
last cold stage.
Where free fall of individual particles occurs, screes may accumulate. In
cool and arctic environments, as on Nevy may be a
Island, Antarctica, screes
hundred or more metres high. Average angles for entire screes are usually
between 32 and 36 degrees. Tinkler (1966) found that 40 per cent of mea-
surements made on limestone screes in the Eglwyseg valley near Llangollen,
North Wales, were 35 degrees. Other measurements on some steep-fronted
moraines and on the lee-side face of sand-dunes have averaged 34 degrees.
C. Turf-fronted solifiuction terraces on Slaetharatindur, the Faeroes (from C. A.

Lewis and G.M. Lass, 1965, Geog, Jour., Vol. 131). Species of Rhacomitrium
form the riser of these terraces: both are at 400 m; (i) south-facing, and (ii) north-
facing.
D. A stone-banked terrace on Niwot Ridge, Colorado Rocky Mountains (from

Benedict. 1966). The terrace is at 1065 m on a 10-15-degree slope.
Angles steeper than 3 8 degrees are not usually maintained for more than a
few metres. An example at the other extreme is the declivity of seree slopes in
the vicinity of the Kalagarh landslip. Uttar Pradesh, where quartzite blocks
accumulate at an average angle of 31 degrees. One factor influencing the
steepness of screes may be the tendency, as observed in experiments by van
Burkalow (1945), for more angular material to bank up at slightly steeper
gradients. Another major influence is the shape in plan of the supporting
valley wall, with screes banked up more steeply where contours are concave
in plan. Many screes, like those on Palaeozoic rocks in parts of Snowdonia,
which average 34 degrees, are slightly concave in profile.
2. Depositional landforms in fluvial environments
Although they are of comparatively rare occurrence, there are reasons for
discussing mudflows as the first depositional form in fluvial channels. Mud-
flows have abrupt, well-defined margins, are convex in profile and decrease
regularly in thickness downvalley or down-fan. At the toe there may be a
radial arrangement of surface ridges composed of rounded boulders. Mudflow
material is poorly sorted due to the presence of large boulders and abundant
smaller pebbles embedded in the silt and clay matrix. Occasionally debris
includes organic material, which, because of its light weight, would be carried
away in water flows. Some indistinct stratification in beds 1-5-IOm thick,
sometimes outlined by layers of water-laid material, may be visible in large
exposures. A fluid mudflow shows graded bedding with the larger rock frag-
ments sometimes orientated in vertical as %vell as in other positions. Apart
from the surface form of the mudflow and its valley-floor position there is
often little to distinguish the material from boulder clay.

The shape of an alluvial fan is a result of deposition by a stream that swings
back and forth over the accumulating material but is fixed at the apex by the
position of its bedrock valley upstream. Deposition may result in some cases
from a decrease in slope where a stream enters the apex of the fan, but more
generally might be related to a decrease in depth and velocity of flow as a
stream spreads out downvalley from the confines of its headwater valley. In
the initial growth of a fan a powerful self-enhancing factor is reduction of
surface runoff by infiltration into the porous accumulation of debris. There
are two types of water-laid sediments on alluvial fans, for reasons somewhat
similar to those which explain the contrast between river channel and flood-
plain deposits. On fans most of the water-laid sediments consist of sheets of
sand and silt deposited by a network of braided streams which are rarely
more than 30 cm deep. These sands may be well-sorted, with a quartile
deviation of 1-6 phi units being reported from fans in western Fresno county,
and contrast with the interstitial sand and gravel deposited in the more
permanent beds of the main stream channels. Efficiency of sorting and some
rudimentary stratification may increase down the fan. Fig. IV.28E and F
illustrates the downslope decrease in size which accompanies an increase in
pebble wear and the decrease in gradient. On some fans there may be a
contrast between mudflow deposits descending from the fan head and water-
sorted material deposited nearer the toe.
Fans are widespread and are most common in arid and semi-arid areas of
the world. About one-fifth of California is covered by alluvial-fan deposits.
Coarse gravel fans on the western front of the Black Mountains in Death
Valley vary from a few hundred metres to 1 or 2 km in radius with inclina-
tions ranging from 3 to 25 degrees. The steepest fan in the Tucson area is
about 10 degrees. On a strip of coalescing fans 20-30 km wide in the foothills

of the San Sanquin valley the inclination of fans decline from 1 :35 to 1:530.
Of much gentler inclination than valley-fed fans are those which impercept-
ibly extend downslope the surface inclination of gravel- or sand-veneered
pediments. These alluvial fills, made up of confluent fans of poorly sorted clay,
silt and sand, known as bahadas, may thicken downslope to depths of 5-60
m. Their declivities in the Tucson area vary from 0- 5 to 2 degrees. In the Ajo
region, declivities are about 0-5 degrees. Fans are not confined to semi-arid
areas. Coalescing fans, for instance, form a tilted plain on the west side of the
Mackenzie delta in north-west Canada. Here the average frost-free period is
only 66 days but there are annually twenty -five freeze-thaw cycles of a 34°-
28° F amplitude which might be an important factor in the disintegration of
bedrock. Two fans that have been studied in detail are of similar dimensions
being 2-8 km long and about 4-0 km wide at the base. The inclination of the
first fan starts at 1 degree declining to 0-6 degrees at the toe; the second from
3 degrees to 1 • 1 degree. There are sands and pebbles at the apex, passing
down into material less than 2 mm at the toe where silts predominate.
Faulting is often an important factor leading to sedimentation of a fan surface.
Downstream from the steeper headwater reaches where fans may be
present small but significant features are the traction clogs which may form
spool-bars. These diminish in frequency downstream as the critical concen-
tration of bedload particles for jamming to occur is seldom reached.
Farther downvalleys a distinctive feature of deposition on the level flood-
plains is that, even over spans of thousands of years, building up the flood-
plain may be a progressive process. It is difficult to decide whether long-
continued accretion is a response lagging behind postglacial sea-level rise, or
is due to man-accelerated erosion or whether some other factor is involved.
In part at least continuous alluviation appears to resultfrom deposition
of sediment trapped by vegetation at occasional high flows. Parts of
the
Nile floodplain build up at a rate of 10 cm/ 100 years. An average
of 45
cm/100 years overbank accumulation seems possible in the Chemung valley.
a tributary of the Susquehanna. In the Little Missouri valley, where a flood-

water height over 6 m may occur once every 50 years, floodplain build-up may
be by 1-2 m/100 years. In the Tesuque valley, which rises in the Sangre de
Cristo Range in north-central New
Mexico, deposition averages between 19
and 29 cm/ 100 years. In Eastern Europe, parts of the Danube floodplain are
4-5 m above average low water, and comparable figures for the Dneister.
Southern Bug. and Ingul are 3-0 m. 12-0 m. and 21-0 m. In humid tropical
areas the clays decanted into hollows may build up rapidly, particularly in the
stagnant water of mangrove swamp areas where accretion rates may be several
centimetres a year. In cooler areas, floodplains like the Russian plain are in
part relict features produced during thousands of years of flooding by fiuvio-
glacial melt water from a receding ice-sheet and settling in huge temporary
backwater lakes at the edge of the ice. In fact, in higher latitudes and in areas
near the margin of former ice-front, alluvial plains are described as outwash
plains. Where they are confined as a valley train their scale may make these
flat depositional features one of the most striking landform characteristics of
these areas. Scandinavian workers use the term sandur to describe these
depositional plains flooded by melt water and distinguish between plain and
valley sandurs. In Sweden, one of the thickest Quaternary deposits fills the
Petsaure outwash plain to depths exceeding 100 m. Even as far south as the
headwaters of the Mississippi, valley fills of 60 m or more are outwash
gravels, relics of their former marginal position to an ice-sheet. Similarly, in
the Forth valley, south-east of Stirling, there are extensive near-level spreads
of gravel which become very coarse towards the north-west upvalley end. As
this surface leads upvalley into areas of ice-contact forms and has a down-
valley gradient of 5-5 m/km. which is ten times that for raised beaches with
which it merges. J. B. Sissons ( 1 963) recognizes a relict glacial outwash plain.
Outwash plain material is generally coarse sand and gravel, contrasting with
floodplain fine silts and clays, except where fine silts, settling in lakes, escape
wind transport. In some former lake basins, as in Osterdal in central Norway,
an infill of lacustine silts and sands may eventually be covered by coarser
outwash material.
Another characteristic of the floodplain environment in areas where flow is
irregular is the scale of accretion that may follow the extreme flood. In 1965
in eastern Colorado, when most of a 200-300 mm downpour fell in an hour
and the subsequent flood was nearly twice volume of the previously tlie
recorded maximum flow, the floodplain sands deposited within hours were
many hundreds of metres in width and up to at least 3-6 m in thickness.
In a sinuous channel, deposition takes place on the inside of the river
curves of material coarser than that of the flanking depositional plain. The
ridges, or point bars, that form, consist mainly of fine-grained, well-sorted
sands with interbedded silts. Their mode of deposition is not well understood,
1
although it is possibly related to a Hjulstrom effect where the slackening flood

velocities, at which deposition begins with traction clogs building up, is
insufflcient to entertain material of the same size. Gravelly sands, with coarse
fractions ranging from 2 to 1 1 mm in diameter were found in deeper parts of
the bar shown in flg. IV.15.D. where trough cross-stratification is charac-
teristic. Each trough-shaped set consists of an erosional scour, its long axis
parallel to the local stream direction, 0-5-1 m wide and filled with scoop-
shaped layers less than 0-3 m deep. Horizontally laminated sand, although
not abundant, is found in layers 0-3-1 m thick at various locations and at
different levels in the point-bar deposit. Also trough cross-bedding is not

characteristic of all point bars. From the study of a 320-m-long point bar on a
Mississippi meander where finer materials in silt and silty clay sizes ranges
are inevitably involved, the channel on the toe of the slip-off slope appeared to
be related to a portion of the current returning upstream rather than to a
thread of slackening floodwater crossing the toe of the slip-off slope. As point
bars grow through accretion of sediment outward from the bank and as river
undercutting leads to continuous shifts in the position of the outer side of a

river meander, a series of point bars, or scroll, may be stranded on the slip-off
slope. With higher floods than those which can form point bars, overbank
flows may form levees due to the accumulation of sediment on top of the
channel banks. These asymmetrical ridges are often lense-shaped in cross-
section and may decrease in size downstream. The crest is close to the
channel because the quantity and calibre of sediment deposited is normally
greatest near the channel and decreases in a lateral direction (fig. IV.15.E). In
general, levee deposits are on the whole finer than most channel deposits and
may contain plant remains. Their structures show evidence of their sub-aerial
exposure during the greater part of the year, particularly disturbances due to
organic activity, and may be reddish brown in colour due to oxidation.
One final aspect of those forms and features, due wholly to deposition from
rivers, involves chemical precipitations. The floodplain and its sediments,
particularly if these include colloids, form a geochemical barrier to the migra-
tion of chemical elements which does not exist higher upstream. Of those
most commonly found in the natural environment the Al, Fe, and Si com-
pounds are the least soluble and are the first to be precipitated when the water
moves. As temperatures become sufficient to induce evapotranspiration losses
from the floodplain water-table, deposition of these compounds becomes
increasingly marked. The soils of the Vikhra river floodplain, for instance,
contain five to eight times more iron and five times more manganese than do
soils of the adjacent watershed. The deposition of chemical elements in flood-
plains or swampy ground adjacent to streams is an important factor in
landform study as a whole and of crucial significance in humid tropical
environments. However, it is probably only the deposition of carbonates
5
B
Horizontal ood
vertical scote
Approx t tS
Q )eel ^
Q metres
Mudllow deposits Sand pebbly sond Grovel layers Gravels

unstrotilied granules strotified ono lerises imbncoted
Figure IV. 1 Fluvial deposition forms and their sedimentary structures.
A. Formation of spool-bars and spool-like islands in sandur streams (from

Krigstrdm, 1962). From left to right the diagrams indicate - the embryo (C-D)
and hypothetical growth of a spool-bar; an almost submerged spool-bar; the same
spool-bar at a lower water level; and a spool-bar transformed into a spool-like
island.
Figure IV. 16 The Garcia cave, north-west of Monterey, Mexico, two-thirds filled
by stalagmite {after H. Enjalberi. 1968, Mem. et Doc. Centre Rech. Doc. Cartog.
Geog., Ed. C. N. R. S., Vol. 4).
which assumes distinctive forms. Organisms like algae, by extracting carbon
dioxide from the water or from the dissolved bicarbonate, may create large
volumes of calcareous tufa which forms dams across streams and smooths the
outline of waterfalls. Calcareous material may also be deposited in floodplain
soils. 60-m section of a stream channel below a spring in Virginia, with a
In a
flow rate of 3901/min approximately 2900 kg of calcium carbonate were
deposited annually as tufa. Along the Kaap escarpment, north-west of
Kimberley, there are remarkable accumulations of travertine.
3. Deltas
While some transported material may remain at the base of slopes for long
periods, much of that entering rivers may similarly accumulate at the river
mouths where deltas form. Approximately half the amount of the material
B. Sedimentary structures in an alluvial fan {from Blissenbach, 1954). Main

direction of transport from right to left; based on fan deposits in the San Catalina
Mountains, Arizona,
C and D. Point-bars. Because of the convex cross profile on the spur of the ‘slip-
off’ slope(C below), currents at rising water are deflected towards the inner bank
(from Krigstrom, 1962). (D) shows the sedimentary structures of the Beene point
bar, on the Red River near Shreveport, Louisiana {from J. C. Harms et al.,
1963,
Jour.-Geol., Vol. 71).
E. Form and structure of flood deposits in the channel of a small
mountain stream
{from H. Stewart and V. C. LeMarch, 1967, U.S. Geol. Surv. Prof. Paper 422-
J.
K). The deposits are the product of a catastrophic

100- year interval flood in
Coffee Creek in Trinity County, Northern California.
transported to the Yangtze-Kiang or Nile deltas is deposited there. However,

the loss from the clay fraction may be substantial. A delta is a partly sub-
aerial deposit built by a river when spreading of flow into a body of little-
disturbed water checks its velocity. Postglacial rise of sea-level has accen-
tuated this braking effect. However, a small tidal range is not a pre-requisite
for deltaic accumulation, as a broad shallow area can diminish wave action
sufficiently. Thus the Ganges and Irrawaddy deltas continue to advance into
coastal areas where the respective tidal ranges are 4-5 and 5-5 m. In general,
delta shorelines continue to prograde as long as the supply of sediment,
mainly from the river, exceeds removal by wave and current action, particu-
larly in shallow seas. The Volga delta’s spectacular advance is by 170 m/year.
The huge volume of the Mississippi’s load is a major factor contributing to
the 400 m/year mean growth of the south-west part of its delta. The Orinoco
delta progrades by 200 m/year. Less striking are figures of 4 and 4-5 m/year
for the River Meander delta in western Turkey and the Gulf of Thailand
respectively, but which still represent huge rates of change in the context of
Examples of intermediate rates of advance include 10 m/year for
millennia.
the Don. 20 m/year for the Po. and 27 m/year for the Kilia delta of the
Danube. Extrapolation of rates of accumulation suggest that the bulk of
deltaic series is the product of postglacial times. For instance, the Krikkjokk
delta in Sweden could have accumulated in 7200 years if the rate of growth
for the last 80 years had been constant over this period. Conversely aban-
doned portions of a delta mouth may be cut back. Since the outflow through
theRhone delta has been through the Grand Rhone, the banks of the Petit
Rhone have receded by 2 km since the seventeenth century. In relation to its
mode of accumulation the delta is the last downstream in a series of subaerial
lobate depositional forms produced by the shifting and splitting of streams
supplied from a relatively fixed source. However, the possibility of shifts in
the supply point, restrictions imposed by the position of adjacent uplands,
makes the plan of deltas the least regularized of the fan-shaped depositional
forms. The slope of the shelf offshore is a further diversifying factor and it
would be as realistic to enumerate the unique characteristics of individual

deltas as to search for generalisations. The Orinoco resembles the Mississippi
delta in the predominance of fine sediments and similarity of facies types.
However, the processes of delta growth and distribution of facies types are
fundamentally different. Whereas the Mississippi delta advances by accretion

of bar and channel sands and by levee construction at the radiating distribu-
tary months, a strong marine current deflects material from Orinoco distribu-
taries abruptly north-westward and the delta advances by littoral accretion.
Turbulence of Atlantic swells is another factor preventing sub-delta develop-

ment. In some situations delta growth is spasmodic rather than continuous;
half the total annual growth may take place during a single week in the
-
1 Subaerial Subaaueous
deltaic plain deltaic plain
Lower Upper i Lower Della
alluvial volley deltoic plQin deltoic plain fringe ProdellQ marine
Meander belt Channel sond, noturol 1

Brackish woter clay Sand Richly orgonic cloy
sand, back -
|
levee silt, fresh woter ond silt, distnbutory silt, ond silty cloy
loke ond swomp cloy
,
swomp cloy ond tidol chonnel plant

morsh peat SQnd« peat debris
sea
level level
[
Gravel
Sand
S — 1
Cloy
Corbonoceous
^ 1 1 • 1 matter
Sill and silty cloy 1

1
Peal
Longtiludinal section
Coastal Coastal
deltaic sediments sediments
interdeltaic sediments interdeltaic
Figure IV. 17 Section through a deltaic mass of sediments {after E. H. Rainwater

in Shirley and Ragsdale. 1966).
Laitaure delta in northern Sweden. One general rule, however, is that the
thickest section of a delta tends to be in the proximal pro-delta region where
silt and clay are dominant. Landward, seaward, and coastward, the deltaic
mass tends to thin. Also the grain size of the bed material tends to decrease
slowly downstream in active deltaic channels. In the lowest sub-aerial parts of
the Godavari delta, India, sizes decrease by about I phi unit/25 km. There are
lateral and vertical gradational changes and sharply delimited scour and fill
structures. Organic matter may accumulate in stretches of stagnant water and

water-logged depressions which form substantial portions of the area of the
sub-aerial deltas. The Ila river delta, which widens to 210 km at its outer edge
along the Lake Balkash shoreline, includes 12 per cent of open water in its
total 9000 km^ The principal structural features of coarse-grained deltas

differ considerably from those of fine-grained deltas. Where bedload is
carried into the delta area this material gives rise to more rapid changes in
deltaic channel pattern and if it reaches the delta front may be deposited as
foreset beds. In fine-grained deltas, where accumutions are essentially
deposited from the suspended load, distinctive sets do not develop. There are
also contrasts in the average inclination of the sub-aerial parts of a delta,
those on small coarse-grained deltas being rather steep, up to several metres
per kilometre and overlapping with the order of gradients on bahadas.
On large fine-grained deltas the inclination is much lower, of the order of
5 cm/lOOm. Also the emerging bars of bedload origin are less numerous
in the fine-grained delta, so the degree of braiding is low.
4. Estuaries
On some coasts tides and other factors like river velocity may combine to
prevent large-scale deposition in river mouths where many well-known
estuaries funnel out to the sea. Within estuaries the electrolytic effect of
sodium chloride in sea-water flocculates the clay suspended in fresh water.
The neutralized clay colloids by drawing together into larger aggregates settle
faster than individual particles. Velocities greater than 0-7m/sec tend to
break up aggregates but at velocities less than 0-28 m/sec flocculation is
rapid. It is primarily the rising tide which carries this material, first, up the
channels of the estuarine fiats, then like a river at bankfull stage, spreading out
on to the bordering fiats, which may include parts of deltas and river flood-
plains as well as estuarine salt marshes. Due to the Hjulstrom effect the
receding ebb tide does not normally have velocities sufficient to dislodge the
deposited clay aggregates and rates of accretion may be appreciable. In the
Wash and other shores of the North Sea and Baltic, like the north German
coast, accretion may be up to 3 m/year. Accretion of the substantial clay
fractions in the loads of tropical rivers may also be rapid. Following the
deposition of a sandy marine horizon, homogeneous clay layers some 20 cm
thick have been observed to form within weeks on the east coast of Malaya.
Indeed, tidal inlets are among the most variable and mutable of land surface
features. Of the features created by estuarine deposition, the marsh or swamps
may have a negligible expression in cross-section but introduce profound and
rapid modifications to the shape in plan of the river mouth and adjaeent eoast.
The shores of the Wash embayment on the east coast of England where tidal
range is 6-7 m in spring tides and 3-4 m during neap tides, provide a good
example of a salt marsh environment in mid-latitudes. The gradual decrease in
the velocity of tidal currents as they move over inter-tidal flats causes a
reduction in their competence and results in size differentiation of the

deposited materials which, because of the Hjulstrom effect and the braking
and trapping action of vegetation, tends to accumulate. However, the ways in
which ebb of water draining back down the salt marsh creeks might contri-
bute to a redistribution of fines are not clearly understood, partly because of

the hazards of field work in such areas. The marsh in the Wash is flooded
only at high water spring tides when silty clays, clayey silts and small
amounts of sand are laid down in well-defined laminae. Seaward are the
higher mudflats which in turn give way to broad sand flats, a small cliff a few
inches high often demarcating the transition. Lower down the sand flat bur-
rowing organisms like Arenicola marina destroy initial stratification, but the
sand surface is usually a distinctive pattern of straight-crested ripples, of 0-5-
1 cm 0 amplitude and spaced some centimetres apart. In section the ripples
may be rounded, truncated by wave action, or remain sharp. A fourth zone,
the lower mudflats, coincides approximately with the more steeply sloping
part of the intertidal zone, and occupying a flatter zone between this and the
low-water mark is a fifth zone, the lower sand flats. Here material is the
coarsest in the area and megaripples are 25-35 cm high and spaced 0-5-
3 5 • m apart. In tropical areas or in cool or cold latitudes, estuaries are not
well-developed; this is possibly related in part to absence of extreme flood
discharges in tropical areas and unfavourable conditions for vegetation
growth and clay weathering in cooler environments. In the latter context an
unusual feature characteristic of the St Lawrence salt marshes is the abund-
ance of boulders 0-5-2-5 m in diameter deposited from floating ice and
forming ridges or pavements on the mud surface. These areas are also pitted
with numerous irregular hollows 0-5-3 m in diameter and 25-60 cm deep.
These are due to lifting of grass rafts frozen on to the underside of ice blocks.
5. Depositional landforms along coasts
Although it is difficult to ignore the fact that the shore is a highly dynamic
environment with the continual shifts, readjustments, and replacements of
beach materials scarcely separable from the unceasing motion of the waves,
certain aspects are sufficiently unchanging to be described as for those of
static depositional landforms. One such feature is the highly characteristic
‘step’ which occurs in the surf zone close to the mean low-water line, and for
many workers marks the seaward limit of the beach. Spring-sapping by
groundwater draining out of the beach at low water during times of maxi-
mum tidal range is an important agent in steepening the base of the beach
face. Immediately upbeach from the step there is a small convexity but the
cross-section of much of the remainder of the beach is somewhat concave
upward. Backshore beaches do not occur on coasts where a wave-cut cliff
occurs directly upbeach from the foreshore. Frequently the summer berm
forms a noticeable near-level crest just above the upper limit of the zone of
swash and backwash on the beach face, landward of which there may be sand-
dunes.
Despite the continual shifting of beach sands some sedimentary properties
remain relatively unchanging. The step is usuallycharacterized by the
Seawall
High marsh Low marsh with B(i)
Mud flats with
Sand Channel
\ with deep creeks Spartina, Salicornia, mussel bonks
\)f rTryyrvyTiX
Mean level of high tides
Mean level of
Holocene marine
shelf facies mad
Holocene (modern)
aKavium
Holocene littoral
sands
Pre - Transgressive
(pleistocene) alluvium
Figure IV. 19 Profile of a beach and nearshore region {after D. Inman, 1962, in
McGraw-Hill encyclopedia of science and technology).
presence of coarser sediment such as gravel, in rather sharp contrast to sand

above and below, with the best sorting frequently occurring on the beach side
of the surf zone. On Sandy Hook beach. New Jersey, the material at the step
rapidly approaches a maximum diameter in the 4-6-mm range. Here the
gravel is rather well-sorted. Immediately seaward of the step base the sands
are about 0-5 mm in diameter and decrease in size offshore. A narrow range
of grain sizes in the zone of the swash is a characteristic feature on most large
Figure IV. 18 Estuaries and beaches.

A. The shape of the lower Humber estuary and the ebb-flood channel pattern of
the bottom {from A. H. W. Robinson, 1968, Zeit. f. Geomorph., Suppbd. 7). Spurn
Point spit modifies somewhat the funnel shape typical of many estuaries.
B. Section and plan of an estuary bank, Wadden Sea area {after L. M. J. V. Van
Straaten, 1954, in Guilcher, 1958).
C. Cross-section of a Tabascan beach ridge {from Puts'y, 1965). At the seaward

end of the section there is a berm on the foreshore. To the right of the centre
of the
diagram there is landward-dipping foreset stratification with gently inclined com-
ponents merging inland with a more steeply inclined component. At the inland
end
of the section swale stratification with complex near-surface
component overlying
seaward-dipping stratification common to the beach.
D. Cross-section through coastal plain and inner continental

shelf {from J. R.
Curray and D. G. Moore in Van Straaten. 1964) based on observations
along the
regressive Costa de Nayant, Mexico. Successive accretion of submerged longshore
bars creates ridges. During periods of low waves,
a bar is built to the surface and
becomes the new beach, isolating the former beach and creating
a narrow lagoon.
sand beaches, predominantly in the medium- to fine-sand range. This is
because sand of approximately 0-2 mm is generally too fine to roll along the
bottom and too coarse to be kept steadily in suspension. A median figure of
0-165 mm has been suggested as representative of Californian beaches,
0-27 mm from the Scanian coast in the Baltic and a median of 0- IS-
C' 30 mm at Recife. Offshore from the step there only transport and deposi- is
tion of fine particles removed from shallower areas. Upbeach from the top of
the step the grain size decreases rapidly towards the midpoint of the swash
action. The sorting may also become better, but the grain size again increases
before gradually and steadily decreasing to the top of the zone of swash and
backwash where sizes are often least. The occurrence of coarser particles
within the swash-backwash zone is particularly marked on beaches with
plunging breakers where sediments may in fact be coarsest at the plunge point
seaward of the mean-tide level. Immediately above, coarse material on the top
of the summer berm may constitute a secondary size maximum. From this
point sizes diminish towards the dunes.

Although sedimentary characteristics on each part of a beach are distinc-
tive because of particular combinations of factors that control its develop-
ment at these points, the advance or retreat of a beach with time may develop
the essentially static depositional characteristic of sheet-like stratification. For
instance, the seaward inclination of the foreshore beach on Mustang Island
near Corpus Christi, Texas is 2-5 degrees. The inclination of backshore
slopes is as much as 2 degrees inland. Inclination of stratification is 5 degrees
on the upper foreshore, and on the backshore is irregular and characterized

locally by beds inclined at 6-12 degrees.
One of the beach characteristics of most immediate geomorphological
significance is the actual slope of the beach which tends to be steeper the
coarser the median grain size of the beach material. Along the Californian
coast beach, declivities may be up to 10 degrees, commonly are 7-8 degrees
and contrast with Texan beaches where declivities are usually 5 degrees or
less. On the east African coast between Kenya and Tanga, the narrow coral
sand beach at the base of coastal cliffs has moderately steep inclinations
ranging from 1 ; 10 to 1 ; 20.
A second significant form-characteristic of beaches is bars. These include
in a genetic sense, those that may form seaward of the plunge-point of
breakers from the accumulation of coarse material moving landward outside
the plunge-point together with that of sand moving seaward from points
upbeach from the plunge-point. The identification of a sub-aerial ridge as a
former plunge-point bar. however, is very difficult, partly because both the
plunge-point bar and the berm formed at the upper end of the swash zone may
be driven inland by rising water to produce ridges, like the Tabascan beach
ridges which attain heights of 1-2 m above the swales. Spacing varies be-
tween 20 and 90 m. Conversely it is possible to mistake sub-aerial dunes that
have been submerged by a progressive sea-level rise for plunge-point bars.
The interpretation of beach bars or related relict features on the shore is
complicated by landward movement of barrier islands. The development of

any form of bar, barrier or berm is favoured by abundant sediment supplies
from offshore or transferred from rivers mouths by longshore drift, as sug-
gested by G. K. Gilbert in 1885. On broad, gently inclined coastal plains,

where the occasional high-water level is superimposed on a sustained gradual
rise of the relative level of sea in relation to the land, long ridges may develop
parallel to the shore and attain a substantial size. Barriers are best developed
where the tidal range isrelatively low, as on the south-east coast of Australia.
A large tidal range, by generating strong ebb and flow currents, prevents
accumulation in a gap through a barrier. Some elongate barrier islands may
range from a few kilometres to more than 150 km in length, separated from
the mainland by a bay, lagooon, or marsh area; the Pleistocene and Holocene
barrier islands which border the Georgia coast area are 1 1-25 km long and
3-6 km wide and separated from the mainland by 6-10 km of salt marsh. The
predominance of fine-grained horizontally stratified sediments landward of a
barrier may indicate that, provided an origin as a submerged sub-aerial dune
appears unlikely, barrier development was essentially an onshore movement,
whereas lagoonal conditions would develop landward of a barrier extending
longshore only in a comparatively late stage. The main effect of migration on
sediment properties is the modification of stratification. The littoral and
nearshore neritic sediments deposited along the fronts of barrier islands of the
Georgia coast have low depositional slopes, rarely steeper than 6 degrees.
Because of postglacial sea-level rise enormous Pleistocene sand ridges with
smaller, present-day longitudinal dunes in the inter-ridge areas are a feature of
many coasts with gently inclined shelves, particularly in lower latitudes; such
a zone on the Senegal coast is about 92 km wide.
So far this discussion of coastal depositional forms, although stressing the
complications introduced into the interpretations and classifications by sea-
level changes, has not yet included changes alongshore. Deposition on a shore
with rocky headlands often occurs as in many sub-aerial environments, with
a divergence in the flow lines of the transporting agent, reducing its com-
petence. On the shore this occurs as waves enter the relatively concave sectors
of a coastline with most flow lines focused on the headlands by refraction,
and
the form of bay-head beaches, concave-seaward in plan, is well known. With
prevailing winds blowing obliquely onshore, reduction of energy down the
Figure IV.20 Coastal features involved in cuspate spit formation (from Zenkovitch,
1959). A previous position of the spit appears on the left of the diagram, as spits
may be dynamic depositional landforms (see Chapter IV. D.6).
length of a beach usually results in deposition of the longshore drift load.

Where sediment supply is abundant and the declining energy of longshore
drift the dominant process in coastal deposition, and the trend of the coastline
changes, a depositional ridge will continue the updrift trend of the coastline as
a spit, to a point where the depth of water in the bay or estuary is too great for
waves to be effective. The strength of currents in or out of an estuary may
also be a limiting factor. The five examples on the northern shore of the Asov
Sea have lengths of up to 40 km. In areas of active spit formation rates of
advance may be by several metres per year, as mouth of the
at the
Courantyne river on the British Guiana — Surinam border. The spit, exposed
to the same north-west drift that dominates the form of the Orinoco delta,
advances at 62 m/year. North of Benguela, the Lobito spit advances by

20 m/year.
If the end of a spit is limited and if material continues to arrive there by
longshore drift, a hook may form. The direction of the hook will depend on
whether the force acting at an angle to the longshore drift is dominantly
seaward or onshore. Usually, refraction at their distal ends produces a re-
curve which doubles back towards the coast. On the northernmost tip of the
United States a spit extends 8 km north-eastwards from Barrow and then
hooks sharply and extends south-east for 5 km. In width it varies from about
10 m to almost 1050 m at Point Barrow, it is 4-5 m high and 1000-4000
years old. If a recurve joins the coast, the landform, now triangular in plan, is
termed a cuspate foreland, which recurves may continue to join on its down-
Figure IV. 21 Representative reef structures. Ten Thousand Islands, Florida (from
D. E. Shier, 1969, Bull, geol. Soc. Am., Vol. 80). (i)The reef has rapidly expanded
to cover a shoal area during a period of
little or no sea-level rise, (ii) The reef has
upward and expanded slowly in area during a period of sea-level rise, (iii)
built
Growth on the periphery of this large reef has kept pace with sea-level rise and
increased the area of the reef. Poor water circulation and the settling of fine
sediment have prevented growth in the central part of the reef.
drift side. The enclosing Lake Bumbe on the east shore of Lake Albert
tip
turns through 10 degrees. In Dungeness the highest levels consist of uni-

1
modal pebbles. 1-1-5 m below the surface the closely packed pebble frame,
with larger particles 7-5-40 mm
in size, is filled with sands. Bedding planes
dip constantly 8-10 degrees in the direction of inclination of the shore face at
the time of deposition. Advance by accretion on the eastern side may be about
0-75 m/year.
In the formation of reefs, chemical and biochemical processes
produce one
of the most distinctive depositionai forms of the marine
environment. They
are found between 25^ N and 25°
S. particularly in the Pacific and Indian
Oceans. Shapes which may
be altered by structural weaknesses range from
circular to very irregular or ribbon-like. A ring-shaped
reef enclosing a
lagoon is termed an atoll. Biohermal reefs consist of
a growing network of
intedinked organisms, mainly corals and coralline
algae, and contrast with
detrital reefs; fragments of older consolidated
reef rock, eroded from the
growing biohermal reef on its windward side, are a common constituent of
many reefs. The largest detrital deposits accumulate along the leeward side.
Substantial vertical thicknesses of reefs,
which may be up to 1500 min deep
oceans, are due to the foundering of
the submerged volcanic cones which
provide the basement from which most deep
ocean reefs grow. Reefs contrast-
ing in origin with the atolls of the deep oceans rise from continental shelves in
the tropical zone. The Great Barrier Reef ofT the Queensland coast of
Australia is the most extensive example. It consists of an outer barrier made
up of a line of innumerable ‘ribbon’ reefs separated by shallow passages less
than 1 km wide, with countless ‘platform’ reefs between the outer barrier and
the mainland.
6. Depositional landforms in eolian environments
Although sand covers only a small portion of most deserts the areas of wind-
blown sand deposits can be huge. The area of sand in the Sahara is about 7
million km% in the Gobi desert about 2 million km^ The extent of some vast
sand spreads, like the Kalahari in interior South Africa, was considerably
greater in the Quaternary and in other areas formerly active tracts of sand
movement are now largely stabilized like a 20 x 1 15 -km strip in the Las Bela
valley, inland from the West Pakistan coast. Here the maximum depth of sand
is 120 m; in parts of the Libyan sand sea depths exceed 300 m. Eolian
deposition is also a feature in many localities in humid areas where an
un vegetated surface is made up of unconsolidated sands. In sub-aerial parts of
deltas, fluvial deposits are reworked and redeposited by wind and identifiable
by bedding characteristic of wind-laid sediments, like the fore-dunes built up

at the back of a beach or on the crest of a beach ridge. On the plains adjacent
to Pleistocene outwash areas in Poland and northern Germany, long belts of
wind-deposited sands remain today as strips on the present ground surface up

to 30 km in length and as much as 1 km in width. They are characteristically
unstratified and not as well-sorted as many wind-blown sands due to their
mixture with snow during winter blizzards. These sand covers are usually less
than 20 m deep.
The distinctive feature of wind-deposited sands is cross-bedding stratifica-
tion with some 30 degrees or more if they formed by gravity

sets inclined at
sliding on a steep lee-side slope. Angles on some beach dune slip-faces,

formed by grain by grain deposition and accompanied by partial cementation
by salt and water, are stable at 42 degrees. Individual laminae in most dune
deposits average 2-3 mm in thickness, but extremes range from 1 to 100 mm,
with single lengths of up to 15m or more. In some unusual localities, but
including examples from five continents, where clay includes salt, clay aggre-
gates may be transported, stratified by wind and even form conspicuous
elongated ridges.The south part of Bonneville Flats. U.S.A.. is one example,
the north and west side of Oso Creek estuary near Corpus Christi, Texas,
another, and other instances were first recorded in Western Australia, 130
j’ears ago. Downwind from sandy areas on desert margins wind-entrained fine
silt-sized particles are eventually deposited as loess mantles. Those wide-
being
spread on the plateau and upper valley slopes of the Dneiper uplands,
5-35 thick, give an indication of the blanketing effect of a loess mantle. As
m
vegetation cover is vital calm ground surface conditions for fine
for creating
silt to settle permanently, the removal of

vegetation has had particularly
severe repercussions in these areas. As early as 1922 the resettling rate of

dust in Idaho was 10 cm/ 100 years.
In Antarctica, where wind is the only transporting agent over much of the
non-glacier area, dunes demonstrate their azonal characteristics. Those in the
Victoria valley are 200-600 m long and 10-15 m high. In some areas the
pattern of sand-dunes is not necessarily regular. In the eastern Mojave desert,

for instance, where sand-moving winds blow from all directions in addition to
that of the prevailing westerly winds, large parallel ridges have developed up
to 6 km long and 170 m high, with short reaches diverging by as much as 35
degrees from the prevailing trend. R. P. Sharp (1966) considers that these
dunes, now showing little systematic movement, might be relict features
dating back several thousands of years, perhaps even 20 000 years. In other
areas sand ridges and intervening troughs are remarkably straight and nearly
parallel over broad areas. In Egypt they are hundreds of kilometres long;
some Australian examples are longer. The remarkably clearly defined cor-
rugations in southern Arabia are at least 30 km long and are commonly 200-
300 m from crest to crest. Stabilization of areas of relict dunes has occurred
insome sub-humid areas, as in the High Plains or in Rhodesia where ridges
may approach 130 km with many lengths unbroken for 50 km. Spacing is
about 1-6-2 -6 km and occasionally as much as 5 km. Although these dunes
have preserved their pattern in plan they are characteristically very low due
to sheetwash reducing the ridges to heights a mere 2-4 m above the interven-
ing troughs. The sands here are characteristically very well sorted with a
median size of 0-2 mm.
Barchan dunes, due to the smoothness of curvature in their crescentic
shape and the regularity of their asymmetric cross-section tapering to
in plan
two horns downwind (fig. I.4.C), are one of the landforms approaching most
closely a geometrically ideal shape. In some instances the length and width
are approximately the same and the height about one-tenth this size. Measure-
ments of height, width, and length average 3-12, 40-250, and 60-210 m,
respectively, for the larger Salton dunes in Imperial Valley, California, 1-6,
37, and 1 20 m for dunes in southern Peru, and 30, 400, and 400 m for the
maximum sizes in the Egyptian sand sea. Almost invariably slip-faces average
degrees in inclination, with a
maximum of 34-5 degrees Just beneath
the crest.Barchans occur on desert floors downwind from fields of more
amorphous dunes. Dowmwind summits are aligned en echelon with
the
upwind horns. In southern Peru, where the mean density of
barchans is
1/7-1 ha. their distribution is relatively even. There
is sometimes a tendency
for barchan field patterns to diverge downwind perhaps related to reduced

velocity of the wind due to lateral spreading of the current after wind-
funnelling through a constriction. The downwind spread forms a pattern
similar to that of some drumlin fields. Only coarser material remains in the horns
and near the crest, the sites from which material is most readily removed.
Seif dunes are remarkable single continuous ridges which run straight
across relatively featureless desert areas. Leeward they dwindle to a sharp
point and appear to follow the direction of the long-period wind regime.
According to local conditions, however, the overall plan of seif-dune patterns
may vary considerably (fig. IV.22). Individual dunes may be perhaps 60-
100 km in length. In the Egyptian sand sea many dune summits are 100 m
high and in southern Iran over 200 m high. The transverse width at their base
is about six times their height. In cross-section the seif dune stratification
reflects the important distinction between deposition by accretion, at
Figure IV,22 Some seif dune trends (from McKee and Tibbitts, 1964), The
patterns were traced from air photos of Libya by L. C. Conant.
Figure IV.23 Stratification in sand-dunes.
A. Profile of a barchan dune near Leupp, Arizona, taken through the crest in
direction of dominant wind {after E.D. McKee. 1957. Bull. Am. Ass. Petrol.
Geols., Vol. 41).
B. Cross-section of a high transverse dune. Mustang Island. Texas, anchored by

vegetation {from McKee. 1957).
C. Cross-section of a seif dune, west of Sebhah, Libj'a {from McKee and Tibbitts.
1964).
angles of 15 degrees or less, and by gravity flow creating dips of 24-34

degrees. These inclinations may flatten out near the base of the deposit and
bottom zones may represent the accumulation of relatively coarse sand from
the nearby desert floor (fig. IV.23.C). The slip-face, instead of lying transverse
to the prevailing wind as in a barchan dune, runs parallel with it and on the
side away from that of the prevailing cross-wind. In the case of the sym-
metrical seif dune, the slip-face is a temporary phenomenon which is formed
and reformed on either side according to the direction from which the cross-
wind happens to blow.
7. Landforms due to glacial deposition
A characteristic of glacial deposition, particularly of basal tills, is the large

amount of clay involved, for many workers the product of interglacial and
preglacial chemical weathering rather than the result of bedrock erosion by
the ice itself. Nonetheless, the amount of clay that accumulates in glacial
deposits still poses the problem of understanding the depositional mechanism
involved. One suggestion is that intergranular melt water, arising from fric-
tional heat of abrasion and ice-movement processes might possibly facilitate
the downward migration of the fine products and their deposition as the sub-
glacial matrix of the boulder clay. Preferred pebble orientations are usually
aligned in the direction of ice movement, and the flow of a fluid or semi-fluid
till appears necessary for the production of such patterns. Thus in addition to
the resemblance of the material of sub-aerial mudflow to that of sub-glacial
tills, the condition of two materials at the time of deposition may also have
certain similarities. Most of the development of preferred orientation in clay
compacted artificially under pressures of 0-100 kg/cm’ takes place very early,
at pressures near 1 kg/cm-, the most critical factor possibly being the amount
of water held by the clay. In addition to lack of any precise knowledge on the
mechanisms by which ice lays down its load, there are problems of dif-
ferentiating and interpreting landforms produced by glacial deposition. Dis-
tinctions between forms in the field are often more difficult to grasp than
between idealized-type examples. Generalization about typical dimensions
and shapes is not easy, and the description of some ice-pushed ridges as
end moraines is just one instance where mistaken identifications have arisen.
The main contribution of ground moraine is to the appearance of the land
surface.The smoothing effect is due more to the lodgement or basal till which
was plastered on to the bedrock contours. The coarser ablation till settles on
to the ground surface with little regard for the relief beneath the ice and with
thicknesses of 1-5-3 m is usually an inconsequential aspect of the shapes of
morainic terrain. In some areas an additional factor in the laying down of till
is sub-glacial processes beneath active ice which produce aligned ridges, and
m
grooves parallel to ice movement may corrugate the till surface. In higher
mountainous areas of Scotland, such features may occur on cirque and valley
floors, ranging in height from 0-3 to 6-0 and in length from 20 to 300 m.
An end moraine may be one of the most clearly defined landforms made up
of till, usually regarded as marking an extreme position reached by a glacier.

It is not simple to decide if the ice remains in this position for any length of
time, maintained by an equilibrium between accumulation and ablation and
continuously supplies a load to pile up as a ridge of increasing size at its
margin; whether ice could exert pre-existing force to push material forward
or would tend to override it — if the inward-facing or proximal slope is still as
"steep as 30-35 degrees, push might be indicated whereas a long proximal

slope might indicate overriding; or to what extent an end moraine once
formed provides an outer limit for subsequent minor readvances. Regardless
of exact mode of origin, a characteristic usually observed in well-defined end
moraines is a contrast in morainic material between the inner and outer slopes.
On the inward-facing slope the material is similar to ground moraine. It is
unsorted and with rare or indistinct stratifications, but may include irregular
lenses of well-washed pebbles laid in melt-water channels cut during previous
retreat stages of the ice. On the gentler outward-inclined slope the former
action of running water, including sheetwash, means that bedding is common,
and fragments are more rounded and better sorted. Sizes may decrease down
the slope. An end moraine is not necessarily elongate and narrow but may be
a complicated mosaic of round and broad hummocks, perhaps 150-300 m in
length, and lacking any systematic pattern. Of various possible origins for
such complex end moraines, deposition by stagnant ice is often favoured. Nor
is a distinct end moraine necessarily continuous but when traced laterally
may become more poorly defined or may even temporarily disappear. For
instance, the Illinoian drift sheet has no continuous end moraine, and the
margin of many drift areas is without definite morainic ridges. If the discon-
tinuity is due to washing of till by melt water during and after deposition,
rather than non-deposition, a trace of large boulders without fine material
may remain. In valleys end moraines are usually higher and broader in cross-
section on or near valley floors, but they often extend a hundred or so metres
up valley slopes. Amplitudes of 400 m have been reported from Greenland.
In general, heights are very variable. Where several are closely spaced, indivi-
dual heights tend to be less, as along sections of the Labrador coast where the
heights are 4-5-9 m. Some end moraines in north-west Scotland are up to
40 m high, some in Iceland are about 20 m high. A distinctive type of
morainic deposition is the cross-valley, washboard, or de
Geer moraines
which occur where deglaciation took place in a subaqueous environment
like
south-central Sweden. In Baffin Island they are 3-20 m high. In 70
km of the
Isortoq valley. Labrador, there are at least 2000 moraines
spaced on average
50m apart. In Manitoba examples are 5 m high, about 100-150 m apart and
may be 1-5 km in length. J. T. Andrews (1966) suggests that cross-valley or
washboard moraines may be formed by the squeezing of plastic water-soaked
ground moraine into basal crevasses or to the ice-front at the edge of a
proglacial lake. The weight of the ice itself could exert the necessary pressure,
but some slight ice-flow, possible along basal thrust planes might have con-
tributed in certain places.
The terminus of a stagnant glacier is quite different from that of a receding
glacier because a large loss of volume results from vertical shrinkage rather
than from terminal recession and because material no longer transported to is
the ice-front. The surface becomes covered with englacial debris and a chaotic
surface of stagnating ice with debris knolls and ridges separated by ponds and
streams emerges. During stagnant stages of glacial lobes linear ridges may
develop superficially moraine-like in appearance but composed essentially of
fluvioglacial material. Those on the Plains of North America range up to
10 m in height, may be 7-5-100 m broad and from a few metres to 10 km or
more long. The most prominent features of these ice-disintegration ridges is
that they are essentially straight and lie normal, parallel or at an angle of 45
degrees to the direction of flow and therefore suggest a genetic relationship to
crevasses. The ridges may be related to the filling of a massive crevasse
Figure IV. 24 Landforms due to glacial deposition.
A. Section through a typical lee-side lense of till from central Sweden {after H.
Mailer. 1 960, Geol Foren. Stockholm Forh., Vol. 82). On the stoss side till is hard
and foliated. On the lee-side a loose till, with many lenses is covered with foliated
till. A layer of very fine sand often occurs between till and bedrock.
B. Cross-section through an ice-marginal ridge (Salpausselka I) in southwest
Finland {after K. Virkkala. 1963, Bull. Comm. geol. Finlande, No. 210), observed
between Ojakkala and Otalampi. It seems probable that parts of the ridge were
deposited in lakes at the edge of the continental ice-sheet while other parts were
deposited sub-aerially, according to local differences in relief.
C. Profile levelled across some small moraines in Little Cataraqui Creek, south-
eastern Ontario {after O. H. Ldken and F. J. Leahy, 1964, Canadian Geogr.,
Vol. 8). Figures indicate median particle size in microns.
D. Plan of Blackford Hill ‘crag-and-tail’. Edinburgh, with till fabric {after R. P.

Kirby. 1969, Geografiska Annaier, Voli 51 A). Compare with A, above.
E. The conversion of gravel deposits, a and b, laid down against ice, into kame
terraces after ice-melt (from T. F. Jamieson, 1874, Quart. Jour. geol. Soc.,
Vol, 30), The removal of lateral support often induces slumps in the iterraces.
F. Cross-section through an ice-cored end-moraine at Grasubreen, Norway {after
Ostrem, 1964). For each group of figures the three columns represent, from left to
right, depth at which soil sample was recovered, median particle
size in milli-
metres. and moisture content as a percentage of dry soil.
complex and associated knolls 20-30 m high to the infilling of melt-holes at

crevasse intersections.
In many glaciated lowlands, particularly where ice recession was rapid,
like much of lowland Sweden, eskers are the most common type of deposi-
tional landform. Whether they are better regarded as ice-contact features or
as essentially fluvioglacial in origin may depend on local conditions. They
consist of well-worn, sorted, and stratified material with pebbles in the 5-20-
cm range and often containing maximum sizes larger than a metre, although
some eskers consist entirely of sand. The inclination of bedding is usually
away from the axis of the esker at dips of 1 0-20 degrees or even coinciding
with the surface slope. Like crevasse fillings eskers can ascend reverse
gradients and the disregard of their trace for underlying relief is a noted
feature. Unlike the linear crevasse fillings, eskers are characteristically sin-
uous in plan and always narrow in cross-section, 3-15 m in height and up to
50 m as a maximum. Angle of slope of esker sides is usually 25-30 degrees.
An esker is not necessarily a continuous ridge. It may have gaps in some
places and in others expand into a form with a more bulbous ground plan, the
beaded esker, with many kettle holes in the wider section. Lengths of single
continuous ridges of 1 - 5 km are reported from both the Kennebec river area,
Maine, and the proglacial area of the Casement glacier in Alaska. The fact
that meltwater under hydrostatic pressure could flow up reverse gradients has
favoured an explanation of eskers as the deposits of the flow of sub-glacial
streams with competence reduced as they emerge from the confines of an
open crevasse or tunnel. Tunnels are a feature of stagnating ice as actively
moving ice would tend to close them up. However, in some areas the begin-
nings of eskers appear to develop on the surface of the ice. The preservation of
steep sides of eskers is a somewhat enigmatic feature in easily reworked sands

and gravel. Perhaps the binding effect of roots of the vegetation which rapidly
colonizes the debris could be sufficient to preserve even the form of a super-
glacial debris ridge. In many areas like southern Norway and the Alps forests
grow close to the glaciers. R. J. Price (1966) records the rapid and consider-
able colonisation by alders of eskers at the Casement glacier ice-front.
A feature of glacial deposits when chiefly composed of water-sorted gravel

and sand is that they may have been deposited against ice or on ice. When the
ice melts away ‘ice-contact’ landforms remain. As with the melting of buried
ice, the loss of lateral support in ice-contact landforms leads to collapse and
the associated deformation and contortion of structures. Mounds produced in
this way are sometimes described as karaes. Material deposited in an ice-

marginal lake and banked against the ice is termed a kame terrace, relict
forms being distinguishable from fluvial terraces by their irregular altitude
and irregular surface. In higher latitudes the coarser debris may similarly
accumulate to form a block terrace.
Figure IV.25 Block terrace. Marguerite Bay area, Antarctica (from Nichols,
1960).
Another feature of glacial deposits is that in areas adjacent to present-day

glaciers they may still contain blocks of buried ice and that in areas of
Pleistocene glacial deposition the subsequent melting of such buried ice
blocks has given rise to a distinctively irregular land surface due to numerous
‘kettle-holes’ depressions created. Ice may also persist in the cores of some
moraines as in the Jotunheim area of Norway where the distribution appears
to be related to the altitude of the temporary snowline. A characteristic of
these ice-cored moraines is a rounded surface and outline with minor ridges
superimposed on the top of the larger feature.
In contrast to ice-marginal features the emplacement of drumlins appears
to be the work of active ice. but the nature of the mechanisms involved is
obscure. Whatever they may be, drumlins appear to be favoured by flatter,
more open terrain where ice-flow is spreading out, possibly losing some
competence in developing longitudinal crevasses in the process. Drumlins are
low elongate domes usually strikingly symmetrical in appearance. They are
not usually more than 50 m high or 0-8 km in length. Widths vary and some
drumlins are so narrow as to be virtually linear ridges. There are many
contrasts in the ratios of the three main dimensions from one drumlin field to
another. The symmetrical drumlins in the Wadena field. Minnesota, are only
6-15 m high, yet are 1 -5-6 km long and even 1 1 km on occasions, widths are
0-1-0-6 km. More typical are the Glasgow drumlins which are about 30 m
high and 0-6 km long. Those on North Uist, are as small as 45 m long and
8 m high. Forms may change within one drumlin field. Near Long Prairie, in
the Wadena field, the characteristic elongated form gives way to a broader
oval or almost circular form towards the terminus of the former ice-sheet. In
some drumlin fields there is a many drumlins
tendency for two size modes;
near Syracuse. New York 600-900 m long, and
State, are 30-45 m high,
200-300 m wide, with smaller more elongate drumlins commonly occupying
the intervening areas, 3-12 m high. 250-750 m long, and 60-90 m wide.
D. Inter-relationships between Rock Breakdown, Transport, and

Deposition
1 . Time-lags
A fundamental characteristic of inter-relationships between landforms, land-

forming processes and time is the time-lag between some antecedent perturba-
tion and subsequent adjustments to the changed condition. This makes the
view of simultaneous interactions unrealistic in many geomorphological
situations and the multiple correlation of several associated factors often
unsatisfactory. For instance, even on the shortest time scales, a relatively high
level of unassigned variation in the multiple correlations in the beach environ-

ment of Brodie Island, Carolina, was thought to be partly due to the time-lag
between the changes in process intensity and resulting beach-face adjust-
ments. One complication is that shoreward flow transports largest and
heaviest material to high beach levels early in the declining phase of wave
heights. In cool environments delayed thawing explains some apparently
anomalous late summer or autumn features, like the occurrence of rockfalls
observed at that time of year in Scandinavia and elsewhere (Table IV.9). In
streams, although correlations between discharge and sediment load are
always very highly correlated because of the statistical effect of sediment load
being itself the product of sediment concentration and discharge, the higher
discharges, despite their greater competence, may be only poorly correlated
with sediment concentrations. One reason for this is that quite often, particu-
larly in spring in temperate regions, the maximum sediment concentration
precedes the peak of the flood. F. Hjulstrom (1935) attributed this to the silt
reaching rivers from the interfluves more rapidly than the water-level in the
river rises. A more likely explanation is that the first sediment entrained
during a rising stage is that loose channel bed material which was deposited
during the preceding falling stage. During summer the lag between sediment
concentration peak and flood peak may be less marked as the interfluves and
banks are better protected by vegetation. In glacial outwash some observa-
tions indicate that the peak of silt concentration may be 2-3 hours before the
daily runoff maximum. Similar time-lags between the initial loosening of
material and its subsequent removal may span millennia, where the weather-
ing of bedrock is involved. Much of the load of many mid-latitude streams is
cut mainly from river banks of glaciofluvial material and till, which before its
movement by ice might have taken millions of years of weathering to become

detached from solid bedrock.
Table 1V.9 Frequency of mass movements in Norway (from Rapp, I960 and O.
Holtedahl, I960, The geology of Norway).
J F M A M J J A S 0 N D
Frequency of rockfalls near Bergen. 1920-58 13 11 II 38 9 8 II 6 3 13 7 10
Frequency of quick clay slides, southern
Norway. 1900-60 5 1 4 9 9 5 4 3 4 9 6 5
The time taken for changes to be transmitted through the ground is another
significant source of a lag between an antecedent condition and a subsequent
change. Because of slowness of penetration of thermal waves, residual perma-
frost layers can remain for a long time. The characteristics of spring water on
emergence may reflect the influence of soil conditions at some previous time
when the water entered the soil. Fig. IV.26.A shows how water temperature
and amount of dissolved calcium carbonate, the latter reflecting the tempera-
ture-controlled biological production of soil carbon dioxide, in a cave pool
show an early autumn peak and general seasonal lag of about 2 months
behind the controlling temperatures that would exist at the surface.
There is also the possibility of a geographical distribution in the degree to
which some points lag behind the time of change at others, even though the
lagged responses may be a reflection of the same perturbation. In some cases
runoff peaks from tributaries on either side of a valley do not coincide in time
after precipitation. On a broader scale, it is possible that the date of the
transition from Hypsithermal to late postglacial conditions in north-west
North America lagged according to latitude, due to the greater influence in
the north of the south-flowing cold polar air. This
change may have occurred
about 4000 years ago along the western Gulf of Alaska. 3500 years ago in
south-east Alaska. 3000 years ago in southern British Columbia, and 2500
years ago in California. There is also the possibility of different responses to
the same perturbation lagging one behind the other: one of the major difficul-
ties in correlating glaciations with pluvial episodes is the possibility of pluvia-
tion lagging behind glaciation.
Finally, as rates of tectonic movements may greatly exceed erosional rates

Figure IV.26 Lagged efTects in time and space.
A. Calcium carbonate content and water temperature of Lake Styx, Peak Cavern,
Derbyshire, lagging two months behind air temperatures {from A. F. Pitty, 1968,
Trans. Cave. Res. Group Gt. Britain. Vol. 10).
and with an approach towards some equilibrium therefore unlikely, much of

the broader-scale erosional history of landforms reflects the comparatively
slow response of erosional processes lagging behind tectonic movements.
Therefore in landform development there is a tendency for rock break-
down, transportation, and deposition to succeed each other in distinct phases,
with measurable time-lags between the antecedent conditions and the sub-
sequent change. This is no more than a tendency and the definition of what
would constitute a measurable time-lag varies according to the interactions
under investigation and the subjective judgement, if not merely the taste, of
the investigator. Therefore the consideration of the interactions between these
phases and with landform in the following section is as much a change in
viewpoint as a specific consideration of those areas and environments where
the interaction between rock breakdown, transport, deposition, and landform
is essentially inseparable.
There are five contexts in which it is essential to consider the inter-
relationships between two or more of these four variables. The first is situa-
tions where certain local features put inevitable and distinctive constraints on
the way in which some response might operate. Secondly, there are co-
variances where two or more phenomena occur together in either space or
time but where the mechanism of their interaction is often obscure. Thirdly,
there is the rearrangement of weathered products which may involve essen-
tially a reworking or recombination of weathered products in situ. Fourthly,
there are dynamic depositional landforms. and lastly, there are situations
where the characteristics of some previously weathered material appear to
influence the way in which subsequent weathering processes operate.
2. Influence of source
In many ways rock breakdown influences transportation, deposition and

further rock breakdown simply because the results of its action are the
characteristic products from distinctive environments. The supply of abra-
sives is an obvious example. For instance, in northern Alaska and
Saskatchewan, areas of sand-blast striations on outcropping sandstones are
B. Greater lengths of estimated flow-through time of karst water in the southern

and central Pennines leading to greater solute concentration ij'rom A. F. Pittw
/ 9(55. Nature. Ko/. 27 7).
C, Deposits downwind from the salt lake system at Hines Hill, Western Australia
{after Belteiiay. 1962). The cross-section {below) is not to scale.
D and E. Petrographic changes downstream from changes in bedrock
{from
Tricon. 1 965). These observations on the Ceze river by
J. Capolini show that the
persistence of granite near the ancient massif at Peyremale,
D. falls off down-
stream at St Ambroix. E, as do the schists and quartz.
A ® Kaolmite
Mississippi fîver
Percentage l»mestone >n catchment area

©
Mobile Rivee
^DQ/acft/colo diver
Figure IV.27 Influence of drainage basin characteristics on geomorphological

responses.
A. Increased concentration of calcium carbonate in streams draining from catch-

ments made up of greater proportions of limestone {from Fitly, 1 968) as observed
in the southern Pennines.
B. Composition of three suspended clay loads reflecting an easterly increase in
kaolinite relative to montmorillonite in the soils of their river basins (from G. M.

Grijfen, 1962, Bull. geol. Soc. Am.. Vol. 73).
C. Greater alluvial fan areas at the outlets from larger drainage basins (from Bull,
1964). observed in western Fresno County. California.
well displayed perhaps because of the abundant abrasive sand in glacial
deposits about60 km to the north and north-east. In Sweden in sediments on

thebottom of the Gota alv river there are erosional grooves downstream from
a moraine, whereas none exists upstream where there is almost no bedload.
The paucity of material coarser than sand in tropical rivers, due to rapid
chemical decomposition and the limitations imposed on erosional activity in
streams, has been much emphasized in recent years. In fact more generally,
and in a negative sense, one of the most significant controls on transportation

is that of very slow weathering providing little load. Rivers draining from the
west and south slopes of the Ukrainian Carpathians carry little suspended
sediment, despite their turbulent flood regime because few fines are available
for transport. In Western Australia, gypsum dunes, lunettes, and sheet
deposits are almost entirely restricted to semi-arid areas where the extensive
bare desiccation-cracked surfaces of salt lakes are ideally suited to wind
deflation and provide a ready source of gypsum, sand, silt, and clay. Along
the eastern coast of South America, the highest concentrations of kaolinite
occurs at the mouth of the small rivers which drain in the coastal areas where
lateritic weathering is well developed. There is relatively less kaolinite in the
clays at the mouths of the large rivers which drain from mountainous Andean
areas where chemical weathering is less significant in processes of rock
breakdown, and the absence of a pebble strand is typical of this as of other
tropical coasts. From fig. IV.27.B it is apparent that the dominantly mont-
morillonite clay from the Mississippi river will dominate the clay mineralogy
of the eastern Gulf of Mexico, whereas the Mobile and Apalachicola rivers
will introduce only local modifications.
The effectiveness of transportation depends on the degree of prior size
reduction and on the properties of the parent material which affects the
internal friction of debris. Deeply weathered or poorly consolidated material
if falling from a free face might disintegrate on impact and like the ‘rotten’
rockfalls in south-west United States add little coarse material to lower
slopes. The development of screes, and as a result, that of scree-controlled
bedrock features too. is in consequence reduced. Conversely, in cooler envir-
onments forms like rock glaciers develop only where rocks weather to blocks.
In the central Alaska Range, rock glaciers developed in granodiorite are
absent on higher walls of the canyon where schist outcrops. The proportion of
mudflow material in and on may be absent on
alluvial fans varies widely and
one fan yet predominant on an adjacent one. These differences depend on the
control of local lithology
which limits the amount of fine material that might
be available for transport. Initial size is also a factor in the amount
of
subsequent chemical solution of weathered particles, the rate of attainment of
equilibrium being theoretically strongly dependent on particle size. A
final
point, so obvious that departures are perhaps more intriguing
than examples
Inter-velationships between processes and landforms 263
that of abrasion which involves disordering and hydrating the

solution
surface layers of quartz by continued impact of other grains to produce a
finely divided amorphous silica, and results in the probable rounding of par-
ticles. Similarly, solution may follow abrasion, as on the hydrated silica
crystals and newly exposed metal cations on the separating surfaces of silicate
particles in the rock flour suspended in glacial melt waters. Hydrogen ions are
immediately available for replacement of the newly exposed cations on the
broken surfaces of the silicate minerals. Conversely, in sand-laden swash, size
reduction is predominantly an abrasional process and any rounding of river
gravels might similarly be the result of wet sand-blasting. The importance of
crushing and abrasion of stones during glacial transport has long been
recognized. Much of the abrasion occurs on flat surfaces originally related to
stratification, jointing, and fracture planes, but amounts bevelled off corners
and edges are more significant.
Downstream there may be progressive changes in mineralogical composi-

tion, with climatic conditions playing an important part in determining the
initial composition as well as their progressive changes during transit. As
with size changes, composition changes below tributary junctions or points
where rocky outcrops are eroded. J. Tricart (1965) records that in many
humid tropical rivers as in Colombia. Brazil, and West Africa, the proportion
of siliceous pebbles is always very high in coarse alluvium, from 85 to 100
per cent. Individual quartz fragments 4-6 cm in size may travel 10-15 km
Figure IV. 28 Downstream changes in properties of fluvial sediments.
A.Mean particle sizes of bed-material along the Mississippi {from Leopold

Wolman and Miller. 1964). The data, collected by the Mississippi River
Commission, runs from Cairo, Illinois, to the river mouth.
B. Median particle sizes of bed-material in a North African wadi {from 1. Douglas.
J. D. Leatherdale. and A. F. Pitty. I960, itnpiibl.). The example is the Wadi Kuf in
Cyrenaica, Libya, along which flow is intermittent. The surveyed longitudinal
profile is also shown.
C. Diminution in median particle size with increasing distance below tributary
valley junctions in the middle reaches of the Wadi Kuf (ibid.).
D. Relation of roundness and sphericity to distance from head of terrace {from R.
F. Hadley. I960. U.S. Geol. Surv. Prof. Paper 352A) in Fivemile Creek,
Fremont
County, Wyoming; pebbles are in 16-32-mm-size class.
E and F. Morphometric properties down an alluvial fan in the

Santa Catalina
Mountains, Arizona {from Blissenbach. 1954). (E) shows the distribution of
maximum particle sizes (dashed line) and surface angles (solid line), and (F) the
distribution of roundness and sphericity indices.
G. Downstream changes in the quartz/felspar ratios in the 0-25-0-50-mm par-
ticles sizes in the lowest part of the Godavari river. India {from A. S. Naidu. in
Shirley and Ragsdale. 1966).
downstream before penetration of hydroxides along fissures during low water

leads to their fragmentation. Crystalline debris may disappear 1-2 km down-
stream from its source in a rocky bar. In Dutch Guyane. a distance of 5-
6 km has been observed. In contrast, in the Lower Colorado river, Texas, a
granite proportion of 24 per cent of bedload pebbles, decreasing to 2 per cent
250 km downstream only disappears completely after 400 km. Rocks like
gneiss, schist, and sandstone decreased rapidly from a combined proportion
of 23 per cent to about 1 per cent 150 km farther downstream. Limestone, as
in the Black Hills, appeared to wear at about the same rate as the granite,
suggesting that in these environments felspars and limestone tend to be

equally durable. As such studies of changes in pebble composition yield a
natural index of the relative resistance of a range of rock types in a given
drainage basin, and an index independent of the rock’s role as a relief former,
the investigation of the sedimentary properties of present-day streams should
repay more detailed geomorphological study. These include changes in indi-
vidual minerals as well as in the composition of pebble suites. Felspar breaks

down along twin composition surfaces. Disintegration is fairly rapid in high
gradient streams but lost more slowly in large streams of low gradient. In the
Findhorn river. Morayshire, a feldspar content of 42 per cent is halved 50-
65 km downstream, whereas in the Mississippi feldspars are about 25 per
cent of the sandy material at Cairo. Illinois, but still 20 per cent 1750 km
downstream in the Gulf of Mexico. It is possible that heavy minerals that
increase in roundness downstream may be less than a specific hardness
depending on the abrasional rigour of the stream. In the south Canadian river,
only heavy mineral particles of hardnesses less than 6 were rounded. Quartz
and feldspar show a downstream decrease in roundness in larger size grades,
perhaps due to chipping and cleaving respectively.

Changes in debris moving by gravity downslope are as intricate and vari-
able as those involving other transportational agencies. One distinctive
feature brought out by a study of nearly 500 soil samples from varying litho-
logies in the Oxford area is that the sorting of mineral particles was very
similar regardless of whether the underlying rock was homogeneous, like the
Oxford Clay or whether it was heterogeneous, like the contrasted beds in-
cluded in the Inferior Oolite. This result demonstrates how downslope move-
ment of soil is of fundamental and widespread importance in its influence on
the nature of the debris mantle. On a granite slope in Rhodesia, sorting
appeared absent in the summit area Downslope sorting in the
(fig. IV.29).
upper 5 cm was better than in underlying horizons at most sites, but there was
little systematic change downslope until the lowest site. As in stream chan-
nels, debris tends to move more rapidly on steeper gradients and the likeli-
hood of breakdown of coarse debris during transportation diminishes. Where

free falls are involved size gradation on the scree may be related to the height
of the cliff-face. If large boulders have a substantial initial fall their range is
greatest, whereas with little critical momentum in falling off a low free face
they may instead remain near the top of a scree. Quarter of the way down a
scree beneath the high rockwall in Insigsuin fiord, mean dimensions of frag-
ments are 49 x 34 x 22 cm. with a smaller set averaging 41 x 28 x 20 cm
slightly lower down At the base blocks are up to 10 m in diameter.
the cone.
By contrast the fragments on some screes in Tasmania do not change signi-
ficantly in size downslope. Once again conditions in intertropical areas appear
dilTerent. with the absence of scree at the base of cliffs being striking. Blocks
may disintegratewithin 200-300 m
beyond their source outcrop, with only a
few small pieces of quartz remaining. With the more rapid breakdown in
material during transport on less steep slopes a diminution of sizes downslope
Figwe IV.30 Downstope changes in solifluction products.
A. Types and distribution of head on slopes above the Axe valley, Devon (Jrom
Dines et al.. 1940).
B. Fabric diagrams on a Schmidt net showing changes m strength of particle

orientation down a slope in the SouthernUplands of Scotland {Ragg and Bibby.
1966): (i) near the summit, degree, (ii) 16 degrees, and (iii) 9 degrees, near the
I
base.
may be marked. This is the case on many alluvial fans and on some slopes. On
slopes in stony deserts of Central Australia, material downslope from boul-
der-strewn areas was less than 5 cm in size when it reached the footslope
pediment. Similarly, around the steep isolated rocky hills which characterize
the savanna landscape in central Sudan, the surface sequence of solid rock,
boulders, sand, and clay corresponds with the topographic sequence of buttress,
hillslope, footslope and plain. In south-west United States, J. Mam-
merickx (1964) observed sizes grading from boulders 3-6 m in diameter
on mountain slopes, which were absent at the foot of the slope, to coarse sand
and granules containing fragments 10-15 cm in diameter. Talus, above pedi-
ment slopes in dissected Triassic sandstone tablelands in New South Wales,
also showed a similar pattern, but across the pediment surfaces the 200-
600 p weathered material was relatively uniform in its textural parameters.
The degree of orientation of coarse material is another physical property that
may change downslope. On the north west flank of Broad Law in southern
Scotland, petrofabric analysis at site I (fig. IV. 30), where the slope declivity is
only 1 degree, showed that the degree of orientation was very low at all three
of the depths examined, with a tendency for stones in the subjacent layer to be
aligned vertically. At site II, on a 16-degree slope, the alignment in the rubble
layer was marked with subjacent layers less strongly orientated. Where the
steepness of the slope decreased to 9 degrees at was still
site 4, the orientation
strong in the rubble layer but fell away more rapidly with depth. A number of
steep axial dips were recorded, a feature uncommon at the surface. Most
subjacent layers in this area have a secondary concentration of stones dipping

into and upslope. Mineralogical changes in the detritus may occur downslope
due to the weathering out of less stable compounds (fig. IV.31). Changes in
soil water chemistry may also occur downslope but these have scarcely been
studied and so their possible significance in landform studies is not well
known. However, springs at the base of some savanna slopes are known to
contain enough iron to form crusts on the basal portions of the slope, and
similarly, caliche is a prominent feature of pediments in the limestone (fig.
IV. 30) ranges of the Fitzroy basin in semi-arid tropical Western Australia.
Longshore drift may grade beach and nearshore sediments laterally, as a
general decrease in grain size can be expected from areas of high energy
towards areas of low energy, seen most clearly in the contrast of material
between exposed headlands compared with sheltered bays. As in other en-
vironments residual material only remains in the most exposed areas. On
shingle beaches differentiation of particle shapes may take place downbeach.
Spherical particles, on some beaches at least, move more quickly through the
interstices than do other shapes. In fig.IV.32 the large disc zone has an apron
of spherical pebbles a few feet seaward of the bands of seaweed marking the
turning-points of the flood-tides. The spherical pebbles may have moved
O 100 200 300 400 500 metres
Gently convex rock pediment C « caliche

With brown, heavy (extare so»/ CR* caliche nb
covering the 30-60% of area Sz so»f
reef fc^fween strike ribs
limestone^,. CR C Cff
S ^ ^
£_£
350 400 metres
Figure 1V.3I Downslope changes in soil characteristics.
A. Median particle size on a pediment slope at Broken Hill, Australia {after G. H.

Dury, 1966, Aust. Geog. Studies, Vol. 4).
B. Decrease in percentage of felspar in the sand-sized fraction in the surface soil
below a gritstone escarpment caprock {from P. Zalasinski, 1970, unpubl. B.Sc.

dissertation, Univ. of Fluil). Samples were taken at 4-5-m intervals, the first being
detritus from a bedding plane in the Upper Carboniferous Rivelin Grit outcrop on
Curbar Edge, Derbyshire.
C. Decrease in percentage heavy mineral residue in the fine-sand fraction of the
top 15 cm of soil on a slope at Prospect Hill, New South Wales {from R. Brewer,
Figure IV.32 Inter-relationships between lithological composition, shape and

sorting of material on a gravel beach {after Bluck, 1967). The section was
obtained from trenches at the Newton beach, near the mouth of the Ogmore river.
South Wales. Sg = subgreywacke, L= limestone. Q = quartzite and ORS = Old
Red Sandstone; S = spherical, R= rod, D= disc, and B = blade.
through the cobble frame of the large disc zone. The same lateral filtering of
particles according to their shape also takes place in the outer frame with the
spherical particles moving more readily through the cobble frame than the
rod-shaped pebbles.
Changes occur downwind in the size of wind-blown materials. Wind-blown
silts become finer away from source areas at the edge of glaciers, temporary
lake floors or river beds. Similarly, in dunes climbing over an irregular
bedrock surface, fine-grained sand is winnowed out on the windward slope,
making the remaining material relatively coarse, particularly on the lower
slope. In dunes descending on the lee-side, relatively constant finer sizes are
likely to predominate. However, the size of wind-blown sands in some situa-
tions, as in the Peruvian desert, may become coarser downwind possibly due
to additional coarser material being supplied from the desert floor.
There are not many studies of changes in debris sizes and shapes in glacial
1 950. Jour, and Proc. Roy. Soc. N.S.W., Vol. 82). Basalt is the bedrock at the
summit of the slope, the underlying sandstone becoming the soil parent material
for the lowest third of the profile.
D. Caliche accumulations on the lower part of a pediment slope in the Fitzroy

basin. Western Australia (from J. M. Jennings and M. M. Sweeting. 1963, Bonner
Geog. Abh., Vol. 32).

material. Some end moraines contain sharp-edged boulders and stones, but
the general tendency transport to reduce somewhat initial irregu-
is for glacial
by abrasion. Southward in the direction of ice-movement in central

larities
New York State, rhombohedral-shaped fragments decrease by about one-fifth

and wedge-forms by about one-sixth, with increases in ovoid and discoidai
tabular forms.
In flat areas a lag of coarse material may develop as a signiflcant feature,
involving weathering, removal of fines, with the residue of coarser material,
although it has undergone negligible transport, remaining in situ as a distinc-
tive type of ‘deposit’. Commonly, on summit plateaux in cooler environments,
boulder fields develop where gradients are too slight for the evacuation of
coarse material. If the rubbly material is thought to be a residue resulting
from weathering in situ, the term ‘mountaintop detritus’ describes its autoch-
thonous character. In many formerly glaciated upland areas weathering pro-
cessesmay have started only in postglacial times, and the depth of mountain-
may be very variable. The blockfield cover in the Skojem area,
top detritus
Norway, is 10-20 cm and up to 2 m in thickness. In north-east Scotland it
may be as much as 45 cm deep, with stone sizes of 5-10 cm, and best
developed at altitudes between 600 and 900 m. Rocks which, unlike granite,
resist disintegration into gritty particles, provide the most typical layers. The
cover is thickest on norite in the Skojem area and less on granite. However,
stone-pavements of frost-terraced ground near the summit of Hallival, on the
island of Rhum, west Scotland, have the stability of a relict feature. In
Scandinavia, frost-splitting of similar material is certainly not rapid. On
gentle slopes or on valley floors blockfields, superficially similar to mountain-
top detritus, may reflect the selective removal of fines. In 1906. Andersson
observed the development of blockfields from solifluction deposits in this way
on the Falkland Islands, but there, as in similar cases where blocks and
boulders accumulate on a slight slope due to the washing-out of fines and
hollows, sometimes a crude branching pattern may appear on the blockfield
surface. In tropical areas stones and ferricrete fragments may accumulate in
similar sites to the mountaintop detritus of cooler areas, although the round-
ing of material, as observed in Rhodesia, may indicate their accumulation
during millions of years of erosion. In the absence of frost, insects may be
responsible for bringing stony material to the surface. Desert pavements are
smooth, gently sloping surfaces composed of closely packed rock fragments,
commonly coated with desert varnish. The stones range in diameter from a
centimetre or less to a metre or more. Many coastal sediments are also lags.
Even a beach sand can be considered as a lag deposit with the fines removed
by waves and currents and the sand remaining as a relict sediment.
Unlike the covariances considered so far which illustrate changes with
distance, a second type of correlation is sometimes recorded between gradient
A O
Lower slope
debris
sized
Slope angle (»n degrees)
largest
ol
Mode 4 — o
O1 ii so
Q.
I 1 r T" 1
O
60
5 10 15 20 25 e
Average break -m-slope c
2 40
Intrusive Igneous u
ns
u
Extrusive Igneous ^ 20
Metamorphic 0 4 8 12 16
i/>
i Sedimentary Slope angle (in degrees)
Longitudinal profile
Length m metres
Figure IV. 33 Relationships between slope-angle and characteristics of debris and

soil.
A. Relationship between the mode of largest-sized debris and the average break-in-
slope around an inselberg in south-western Arizona {after Rahn. 1966).
B. Relationship between organic carbon percentage and slope-angle on a

Cotswold scarp {from Furley. 1968).
C. Relationship between silty fraction in the soil and slope angle {from A. Jahn in
Macar. 1966) in a plateau area made up of moraine clays and fluvioglacial sands
in the Upper Odra basin, Poland.
D. Longitudinal profile of a dry trough-like denudational valley in the Upper Odra
basin and the silty fraction in the soil (from A. Jahn in Macar. 1966). Angles in
degrees.
Figure IV.34 Inter-relationships between morphometric properties of drainage
basins and of other landforms.
A-D. Hortonian analysis of Watts Branch above Glen Hills, Rockville

Quadrangle, Maryland {from Leopold, Wolman and Miller. 1964). (A) Map show-
ing stream orders. (B) Relation of stream order to number of streams. (C) Relation
of stream order to stream length. (D) Relation of stream order to drainage area.
(E) Relation of channel order to drainage basin area {from M. J. Selby, 1967,
Proc. 5th New Zealand Geog. Conf.) showing contrasts between rock types and
between the same basins using different basic data. To the left are pumice drainage
basins according to map-based data (dots) and air photos (open circles); triangles
are greywacke basins.
and associated variables. These results are often less useful than those de-
scribing the first type of change in which the order of occurrence is fundamen-
tal. Being removed from their spatial context, they are often difficult if not im-
possible to interpret. Fig. IV. 3 3 illustrates three ways in which this problem
has been reduced. In A, four different rock types are distinguished. In B, where
there is no significant relation between organic content and slope angle for an
Upper Chalk slope as a whole, separation of the observations reveals that there
is a clear relationship between slope-angle and organic content in the upper
part of the slope and the numbers indicate the sample sequence downslope.
In C, the increased coarseness of soil on steeper slopes is shown to be less
clearly developed in a sequence of observations along a longitudinal profile, D.

A third group of geomorphological covariances are derived from the geo-
metry of river basins, rather than from the study of processes and their inter-
relationships with landform. However, some of those who have followed

R. E. Horton’s (1945) lead have concluded that the degree of covariance
among the geometrical properties of ‘normal’ basins indicates that a balance
exists among the associated dynamic processes in the drainage basins. There-
fore some illustration of these relationships is relevant in the present context.

Fig. IV. 3 4 shows how, in a Hortonian analysis, stream-order plots on semi-
logarithmic graph paper against number of streams, channel length, and
drainage area as a straight The slope of the line in fig. IV.34 is Horton’s
line.
bifurcation ratio, expressed as the mean of the ratios of the number of streams
of any given order to the number in the next lower order. However, because
of the statistical effect due simply to the way in which a hierarchy of stream
orders is built up, the bifurcation ratio has little discriminatory power and
inevitably tends towards a constant value for widely contrasted types of
drainage basin. Furthermore, the position of the line on the graph, as opposed
to its slope, is of little value as it fluctuates widely depending on the detail
available on the map or photograph of headstream areas. Fig, IV.34 illus-
trates the difference obtained in Hortonian analyses of the same drainage

basin in an area of pumice in New Zealand, using maps and also air photo-
graphs. The degree of covariance between some other measures is exag-
gerated in instances where both include the measurements of the same vari-
able.For instance, parameters of both drainage area and drainage basin
volume include basin length in their calculation, and part of a covariance of
any pair of these three is due simply to this statistical effect. Some workers
consider these covariances as possibly trivial and probably inadequate and
F-H. Relations between length, width and depth of re-entrants in the Niagara
escarpment {from Straw, 1 968).
J. Relations between degree of protection of a beach from littoral currents,

beach
slope and mean particle size {after W. Bascom, I960, Sci. American, Vol. 203).
inappropriate as a basis for concluding that process and form are in equi-
librium, because when translated into qualitative terms they merely state in-
tuitively obvious relationships. For instance, the bifurcation ratio describes
how the larger a stream, the greater will be the number of its headstreams and
tributaries where the number of headstreams and tributaries has in fact been
the sole basis for describing the larger streams as ‘large’.
Other geometrical covariances in which the dimensions are not statistically
independent also often guarantee some statistical relationship. If the basic

data is open to fewer sources of error than those to which stream ordering is
exposed, the slopes of a series of regression lines may provide useful informa-
tion. provided that the difficulty of interpreting correlations is fully stressed
and that the misleading misuse of the word ‘law’ avoided. Fig. IV. 34 illus-
trates the relationships between the length, breadth, and depth of valleys used
by A. Straw (1968) in his study of the erosion of the Niagara escarpment.
The most valuable geometrical relationships are those between variables
represented by statistically independent sets of data, established in the con-
firmation of careful qualitative observation, or in the testing of shrewdly con-
strued working hypotheses, by measurements. Fig. IV.34 shows the effect of a
headland on the slope of a beach eroded by a littoral current. The increased

slope of the beach on the lee-side is attributed to the increased winnowing on
the increasingly exposed parts of the beach, leaving progressively coarser
material down-current. The example. Half Moon Bay, is on the Californian
coast.
5. Reworking of sediments in stream channels
The inter-relationships in streams between erosion, transportation, deposi-

tion. load characteristics, reworking, and further erosion produce distinctive
features on the land surface. These features can be viewed in plan, in cross-
section. or in longitudinal section.

In plan, one of the organized patterns to appear nearest to a stream’s
headwaters is associated with the small sandy flats that build up behind
boulder clogs, spanning the channel. Eventually floods may notch the boulder
clog at a weaker point with a narrow scour channel, with the sandy level
remaining at a slightly higher elevation. Farther downstream similar diver-
sions may occur, but the drop in elevation is much less than in a boulder-clog
channel, and a channel fork develops with water continuing to flow through
both the old as well as the new channel. Both branches may rejoin farther
downstream, creating the anastomosing or double-branching pattern of the
braided stream because of the likelihood of the creation of a further bar at the
confluences. If the efficiency of one channel decreases in relation to the other
branch channel, the channel will tend to diverge at a greater angle from the
River at average
Marsh flood stage Levee
Stage A bank e Isobaths Current direction

Figure IV.35 Stages in the closing of a distributary channel {after F. A. Welder,
1959, Coastal Studies Inst., Louisiana State Univ. Contrib. 59-7).
trunk stream than the more efficient branch, and with backwater conditions
developing, to become the site of deposition. This process involves a reverse
eddy forming in the side channel opposite the point of bifurcation (fig.
IV.35.B) which, by reducing the effective area of the section in the main
channel, accelerates its flow. The water surface of the constricted
flow
becomes super elevated at the point of bifurcation and a very high percentage
of the bedload diverted into the side channel. Continual splitting, formation,
is
oscillation, and closing of distributaries is a characteristic of

dynamic de-
positional landforms other than the braided stream, like
the outwash plain, the
alluvial fan, intertidal flats or the sub-aerial part of a delta. For any of these
features bar formationand resultant channel splitting tends to be more fre-
quent where comparatively coarser material is involved,
and where there is
strong aggradation raising depositional levels
and thereby increasing the
chances of bank overflow. In dry inter-channel
areas there are abandoned
channels tending to be above the ones active
at any given time.
In many downstream reaches of rivers
the pattern associated with the re-
working of alluvial material is the meander,
not the braided stream. Although
it is difficult to distinguish the independent variables controlling this change

from numerous dependent variables, evidently one of the most important
factors in the origin and perpetuation of floodplain meanders seem to be
which constitute the stream bed and banks. The
the firmness of the materials
bed and banks must be firm enough to allow the maintenance of an open
channel but soft enough to facilitate channel migration. It is possible that if
this intermediate condition varies towards either extreme, braiding or incised
meanders tend to develop. In the latter case there must also be, however, a
certain amount of coarser material for the development of point bars along
the convex bank. In areas where banks are firm but such a supply is deficient,
as in the humid tropics, meander development is less prominent. Even in an
intertidal environment, watercourses traversing the intertidal zone tend to be
shallow and braided where sand is abundant and to meander where clay and
silt predominate. It seems that where banks are easily eroded broad shallow
channels develop in which helical flow is too weak to impose the rhythmic
order of point bar and concave bank and that it is the bank erodibility which
determines whether a meandering or a braided pattern will develop. It is
difficult to dissociate bank erodability from the tendency for sand and gravel
shoals to develop in the stream bed which encourage braiding, as the presence
of coarser material favours both developments. Stream meanders show very
consistent geometric relationships from laboratory streams a foot wide to
those of the Mississippi nearly one mile wide. The meander length is con-
sistently between seven and ten times the channel width. Also, as shown on
the River Elbe, relative depth of water at a bend increases as an inverse
function of the relative radius of the bend. However, for meanders in unstable
material the relative depth may increase as the relative radius increases.
The point of inflection in a river bend is closely associated with a shallow
portion of the reach or a depositional bar on the bed. Material eroded from
one bank tends to be deposited on a point bar downstream on the same bank.
The bar tends to concentrate scouring of the concave bank downstream from
added to the next point
the axis of the bend, the material thus loosened being
bar. In this meander bends move progressively downvalley. Forward
way the
movement of meanders on the Russian plain is commonly as much as 10-
15 m/year. Between 1880 and 1938 bends on the Oka river above
Dzerzhinsk moved at 3-6 m/year, with a maximum value of 7—8 m/year. This
compares with a movement of 3 m/year on the Chemung, a tributary of the
Susquehanna, between 1938 and 1955. The degree to which bank recession is
actually the product of mechanial scour or hydraulic drag is debatable.
During rising-flood stages the thread of maximum velocity tends to shift away
from the outside of the curve towards the centre of the channel, and during
bankfull stage tends to straighten out. scouring across the point-bar deposits
as its stream flow approximates to the line of the downvalley slope. Clearly,
therefore, the main process causing bank collapse cannot be associated

directly with high flows but with the instability of water-saturated banks after
a flood subsides, and with winter freeze-thaw when flood discharges are less
than those in summer. On the lower Mississippi, banks steepen until they are
steeper than 1 in 2, the ensuing caving producing a more stable slope of 1 in 3
or 4. According to C. W. Carlston (1965) meander wave length is controlled
by and 40 per cent of the

the range of flows equalled or exceeded between 10
year when the processes of transportation and point-bar deposition are most
effective. The significance of sand-dunes moving down the channel floor is
unknown, but those observed in Russian rivers are often about eight times the
channel width and therefore have lengths similar to the meander length. Their
movement appears to be related to the sudden local increase in gradient and
velocity that accompanies the cutting through of a narrow meander neck.
Observations on the River Ob suggest that a meander neck might be broken
through once every 30 years.
It is difficult to establish how the observed hydraulic characteristics of
meandering streams are inter-related with the associated geometrically shaped
forms. Although helical flow is characteristic, due to water piling up on the
outside of the bend and initiating a bottom channel flow towards the inside of
the bend, no single parcel of water crosses much more than two-thirds of the
stream width. Regardless of genesis an important fact is that meanders dissi-
pate more energy than a straight channel of similar dimensions and boundary
roughness. Because of the increased length of channel per unit length of
valley, a meandering stream is associated with a low gradient and a straight
stream with a high gradient. S. A. Schumm (1963) suggests that the mainten-
ance of a straight channel during valley aggradation resulted from the need to
utilize all the stream’s energy in overcoming frictional resistance to flow and
in transport of sediment through the channel.
The functional difference between meandering streams and braided streams
might be overemphasized as the former, with its channels scoured across the
point bar deposits and its periodic cut-off diversions is essentially a braided
pattern for certain short-lived periods. Also, when discharges exceed bankfull
stage, both patterns of flow are submerged beneath a sheet-like flow down the
axis of the valley floor. Nor should the possible significance of downstream
change in sediment type as the controlling variable be overlooked. It is readily
observable that watercourses traversing the intertidal zone tend to be shallow
and braided where sand is abundant and to meander where clay and silt
predominate and it has long been recognized that these two materials, in areas
of glacial deposition, give rise to contrasted landforms.
Another aspect of
fluvial deposition which influences the characteristics of
itsreworking has already been mentioned in comparisons between braided
and meandering streams and relates to channel cross-sections. S. A. Schumm
5 , 0-1 ^
^
;
0 110 10 100
Weighted mean percent S>U-cIoy (M)
Figure IV.36 Some inter-relaiionships observed in alluvial channels.
A. Increase of meander wave length with greater discharge (from Carlstom 1965).
B. Relation of discharge to slope, and a line which separates data from meander-
ing and braided channels {from Leopold, Wolman and Miller. 1964).
C. Relation between a width-depth ratio, F, and weighted mean per cent silt-clay,
M (from Scitumm, 1960).

D. An instance of pool depth decreasing with increased meander radius (from E.
A. Kondiferova and I. Hydrology, 1966). This graph by
V. Popov, 1966, in Soviet
V. A. Orhânkin shows a relationship inverse to that observed in some other
situations.
found that relatively broad channels are associated with bank and channel
deposits high in silt and clay with little material coarser than 74 microns. It
seems possible that where coarse material is available it could, as in so many
other situations, protect the underlying material, thus favouring bank erosion
and channel widening rather than channel deepening.
The mechanism of scour and fill is not known at present, but appears to
take place as one continuous process beneath the water surface in channels
which are relatively shallow. One suggestion is that scouring of troughs takes
place by eddies generated at the front of shoals or of advancing sand waves. In

headstream areas floods leave a basic stepped nature to the channel profile
with more level areas of fines between irregularly distributed boulder clogs.
This pattern is best seen in the waterless channels of ephemeral streams. In
higher reaches scour and fill may be inter-related with the channel cross-
section, with scouring by the water moving faster through constrictions and
the building up of riffles and bars where water decelerates into broader cross-
sections. An initial deepening favours further deepening just as an initial
depositional structure, with flow-lines tending to diverge from its summit,

may favour further deposition. Ultimately both processes become self-
arresting. In straight or non-meandering channels the spacing may become

regular at a distance of 5-7-channel widths apart. As with many dynamic
depositional forms, like barchans or point-bar deposits, a riffle is a set of
sedimentary particles retaining a characteristic shape, while individual
particles move intermittently downcurrent from one form to the next. After
devastating floods scoured channels in mountainous areas may not be filled up
and sizeable excavations remain. After a flood in 1947 in Kantz Creek
Valley, the channel scoured out was 7 m wide and 3-20 m deep. In mid-
latitudes a flood of 500 cusecs in a 10 000 km^ catchment could scour a
channel some 3 or more metres in depth.
6. Dynamic depositional landforms
Different from traction carpets or suspended loads are dynamic depositional

landforms in which particles are, at least for part of their transportation time,
incorporated in masses rather than moving as individual particles. The mass
itself may move or remain stationary while some particles arrive, usually by
surface creep, change their relative position within the mass, and others
depart. Where the forms themselves move they are usually wave-like
in
appearance and may
range in size upwards from ripple marks which may be
only 1-5 cm apart. They arise, it seems, because a flat sand surface is
un-
any small chance deformation tending to become accentuated by
stable,
the
mode in which saltation moves sand. To the lee of a dune summit the
reverse
wind up the slip-face is below the threshold velocity
for grain motion. The
Figure IV.37 Equilibrium chart for evaluation of various shoreline features (from
Tanner, 1961). Wave energy is the average breaker height in centimetres, and the
sand supply units are cubic metres of littoral transport per year. The circle plots
the Apalachicola river delta, suggesting that the amount of material supplied by
the river is too much for the waves and currents to redistribute.
existence of the lee eddy and reverse wind causes sand blown over the summit
to be deposited on the upper part of the slip-face. This mechanism keeps the
sand from blowing from dune to dune, and thus is essential in conserving the
size of dunes as they travel, and the slumping is probably the principal agent
in shaping them. The surrealistic appearance of barchan dunes reflects a
quasi-equilibrium between a dynamic depositional form and aerodynamic

forces evident both in section and in plan. The sand coming to a steady-state
dune from upwind serves principally to set the dune’s sand in motion and to
replace the loss from the tip of the horns. Sand flies rapidly across the bare
stretches between barchans, especially to the leeward of the horns, accumulat-
ing in the next barchan which obstructs its path. Although at a given time an
individual barchan can be considered as an aerodynamic equilibrium form
this does not apply to a barchan field as a whole. For instance the size
distribution of barchans from the upwind end of the dune field towards the
OsciIlQlory Wave Waves ot Ironslationdjores),

Ware Swash
Motion wove* collapse Longshore currents, Seoward* collision backwash Wind
relurn Mow.. Rip currents
Dynamic Offshore Breoker Surf Transition Swash Berm crest

zone
Profile
Sediment
size trends
Coorsesl
Grains
Coarser aw— i^ Bi - modol
log deposit BBBB Wind
winnowed
~
tag deposit
Predominant Accretion
Accretron Erosion Tronsportalion Erosion and
action Erosion
Sorting n
ucucr Poor Mixed Poor Uellfri
Energy — Increase — High Grodient — High
Figure IV. 38 Summary of the effect of the four major dynamic zones in the beach
environment (from Ingle, 1966). MLLW = mean lower low water.
north-north west end of the Pampa de la Joya, Peru, suggests that dunes are
formed in this area, grow to a maximum size, and then shrink again and
vanish in the downwind part of the field. It appears that fine sand is imported
into the area not only in the form of dunes but also that the desert floor
represents an additional source. Also sand transport in the downwind part of
the barchan field takes place increasingly in the form of streamers. Some
investigators consider alluvial fans as stationary
dynamic forms with their
size and declivity an approach to a balance between rates of supply
reflecting
and rates of removal from the fan. If a fan receives an increased supply of
debris at its head this will steepen the gradient part way down the fan where
gullying may lead to accelerated removal. The concept could also be applied
to spool-bars.
The best-known example is the sandy beach where according to local

conditions material is provided by longshore drift of fluvial sediment, shore-
ward movement off the sea floor and perhaps a
due to erosion of little is
adjacent parts of the coast. Balanced against this supply are losses
alongshore
and offshore and landward removal of sand by wind. The
alternation of scour
by storm waves and berm-building in calm weather known
is as ‘cut and fill’.
A distinctive feature is that losses may take place in a matter of hours,
whereas accretion to replace some former volume takes
several days to
several months.
Acritical factor is the 9-m limit on the depth
from which waves can supply
sand. Shoreward, the sand lense of the beach
is a profile of equilibrium that
is
balanced against gravity and wave energy.
Despite some difficulties in inter-
pretation and conflicts between laboratory
and field data, average breaker
height at any period appears to be the
most reliable index of accretionand
losson natural beaches.
As a specific example, on Brodie Island beaches,
Carolina, waves higher
than 70 cm occurring with water
levels 80 cm or more above mean
sea-level
resulted in decreases of beach thickness of up to 60 cm and of widths by up to

7 5
- m, with a flattening of the beach-face slope.
Considering a beach in plan, W. F. Tanner (1962) suggests that the curva-
ture and sand prism characteristics of the hypothetical equilibrium beach are
adjusted to each other so delicately that the potential littoral motion provides
precisely the energy needed to transport the detritus supplied at the upcurrent
end. The point of maximum net erosion tends to be close to the updrift end of
the beach, its precise location depending on the rate at which the drift is
moved, on the ratio between the fetch at the downdrift end and the fetch at the
updrift end, the length of time during which the present processes have
operated and on whether at present there is any systematic change in average
energy levels. Beach slope is an important inter-related feature because long-
shore-current velocity increases with increasing volume of water shoreward
of the breaker zone. Although material passes through many beaches and
other dynamic forms of coastal deposition, other depositional forms on coasts
may be essentially closed systems like Cape Cod, made up of material derived
from a source immediately offshore or from the erosion of fluvioglacial
material of the Cape itself. Other beaches can be regarded as reservoirs of
sand and gravel slowly being released to the sea.
7. Reworking of sediments into distinctive patterns
In periglacial areas the reduced efficiency of fines as thermal conductors

appears to be involved in the development of distinctive patterns in loose
debris where an ample supply of water is available, as on former lake floors.
Because thawing temperatures arrive sooner at a given depth below or close

to stones than through finer material, stones on surfaces where vegetation
cover is discontinuous may tend to migrate into thaw depressions. From these
zones finer materials might be removed by eluviation in water which may flow
in the fissure network. As well as frost-heaving, drying and shrinking are
involved in the production of sorted patterns. Absence of continous snow
cover is also necessary and in mountains the upper limit of structural soils
tends to follow the lower limit of permanent snows. Most polygons have from
three to six sides formed by straight to gently curving, orthogonally inter-
secting cracks. An initial random outline becomes more regular with increas-
ing age. One- to five-metre diameter polygons occur on 0-5 degree slopes in
north-east Greenland. Polygons of a much larger order of size develop in
permanently frozen ground. In north Baffin Island, examples are up to 50 m
in diameter. Comparable in size are some large ice-wedge polygons still
discernible from the air or in vertical sections in former permafrost areas, as

in south-central England. On the Three Pigeons plateau, 13 km east of
Oxford, the diameters of relict polygons range from 40 to 150 m.

Despite the fact that polygons may dominate areas of frozen soil, there is
nothing unique about a polygonal fracture pattern. Apart from soils in peri-
glacial environments it may be seen on mud, basalt, concrete, varnish, paint,

and other media which undergo volume change, resulting from the contrac-
tion of a layer of homogeneous material perpendicular to the cooling or
shrinking surface. Desiccation polygons in semi-arid playa floors develop in

most clays apart from kaolinite, due to exceptionally great volume changes.
Widths range from 15 100 m and those of Guano Lake, Oregon, are up to
to
300 m in width. Like frost and ice-wedge polygons there are two orders of
size with giant polygons rarely less than 15 m in diameter. Smaller features
appear on many types of surfaces, due to day to day changes, if drying

conditions follow rain.
Unlike polygons on flat ground, surface patterns on slopes are much more
restricted to periglacial environments. Downslope from upland plateaux poly-
gons are drawn out on gentle 2-5-degree declivities until the sorted material
lies essentially in stone stripes. There is no general critical declivity, the
transitionfrom polygons depending on the mobility of the thawed material. In

the Antarctic, for instance, polygons develop on some essentially immobile
35-degree slopes. Pipkrakes are probably involved in the sorting process that
produces stone stripes as these features are best developed where night tem-
peratures are very cold. T. N. Caine (1963) considers that periods of more
than 24 hours may be required before the differential frost heave of stripes
becomes apparent and suggests that periods of freezing of more than 3-4
days are more effective as far as the mechanics of sorting are concerned.
Artificially disturbed stripes have reformed within a winter in Japan. The
longer the duration of movement the more distinct becomes the boundary
between stripes made up of different types of material. In observations in New
Zealand, at an altitude of 660-800 m, twenty-one of sixty-eight marked
stones moved from fine to coarse stripes in 2 years, and comparable rates of
sorting have been recorded in the Lake District Hills. Rillwash, utilizing the
coarser bands as natural drains, is an associated process accentuating sorting
by removing fines. In fact in drier environments or on hillcrest sites in
periglacial areas where rillwash is not important it is the portions of polygons
parallel to the slope which
termed soil guirlands.
persist,
where permafrost is absent, frost processes may still play
In cooler areas
an important part in the development of ground surface patterns. The earth
hummocks or ‘thufurs’ are well-developed under relatively mild conditions in
Iceland. Somewhat similar mounds have been identified in some higher
areas
in Europe, as on Hohneck in the Vosges, but according to K. F. Schreiber
(1969), those in the Swiss Jura are anthills. A
peat or grass mat covers the
loamy or clay interior ofthe thufurs. They are about 25-30 cm high and 1-2
m
in diameter, and if artificially levelled are known to re-form within a few
teet
in
height
Wawe
years, but by a process that remains obscure. In dry areas of Australia

volume changes in clay soils, related to desiccation and occasional flooding
produce the gilgai soil pattern which resembles the thufur in its external form.
Pimpled plains western United States, consisting of low domes 60-

in the
120 cm high and 15-30 m

across, may be related to the desiccation cracking
of clay pans which they usually overlie.

If tidal range is small and if there is a balance between beach profile and
wave characteristics, beach cusps may form a regular pattern on the seaward
side of advancing and broadening berms. Initially patches of coarser material
accumulate at irregularly spaced intervals along the upper limits of wave

uprush and gradually develop flat tops. Regularization of form and spacing
appears to be related to rotating vortices developed between the coarse
patches by longshore drift, with water swirling into the gaps to pile up
against streams from the developing cusp on their opposite sides. Occa-
sionally, similar features of considerably larger size form in lagoons as
cuspate spits of very regular and symmetrical outline, like those of the St.
Lawrence Island in the Bering Sea or on the shores of the Black and Asov
seas (fig. IV.40).
8. Recombination of weathered chemical elements
One of the most distinctive products of the combined action of rock break-
down and deposition is the development of weathering crusts and other
products involving the recombination of elements released by weathering.
Transportation, apart from vertical movements within the weathering profile,
is not necessarily involved. However, the interpretations have involved some
confusion because similar recombinations may be produced by solutions that
have travelled some distance downslope or downstream as well as by dis-
solved elements reprecipitated essentially in situ. Another source of confusion
is the development of a crust rich in one particular element either because it is
precipitated from a solution rich in that element, a process termed secondary
Figure IV.39 Microforms and distinctive patterns produced by the reworking of

unconsolidated surface materials.
A. Cross-section through striped soils {from Lliboutry, 1965) observed in the

Santiago Andes at 3800 m
on a 36-degree slope.
B. Cross-section through material filling a relict Pleistocene ice- wedge on the
North Yorks Moors {after G. iV. Dimbleby, 1952, Jour. Soil Sci., Vol. 3).
C. Cross-section through a thufur in the Tatra mountains, Poland {from A. John,

1958, Biuletyn Peryglacjalny, Vol. 6).
D and E. Beach cusps. Relations between mean cusp spacing and wave height
{from M. S. Longue t-Higgins and D. W. Parkin, 1962, Geog. Jour.,
Vol. 1 28) and
(E). a block diagram of beach cusps {from Kuenen, 1950).
ocean
Figure 1^40 The development of stable spits and lagoons (from Zenkovuch,
1959) In the stable stage the wind rose is nearly round; key as for fig IV. 19
enrichment, or alternatively essentially the same end-product might arise by

residual enrichment where an element remains relatively undissolved while
other elements are leached away. Another problem is that while organisms
are known to be a predominant factor in the reprecipitation of some elements
like calcium, their relative importance in other processes is unknown but
suspected to be highly significant. A final problem is in deciding whether the
weathering crust is in active formation under prevailing conditions.
One of the most immediate reasons for including the discussion of the
complicated topic of weathering crusts within landform study is that many
crusts have become distinctive iandform features. Ferruginous crusts in
Brazil, for instance, preserve ridges and plateaux because the underlying
softened rock is readily eroded. Their resistance can be gauged from some
areas where fragments from ancient crusts, like the early Tertiary siliceous
layers in North Dakota, have been lowered 100 m or more, while the under-
lying rock has been eroded away. Similarly, some beaches in lower latitudes
may be protected from erosion by beachrock encrustations. Secondly', crusts
have been used as stratigraphic horizons marking stages in erosional histories
in some tropical areas. In this context plateau top residual enrichments have
sometimes been erroneously linked with valley floor secondary enrichments.
Beachrock originating close to the strandline, is an excellent indicator of
recent changes in the relative level of land and sea. Thirdly, some crusts, par-
ticularly several varieties of calcareous crust, have been used as paleoclimatic

indicators. Gypseous crusts are characteristic of a desert climate, whereas
others are associated with a humid climate supporting a forest vegetation.
Three factors are involved in the production of crusts which favour their
development in lower latitudes. First, mechanical erosion of the surface must
be less than rates of chemical weathering for residues to accumulate.
Secondly, for cementation of the residues high temperatures are necessary to
evaporate the mineral-bearing solutions while these are still in the weathering
profile and for intense biological activity that might be involved in precipita-
tion processes. Thirdly, for the preservation of the porous textured concre-
tions, freezing temperatures must be absent. Thus in mid-latitudes it is only
calcareous encrustations which may develop where flowing water, retaining
some warmth from an underground source, always saturates the deposit.
In a semi-arid or arid climate where evaporation exceeds precipitation soil
moisture may, having dissolved calcium compounds from soil and rock in its
downward course during a rainy spell, move upward towards the surface
during the ensuing dry period, where evaporation leaves behind as caliche a
somewhat spongy calcium carbonate crust around pebbles or as layers just
below the surface. An example of this type of duricrust or weathering crust is
the ‘caprock’ of the High Plains in Texas and New Mexico. Relict caliche
horizons are often better preserved today where they have remained beneath
an alluvial cover. Recent caliche begins to form as isolated nodules around
roots and pebbles beneath the surface. Eolianite is preponderantly sand of
eolian origin also owing its degree of induration to the amount of cementing
calcium carbonate present (fig. IV.41). Cementation might occur during the
evaporation of rainwater and is favoured by the alternation of wet and dry
60 -1
[ i I
;
T”
10 20 30
Percentage calcium carbonate
Figure IV.41 Cementation as a function of calcium carbonate
content in partially
consolidated eolianite (from Yaaton. 1967). The Wingate dune.
Israel, provided
this example of induration.
seasons which characterize the Mediterranean climate. Cementation is

greatest near the surface where drying is quicker and more pronounced.
Dune plants and shrubs might be involved in the process, but formerly held
ideas on the capillary rise of lime-charged ground water are not now thought
to be physically possible because of the coarse pores of sandy material. Indura-
tion is relatively rapid extending over tens or hundreds of years. Eolianite may
be found close to shores with beachrock but is distinguished from the latter by
better sorting and by distinctive aeolian bedding. Depths of 43 m have been
proved along the Egyptian north coast, suggesting that formation started
when was lower. Beachrock. incorporating sand, shells, and other
sea-level
residual materials notblown off the beaches may vary in composition over
very short distances. The beds trend parallel to the coast and dip gently
seaward. The cement is calcite and the beachrock forms near the shoreline
where a dense algae population may play an important part in the chemical
precipitation processes. The cementing calcite appears to be precipitated from
percolating ground water encountering carbonate-saturated sea-water,
although repeated wetting and drying as the water-table rises and falls with
the tide may be a more significant factor. Where the cement is aragonite, as on
British Honduras, the origin of beachrock is probably marine. A precise
climatic control, possibly related to algal activity or to temperatures needed
for evaporation limits active beachrock development to 33“ N in California
and within the Mediterranean. Some workers consider that beachrock is
essentially a characteristic of retreating beaches.

Crusts due to the reprecipitation of iron are sometimes referred to as
ferricretes. Boulders of this material on hilltops near Salisbury, Rhodesia, are
probably related to a former more extensive spread of Kalahari sands
although their nearest deposit is now 60 km distant. In Brazil, the superficial
weathering product of an iron-rich rock formation is 89 per cent FeÔj and

2-9 per cent AljO^, compared with 54 and 0-5 per cent respectively in the
unleached rock. Estimates based on silica removed in spring water suggest

that similar iron-rich crusts in Venezuela might have taken about 20 million
years to form. Silica, once dissolved in percolating water is not easily repre-
cipitated. whereas the iron in hematite is readily reprecipitated as relatively
insoluble hydrous ferric oxide. This is, therefore, not easily remobilized with
the water-table rise in the succeeding wet season and the progressive enrich-
ment in iron takes place. The source of renewed supply of iron appears to be
solutions brought up to the surface. Examples of iron-rich weathering include
crusts along the Niger, where quartz pebbles are cemented by iron oxide to a
depth of 5 m at least. In the Llanos, older terrace deposits include a ferru-
ginous hardpan at the base of gravels which cap the terraces. These examples
show how widespread iron-rich crusts are in seasonally dry tropical environ-
ments where they form important landscape elements. They may also form
Figure IV.42 The effect on a tropical deeply weathered profile of an environment

propitious for long-continued leaching of quartz and of hematite and for repre-
cipitation of iron (J. V. N. Dorr, 1964, Econ. Geol., Vol. 59). The profile is that of
the Casa de Pedra itabirite deposit, Brazil. Itabirite is the local name for the
residually enriched intermediate grade ore.
in more arid environments where the water-table is near the surface, but
might equally be relics from a seasonal wet climatic phase in the past. For
instance, in the desert in Upper Egypt, ferruginization of the Kharga oasis
deposits is a characteristic feature, with the iron for the crusts and concre-
tions believed to have come in part from ground water.
Bauxites are another partly residual product due to leaching out of other
elements, but some are too thick to be produced by the solution of limestone
alone and suggest the supply of siliceous and aluminous materials from
allogenic sources. Bauxites are widespread in Jamaica. On resistant Oligocene
and Miocene limestones in the centre of the island, the AI 2 O 3 percentage is
43-45, Fe 203 is 20-22 per cent, and SiOj 3—4 per cent, the proportions of
iron and alumina being rather higher than for many bauxites.
In cooler environments the differential downward movement of A1 and Fe,
termed podzolization, is less pronounced and occurs within even shallower
depths, but leads to the gradual development of an indurated horizon as the
clay is cemented by small amounts of translocated Al. At a Roman camp-site
near Forfar in Central Scotland, a 15-cm indurated layer appears to have
formed in about 1000 years. In Vermont, hardpan depths in a loam were
silt
largely restricted to depths of 30-40 cm below the surface. Although these

cool environment concretions do not compare in extent, thickness, and resist-
ance to the rock-like crusts of savanna areas, they do have some influence
on land-forming processes by restricting drainage and tree-root penetration.
On bare rock surfaces in tropical areas thin veneers may appear, the result
of processes similar to those which produce crusts in weathering profiles.
Like crusts, veneers are often allogenic rather than composed of elements
drawn up from within the rock. For instance, the manganese oxide veneers a
few millimetres thick which add greatly to the resistance of rapids to erosion
are deposited from stream water where seasonal variations of flow exist. The
rocky domes of inselbergs are often smoothed by a similar ferromagnesium
veneer which has the significant effect of preventing the entry of water into
fissures and aids rapid drying on the surface. Silica veneers on granite out-
crops. as in Surinam, have been observed to have a similar protective role.
Algae may be Involved in the reprecipitation processes.
Clay mineral synthesis in tropical environments may accompany the for-

mation of weathering crusts and is a geomorphologically significant process
because of the depth to which soft weathered material is produced, the
distinctiveness of slope form where it is eroded from beneath a weathering
crust caprock, and the significance of differences between the properties of the
various clay types. Clay mineral formation is also significant in that clays are
usually the result of the same processes that produce weathering crusts in
lower latitudes. Finally, an understanding of how dissolved elements are
reincorporated in clay mineral synthesis is necessary in appreciating why net
denudation rates in tropical areas are usually relatively low. The uptake of
silica during the formation of new silicate minerals is one example.
Clay minerals of approximately similar compositions may be produced
in more than one way. W. D. Keller (1964) recognizes two directions of
approach towards a given end-point or equilibrium conditions between the
quality of percolating waters and the residual clay minerals. On the one hand
anamorphism describes the synthesis of clay minerals from smaller or
simpler units such as alumina and silica gels. An example would be illite
formation by the absorption of potassium and magnesium ions from drainage

water derived from an allogenic source rich in these minerals. Conversely
katamorphism describes the degradation of clays. This type of clay formation
is widespread but not as significant as has been thought in the past. It is
restricted to areas of free drainage and to where ground water does not
participate in the soil-forming processes because it is essentially beyond the
reach of plant roots. The relation between the exchangeable cations on the
clay minerals and drainage water separates soils into two main groups, a
contrast which may be as much in evidence between top and bottom of a
tropical slope, or in sites adjacent or removed from a stream, as in a contrast
between humid tropical and seasonally dry tropical zones. The two major
tendencies in clay mineral formation are, first podsolic alterations in an acid
environment, with the development of silica-rich superficial horizons.
Secondly, in a near-neutral environment lateritic alterations produce a zone of
aluminium and iron concentration.
The synthesis of crystalline alumina and silica into kaolinite occurs where
1
landform. climate, or both produce an acid environment with the free removal
of calcium, magnesium, iron, sodium, and potassium, and with hydrogen ions
being supplied in abundance. Thus with a pH of 4-5 the soil clay minerals
tend to become H clays. For iron to be removed in its soluble ferrous
bicarbonate form, air must be excluded. In practice, some precipitation in the
ferric iron form also takes place, and this association of iron oxides with
kaolinite is one of the commonest features of tropical soils and accounts for
their predominantly red colour. The iron oxide is precipitated as a coating on
the surface of clay minerals formed in an acid environment, reducing the

exchange capacity of these clays like kaolinite or halloysite, and protecting
them from further decomposition in percolating waters. Aggregation of iron
hydroxides may give the kaolinite a certain permeability. Desilication of kao-
linite may yield gibbsite under conditions of strong leaching where good drain-
age makes the oxidation potential high. Gibbsite precipitates from percolating
waters during drier periods within the weathering mass, fissures in the soil
providing a drying surface. Root surfaces in addition to providing a drying
surface also form a zone of relatively high carbon dioxide saturation which
favours the precipitation of aluminium oxide and the formation of gibbsite.
In less severely leached situations than those in which kaolinite forms and
where the water has an alkaline reaction, cations in the exchange position of
clays may
be Ca, Mg. Na, K. or others and favour the formation of montmor-
The concentration of hydrogen ions is correspondingly low. This
illonite.
may occur in a semi-arid environment where percolating waters carry little

organically produced acid due to the very rapid oxidization of organic
material and where hydrolysis of the silicate occurs in the presence of
almost neutral water. Under such conditions of slight alkalinity the silica tends
to be dissolved and removed, with an associated concentration of alumina
and ferric oxide in the upper part of the weathering zone. During the drying
which follows the seasonal rains the solutions originally dilute in cations
become saturated with magnesium, calcium, iron, and sodium which com-
bine with oxygen, silica, and alumina to produce the montmorillonite, often
beneath an alumina-iron weathering crust. However, the same clay-mineral
synthesis can occur in humid environments where poor drainage leads to
a build-up of cations.
Illite is a term describing clay-mineral constituents belonging to the mica
group where the magnesium content is low. Illite is structurally similar to

montmorillonite except that 1 5 per cent of the Si^'* is replaced by AF^ and the
resulting charges balanced by K* ions. It tends to be characteristic of soils in
drier environments. Combined depotassication and desilication of illite may
yield kaolinite, the alumina content in the latter being relatively higher.
There are many variations of the three main types of clay mineral and even
these are not clear-cut entities. The synthesis of elements in clay-minerals
should be regarded as dynamic recombinations rather than static end-points,

susceptible to continuous evolution according to the climatic conditions, site,
and the composition of solutions moving through the weathering profile. The
degree to which organisms are involved if only indirectly in these processes
has not been established, but it is known, for instance, that most natural
surface and near surface waters do not have enough silica dissolved to pre-
cipitate amorphous silica inorganically.
9. Influence of weathered material on subsequent weathering
One feature of processes of rock breakdown expressed in innumerable differ-

ent ways is that the process becomes self-arresting by producing a sufficiently
thick cover of material in which much of the energy reaching the ground
surface is dissipated without reaching bedrock. In other areas vegetation
cover may, in an analogous way to the debris mantle, intercept incoming
energy. The peat cover of humid cool temperate latitudes in which the entire
supply of incoming energy is absorbed in evaporating the excessive moisture
content is one example that has already been considered. Similarly, in humid
tropical areas weathering and either subsequent removal or recycling of
elements is related to the intense biological activity and the degree to which
this merely rapidly recycles the same mineral elements without appreciable
losses to runoff water and significant gains from weathering bedrock. Such a
state of biological equilibrium between soil vegetation and climate is termed
biostasis. With a breakdown in equilibrium, rhexistasis, the forest progres-
sively disappears, perhaps due to the climate becoming drier. In other areas a
weathering crust affords immense protection from weathering to underlying

materials, but in many cases, regardless of vegetation cover and without
consolidation, the mantle of weathered debris arrests the penetration of rock-
weathering processes. The significance of soil depth in protecting underlying
bedrock is particularly marked in relation to frost penetration in cooler
environments. On Kosciusko summit, the stony soil 1 m deep and an over-
lying stony layer appear to limit the frost-shattering destruction of bedrock
outcrops today. In contrast to the self-enhancing immunity of bedrock outcrops
in tropical environments, the exposure of bedrock above a debris mantle in
cold climates hastens its disintegration by frost action. Depth of soil may
also be significant in warm seasonally dry areas where there is an upward
return of moisture before it reaches down to the bedrock. There is also the
constancy of soil temperatures at depth in a soil compared with seasonally
high temperatures near or at the surface, accelerating biochemical processes.
J.A. Mabbutt (1966) describes how the mantle of debris on some Australian
pediments imparts a general levelling to an otherwise uneven and hetero-
geneous bedrock surface directly through its base-level control of ground-
level ‘sapping’. Gilluly (1937) suggested that rocks yielding relatively coarse
and resistant debris have steeper pediments than those yielding relatively
coarse and friable materials. In cool temperate environments a residual sur-
face-lag deposit, by making the soil more permeable, reduces mechanical
erosion at the surface. Also weathering at depth, despite the dampness of the
soil environment, is slow because of low temperatures. It protects the surface
from puddling and the dislodging action of raindrop impact. Field experi-
ments in Maine indicate the importance of this factor in a temperate climate
(Table IV. 10). In semi-arid environments the surface-lag deposit is more
important. Studies in Israel showed that with the removal of stones from areas
Table 7F.70 The effect of surface stones on soil erosion {after E. Epstein et al., 1966,
Soil Sci. Soc. Am. Proc., Vol. 30).
The soil is developed on till, Presque Isle, Maine. Averages are for the four years
1961-4, during which time annual soil loss was greatest in 1961.
SOIL LOSS RUNOFF
Stones Stones Stones Stones

Natural removed crushed Natural removed crushed
Kg/ha Millimetres
1961 5954 7574 5711 205 245 220
1961-4 3224 4062 3224 153 191 167
where they made up 28-62 per cent of the surface led initially to a twelve-fold
increase in erosion. As early as 1927, H. Mortensen described how such
hamada covers stabilize surfaces in semi-arid areas by reducing the erosion
on underlying soil to a minimum. In arid areas a lag gravel of small rounded
pebbles accumulates which may constitute a continuous sheet or serir, pro-
tecting underlying sand from deflation. In many other areas, including some
shingle beaches, there is a lag deposit cover due to the removal of the fines.
Thus the beach material that would provide an abrasive for waves impinging
on cliff faces may, if supplied in profusion, provide a protective ramp on
which all storm wave energy is expended. Similarly, nearshore shoals have an
important effect in protecting beaches from erosion by absorbing storm
energy. The most rapid retreat of coasts
is in areas where there is not enough
residual material and where the abraded masses are continuously lost by
beach drifting. On flat sites inland a relict sediment like mountaintop detritus,
may develop and in warmer, damp areas certain mineral elements may slowly
become more concentrated in surface layers due to the progressive leaching
of more soluble compounds.
Although a soil is in some senses the end-product of rock breakdown and
may in certain circumstances arrest the effectiveness of some weathering
processes in other situations, it may be the starting-point of further rock

breakdown. One reason is that some weathering products,
like an abundant
supply of silts, will choke open pore spaces in the surface horizons of the soil
and tendto favour sheetwash. Conversely, the sandy quartz residue produced
on acid magmatic rocks increases permeability. This characteristic may
become a self-enhancing element in weathering processes as soils of a certain
depth, particularly in lower latitudes, will tend to retain moisture longer than
exposed rock surfaces, and thus sub-soils at the periphery of bare rock
surfaces may even tend to be especially moist due to drainage off the bare
rock. This contrast in chemical weathering may explain the sharpness of the
break at the head of some pediments and the steepening at the base of bare-
rock domes. Accelerated chemical weathering in the moist sub-soil environ-
ment seems assured as long contact of moisture with rock is necessary for
water to penetrate into capillary-sized spaces between crystals. Many writers
have suggested that the chemical weathering of granitic rocks, for instance, is
much slower on exposed outcrops which dry after each rain than on buried
surfaces. In Hong Kong, granite corestones at the surface and at depth
display this contrast and there are similar contrasts in granite and gneiss
weathering in southern South Australia. In older tills a contrast between
sound granite boulders on the surface and disintegrated boulders within the
soil is frequently reported. In theTahoe Till. 25-100 per cent of boulders

have now disintegrated to grus, and although in this case and in others, such
as the weathering of schist on Bjornefjell, near Narvik, there is no great
demonstrable chemical change. On a much larger scale J. Biidel (1957)
accounted for the contrast between broad plains and residual bornhardts in
the tropics by emphasizing, in an hypothesis of a double surface of planation,

the contrast between the essentially arid environment of exposed rocks and
the warm humid sub-soil environment. Although Biidel was one of the first to
appreciate the significance of different self-enhancing processes operating in
adjacent areas to produce contrasted relief effects, both surfaces are usually
too irregular for an hypothesis of a double surface of weathering to account
for planed features.
Clays, apart from retaining moisture, may increase the destructiveness of a
soil environment both by chemical and by mechanical means. An acid clay
may by cation exchange act as an agent of weathering to produce more clay
from silicate parent mineral and rock. Colloidal clay may fill fissures in some
grains and the swelling of such colloidal infilling may accelerate the dis-
ruption of an already fissured crystal. The deposition of iron hydroxide in
networks of microfissures smaller than 3 microns is possible and can exert a
similar disruptive effect, as may the crystallization of caliche.
Finally, the influence which debris may exert on snow or ice can either
tend to accentuate or to reduce some processes just as the surface soil
influences weathering at greater depths. A very thin layer of debris on a

glacier surface will increase melting owing to a lowered albedo. The minimum
thickness of complete protection of underlying ice is 1-2 m. Similarly in-
dividual blocks, if thin enough to be warmed to their base, melt their way into
the underlying ice and form a hole. Thicker blocks act as insulators and form
rock tables as adjacent exposed ice melts down. After nearly one complete
season of ablation the lower end of the Sherman glacier avalanche rested on a
platform of ice 7 m high. The effect of this mantle will be to cause the glacier
to advance beyond the point where the climate of the region can maintain it.
The insulating effect is reversed if subjacent material is unfrozen soil and

pipkrakes develop beneath and lift up particles small enough to transmit
freezing temperatures to the subjacent soil. The presence of an impervious
shield of moisture-saturated ground in a frozen condition may also have very
different effects according to local circumstances. On the one hand there is a
very limited infiltration capacity for surface water, yet if ground below is
unfrozen a river at high stage may build up substantial pressures within the
unfrozen material, particularly if this is sandy favouring the lateral trans-
mission of the pressure. In low-lying periglacial areas surrounded by poorly

drained flats, points on the surface may
domed up by the trapped water
be
under hydrostatic pressure. Uplift rates of small domes in Russia have been
measured at 1-5-2 m/20 years. Mounds entirely in arctic peats are a few
metres high and termed paisas. Pingos are similar domed-up forms of sedi-
ments, usually at least 95 per cent sand and rarely more than 1-2 per cent
silt, which may The degree to which the formation of an ice
include ice layers.
core precedes, accompanies or follows doming is not clear. The overburden is
usually half to a third of the total pingo height. Their sides are usually as steep
as 20-30 degrees but of different orders of size. Some are only 2-4 m high
like paisas, but commonly they are 12—25 m high and 30-60 m long. If the
summit of a pingo bursts the form gradually sags back to form a ridge
encircling a small pool.
10. The concept of quasi-equilibrium
The concept of a quasi-equilibrium between rate of supply of material, its

temporary incorporation into a dynamic depositional landform and its subse-
quent removal from this form, is being applied increasingly in the study of
some bedrock landforms termed surfaces of transportation. The form and
inclination of these surfaces appears to be closely related to the form and
inclination of the mantle of debris where there would be a quasi-equilibrium
between supply and removal. The implication is that the processes of rock
breakdown, the transport and deposition of sediment and the ways in which
depositional landforms develop, may have some decisive influence on
landform development as a whole. However, it is extremely difficult to decide
at present whether, over longer periods of centuries or millennia, aggradation,

degradation, or balance might become increasingly well-established. There is
also the probability of widespread changes during the last few millennia
accompanying man’s removal of natural vegetation covers and in considering
the possibility of balance or grade in streams, the profound implications of
the reality, scale and recency of postglacial sea-level rise are too often com-
pletely ignored. Even where a surface of transportation exists with a formal
resemblance to the overlying or adjacent debris mantle it is very difficult to
specify in detail themechanisms by which this similarity arose. To explain

their development it seems that the most accurate statement which can be
made as yet is simply that some balance between processes may be such that
any reduction in thickness of the debris mantle leads to accelerated processes
of rock breakdown on the exposed bedrock, rather than to self-enhancing
mechanisms which lends progressively greater immunity to the area of
exposed bedrock. Where there is an even solifluvial sheet flatter slopes tend to
be associated with finer mantling material. In the eastern parts of the Paris
Basin, slopes on chalk, where debris is less than 7 cm in size, are only 2-3
degrees, whereas on more resistant limestones furnishing fragments 10-
20 cm in size, the slopes are about 15 degrees. In areas where debris is fine or
at the base of slopes, if diminution of size takes place during downslope

movement, the excavation of debris requires the maintenance of progressively
less steep declivities. Thus current interpretations of pediments regard these
geometric rock-cut surfaces as a transportional or equilibrium surface where
the prevailing hydrological conditions and that declivity provide just suf-
ficient force to keep material moving downslope. A decrease in sediment yield

or in grain size would favour erosion as would a change towards greater
efficiency in runoff regime. If any of these three conditions changed in the
opposite direction aggradation arrests the erosion of the buried rock sur-
face. A great deal of work on actual pediments and slopes will be required
before more specific statements can be made about slopes as transportation
surfaces.
In some areas a theoretical quasi-equilibrium between inter-related pro-
cesses and the development of an area as a whole may be postulated. For
instance, if deflation produces a narrow basin in weakly consolidated rocks it
might be assumed that cannot grow deeper as the velocity of the
this feature
airstreams diminishes rapidly close to its floor. On the other hand, the
broadening of the basin by sheetwash cannot proceed indefinitely unless
excavation of debris by deflation continues. Ultimately, deepening may be
self-arresting because, when lowering reaches close to the water-table level,
increased dampness on the playa floor increases the resistance of the particles
to dislodgement by wind. Similarly, the recession of a coastal clifT depends on
fragments being eroded to sufficiently small sizes to be moved down the wave-
cut platform.
The same conclusion might apply to the similar problem of adjustments
between stream gradient and the entrained sediment load for which theoreti-
cal inter-relationships have been discussed for nearly a century. Again, for
the transportation of coarse debris a critical gradient would seem necessary to
provide the velocity required both to move the material and also to overcome
the increased channel roughness created by coarser debris. Thus it has been
reasoned that in a given segment of a river there is a tendency for gradient to
adjust to a load of given quantity and calibre. However, the controlling
factors are likely to change, and stream diversions, abrupt structural move-
ments, climatic change, or man-accelerated erosion might introduce into a
channel debris of quantities and calibre beyond the transporting competence
of the stream. Deposition then takes place in the upper part of the stream
segment until the steepened gradient provides the velocity required to move
the coarser material. Downstream, as seen on alluvial fans, any reduction in
particle size would require less steep slopes to ensure their removal. These
arguments have been used to explain why the longitudinal profiles of many
streams tend to be concave upward.
The value of theoretical discussions on the concept of the adjusted or
‘graded’ stream is seriously impaired by lack of actual measurements, par-
ticularly of the effects of heterogeneous sizes of particles, the influence of
channel cross-section and the possible role of channel vegetation. If ‘load’
does influence channel gradient, it has yet to be established which parameter
is most significant. It is more likely to be debris of such a size to remain as a
channel-lag deposit, unmoved by the extreme flood rather than the total load,
which, if made up largely of material no coarser than fine silt could, as is the
case in many tropical rivers, be evacuated on the gentlest gradient by compara-
tively small fiows. Some studies have shown that dislodgementand amounts of
transportation in stream channels rather than being correlated with the
factors commonly assumed to be the dominant influences, such as gradient,
velocity, or bank material are closely correlated with the inflow or outflow of
ground water. Substantial if not total inflow may characterize a flood in a
semi-arid area; meander bank caving appears to some investigators to be
related in part to ground-water outflow after a flood. In other areas, par-
some limestone terrains, deep chemical weathering of the sub-
ticularly in
channel zone might reduce the significance of purely mechanical processes.

Apart from neglect of the influence of ground-water disposition, theoretical
discussions on grade are also made less realistic by inattention to the actual
events of the recent geological past, particularly the postglacial rise of sea-level,
and to actual examples which, apart from showing some approximate con-
cavity, inevitably the result of flowing from mountainous source regions on
to surrounding plains, show

resemblance to the smooth curve of the
little
theoretical equilibrium profile. For instance, in contrast to some rivers in

central Africa which have pronounced falls comparatively close to the sea,
the Orinoco flows through an alluvial plain as a broad braided stream for
more than 2000 km of its 2500 km length at gradients generally less than
1 15 000. The gradient of the lower Amazon is 1 -.50 000 and even in
:
headstream areas in north-east Bolivia, gradients are similarly very low and
aggradation by the silt-laden waters has forced streams like the Beni and Rio
Grande into major diversions. On a smaller scale, rivers in western Malaya
have gently sloping often swampy valley floors penetrating far inland.
It seems that in view of the many sources of variation and the relative
paucity of actual observations any attempt to generalize here about the inter-
relationship between hydraulic and sedimentological parameters of a stream
and its gradient would be premature speculation.
E. Inter-relationships between Form and Process
The Lake District mountains. '. . . were it not that the destructive agency must
abate as the heights diminish would, in time to come, be levelled with the plains."
William Wordsworth ( 1 822)
In many situations one of the major influences conditioning the course of

erosional and depositional processes is the characteristic of the existing land-
forms themselves, posing a fundamental problem in the construction of con-
cepts of landform development. On the largest scale, and within the pattern of
the general atmospheric circulation, it is the major, structurally generated
relief forms which largely control the distribution of glacial, humid, and arid
environments. Most arid regions lie to the lee of mountain ranges which bar
the passage of moisture-laden air currents and the detail of the course of
climatic change may reflect similar influences. R. F. Flint (1943) believes that
glacial advances in North America terminated farther north in the Dakota
area than in the Mississippi lowland, not only because increasing altitudes
hindered advance into this western area but also because of the rain-shadow
effect of the Rocky Mountains. Even on the smallest scales, forms and
features may show a certain inertia to changes and persist to influence the
effectiveness of processes established under the new conditions.
The greater the available relief the greater the hydraulic potential and
kinetic energy. Also, where horizontal distance is limited the propensity for
streams to deepen their channels is also greater. Conversely, ‘the degradation
of the last few inches of a broad area of land above the level of the sea would
require a longer time than all the lOOO’s of feet that might have been above it’
(J. W. Powell, 1875). It is therefore widely assumed that a rate of erosion
tends to be proportional to the average elevation above sea-level. It is,
however, often extremely difficult to identify, where some correlation between

altitude and a geomorphological phenomenon exists, the exact nature of the
mechanisms which the factor of altitude changes. It is even difficult to general-
ize about the well-understood effect of altitude in producing orographic

rainfall on the uplands and in leading to greater evapotranspiration losses
from lower areas. One complicating factor is that precipitation amounts

depend very much on the closeness of the mountainous areas to the sea, like
the situation producing the intense downpours in Central America and Java.
most obvious controls of landform distribution reflects the
In fact, one of the
progressive eastward rise of snow-lines in mountainous areas on western
maritime fringes in higher latitudes. In many such areas the elevation of
cirques show a marked and consistent eastward At 41°N cirque eleva-
rise.
tions in northern California rise eastward at about 3-4 m/km. At 39—41° in
the Argentine Andes similar gradients range from 9 to 16 m/km. In the Ruby
Range in Alaska the inland rise is about 7-3 m/km (fig. IV.43).
There are some actual measurements which suggest that greater mechani-
cal erosion does take place with higher elevations. A. Cailleux (1948) sug-
gested that denudation in mountainous areas might be up to ten times greater
than that from adjacent plains. It is never easy, however, to decide whether it
is primarily due to greater kinetic energy of available relief, that higher areas
tend to be steeper so that altitude is merely an indirect expression of steep-
ness, whether disintegration processes are more intense at greater altitudes, or
whether the significance of chemical weathering might be greater in warmer
lowland environments. A classic problem is that of an area of valley glacia-
tion and whether erosional activity is greater on the ice-free peaks exposed to
freeze-thaw than on the valley floors beneath ice tongues. On the other hand,
water equivalents on watershed divides may be minimal in late spring thaws
because snow depths are minimal on such sites. Ridges therefore might be ice-
free because they are unsuitable for ice-cap accumulation not because they are
the product of ice erosion. One example of a measured relationship between
altitude and denudation comes from south-west United States, where some of
the thickest alluvial fans occur at the outlets of basins which in parts exceed
2250 m in maximum elevation.
Rivers on the flat swampy lowlands on the left
bank of the Dneiper have sediment concentrations of less than 20 gm/m^,
whereas in headwaters on the north-east slope of the Carpathians turbidities
increase to 250-500 gm/m^. A further point is that much of the material
eroded from upland areas is evacuated whereas the amount of material that is
subsequently removed from flat terrain may be only about half of that dis-
lodged. One effect of rugged mountainous relief is to make chemical weather-
ing relatively negligible, partly because debris is removed
too rapidly by
kilometres
Figure IV.43 Altitudinal relationships of cirques.
A. Scotland, south of National Grid line 700 {from Linion, ]959); distance
measured eastward from the National Grid line 100 east.
B. St Elias Mountains, from the Malaspina Glacier to Ruby Range {from Denton
and Stuiver. I966)\ distance measured eastward from the Gulf of Alaska,
C. Chile-Argentine Andes, in the latitude belt 4 1

° 00'-4 1
° 20' from Llao-Lao to
Bariloche {from Flint and Fidalgo. J964); distance measured eastward from long.
72° 00' W.
mechanical processes, even where there may be a combination of climatic

factors favourable to chemical weathering. Springs in the highland interior of
Hawaii may have only 2-5 ppm silica in solution, whereas in shallow wells on
the coastal plain values may rise to 30 ppm. In steeper areas water runs off
quickly and in cooler environments snow cover provides an increasingly large
proportion of the runoff, adding to the reasons why the relative importance of
solution should decline as gradient and altitude increase (fig. IV.44). Also,
reduced vegetation growth at higher altitudes leads to slower rates and
amounts of biological activity and leaves the ground more exposed. On
exposed summits above 500 m in the Southern Uplands of Scotland, the
vegetation consists of Rhacomitriiim heath and contains stunted grass species
such as Festuca vivipara, offering little insulating cover. In less severe condi-
tions below 500 m ericaceous species like Calluna vulgaris and grass species
like Nardus stricta form a much tougher surface mat.
There have been many speculations on how steepness of slopes might affect
erosional processes. W. Penck supposed that the loss of soil mass on a slope
was exclusively a function of gradient, and L. D. Baver observed the relation
E — 0-065 5'“*’, where E is total soil loss in tons per acre and S is slope
3 5-
y/ — 600
'
900
'
1000
altitude (metres)
Relative altitude (rnetres)
D
rng/l 1 Screes
2 Avolonches &. high mountom lorms
in 3 Sheetwash
4 Gullying
Mineralisation
'-7
7. I-
- '
'
1000 1500 2000 2500 3000

altitude (metres) altitude (metres)
Figure IV.44 Changes in bedrock, soil, solution, and surface form characteristics
with increasing altitude.
A. Increases in porosity of granites in central North Scotland {after Godard,

1965).
B. Increase in pH at the mineral surface (solid dots) in alpine soils in north-east

Scotland {from J. C. C. Romans et al.,1966, Jour. Soil Sci., Vol. 1 7). There is no
progressive change in pH 60 cm below the mineral surface (open circles).
C. Decrease in mineralisation of stream water in eastern Georgia, U.S.S.R {from
Z. I. Chantladze, 1965, Soviet Hydrology). Altitudes are for the average elevation
of watersheds, I, for the rivers Khadis-Khevi, Black Aragvi, Tama, and Lekhura,
and II, for the Ktsiya-Khrami, Alazani, and Aragvi,
D. Distribution of certain processes and associated forms in the Barcelonnette
area {after L. Oilman, 1961, Mem et Doc. Centre Rech. Doc. Cartog. Geog., Ed.
C.N.R. S., Vol. 8).

gradient in per cent. One difficulty in generalizing about the effect of slope
steepness on erosional processes is that beyond certain declivities certain
factors diminish in significance or cease to operate. Snow, for instance, will

not remain on very steep slopes exceeding perhaps 40 degrees and in conse-
quence nivation processes are less effective in these comparatively dry en-
vironments. On steeper slopes the effectiveness of precipitation may tend to
be greater because more circulation of surface and underground water reduces
water losses by evapotranspiration. A more pronounced tendency, measured
on slopes in Japan with declivities between 1 5 and 45 degrees, is for infiltra-
tion to decrease with increasing slope gradient. Landslips inevitably occur
only where gradients are steeper than a critical value. During late autumn
storms in 1965, landslips occurred in the San Gabriel Mountains in
California only on slopes steeper than 38 degrees. In coastal areas the off-
shore gradient can be one of the most significant factors influencing erosional
processes. When Hurricane Audrey flooded west Louisiana coasts in 1957,
the width of inundated land was as much as 50 km because the continental
shelf here is 1 60 km wide, whereas flooding along a coast flanked by deep
water would have been insignificant. Another way in which the shape of the
water body being filled is significant is in its on the form of a delta.
influence
Apart from ways in which steepness and elevation of relief form can
influence their own erosion, the pattern of an existing landform may have a
profound influence on depositional processes due to perturbations introduced
into flow-lines of transportational fluids. Flow-lines are greatly influenced by
slight variations of ground surface and therefore become erratic and unpre-
dictable. An obstacle decreases the cross-section through which the lower
layers of the atmosphere or water-flow must pass. When flow-lines converge
in by-passing an obstacle the fluid accelerates and the drag velocity is greater
close to the ground surface than higher up. Conversely, on the lee-side of the
obstacle, flow-lines diverge, the fluid motion is retarded, standing waves of air
or water may be set up, and deposition may result. When transportation is
particularly susceptible to deceleration in reduced fluid velocities, deposits

elongate in shape often taper downcurrent from obstacles. Linear hills in
the lee of pre-existing hills, termed zavieja occur in loessial areas of

Czechoslovakia. In some areas, a systematic linear pattern may cover
100 000 km^ and influence drainage patterns, as in the Palouse region of the
Columbia basin.
The same three characteristic features often occur where an obstacle inter-
feres with flow-lines, even down to the smallest-scale features generated by a
pebble on the beach, desert floor, or in a stream bed. There is, first, a resistant
object standing above the surrounding surface. Secondly, a crescent-shaped
furrow is cut in front, or to one or both sides, of the obstacle by accelerating
currents. Thirdly, there is a tapering ridge of sediment deposited or left as a
protected remnant in the lee of the obstacle. Material may start to accumulate
on the side exposed to the current if the obstacle is low and squat, as the
horizontal divergence of airstreams is stronger than the vertical convergence.
A large number of depositional landforms appear to string out behind a rocky
protuberance. All the complex esker systems near the Casement glacier are
on the lee-side of a series of solid rock ridges 60-90 m where ice might
high,
have been stagnant for a considerable period. In the vicinity of Sebhah, in the
Libyan desert, seif dunes appear to originate where either vegetation or small
rock masses, or both, form sand-traps rising above the general level of the
sand sea. Arrow spits may develop in the lee of an island, like the ‘comet-tails’
off the Brittany coast. In areas of lowland glaciation ‘crag and tail’ features
form one of the best examples of the lee-effect depositional landforms and
small morainic ridges may even form in the shelter of individual boulders. In
Kluane drift in Alaska, till ridges to the leeward of moulded bedrock knobs
and ridges may be up to 10 km in length. Such distal accumulations are
sometimes termed lee-moraines. Relief prominences on the sea floor can, in
some instances, form a nucleus for barrier island development during emer-
gence or sediments may accumulate in intertidal areas protected by offshore
islands.
In some instances an obstruction may be sufficiently extensive to be a

barrier rather than an isolated nucleus precipitating a lee-side effect. In
Hardanger fiord, the position of distinct terminal moraines of late-glacial
readvance appears to reflect the position of the rock threshold acting as a
barrier.
Some initial relief irregularities often influence the track of transportational

movements without, particularly on larger scales, producing a clearly defined
depositional form. Such irregularities often influence the splitting and closing
of channels and the formation of central and marginal delta basins. If the
position of an obstacle in a stream is asymmetrical, an obliquely orientated
line of scour starts upstream due to deflection of floodwaters. Similar diver-
sions of ice-flow may lead to fluting and grooving of adjacent surfaces by the
deflected ice. On the smallest scale the processes of ice-flow round an obstacle
were observed in relation to a 2-5-cm cube bolted on
bedrock at the to the
base of the Casement glacier. Ice flowed over and around the cube, revealing
a thin la3'er of clear ice and a cavity on the down-glacier side 1 cm narrower
than the block itself. On a larger scale, comparatively high ground has often
provided an obstacle in East Y orkshire, forcing ice-flow mainly to the east of
Flamborough Head. In was able to fan out
the lee of the headland the ice
westward over the low-lying ground of Holderness, though in southern
Holderness the south-south-west regional direction of movement was main-
tained (fig. IV.45). On a larger scale still, many of the lobes of ice that
advanced in eastern North America were influenced, in position and direction
Figure IV.45 Features produced by erosional reworking and depositional pro-
cesses in the proximity to obstructions and in their lee.
A. Scour marks in snow (from J. R. L. Allen, 1965, Jour. Sed. Petrol., Vol. 35).
(Left) View from above of flow pattern in symmetrical current crescent of snow;
M = resistant snow mass, R = ridge of deposited snow, F = furrow eroded round
resistant snow mass. (Right) Flow pattern in longitudinal section; S, = first fall
snow, Sj = second fall snow.
B. Creation of current crescents of sand by a pebble deflecting the direction of

stream-flow (after S. Sengupta, 1966, Jour. Sed. Petrol., Vol. 36).
C. Secondary ice-scour directions due to deflection by unevenness in the bedrock

surface (from M. Demorest, 1938, Jour. Geol., Vol. 46).
D. The effect of the Chalk escarpment on directions of ice movement in East

Yorkshire (after Penny and Catt, 1967). This was sufficiently pronounced
effect
for successive ice-sheets invading the area to follow the same pattern of movement.
of movement, by the major bedrock relief of the area. Many submarine

canyons occur on the upcurrent side of points of land, the decrease in
transportability of sediments by longshore currents leading to accumulations
which eventually become unstable and move seaward. Some lagoon entrances
are ‘rock defended’, being close to offshore reefs or fore.shore rock outcrops
which break-up constructive waves and prevent the completion of a barrier to
the lee of this point. E. C. F. Bird (1968) cites the example of Lake Illawarra,
protected by Windang Island, immediately offshore from the New South
Wales coast.
A gap in a barrier produces an analogous effect on flow-lines to that of the

local acceleration around an obstacle followed by deceleration in its lee, with
acceleration and increased competence through the constriction and subse-

quent deposition in the diverging flow downcurrent from the constriction
(fig. IV.46). In many rocky areas partially covered by moving sand, like the
Cronese mountain area in California, the sand may assume unusual shapes,
mainly due to the funnelling influence of local relief features. Sandy areas
themselves may in turn be due to the funnelling of sand on to a small area by a
local constriction in the relief barrier, like the Salton dunes in the Imperial
valley. The frictional forces of valley walls on moving ice is often noted, and
A. P. Crary’s (1966) measurements in Antarctica suggest that the rate of
thickening inland is a function of the width of the downvalley end, thickening
inland from a 10-km-wide fiord being at a rate of about 20 m/km. An initial
depression ip the ground surface may tend to be the site of progressive

enlargement of a hollow form as appears to happen in the enlargement of
dolines in karst areas. Instances of the opposite effect have also been
recorded. A striking feature of parts of the Columbia lava plateau, stripped of
a former loess cover, is the small mounds 3-10 m across and approximately
1 m high. They occur above depressions in the basalt surface which appear to
have trapped dust, favouring in turn the growth of vegetation which continued
to gather and hold more dust. After the same process had operated for
thousands of years the mounds gradually rose above the rims of the original
depression. .
A finalinfluence of landform on land-forming processes is seen where

certain forms develop only in specific geomorphological sites, although in
They were all forced to flow mainly to the east of Flamborough Head, north-east
of Bridlington.
E- A gorse tussock
building up in the lee of a granite block in the Scilly Isles
{from E. Lenze, 1966, Erdkunde, Vol.
20). Similar features may also be produced
on slopes (see fig. IV.
14), there is classic ‘crag-and-tail’ (fig. IV. 23) and in contrast
to (E), above, fig.
III. 14 shows how vegetation can obstruct as well as flourish in
shelter.
Figure IV.46 Snowdrift ice slabs to the lee of a discontinuous bedrock obstruc-
tion {after Nichols. I960). The slabs, observed on the east side of Stonington
Island, Antarctica, form by the accumulation of wind-blow snow like sand
shadows.
many instances such associations may pass unnoticed. Such restrictions apply
to many glacial features. In the north European plain, ice-pushed ridges show
a marked correlation with river valleys, having been pushed from the valley
axis outwards. In the Illinois drift area, bulky crevasse deposits and narrow
segmented aligned ridges occur on plateau tops but not where the ice over-
rode a rolling relief like that of the Mount Vernon hill-country. Conversely,
because the length of sand-grain trajectories increases over a rough surface,

sand-dunes may fail to develop in such places, although much sand may move
across the area. Cuspate spits, dependent on currents from two directions,
occur only in narrow bodies of water where the width is about half or less of
their length. At the most obvious level, a hillslope facing a plain is more likely
to have a pronounced concave footslope than one facing the opposite side of a
valley.
F. The Significance of Changes in Weather and in Climate
1. Local changes due to differing aspects
As much a continuation of the discussion of the influences of landform on

process, as a starting-point for considering changes in climate, is the changes
in land-forming processes accompanying shifts in the compass direction

towards which land surfaces face. A
major influence is the orientation in
relation to the sun’s position. A secondary factor, frequently superimposed on
the effects of the first is orientation with regard to the direction of prevailing
moisture-bearing winds. Thirdly, wind direction may be important where
conditions are dry enough and winds sufficiently strong for transportation of
material by wind to be significant on land, and on all shores the length and
direction of fetch are always major considerations.
On the 10-degree slopes of a valley in the lowlands of southern England,
summer temperatures tend to increase up both north- and south-facing slopes.
However, while the north-facing slope may increase by 4°-8°C, upper parts
of south-facing slopes may be 5°-20°C higher at the ground surface. In
winter temperature contrasts between valley floor and sides or between
opposed aspects may be small, although on brighter days temperatures may
be as much as 6° C higher on the south-facing slope. For inclined surfaces in
higher latitudes it is more
variations in intensity of cold that might be
significant, but it is difficulthow mechanical weathering might be
to specify
affected. At Valday in the Russian plains, with a December-March average
temperature of — 7- 7° C, soil froze to a depth of 3 cm according to measure-

1
ments on the north-facing slope of a ravine a depth of 22 cm on

in contrast to
the south-facing slope, and the water stored in driftingsnow was 30 per cent
greater. The consequent greater spring runoff amounted to an extra 50 mm on
the north-facing slope. In the Mackenzie delta area, disintegration by frost in
east-west gullies appears to be more intense on north-facing slopes. In the
Jura, by contrast, greater accumulation of debris at the foot of south-facing
slopes might be due to greater amplitude of temperature changes and more
diurnal freeze-thaw cycles. Yet in the Kenai peninsula in Alaska, south-facing
slopes are steeper than longer, highly dissected north-facing valley-sides
where winter snow lasts into June. At Resolute in North West Territories,
asymmetry is reversed, perhaps because the maximum
thawing of permafrost
on south-facing slopes favours solifluction but without conditions becoming
sufficiently mild to encourage vegetation growth.
Although the relationship of land surface to the direction of sun’s rays is

more rigorously invariable over spans of geomorphological time, situation in
relation to prevailing winds is an equally important inter-related factor. This
is mainly as a control on the supply of moisture on a
regional scale. The
rainfall range in Puerto Rico is from 750 mm on the south coast to 2500 mm
in the north-east, due to the prevalence of the north-east trades,
and appears
to correspond with the transition from knife-edge ridges with 30-45-degree
slopes, clayey soils, and the sheetwash and gullying processes which accom-
pany torrential rains, to the rolling hills and more rocky soils of the drier
south. Although it is possible to point to many examples where a leeward

position is arid, like north-east Iceland, eastern southern Italy, or east South
Island, New Zealand, it is easier to indicate instances in cool environments

where the influence of snow drifting into a leeward position appears to
produce the opposite effect, with clear morphogenetic results.
Within a given small area aspect is a crucial factor in its control on soil
moisture as exposure to sunshine accelerates evapotranspiration processes.

In the Salskaya steppe, the contrast in an undisturbed area was 97 per cent
average soil moisture for a north-facing slope and 80 per cent for a south-
facing one. Under a 75 per cent afforested cover soil moisture was slightly
higher but the contrast was reduced, comparable figures being 102 and 90 per
cent. In sub-humid areas there is sometimes a correlation between steeper
north-facing slopes, their soil moisture conditions and denser vegetation. In
San Diego county, vegetation reduces the amount of effective precipitation,
runoff, and the erosional impact of falling rain, and exerts a binding force on
the soil. Generally, however, the effect of shade on a northern exposure is to
favour higher soil moisture contents despite the concomitant increase in
vegetation growth. In meadow chernozem soils the spring moisture deficit on

a north-facing slope may be less than a third that on south-facing slopes. In
cooler environments where increased exposure to sun’s rays does not lead to
critical losses of soil moisture, biochemical processes, at least, are more
active on the south-facing slope. In the Medicine Bow Mountains. Wyoming,
at 40°N and at an altitude of more than 3300 m, productivity on a south-
facing slope with 73 per cent plant coverage was measured at 585 kg/ha.
compared with 224 kg/ha on the north-facing slopes with 54 per cent cover-
age. In parts ofSweden under natural pasture, south-facing slopes, whether
20 degrees or 4 degrees, are subject to less leaching, have higher pH values,
greater pore space, water capacity, and infiltration rates, and brown earths
may form, whereas on the north-facing slope, where evapotranspiration is
less, podzolization prevails.

Added to the complicated contrasts between processes operating on
opposed slopes is the possibility that greater sediment production on one side
might force the valley floor stream against the opposite bank and the under-
cutting leading to its steepening. In other instances it is difficult to decide
whether present-day asymmetry of form is attributable to contemporary
contrasts in processes or whether the greater steepness of one side is a relict
feature of past conditions. For instance, in the Bitterroot Range in western
1
Montana, the north-facing wall is much dissected by cirques and hanging

valleys, whereas there is evidence of only gullying on south-facing slopes.
C. B. Beaty (1962), by counting the mass-movement phenomena, active rill-
and-gully systems, and nivation depressions estimates that contem-

porary gradational processes are perhaps two to three times as active on
the shaded north-facing slopes. Table IV. 1 1 lists the contrast in mean slope
angles.
Table IV. 1 Slope inclinations of selected trunk canyons in the Central Bitterroot
Range {jiom Beaty, 1963)
DRAINAGE SOUTH WALL NORTH WALL

(facing north) (facing south)
Bear Creek 17 23
Fred Burr Creek 22 27
Mill Creek 21 27
Blodgett Creek 18 25
Sawtooth Creek 17 24
Roaring Lion Creek 21 35
In cold environments aspect may be critical in determining where ice

accumulates. As well as exposure to sun’s rays, position in relation to prevail-
ing winds is again significant because of the lee-side accumulation of drifting
snow. For example, on the Decade glacier, Baffin Island, there is a mountain
ridge running almost parallel to the south-west margin of the glacier which
probably leads to the snow accumulation maximum in its lee (fig. IV.47). In
the Alaskan Range the last remnants of the dissipating glaciers are usually
preserved only on slopes descending north and north-westwards, although in
terms of climatic statistics the southern side is wetter and cloudier. Between
46° and 5 1°S the Patagonian ice-fields show a marked contrast between the
western slopes with a maritime climate and the abrupt transition eastward to
continental conditions.About 1945 the contrast was particularly marked
with western glacier fronts close to the forest edges, whereas virtually all
eastern glaciers had retreatedmarkedly from recent end moraines. In east-
central Africa, former glaciers descended loweron the eastern than on the
western flanks of Mount Elgon and Mount Ruwenzori. A similar relationship
appears to have existed on Kilimanjaro during the last major glaciation,
but altitudes reached on the southern flank were even lower than
those on the
east
There are many landform contrasts which reflect the cumulative
effects of
contrasting processes on land surfaces of different
orientation. Some are
perhaps too readily overlooked, some studied only in
one or two instances,
Figure IV.47 Influence of aspect on glacier accumulation and disposition.
A. Ice accumulation map I964-S of the Decade glacier, Baffin Island (from G.
0strem et al., 1967, Geografiska Annaler, VoL49A) showing the effect of a
mountain ridge, not shown on the map, running almost parallel to the south-
western outline of the glacier in causing the snow accumulation maximum in the
leeward position along the left-hand part of the map.

B-D. Polar diagrams showing the relation between cirque floor altitude
(concentric circles) and the preferred north-easterly aspect (radii).
B. Northern Nain-Okak part of Labrador (/rom A T. Andrews, 1965. Geog. Bull.,
Vol 7) circles 0-9 5 m.

1
C. North Caernarvonshire, North Wales (/rotn Seddon, 1957) circles 270-750m.

D. West-central Lake District. North-west England (from Temple, 1965) circles
300-780 m.
1
but for others, like the glacial cirque, the influence of aspect has been widely
studied and its striking degree of control clearly established. One of the first
observations was made nearly a century ago when approximately 70 per cent
of the cirques in part of the Jotunheim, in Norway, were seen to lie on the
north side of the massif. For large parts of the mountains in northern
Scandinavia the orientation of glaciers and snow patches, as well as cirques, is
generally eastward. Near Nain, Labrador, the comparable figure is 89 per
cent, with 72 per cent of the cirques orientated between north and east. In the
Trinity Alps. California, 45 per cent face north, 20 per cent north-east, and
1
5 per cent north-west. Comparable figures for the north-west Highlands of
Scotland are 24, 31, and 5, with 16 per cent facing east. Generally, cirque
development in the northern hemisphere tends to be on north-facing slopes
with few facing south. In addition to examples mentioned already, other
instances include the Uludag Massif at 40° N in Turkey, the high land above
the South Fork of the Coeur d’Alene river, similarly in the Bitterroot Range
farther to the east in Montana and in many valleys in the Scottish Highlands
like the Great Glen and Glen Coe, or in the Vosges. There are local variations
in the preferred orientation as in Newfoundland where eastern cirques are
neither as well developed nor as large as those in the west, but nonetheless
fig. IV.47.B-D illustrates typical patterns.
The influence of aspect on the pattern of glacier alimentation is sufficient
for its possible influence on the direction of ice-orientated features to be
considered. Leverett (1929), for instance, suggested that the ice-sheets of the
North American glaciations tended to advance westward because storms
came from the south-west. Other features associated with ice may show a
preferred orientation, particularly towards the north-east. Nearly all the end
moraines in the north-east part of the Jotunheimen are ice-cored, whereas all
end moraines on the south-west side are ice-free. Another example, again due
to snow drifting by westerly winds, is the strong predominance of avalanches
on lee-side slopes. These are mainly east-facing in northern Lapland. In south-
central New Mexico, rock glaciers in the San Mateo mountains occur on
slopes with north or north-west exposures, protected from the sun’s rays. On
Niwot Ridge, 24 km west of Boulder, Colorado, the restriction of stone-
banked terraces to lee-slopes suggests that wind-drifted snow supplied the
moisture which facilitated movement. However, in highland areas in eastern
Australia, similar features are restricted to the upper portions of windswept
south-west to north-west exposures, and the steps may even have an ofT-
contour dip towards the wind. Landslides are often characteristic of north-
facing slopes in the northern hemisphere. In some valleys in eastern Montana,
nearly all the landslide blocks are on escarpments facing north-east.
Farther
south and west the thickest alluvial fans tend to be associated with
basins
having a north or north-east aspect. As these last few examples
show, milder
Figure IV.48
A. Profile across a ravine in a chernozem area, Central Russian Highland {after
V. V. Gertsyk, 1966, Soviet Soil Science)
and rock glaciers in two areas in the Alaskan range,

B. Altitudes of glaciers
showing contrasts according to aspect {from Wahrhaftig and Cox, 1959).
C. Contrast in depth of blanket peat according to aspect, north-east Scotland

(from R. Glentworth and J. fV. Muir, 1963, The soils of the country around
Aberdeen, Inverurie and Fraserburgh).
'
ice-free environments still show strong correlations between landform and
aspect, although the processes involved are usually inferred rather than
separately and specifically studied.
Exposure to wind itself be a further factor where wind action causes
may in
erosion, transportation, and deposition of weathered materials. After 6 years
of exposure to natural sandblast, the exposed face of a brick in an experi-
mental site set up by R. P. Sharp (1964), was cut back a maximum of 4 cm on
its westerly face, whereas the north and east faces showed no wear. In wooded
areas certain exposed points and ridges can be particularly susceptible to
root-disruption by wind-thrown trees. Perhaps fetch in relation to sea and
lake shores provides the clearest controlled effect of exposure to wind direc-
tion.On coasts with rocky headlands, the more extensive abrasional plat-
forms may develop where there is some shelter from the direction of domi-
nant wave activity. In intertropical atolls the importance of aspect is seen
most clearly, with the more irregular, open side on the protected leeward
shores. J. A. Steers (1937) observed that the Australian sand cays are orien-
tated to the east, at about 45 degrees to the prevailing south-east wind. Steep
ridges of coral rubble occur on the exposed sides of sand cays facing the
heavier waves, with finer material and depositional forms on the leeward side.
Wind direction may control lake shape in certain circumstances. In the
coastal plains of Alaska, the remarkably systematic elongation of lakes at
right-angles to prevailing wind directions appears to be due to differential
thawing of permafrost by wind-driven waves and eddies.
Many depositional landforms influenced by wind direction have already
been discussed. If there is any geomorphic significance in wind direction
and the transportation of airborne salts, such as sodium and chloride
from windblown sea spray in coastal areas, this awaits investigation. But
with reference to sources of moisture and heat, the factors of aspect and
exposure in landform studies cannot be underestimated nor too carefully
investigated.
2. Changes in the effectiveness of processes with time
Study of short-term hourly, daily, or seasonal changes in the intensity of

processes is an integral part of landform study. Whether the time interval is
minutes or millions of years the relative intensity of processes changes, if not
the nature of a process itself. Some patterns of change tend to be
self-
regulating. like the summer replacement of beach sand removed in winter
storms. In other instances the very effectiveness of a process depends
on the
cor^lementary effect of variîons in processes. Pipkrake^deyelopment
dep ends bn fros t at night and thaw in the mo rning. .and is most effecfive^when
similar cycles last a few days. The elTectiveness of erosion may well be
9
(inches)
precipitation
Cumuiotive
Figure IV. 4 Variations in the intensity of transportational and erosional pro-

cesses.
A. Variations in suspended clay suites of the Arkansas river at Ponca City,

Oklahoma, during a major storm (from C. E. Weaver, 1967, Jour. Sed. Petrol.,
Vol, 37),
B. Cumulative movement of Threatening Rock (solid line) and cumulative pre-

cipitation (dashed line), Chaco Canyon (from S. A. Schumm and R. J. Charley,
1964, Am. Jour. Sci., Vol. 262). Vertical lines indicate movement during periods
of freeze-thaw, f.
C. Seasonal contrast in amounts of material transported off grassland slopes by

rillwash (from T. Gerlach in Macar, 1966).
incr easedby the dominance of mechan icaLw-eathering in coo ler parts of the
year or in a glacial perio d,_and_, ch emica l weathering iri warmer par ts of the
year or during an jnter- gl acial .
tKb sFudyof short-term changes in the intensity of processes is necessary,
gauge the representativeness of observations contribut-

in the first instance, to
ing to an annual mean. Secondly, a clarification of the nature of short-term

changes may be the most reliable approach from which to comment on more
scanty data related to changes over much longer spans of time. Despite some
obvious differences, the changes in a 24-hour day have some similarity to
those within a year; the year might be a useful model for a part of Pleistocene
time, as might year-to-year changes in considering the order of change from
one millennium to the next.
Small-scale features in unconsolidated material may be changed in a matter
of minutes. A
wind of 30 km/hr or more is sufficient to reverse the asym-
metry of a sand ripple in a minute or two. Certain parts of the day may be
more critical than others, particularly where biochemical processes are
involved. Precipitation of calcium carbonate is greater during the day than at
night. Flow of glaciers in northern Greenland is 5-10 per cent more rapid
during the day. For less obvious reasons landslide movements in the Upper
Columbia river basin appear to be most frequent about 3 a.m. J. Tricart
(1965), observing that in intertropical latitudes even small glaciers have
huge terminal moraines, but that fluvioglacial forms are very poorly
developed, attributed this contrast to the reduced effectiveness of melt water
produced by diurnal thaw compared with that of the seasonal thaw in mid-
latitudes.
Day-to-day changes within a given season depend on short-term extremes

of temperature or precipitation. In stream transport changes may even depend
on where, in a large catchment, rain falls. Studies of the Arkansas river at
Ponca City show ed th at stonnsjnr^prthern Oklahorna and central Kansas
increase'd^"tlTie"‘ôunt of mixed-layer ,illite-montmonllonite,in_the river,
whereas .stor ms farther, to the west, in the Rocky Mountain foothills, even-
tually cause a major increase in the lilite in the suspended load (fig. IV.49.A).
Seasonal factors can be among the most significant events of geomorpho-
logical processes. In northern India the silt load other than in the monsoon
season is a trivial fraction of the whole, an important factor being the abrupt
increase in discharge which may still be relatively feeble in June. It is impos-
sible to generalize about the time of year when sediment removal is greatest
D and E. Differences in frost-shatter between summer and winter conditions

in
cavities beneath a Norwegian glacier (from R. G. Bennett, 1968, Norsk geog.
Tidssk.. Vol. 22), as observed under the margins
of Osterdalsisen, Svartisen,
Norway.
6
Figure IV.SO Changes in position, size, and number of rip currents due to
changes in sea condition {after P. McKeitde, 1958, Jour. Geol., Vol.66). Dee
Why beach. New South Wales. (R) indicates positions of rips.
A. Normal positions under steady northeasterly winds.
B. Changes in position as tide falls, before the currents become consolidated in
their channels.
C. Enlargement, joining together, and moving south of rips under heavy storm
from the north east.
D. Joining together and moving north as moderate swell from the south increases
to storm.
because of the differing times at which controlling factors are most effective.
A study of bank cutting near Rockville, Maryland, revealed that although the
largest discharge occurred in summer, soil wetting was more thorough in
winter, which, in combination with freeze—thaw, led to maximum bank ero-
sion. In Baja California, typical summer winds are unidirectional and
onshore due to the development of low pressure over the inland desert. The
winter wind pattern is bi-directional, with the strongest sea-breeze of about
9 m/sec compared with a summer maximum of 12m/sec in mid-afternoon.

Associated movements of sand are about 8 cm/day in summer, but only

2 cm/day in winter. On some coasts the pattern of drifting varies seasonally.
On Boomer Beach, California, waves from the south-west move sand north-
ward summer and waves from the north-west drive it back in winter. A
in
summer beach in many parts of the world is higher, wider, and composed of
finer sediments. In stormier winter conditions high, steep-fronted waves tend
to remove much of the beach and leave low-level flat areas covered with a lag
of coarse material. On Hawaiian beaches the average increase in grain size is
about 0-5 phi units. In general, drift rates in periods of heavy surf in winter-
spring months may be three to ten times greater than those prevailing during
the major portion of the year. In higher latitudes, although seasonal contrasts
remain, the effect of freezing spray and frozen sand is to protect a beach.
However, with little infiltration or surface roughness, uprush extends further

than under summer conditions. Deposits may reflect seasonal changes too,
particularly in cooler environments with a rhythm from summer silt to clay
in winter. The clay settles out when flows are low and lakes ice-covered.
Compactness of surface materials may vary seasonally. The greater activity
summer probably explains why a sandy beach may be
of benthic animals in
significantly less compact in summer than in winter. In contrast, the badland
slopes studied by S. A. Schumm (1956) are compacted in summer by rain-
beat, become less permeable and the proportion of surface runoff increases.
On a larger scale, while washboard or de Geer moraines were assigned too
hastily the genetic name, annual moraines, the possibility of their deposition
annually remains. J. T. Andrews and Smithson (1966) describe some
B. B.
670-790 moraines in north-central Baffin Island where deglaciation lasted
700 ± 180 years.
The disappearance of seasonal snow cover marks a drastic change in the
physical and biochemical characteristics of the earth-atmosphere interface
and the way in which interactions operate. Spring thaw is a time of consider-
able transportational activity, even if the debris entrained was loosened at
some earlier period. In seasonally frozen streams, a distinctive pattern of
sediment transport may follow the spring thaw. Typical of many streams of
this type is the Colville river, Alaska, which is frozen for 7-8 months in a
year. A distinctive characteristic of the first flood is the very low content of
solids per unit volume of snow melt water, even at a high stage (fig. IV.5 1).
During summer the fluctuations in suspended silt correspond fairly closely
with discharge and thus with precipitation, but with the maximum load per
unit volume tending to occur before the peak discharge. Turbidity is greatest
when all the catchment thaws. However, later in the summer, the silt content
is virtually the same whether the stage is rising or falling. Turbidity is
reduced as frost begins to affect alpine heaths and because the first floods,
once the soil melted, had already washed out much of the loose
material.
3 1 8 Introduction to Geomorphology
24 tt> -31st May )st -Bth June
a 100 200 100 200 300
•o
Centimetres above sea level
Figure IV. 51 Changes in suspended inorganic load in relation to water stage at

different seasons {from Arnborg et al., 1967), illustrating the effect of the 1962
thaw on the Colville river. Alaska.
Dissolved loads of streams in cooler environments may also show sharp

changes as snow melts. Fig. IV.52 illustrates the case of the Pyalitsa, a
lowland river in the U.S.S.R. With snow mq'lt there is an abrupt flood with
associated reduction in mineralization of the water (line 2, fig. IV.52) and
rapid variation in the proportion of the mam
Sodium follows the chlor-
ions.
ide curve (line 3) as both are typically present in water from a snow cover
whereas the bicarbonate ions show corresponding falls. As soil and ground
water begin to enter channel systems in signifi.cant amounts as the high water
recedes, mineralization increases and the relative concentration of bicarbon-
ate ions increases at the expense of chloride ions. In middle and higher
latitudes there may also be a relative surge of biochemical activity in early
spring when plant activity recommences. On the steeper slopes and cliffs,
both inland and coastal, mass movements of smaller material loosened by

freeze-thaw and of larger blocks due to the saturation of the ground with melt
Figure IV.52 Hydrochemical regime of the Pyalitsa river, European U.S.S.R..

reflecting changes in origin of drainage water (from B. G. Skakalskiy, 1966.
Soviet Hydrology); (1) discharge in mVsec., (2) total ions, mg/1., (3) chloride,
mg/1., (4) water of slope origin [surface-slope and soil-surface], (5) Water of soil-
ground origin, (6) ground water.
water follow the spring melt. The latter process was probably involved in the
accelerated winter movements of the 30 000-ton Threatening Rock
(fig. The frequency of rockfalls on three railways near Bergen.
IV.49.B).
1920-58. reaches a marked maximum when air temperatures have passed
above zero, and frequency of slides in quick clay in the same area shows a
similar pattern (Table IV.9). Amounts of slope erosion may also vary accord-
ing to season (fig. IV.49.C).
In savanna areas a profoundly significant effect is the seasonal contrast
between wet and dry conditions. The end of a savanna dry season is most
critical, as the screening effect of parched vegetation over a desiccated soil is
at a minimum. The first winter storms which are generally abrupt and sub-
stantial, lead to rillwash on the unprotected ground. Chemical changes are
equally significant with evaporation leading to the precipitation of minerals
hydrolysed in the preceding wet season.
Compared with seasonal changes within a year, fluctuations of their net
average annual effect usually involves changes in degree only from year to
year. The main interest lies in observing the degree to which an individual set
of observations might bias short-term observations. There is also the possi-
bility of observing contemporary forms, mainly in unconsolidated materials,
which are relics of some extreme activity in a previous year rather than a
form inter-related with the intensity of processes usually observed. C. G.

Tuckfield. studying gullies in the New Forest. Hampshire, noted that although
the period April 1960 to April 1961 was unusually wet, with exceptionally
heavy rain-storms in the summer, there was not enough heavy rain in the
subsequent two years to remove the fallen material completely. With the
additional factor of relatively very cold winters in 1962 and 1963, widening
Distance bahmd present front
H*igh\ in c«niimeV«rs
Figure IV. 53 Variations in movement rates of material. (Left) Arithmetic mean

size of wind-transported sand plotted against height above the ground (from
Sharp, 1964); 29/12/1956 collected for 83 days of strong wind regime;
7/12/1958 collection followed 170 days of gentle wind regime. (Right) Rate of
downslope movement of the front of a stone-banked terrace, Niwot Ridge, Boulder
County, Colorado, based on radiocarbon dates (from Benedict, 1966).
was more effective than deepening. For larger streams year-to-year variability
may be less marked. Annual runoff in the United States over a 25-30-year
period is 10-20 per cent of the median. Sediment loads,
usually within
however, show greater contrasts. For example the 560 000 m^ of sediment
deposited in Lake Constance in 1949 was only one-tenth that of the
Water equivolent Precipitation Surface runoff

of snow
Figure IV.S4 Year-to-year fluctuations in surface runoff in spring following snow-

melt, Donets River area, Ukraine (from A. V. Plashchev and L. A. Uvarov, 1966,
Soviet Soil Sci.). For each year the column on the left is for a dense oak plantation,
that on the right for a maple plantation which stands on the lower part of the same
slope as the oak plantation. Some of the factors causing the fluctuations is protec-
from freezing by a thick snow cover with infiltration of snowmelt water

tion of soil
consequently unimpeded, e.g. 1964; small amounts of snowfall, e.g. 1961, also
reduce runoff. In 1960 deeply frozen sub-soil caused runoff under the oak forest;
measurements were not made in the maple plantation in 1960.
maximum during a 30-year period. This maximum, which amounted to 5
million m\ was observed in 1935. Annual changes in dissolved loads are
much more closely a function of discharge, overriding the effect of retarda-

tion of rates of biochemical processes in drier or cooler years. The direction
of dominant winds may vary from year to year. For instance, records of
onshore winds at Blakeney Point indicate that north-west winds were domin-
ant in 1958, and north-west winds in 1960. There may be a slow shift over
several years as appears to have happened on small reef patches in Djakarta
Bay where coral ramparts have changed position over at least a 75-year
period. The mean annual velocity of movements in dynamic depositional
landforms varies also; for instance, dunes in Peru travelled 41-226 m/year
faster in 1958-64 than in 1955-8. An example of the variability of beach
sand transport is the accretion behind the Santa Barbara breakwater,

California, ranging from 225 mVday to 1095 mVday during the 1938-50
period. Ice accumulation varies, different combinations of atmospheric condi-
tions producing contrasted or similar results. The average ablation gradient
for the Storglaciaren, Kebnekaise, Swedish Lapland, is 55 cm/ 100 m, but
over a 20-year period it ranged from 40 to 70 cm. On a larger scale and of
profound significance is the change in the ice boundary in the North Atlantic.
This is one of the most variable physical features of the earth’s surface,
varying from year to year by several hundreds of kilometres. Seasonal
changes may also influence glacial erosion and transportation (fig. IV.49.D-
E). The degree of frost penetration into the ground varies. In Alsace the
average maximum depth reached annually is 0-2-0-3 m, increasing to half a
metre in more severe winters such as the one of 1955-6, and the exception-
ally persistent frost with a recurrence interval of about 100 years may reach a
metre in places.
3. Examples of past climates
Glancing over longer spans of time consistent trends are evident in the last
century in the levels of enclosed lakes. Those in Oregon and Washington were
at high levels in the
1870s and 1880s, and again in the early 1900s, followed
by extreme desiccation until the middle and late 1930s. In many areas the
results of human activity, such as pumping and the indirect effects of
accelerated erosion are difficult to dissociate from natural changes. It is also
difficult to decide whether trends in levels are global or local, for example,
whether the pattern of years when water levels in small lakes in the south
Ukraine were high - 1882. 1906, 1911, 1912, 1924, 1927, 1928, 1931,
1932, 1936, 1939, and 1947 - is significantly similar to conditions in
Oregon and Washington. This instance illustrates a general problem of inter-
preting climatic changes and one which becomes increasingly difficult to
resolve the farther back in time that investigations lead.
/acr^>
tons
oroduclion
ctcbris
Cumuloitve
- B (i) (ii)
I960 I
1961 1960 |
1961
Figure IV.55 Artificial and natural changes in the litter cover on forest floors.
A. Erosion accelerated by four to sixteen times in the San Gabriel Mountains,

southern California following fire (from J. S. Krammes. 1963. U.S. Dept. Agric.
Table IVJ2 Chronology of Late Glacial and Holocene time (from Fairbridge,
1968).
NORTH EUROPE
YEARS POLLEN ARCHEOLOGICAL
BP ZONES VEGETATION SUB-STAGE NAMES CULTURES
0- 1 000 IXb Beech and heath Late Subatlantic
1 000- 2 300 IXa Early Subatlantic Roman

IronAge
2 300- 3 700 Vlllb Oak and ash Late Subboreal Bronze Age
3 700- 5 300 Early Subboreal Neolithic
5 300- 6 600 Vllb Oak-elm and Main Atlantic Mesolithic
mixed forest
6 600- 7 500 Vila Early Atlantic
7 500- 8 700 VI Hazel-oak Late Boreal

8 700- 9 800 V Hazel-pine Early Boreal
9 800-10 300 IV Birch-pine Preboreal
10 300-10900 III Younger Dryas
(Arctic tundra
climate)
10 900-11 800 II Allerod (Subarctic Late Paleolithic

climate)
11 800-c. 15 000 I Older Dryas

(Arctic tundra
climate)
From A.D. 980 to about A.D. 1540 Norse colonies existed in south-west
Greenland under generally warmer conditions than prevail today. Climate
deteriorated markedly about the beginning of the thirteenth century. From
A.D. 1000 to A.D. 1450, the number of major floods on British coasts in-
creased from seven (a.d. 1000-1200) to nineteen in the latter part of this
period. In South America, ancient Indian sites at points along the Pacific
coast in areas now completely devoid of any permanent fresh water suggests
that desiccation has been progressive even within the last millennium in this
region.
The marked decrease in rate of sea-level rise about 5000-3500 BP may be

due to retardation of ice-cap melting and therefore probably indicates global
Misc. Publ. No. 970). The effectiveness of subsequent precipitation on the burnt
slopes is also shown.
B. Seasonal changes in soil exposure and volume of litter on

a forest floor,
Madison, Wisconsin (from G. A. Nielsen and F. D. Hole, 1 964, Soil Sci. Soc. Am.
Proc., Vol 28) showing: (i) that the weight of leaves and wood debris
is greatest in
early spring, and (ii) that soil exposure is at a maximum in late summer.
cooling. This period marks the end of the warm Hypsithermal interval and the
start of the following cold phase or Neoglacial. Palaeobotanical investigations
indicate that during postglacial time amelioration of climate reached an
optimum in the warm and dry 4500-2500 years BP. At
sub-Boreal, about
that time the timber-line in southern Norway and presumably even the firn-line
was 300 m higher than at present. Previously, a large portion of the Atlantic
Ocean underwent a temperature increase of 6°-l 0° C during a period of less
than 2000 years, with the midpoint of this change within 300 years of 1 1 000
years ago. At this time about half the water had returned to the oceans from
the ice-sheets of the last glaciation. This abrupt world-wide climatic change
was first recognized from deep-sea cores by the study of foraminifera (Ewing
and Donn, 1 956) and of the temperature-dependent oxygen-isotope fractiona-
tion that occurs during the formation of carbonate by C. Emiliani (1955).
Supporting evidence includes glacial retreats and pollen profiles in Europe
and North America and pluvial histories of lake basins. Climates rapidly lost
their full glacial character and became more like climates of today. In some
areas, like Alaska, the climate might have been warmer than at present by
about 8000 BP. the evidence including that of fossil dams built by beavers in
situations beyond their present range. For many parts of the world, however,
the highest postglacial temperatures occurred several thousands of years after
the end of this early Holocene warm interval.
Many theories have been advanced to explain the climatic variations asso-
ciated with the Pleistocene, but none has been widely accepted for long.
Suggestions include variations in the amount of carbon dioxide or of dust in
the atmosphere, continental drift introducing critical changes in the distribu-

tion of land and sea, and the astronomical hypotheses of J. Croll (1875) and
M. Milankovitch (1938) invoking variations in the earth’s orbit. To all these
suggestions there are cogent objections. Neither is it clear whether ice retreats
were an inevitable result of self-arresting mechanisms once a glaciation had
advanced beyond certain limits or whether a separate set of factors operated
to induce ice shrinkage. However, the essential features of atmospheric cir-
culation imply that there is always a tropical moist zone between the north
and south arid belts. The equator is therefore always between two arid belts.
Polewards of the latter lie temperate climatic zones. In this sense the specific
features of Quaternary climatology incorporated no radical change from the

general plan of the present climatic zones. Whatever may have been the
causes of the ice advances or of their subsequent retreats some facts have been
reasonably clearly established. For the Wiirmian glaciation J. Biidel (1963)

suggests that latitudinal variations in the longitudinal strip between 0° E and
15°E included a glacial zone from 90°N to 55°N compared with a present
southerly line at 77°N. The periglacial environments today commonly
encountered between 69°N and 77°N spanned 45°-55°N in the Wiirm,
Table IV. 13 Tentative correlation of Quaternary glacial and interglacial stages tfrom
Fairbridge, 1968).
Glacials are in capitals and interglacials in lower case. Datings and correlations
such as these are under continuous revision; more detailed local sequences will be
found in the appropriate references in the bibliography.
ABSOLUTE
DATING N. AMERICA ALPS NORTHERN EUROPE POLAND-U.S S.R.
0- 67 000 WISCONSIN WORM WEICHSEL (VISTULA) VARSOVIAN

(early W. (early W = Warthe) (VALDAI)
= Iowan)
67 000-128 000 Sangamon Uznach Eem (Hoxnian) Masovian I
128 000-180 000 ILLINOIAN RISS II WARTHE (DRENTHE) CRACOVIAN

RISS I SAALE (DNIEPER)
180 000-230 000 Yarmouth Hotting Holstein Sandoniirian
230 000-300 000 KANSAN MINDEL ELSTER JAROSLAVIAN
I and II (Likhvm)
300 000-330 000 Aftonian — Cromer
330 000-470 000 NEBRASKAN GUNZ MENAPIAN
I and H
470 700-S38 000 Waalian
538 000-548 000 DONAU II WEYBOURNE
548 000-585 000 Tiglian (Tegelen)
585 000-600 000 DONAU I RED CRAG (B UTLEY)
e. 600 000-c. 2 m. Villafranchian
thus reaching to the margin of the present-day Mediterranean climate. In the

Worcestershire Avon valley 38 000 years ago the summer temperature range
was approximately that found today in the high birch scrub of the
Scandinavian mountains. In this environment in north-west Sweden present-
day summers are very short with only 45 days when the mean daily tempera-
ture rises above 10° C. July is the warmest month with an average of 1 1°C,
suggesting that the Avon valley was on average 5-5° C cooler than at the
present day. In central Europe the mean annual temperature in former peri-
glacial areaswas 6°-10°C lower than at present, and the global cooling
might have been, on average, as much as 5°C.
On three or more occasions in most continents an apparent increase in
dampness accompanied a general cooling. C.C. Reeves (1965) lists the
results of several workers in addition to his own from the
southern High
Plains which indicate that the drop in summer temperatures
during pluvial
periods was often very close to twice the mean annual drop. These
observations
might imply that the characteristic of pluvial periods in the
Pleistocene was
perhaps the influence on the hydrological cycle of the relatively
lower summer
temperatures rather than an increase in rainfall amounts.
Some authors
still consider that the comparison of Quaternary pluvials with
glaciations
is full of uncertainties. Particularly for low latitudes speculations concern-

ing glacial and interglacial climates are controversial and sometimes in-
ternally inconsistent. Inevitably, away from the highest mountaintops the

changes were essentially those involving amounts of water, lake fluctuations
being again one of the more reliable records. Farther back in time the climatic
history of even the early Pleistocene is still largely obscure because very few
continuous marine or continental sections of deposits have survived to be
discovered and described in adequate detail. However, it is hoped that in-
creased sampling of the bottom deposits of some larger present-day lakes may
yield some information. In the Cenozoic as a whole the constancy in position
of climatic belts does not signify that the climates remained uniform. Apart
from the moisture increases and temperature decreases in the Quaternary,
there were marked temperature fluctuations even in the Pliocene. In the
Neogene there was a wave of progressive cooling leading to the Quaternary
glaciations and the evidence of fossil plants and animals clearly indicates that
temperatures have decreased more or less steadily throughout the Tertiary. In

the Eocene there was a sharp warming trend extending the boundaries of
tropical flora and bauxite formation to present-day temperate areas beyond

the arid belt. At this time a deep-weathering crust developed in Ireland at
55° N, and in the Oligocene there were pines and firs in Greenland. Flora
indicate warm and moist Canadian Shield in Upper
climates in the
Cretaceous times with temperatures averaging more than 20° C where today
mean temperatures are — 3°C. and about this time there were dinosaurs in
Spitzbergen at 78° N. During the late Mesozoic and Tertiary savanna condi-
tions prevailed in the Sahara and seasonal floods transported huge quantities
of alluvium to the sea on the North African shore. It was only with increasing
aridity at the beginning of the Quaternary that the desert plains of the
southern Sahara started to form. However, in most instances any link between
present-day landforms and palaeoclimates different from the present remains
obscure.
V
Landforms and time
A. Dating
Although there is available an increasingly large number of precise, scientific
measurements of recent geological time, great expertise is essential for samp-

ling, analysis, and interpretation. The sample must be virtually unaffected by
weathering or post-depositional chemical alteration or from contamination in
a borehole by younger material falling from above. The interval between the
deposition of a sample and the geological event to be dated may be wide and
uncertain, partly because the geological history may be incompletely known,
and partly because deposition or growth in some places began only thousands
of years after the drainage of a lake, withdrawal of the sea, departure of the
ice, or some similar event. Usually some independent method of ranking
events in time is sought to confirm the accuracy of an estimate of absolute
age. Conversely the correspondence between the ranks of a time sequence and
a series of absolute ages confirms the efficiency of the ranking method. A
distinction between absolute and relative methods of dating becomes less
useful as radiometric determinations are increasingly used stratigraphically to

infer simultaneity of events in separate areas. Like the estimation of adjust-
ments to absolute ages, methods establishing relative ages also usually depend
on a high degree of sophisticated understanding and the methods are perhaps
and knowledge required
best reviewed according to the specialist expertise
ratherfrom whether the aim is to establish an absolute or a relative age.
Although most dating techniques are specialized skills in themselves, it is
evident to the geomorphologist that he cannot use dates without circumspec-
tion and that therefore some awareness of the nature of these methods and
their limitations is necessary.
The discovery that so-called radioactive atoms have unstable nuclei which
decay at an exponential rate to lower energy states has thrown the entire
geological record into sharper focus. The unit of measurement of the decay
rate is the half-life, the time taken for half of the radioactive atoms in a system
to decay. For instance when potassium-40 decays to argon-40, the crystal
lattice of the mineral traps the argon, an inert gas, and the ratio of radiogenic
argon-40 to potassium-40 is directly related to the time at which the mineral
crystallized. The time-range limits are 300 000 and 100 million years old; the
amount of radiogenic argon is very small due to the long half-life of potas-
sium-40 (1310 million years) and dates less than 0-5 m years are seldom
precise to within 10-20 per cent. Potassium-bearing minerals are usually
found in igneous rocks, like bioxite. muscorite. sanidine, plagioclase, and

hornblende, but glauconite by being a sedimentary mineral greatly extends
the possible usefulness of this method, and there are optimistic hopes for
establishing a time-scale of Tertiary sediments. So far, however, it is Tertiary
volcanic rocks, particularly those in north-west United States, which have
been dated most systematically, resulting in an invaluable control on paleo-
floras and age rankings based on floristic criteria. These ages have also
defined and illuminated time spans for landform development involving the
dissection of the lava flows.
There seems little doubt that extension of results from the expensive and
highly skilled operation of potassium argon dating to the widely scattered
areas of Tertiary volcanism will lend unparalleled clarity to concepts of
geomorphological time. Partly because the decay-rates are too slow, the
uranium-lead and rubidium-strontium methods are usually employed on
rocks too old to provide a significant datum in landform studies.
Thorium-230, formerly known as ionium, is produced in the uranium-238
decay series, and being readily precipitated in sea-water may be incorporated
into ocean-floor sediments. The method gives an estimate of ages up to
300 000 years ago. As Thorium-230 has a half-life of only 75 000 years it is
assumed that when, at a particular depth in a deep-sea core the Thorium-230
concentration is only half that at the top of the core, the sediment is 75 000
years old. A grave restriction on this assumption is the necessity for the
sediments to have remained undisturbed. A related method, which may date
ocean-floor sediments up to 150 000 years in age, is the use of the ratio
Thorium-230/Protactinium-23 1. Pa”', produced by uranium-235 decay,
also precipitates quickly in sea-water, but with a half-life of 34 300 years, it
decays twice as quickly as Thorium-230, thus providing the basis for the
ratio. Although the methods based on progressive decay of radioactive ele-
ments incorporated into deep-sea sediments have provided valuable sequences

of events for late Pleistocene times, the usefulness of estimates of absolute
dates remains uncertain.
C. Emiliani (1955) has used the ratio of the oxygen isotopes O'VO'* in
an indication of the temperatures of the environ-
fossil foraminiferal tests as
ment when they formed. However, reconstructions of past sea temperatures

based on assemblages of foraminiferal species give different results from
those obtained from the oxygen isotopic analyses of the tests. As the basis of
Landforms and time 329
these working hypotheses, the nearly instantaneous heat exchange between

ice-caps and ocean water accompanying oscillations in continental ice,
appears sound. Further work and the devising of further stringent criteria
might help to resolve the discrepancies in interpretation.
A technique currently under investigation which might, if accepted as a
viable dating method provide invaluable dates for landform interpretation, is
the study of thermoluminescence. Thermoluminescence is the light emitted by

many minerals when heated below the temperature of incandescence. If the
temperature gets high enough the thermoluminescence may be completely
annealed away. The amount of thermoluminescence produced by a crystal is
proportional to the number of electrons trapped in some imperfections in the

crystal lattice structure. If temperature is constant the number of escaping
electrons tends to equal those escaping from the traps, thus approaching a
dynamic equilibrium position. To establish such an equilibrium under present
Antarctic temperatures would take at least 2-6 million years. The leakage rate
is a function of a given constant temperature, but any rise in maximum

temperatures leads to the rapid release of trapped electrons. If an equilibrium
value imposed by a former environment of constant temperature can be
estimated, the length of time since the change in temperature in the rock’s
environment can be calculated. For instance, in Spring Cave, near El Dorado,
Kansas, a perennial stream keeps summer temperatures close to the average
annual surface temperature. In the gully outside the cave the temperatures are
much higher in summer and amounts of thermoluminescence are reduced
towards a new and lower equilibrium level in proportion to the length of
exposure to conditions outside the cave (fig. V.l). It has been suggested that
high rates of erosion might expose deeply buried material in a time
substantially shorter than that required for the sample to approach an
equilibrium with the climatic conditions at the ground surface. If the differ-
ence between the sub-surface temperature and the prevailing climatic tem-
peratures can be estimated closely it might then be possible to estimate the
rate of erosion. Although this argument is perhaps as yet somewhat hopeful it
does demonstrate that there are reasonable chances of techniques being

developed which would be focused on some of geomorphology’s most impon-
derable questions.
Despite recent formulation and use radiocarbon dating has provided the
basis of a rapidly accumulating stock of precise
knowledge for stratig rap hers
studying late Pleistocene chronologies, from which landform study has also
benefited considerably. There are, however, still only a few localities for
which there are a series of dates. It is bombardment of nitrogen atoms by the
neutrons of cosmic rays in the upper atmosphere which produces a radioac-
tive isotope of carbon, by knocking a proton out of the nitrogen
atom nucleus.
Because of a delicately poised dynamic equilibrium, diffusion of new C to ,4
the lower atmosphere balances its loss by radioactive decay under natural
environmental conditions. Both radiocarbon, with a half-life of 5730 years,
and normal carbon, combine with oxygen to form COj which is incorporated
into the tissues of living organisms. After death the amount present declines
steadily due to radioactive decay. Sources of organic material are usually
wood fragments in tills and in beach deposits, but may be varied. For
example, small organic particles originally transported on to snowbanks by
wind provided the basis for a radiocarbon chronology of ice-cored moraines
in the Jotunheimen. Norway. Determinations are possible of some bulk shell
samples by a comparison of the Cu/C,, ratio in the carbonate material. With

great expertise C ,4 measurements may be valid for 70 000 years, but very few
results have been obtained beyond 45 000 years. No reliable dates are pos-
sible since 1850 because the combustion of industrial fuel has caused a
decrease in C ,4 activity in the biosphere. In addition to providing absolute
dates, the C ,4 method, like other radiometric techniques, also provides a
means of inter-continental correlation by identifying contemporaneous events
in distant geographical or ecological areas. There are some difficulties here,
however, like the C ,4 date of the Valders glacial maximum being 10 700-
1 1 000 years at Milwaukee, correlating in time with the relatively warm
Allerod in northern Europe. In this situation a few workers prefer to believe
that broad intercontinental climatic conformity is probable, even to within a
Instde Cove Outstde Cove
Figure V.L Amount of thermoluminescence at points inside and outside a cave

{after Roitca and Zeller, 1964). The scale of contrast could offer an indication of
the length of time during which the gully wall outside the cave has been exposed to
subaerial conditions. The example used is Spring Cave, south-central Kansas.
Land/arms and time 331
millennia or two, and to emphasize inaccuracies and untrustworthiness of C ,4

dating.
In all but historical times the traditional cornerstone of time sequences is
the evolutionary changes in organisms, particularly vertebrates, based on

composite aggregates of genera and species, with no genus or species neces-
sarily restricted to the age that it helps to characterize and not by itself
providing a guide or index for that age. Within the range of time which
concerns the geomorphologist the rapid rate of morphological evolution of
land mammals and in particular their extremely rapid dispersal has provided
the basis for a widely applicable chronology of Cenozoic times. A major
recent achievement has been the closeness of correlations achieved in inter-
meshing the North American Land-Mammal Ages with potassium-argon
dates. In their studies of fossils. Quaternary palaeontologists, who include
geomorphologists like B. W. Sparks in their ranks, can distinguish ‘cold’ from
‘warm’ floras. For example, in association with possibly the most definite
of the revised Pleistocene transgressions of the Mediterranean sea, the
Tyrrhenian II level found at 5-10 m on stable coasts, is a varied thermophile
fauna, distinctive because it includes Senegalese species now extinct in the
Mediterranean.
There are several methods not dependent on radioactive decay or evolu-
tionary changes in organisms, which may provide some indication of age over
the past few centuries. The counting of tree rings, dendrochronology, is
particularly useful for the study of certain depositional landforms, but may
depend on an estimate of the time taken for the trees to become established on
initially barren ground. For the Donjek moraines in the Icefield Ranges,
Alaska, the establishment of spruce trees appears to take at least 30 years and
poplars at least 23 years. Fig. III. 17 illustrates the flood-training of cotton-
wood saplings and the ages of various portions of a floodplain based on tree-
ring counts which by no more than 10 per cent for each grove.
differed
Tree damage as well as tree growth may assist in the problem of chrono-
logy. In higher latitudes where ice might damage the bark, cambium, and
outer part of the wood, a dead area of wood knobs forms at the edges due to
the growth of the living cambium cells and subsequently the annual ring here
is dark compared with other layers. This feature may offer some approximate
dating of extreme floods, their height and their frequency. Floods may
damage trees to a far greater extent than the notching of bark by ice. and
some well-rooted trees may contain a tangled record of past geomorpho-
logical events in their branches (fig. V.2).
Like tree-trunks lichens increase in radius with age expressed by their
diameter since growth rates are often constant. J. T. Andrews and P. J. Web-
ber (1964) calculated lichen growth on the north-west margin of the
Barnes
ice-cap. Rhizocarpon geographicum 0-05 7-0- 06 7
being mm/year and
Figure V.2 Sprouts of four ages on a flood-damaged ash tree, Potomac river
floodplain, near Chain Bridge (from Sigafoos, 1964).

Alectoria minuscula 0-40 mm/year. The rate of growth of A. minuscida on a

delta in McBeth fiord, Baffin Island, was 0-41/0-44 mm/year. Despite great
difficulties in allowing for local ecological conditions like varying water and
nutrient supply, differences in exposure and surface roughness of the rock
surface, and problems of understanding the physiology of lichens, some
progress in dating by lichenometric methods has been made since 1950,
particularly by R. Beschel. He has shown that for certain species of lichen the
rate of growth increases ten-fold for a four-fold increase in the ratio of pre-
cipitation to altitude. Reliable dates for individual localities are a prerequisite

for establishing a time-scale, sometimes provided by old photographs or
old dated stones. In uninhabited areas C ,4 dating of depositional landforms
may provide a basis for a lichen chronology. Suitably calibrated, any crusta-
ceous or foliaceous lichen may yield important data, as it spans the last 1000
years under alpine conditions, and in polar regions lichens might be useful for
twice that time-span. In many situations lichenometry may make more
precise statements about events in recent decades or centuries possible. The
amount of lichen cover tends to indicate the degree to which loose rock
fragments in a rock glacier, scree, or similar accumulation are stable. Simi-
larly, it is well known that shore gravel not reached by breakers may be grey
because of a cover of grey crustace lichens.
There are many situations where a uni-directional sequence of geomor-
phological events has taken place and where the vegetation cover gives a clear
indication of relative age. Many abrupt geomorphological events, like glacier
retreat, sand-storms, drainage of a lake, landslipping, or volcanic eruption

expose initially sterile habitats on which the subsequent plant succession is
conspicuous and easily observed, like the moss successions in deglaciated

parts of Scandinavia. Within a few years of ice recession, lichens, ordinarily
crustose types, are the first colonizers of the bare ground, with umbilical
species (rock tripes) appearing later, followed in turn by other foliose lichens,
the so-called reindeer-mosses (a fructicose lichen) and mosses. Cryptogram
communities lead to ground stabilization and humus layer formation. Simi-
larly, successions are well marked in areas of recent volcanic activity where
lichens may appear on a flow 4 years after the congealing of the lava.
Following the steadying in postglacial sea-level rise many submerged coral
reefs have enlarged into areas of dry land, to be colonized in a fairly regular
manner by creepers, shrubs, grasses, and finally trees. Alluvial islands in
some temperate-latitude streams similarly starting with barren sediment
acquire first a willow cover and finally a hardwood forest. There are many
other environments in which the vegetation could be studied to elucidate

changes in landform or process intensity within recent historic times. For
example, the vegetation cover around perennial snow patches may reflect
long-term variations in their extent. With the Chaos Jumbles, a
5-km-square
Figure V.3 Soil and plant evidence in daring depositional landforms.
A. Decline of phosphate content in the surface layer of till with increasing age of
deposit {from Stork. 1963). The morainic debris was left by the retreating
Storglaciar. in the Kebnekaise area, northern Sweden.
B. Progressive increase in number of plant species in front of Storglaciar follow-

ing deglaciation {j'rom Stork. 1963).
C. Changes in species and progressive increases in growth and density of the tree
cover with increasing age of rock avalanche deposits ifrom Heath. I960). The two
transects were laid out in the Chaos Jumbles. Lassen Volcanic National Park.
avalanche deposit of dacite blocks on the north-west part of Lassen Volcanic

National Park, the evidence of the vegetation points to three separate ava-
lanches (fig V.3). On the youngest deposit, a range of colonizing pines is
usually dwarfed and distorted due to reduced root space in the rocky soil. On
the second deposit, the Yellow pines are often normal in pattern, but have
sharply conical trunks that indicate exceedingly slow growth. On the oldest
deposit where there is a soil mantle, the Yellow pine cover is nearly mature or
sub-climax forest. Some of the species which invade the barren areas, like the
Western White pine (P. monticold), surviving as long as the primary competi-
tion is with the elements, are absent on the oldest deposit, where the Yellow
pine becomes well established.
By most powerful botanical technique for estimating relative ages,
far the
spanning times from the recent historical past to well back into Tertiary times
is the study of fossil pollen-rains. These were first analysed systematically by
L. von Post in 1916 and have since flourished to provide one of the most
reliable aids in establishing chronologies of recent geological times. Pollen
shed by trees and by other plants may be carried by air currents and be
washed down into lake sediments, peat bogs, or a similar environment where
the very tough pollen seeds resist decomposition. G. W. Dimbleby (1961) has
shown that pollen may also be preserved in acid soils for several millennia. If
the kinds of pollen are identified and their numbers counted for a succession
of points in a vertical series, the pollen in a given layer may give the skilled
investigator some indication of the type of vegetation growing in the area at
that time. The vegetation gives some indication of the climatic conditions
prevailing at that time, and these characteristics indicate a specific time zone
now usually calibrated by radiometric determinations. A break in the pollen
charts usually indicates that a former ground surface was abruptly buried by
more recently deposited material. The interpretation of pollen diagrams,
however, depends on the intuitive allowances made by skilled ecologists for
several potential imbalances in pollen diagrams for which no absolute correc-
tion exists as yet. These include over-representation by a species dominant in
a restricted habitat close to the sample site, the rapid change of plant species
with micro-environmental changes, such as hydrological conditions over
uneven or sloping ground, the probability that some species contribute pro-
portionately more pollen than others, and that drier, windier conditions in the
past could cause changes in the ranges of pollen dispersal.
Of man’s activities useful in dating events in landform history, the value of
old photographs increases annually and now includes older air photographs.
A photographic survey carried out in 1910 by F. Enquist in the Kebnekaise
massif, northern Sweden, has proved of greatest value to present-day glaciolo-
gists. C. A. Kaye (1964) searched for old photographs of upper barnacle
limits for localities on the New England coast and was eventually able to
demonstrate that while sea-level just over a century ago was comparable to
that of today, itwas about 15 cm lower at the turn of the century. These
fluctuations matched tide-gauge records at nearby harbours. The value of
photographs is not limited to the comparisons that they themselves may
provide, as they may also provide a check on the reliability of a range of
dating methods that might be projected further back in
time. Documents can
provide such a wealth of information about past geomorphological events that

it is difficult to illustrate the range of possibilities. One example is the long-
term records of on Churchill river and the Hayes river entered
ice conditions
in the Hudson Bay Company archives from the early eighteenth century
onwards. For early maps and diaries independent corroborating evidence
usually has to be sought, but again useful information may be gleaned. For
instance, it appears from the Rev. F. Consag’s records that the last eruption
of Tres Virgines volcano in the Gulf of California was in 1746 (R. L. Ives,
1962). For human records, remains and artifacts farther back in time the
study of archaeology is a long-established specialism. An observation perhaps
worth making in the present context, however, is that the implications for
landform study of such evidence may not be directly related to their intrinsic
archaeological interest or significance. For instance in the 1930s several
arrows, with wooden shafts intact, were found Norwegian mountains at
in
sites from which snow banks had recently disappeared. The oldest date from
A.D.400-600 and the most recent from post-Reformation times. As speci-
mens these finds were invaluable, but equally striking is the fact that their
preservation suggests that the associated snowbanks had similar volumes
between a.d, 400 and 600 and again about 1930-40, and that during this
interval they were larger. Another noteworthy point about human artifacts is
that those of very recent origin have immense potential value in the dating of
events and in determining erosion rates over the span of the last few decades.
For example, the incorporation of crown-shaped bottle-tops into tropical
beachrock formations provides one of the best confirmations of the rapidity
of concretion formation in warm environments.
In many ways soil and sediments may help in establishing chronologies
either by providing stratigraphic horizons or by reflecting a time-span in the
degree to which certain characteristics have changed. The possibilities for

absolute dating are normally confined to those glacial clays where fine lamina-
tions of varves can be regarded as annual. C ,4 datings confirming the prin-
ciple of De Geer’s classic method of geochronometry that there might be a
close connection between the periodic laminae of the clays and the annual
ablation of the land ice. Currently there is an extension of the search for
laminations due to the seasonal factors influencing settling of particle^ on to

the floor of sedimentary basins in non-glacial areas.
Of the other soil and sediment evidence, all of either stratigraphic signifi-
cance or indicative of relative ages only, distinctive deposits include wind-

blown silts. On moraines and other glacial features the thickness of wind-
blown silts may indicate the relative age of the forms. Volcanic ash layers in
particular can provide an excellent datum. The Pearlette Ash is a unique
means of correlating late Kansan sediments throughout the Great Plains and
into the glaciated region of the Missouri river valley. The specialist study of
distinctive pyroclastic layers of known age, termed tephrochronology, is
employed in areas of recent volcanic activity like Iceland and Alaska. Other
accumulations occur within soils due to distinctive pedogenic processes.
Caliche accumulations whitening and indurating a soil zone in sub-humid
areas may persist because the nodules are resistant to later elimination by
leaching. These and similar secondary carbonates appear to offer promising
means of identifying relatively dry climates in the Pleistocene, provided that
other influences can be eliminated. However, yet again disagreement in inter-
pretation is well known, due to problems reminiscent of those which compli-

cate the interpretation of pollen diagrams. Some workers like K. Butzer have
concluded that eolianite, for instance, indicates earlier phases of continental
glacial advances, whereas D. Yaalon stresses the importance of local relief,
microclimate, and water-table conditions in their formation. In contrast the

evidence of salt accumulations in arid areas is independent of geological and
biological factors.
Truncated as well as depositional horizons may exist in soils, the former

recording phases of instability and erosion. If not actually visible in the field
section, soil analyses at a succession of depths may, like pollen diagrams,
show breaks as composition changes abruptly at a truncated horizon
(fig. V.4). In Australia, such soils are widely used as records of minor cycles
of erosion separating intervening periods of stability when soil formation took
place.
The study of the degree of soil-weathering is, like the study of varves, a
classic method in establishing chronologies. The depth of penetration of
Figure V.4 Abrupt changes in soil properties, indicating the separateness

of
buried soil layers {H. M. Churchward. 1961. Jour. Soil Sci.. Vol.
12), at Swan
Hill, Victoria. Australia.
pedogenetic processes is often a vital indicator and has helped materially in

differentiating many major drift sheets since A. Penck first studied on a large
scale in 1899 soil formation on moraines in the Alps. In the Puget lowlands
of western Washington, the oldest till, the Shakwak, is distinct in being
oxidized to a minimum depth of 4-5 m, and in the same locality the highly
oxidized Icefield outwash III contrasts sharply with the younger unoxidized
Kluane outwash I which is medium grey in colour. Dune sands may show
similar changes, appearing to acquire a coat of ferric oxide which affects an
increasingly large number of grains and perhaps increases in thickness with
the passage of time, changing the grain colour from that of unoxidized
sand through light yellowish colours to dark reddish-brown or even
brownish-black in very old dunes. Apart from the translocation of iron,
weathering may produce progressive changes in particle size and also in the
composition of suites of minerals in sand grains. Blackvvelder was the first to
use boulder-weathering ratios as an indication of age. Yet again micro-

environmental conditions like dampness, aspect, and position have a marked
significance. There is also the problem of defining when a boulder is
weathered; F. Birman recognizes a boulder as weathered when more than half

of its surface contains grains loosened to the depth of the average grain
diameter. R. P. Sharp (1960) found several dilTerences between boulders on
Early Middle and Late substage moraines in the Trinity Alps. California.
Table V. 1 lists the boulder-weathering ratio. Remnants of mechanically worn

surfaces were found on Late boulders and then only rarely. Spalling and
fracturing along Joints are marked on Late boulders, seen to a small degree on
Middle ones, and almost entirely lacking on Early boulders. Loss of rounding
due to differential weathering is most marked on Early boulders, on which a
Table V.I The ranking in age of Wisconsin moraines by the degree of weathering of
boulders tfrom R. P. Sharp and J. H. Birman. 1963, Bull. Geol. Soc. Am., Vol. 74).
The Mono basin glaciation may be Illinoian.
NUMBER OF BOULDERS
PER CENT weathered/per CENT OVER 1 FOOT DIAMETER
UNWEATHERED GRANITIC BOULDERS IN 1 00 X 20- FOOT STRIP
Glaciation I 2 3 4
Tioga 30/70 30/70 10/90 300
Tenaya 51/49 49/51 50/50 180
Tahoe 73/27 67/33 80/20 115
Mono — — 95/ 5 60
I = Sequoia-King’s Canyon National Parks. 2 = San Joaquin river drainage. Yosemite and
eastern slope of the Sierra Nevada. 3 and 4 = Walker Creek (Bloody Canyon) moraines. Mono
basin.
1
Table V.2 The mean-maximum protuberance of quartz veins and inclusions on

carbonate cobbles as a means for ranking alluvial fan deposits by age. Gorak Shep
fan. Eureka valley. California (from Hooke, 1967).
MEAN — MAXIMUM DEPTH OF

WEATHERING IN MILLIMETRES
Upper segment Lower segment
Recent channel deposit 1-8 2-5
Oldest channel deposit 2-5 4-3
Youngest fan surface 4-0 5-8
Oldest fan surface 12-5 16-3
ground level weathering platform a few inches wide is also relatively common
and where boulder burial by grus is cms
significant. Cracks are etched 2-5
deep on Middle boulders and 10 cms deep on Early boulders. Conversely

inclusions etched into relief project by 2-5-5 cm on Middle boulders and by
more than 5 cm on Early ones. A similar property has been measured on an
alluvial fan where quartz inclusions stand out in varying degrees from
carbonate-rock pebbles, according to the degree of weathering of the frag-
ments (Table V.2). Beaches may be similar to moraines in that older material
may have small weathering cavities, due to the boring activity of algae.
Considering smaller particle sizes, A. R. van Wambeke (1962) believes that
the silt content (2-20/f) of tropical soils derived from identical parent rocks
is amount of weatherable minerals, and on the
closely related to the actual
basis of more than 1000 samples, mainly from the Congo, finds that a silt-
clay ratio of 0-10 appears to characterize soils older than end-Tertiary
(Table V.3). Similarly, the average measured median grain size of heavy
minerals may with the increasing age of sediment, decrease in relation to the
average theoretical median diameter of the heavy mineral suite, due to
chemical decay of less stable heavy mineral species by interstratal solutions.
With the heavy minerals composition changes. In the Jekyll island area, a
general impoverishment in hornblende and epidote and enrichment in
Table K.i Silt-clay ratios of parent materials in some soils from Central Africa
(from Van IVambeke, 1962).
AGE OF PARENT GRAND- MICA

.MATERIAL GRAN’ITE DIORITE SCHIST BASALT
I 2
Recent 0-39 0-36 0-74 0-96 1-03
Pleistocene 0-27 0-17 0-32 0-18 0-32
Pre-Pleistocene 0-11 0-08 0-12 0-1 0-09
sillimanite characterizes the older Pleistocene terraces. Hornblende averages

19-3 per cent of the heavy mineral suite in samples from adjacent rivers,
about 7 per cent along the coast and less than 0- 5 per cent in the Pleistocene
terraces. One aspect of pedogenetic processes, that of clay-mineral alteration,
appears to offer great potential because the changes take ten thousands of
years or more to become clearly established. The time factor is highly sig-
nificant in the formation of kaolinite. Study of a kaolin-rich residual soil near
Thornaston. Maine, suggests a pre- Wisconsin and probably a pre-Pleistocene

age. Clay-mineral alteration data reveal that the post-Wisconsin weathering
profile is distinctly different from those of the Sangamonian and Yarmouth-
ian. The Yarmouth weathering of the Kansas till, for instance, went on for
much longer than examples of the other two tills and led to a complete loss of
feldspars and chlorite from certain zones and the substantial alteration of illite
structures. X-ray analysis by A. Godard of ancient red soils from fissures in
Cambrian limestone found at many points in north-west Scotland suggested a
similarity with Mediterranean terra-rossa soils, a relic of an environment
slightly wetter and warmer than the present-day Mediterranean climate.
In addition to some specific characteristics which promise the possibilities
of quantitative tests,some soils show general structures which may reflect
time of weathering or may add to their usefulness as a stratigraphic horizon.
In the first case, for example. latosols in Natal believed to be associated with
older land surfaces are typically platy vesicular structures with predomi-
nantly red and purple colours, whereas those believed to be developed on
younger late-Cenozoic surfaces are more nodular and pisolithic in structure
with a predominance of colours between red and brown. In the second

instance perhaps the best examples, well-known for their chronological sig-
nificance. are soil structures due to permafrost activity. During the past two
decades. German workers in particular, like J. Biidel and H. Poser, have made
considerable use of periglacial patterns in soil and sediment structures.
Despite the general usefulness of soil structures it should be added that their
appearance may sometimes be misleading. For instance, before analytical
techniques were available to demonstrate the length of the Yarmouth inter-
glacial, it was estimated to be the same as that of the Sangamon because the
t%vo soils appear to have developed to about the same degree.

Another general point is that while several distinct characteristics have
been discussed separately, invariably the usefulness of soil and sediment
studies in the search for chronologies depends on the consideration of more
than one characteristic. For instance, three drifts bordering the Reedy glacier,
Antarctica, are readily correlated by their degree of cavernous weathering,
the amount of staining by ferrous compounds, and by their altitude.
Boulders of the oldest and highest Reedy I drift are stained by ferric oxide
and have deep cavernous weathering pits. Reedy II boulders are similarly
stained, but cavernous weathering is absent or only incipient. On Reedy III
boulders the surface is fresh, with little or no oxidation.

While the range of methods for establishing chronologies has expanded, the
traditional reliance on altitude of land surfaces as a single criterion has been
largely discarded. Difficulties of interpretation, problems posed by the vari-
able amounts of movement of land and sea, the indistinguishability of eustatic

and crustal movements, and the realization that altitudes of former sea-levels
are not necessarily related consistently to their relative ages and the difficulty
of tracing continuous features in the field are some of the reasons for this
change. Even where depositional terrace-like features have clearly defined

altitudes, careful levelling has cast doubt on the reliability of extensive cor-
relations and generalizations. For instance, J. B. Sissons (1963) has levelled
altitudes on a delta remnant in the Forth valley at 35 •4— 35-9 m OD, at 32-8—
33 - 1 m OD, and at 25- 5-26- 8 m OD, none of which appears identifiable with
the former widely recognized 30 m raised beach in Scotland.
Despite the wide range of methods that might provide information for
dating or the ranking in a time-sequence of significant geomorphological

events, some generalizations apply to most techniques. The most important
consideration is that the sampling and interpretation is usually in the hands of
a small number of highly specialized experts utilizing sophisticated and expen-
sive analytical procedures and whose primary interest is often in dating as an
end in itself, and whose secondary interests do not usually include the study
of landform development. The geomorphologist therefore is usually depend-
ent on the findings and progress of these specialists rather than being in a
position to tackle most dating problems himself. On the other hand there are
some relatively straightforward procedures which have been almost totally
overlooked until the last decade. Their disadvantage is that they demand a
great deal of work without necessarily providing a guarantee of a meaningful
reward in return. Another common problem affecting a wide range of pheno-
mena susceptible to environmental change is the degree to which a given
pattern might reflect local micro-environment characteristics rather than giv-
ing a broader general picture.
Few would now agree with W. M. Davis’s firm belief that landforms them-
selves are the best guide to their age. There
is the growing realization that one
cannot tell whether a landform young or old simply by looking at its shape,
is
and that there is a need for some independent indication of how forms and
processes have changed over decades, centuries, or millennia. Intensified
interest in the testimony of sediments, soils, and plants is a current
trend and
likely to lead to a greater precision in the study of sequences
in landform
changes than the scant factual information in the following six sections might
suggest is possible.
B. The Initial Form

‘Many familiar geomorphological features of the emerged lands may be recognised
below the surface of the sea.’ C, Emiliani, 1958
It was only in the 1930s that soundings of continental shelves and ocean
basins became detailed, and it is only within the last decade that surveys
resulting from continuous profiling have appeared in large numbers. As a
result geomorphologists need no longer accept that the constitution of an
initial land surface on which landforms would develop subsequent to uplift
need necessarily be a matter of speculation. The relief of the surface that
would, with a 100-m drop in sea-level, increase the earth’s land area by 2-7
per cent is now better known. In many areas this marginal shelf has the same
degree of gross irregularity as the adjacent land areas. Off the north-west
coast of the United States a rough part of the continental rise is called the
Ridge and Trough province and, off southern California, basin and range
relief is very marked and resembles that of the Basin and Range province. The
continental shelf off the east coast of the United States is much broader than
the narrow Californian shelf, averaging 135 km, but is as little as 8 km off
Palm Beach, Florida. Off Maine it is 420 km wide, but like the shallow
shelves off Quebec, New Brunswick, and west Newfoundland, the surface is
relatively rough. There is also a series of seamounts extending 1600 km from

Bermuda to points off New England, believed to be late-Cretaceous vol-
canoes. Off Vladivostok the shelf is 60 km at its widest, but is mostly less
than 20 km wide and very narrow north of 67° N. Several submarine features
differentiate its surface. Tectonic and structural lineaments as well as sub-
marine canyons are significant features off many coasts. Relatively long and
narrow depressions are found parallel to the main trend of the Norwegian
coastline and similarly on other shelf areas off glaciated coasts such as flank
Spitzbergen, Greenland, Labrador, and Scotland. One on the Labrador shelf
is 400 km long, with an undulating bottom relief including depressions
exceeding 730 m in certain places. H. Holtedahl (1958) considers that the
fracturing may- be related to late-Cenozoic upwarping of the crust. Some

irregularities on shelves show some alignment like the south-west trend of
100-400-m scale of relief east of Madagascar. Similarly in the western
approaches to the English Channel, there appears to be a well-defined WSW-
ENE elongation of structures in the metamorphic and igneous basement
which are traceable to the continental shelf edge. Nearer the mainland there
are isolated trenches along the strike of Jurassic clays and soft sandstones off
the Dorset coast and Bristol Channel. Near Weymouth one depression is
22 km long and 1 - 5 km wide, and 45 m deeper than the surrounding ground.

On the floor of the Baltic a Silurian and Ordovician cuesta system is well
developed. On many other shelves there is irregularity but with little sys-
tematic trend.
Another reason for the irregularity of the sea-floor surface offshore is that
most sediment delivered from many continental areas has been funnelled off
the shelf along submarine canyons. Even in areas where a sediment
cover
exists a considerable amount of buried relief exists, as in the North Sea

between Scotland and Norway. Similarly sediments on the submerged coastal
plain off Buenos Aires province, although averaging 0-5 km in thickness, are
merely fills in two depressions in a pre-Cambrian basement complex. These
depressions run seaward, are at least 700 m long and 150 km wide, and one
includes the Rio Negro basin. A further point about sediment accumulation
on the continental shelves, apart from its general sparsity, is its sporadic
distribution which may assume distinctive trends. In the English Channel the
similarity in orientation of ribbons of sediment and tidal currents is most
striking. Sand waves may be up to 20 m in height with a wave length of
900 m. In the Irish and Celtic Seas, lense-shaped masses are up to 30 and
110 m thick respectively and reflect how sediments pass off the shelf along
relatively restricted paths. In some shelf areas off the eastern coast of the
United States rising sea-level left several terraces of both erosional and de-
positional origin and many submerged bars which once may have separated
long lagoons from the open sea. Within the Gulf of Maine the surface relief
reflects the activity of Pleistocene glaciers. Thus where it exists, a sediment
cover may not provide a smooth-surfaced veneer but may have well-
developed trends. Also, far from blanketing the underlying bedrock surface,
irregularities in the latter are often largely the controlling factor in the
dispersal patterns of sand moving off the shelf.
In terms of their function, age, and size the most striking feature of the
continental shelves is the submarine canyons. Their dominant role is to funnel
sediment from the continental shelf to the deep-sea floor. The age of deposits
partially filling some canyons, indicates their formation in pre-Middle
Tertiary time. Many are huge in size. One canyon between Norway and
Iceland has a depth of 1000 m. Off south-east Alaska, there are three marked
submarine canyons, that off Cross Sound being 22 km wide and 440 m deep.
Oil wells drilled into a valley fill inland and aligned with the Monterey sub-
marine canyon failed to reach basement at a depth of 2380 m. The fill below
900 m may be middle to early Miocene in age. Apart from the remnants of
old aggraded surfaces forming matching terraces high above the present
channel, as in the Newport canyon, many erosional features of
submarine
canyons resemble those on land as off the west coast of Corsica or off
the
Azores. On a smaller scale than submarine canyons are sea gullies
generally
considered to have a relief of less than 60 fathoms which are
common in a
wide variety of environments. Those in the San Clemente area
have average
maxima up to 33 degrees. The long

wall declivities of 15 degrees with local
6^ degrees in the upper portion and 2} in the lower part.
profiles are concave,
Even on the Bering-Chukchi platform which has been described as one of the
flattest erosional surfaces on earth there is a shallow but well-defined sea
valley about 3 km wide.
Recently acquired evidence demonstrates that it is probably unrealistic to
assume an idealized flat surface as an initial stage in landform evolution. It
seems that any relative lowering of sea-level would tend merely to lead the
outflow of a stream, already established inland, into the pre-existing pattern of
irregularities on the exposed part of the shelf. Also the evidence of uncon-
formities, now known in the detail of three-dimensions rather than in the two
of the cross-section, show that many new phases of landform evolution
started in the geological past on irregular surfaces. In the Smith river basin,
Montana, the volcanic ash was deposited on a much dissected pre-Oligocene
relief. Pliocene-Pleistocene erosion both followed the dendritic drainage pat-
tern of the ancient land surface and also initiated several changes in the new
pattern. Similarly it has been suggested that any belief that the parallel ranges
of hills of south-west Ireland is due to the carving out of synclinal valleys
during the Tertiary from a surface continuous at summit levels has been
shown to be no longer tenable. The existence of chalk outliers on valley floors
near synclinal axes suggest that the present landscape is essentially an
exhumed late-Cretaceous surface.
Perhaps the starting-point for the study of the effects of time in landform
evolution could most realistically be a structural surface, and the stages

investigated those of the erosional modification of this initial surface, rather
than the erosional modification of a hypothetical erosional or depositional
surface.
C. Stages in Slope Evolution
‘The secret of landscape evolution lies, evidently, in the mode of development of

hillslopes.' L. C.King. 1953
One of the most controversial topics in landform study is the manner in

which slope-profiles change with time. Disagreement is inevitably the out-
come of discussion in which imagination rather than observation serves as a
starting-point. For instance, soil scientists who have remarked on the con-
siderable conflict of opinion on both the character and the genesis of steep-
slope soils note that detailed studies are few. Similar remarks about the
evolution of pediments are numerous. Amid the uncertainty of the theoretical
assumptions about time sequences due to their limited or non-existent observa-
tional base and the sheer variety of ingenious but unrealistic idealizations of
the problem, it is difficult to isolate some facts which are known with some
One of the main
certainty. practical difficulties is to decide when and under
what circumstances slope angles tend to decline with passage of time and
when periods of parallel slope retreat, as suggested by W. Penck and K.
Bryan (1922), operate on certain parts of the slope. A related problem is to
decide whether a caprock which will inevitably ultimately disappear, may
supply a comparatively steady excess of material to the lower slopes for an
open system to be recognized, in which the lower slope for a significant length
of time, might remain essentially the same in form. Another problem is the
lack of understanding of how, in contrast, changes in the proportions of
bedrock exposure to colluvial mantle may affect the total shape.
Three main factors may influence the course and nature of stages in slope
development. First, changes may be either an adaptation or re-adaptation to
new conditions created by random changes in the morphogenetic environ-
ment, such as a cooler phase of climate in the Pleistocene, or due to some
systematic change related to the erosion of the slope itself like the progressive
removal of a critically strong or weaker stratum or the increasing influence of
some local controlling base-level. A second factor is the different rock types
which introduce fundamental distinctions into the modes of slope form change.
In Central Australia, gradients on granite or gneiss pediments do not generally
exceed 3 degrees, whereas on schists gradients commonly attain 5-8 degrees
near the piedmont junction. A marked lithological control on pediment
gradients had also caught J. Gilluly’s attention in 1937, and more recently
C. R. Twidale (1967) concluded that cannot be too strongly emphasized that
it
structure is the most important factor determining the morphology of slopes

and of the piedmont angle or zone. Slopes in alluvial material are also
distinctive and are therefore unlikely to provide models of erosion on bedrock
slopes. Thirdly, it is necessary to consider carefully some realistic recon-
struction of an ‘initial’ surface. This might be the result of differential erosion
of adjacent bodies of rock of varying resistance as in the Appalachians,
tectonic movement as along the sides of the central African rift-valley, or the
incision by the diversion of a stream into a shorter route to lower ground.
According to S. A. Schumm (1966), most hillslopes owe their existence to the
incision of a terrain by streams.
For most slopes in sedimentary rocks and in many areas of extrusive or
shallow-depth volcanic activity, the caprock, disposition and the degree to
its
which has become dissected into disconnected outliers is the crucial factor
it
in the stages of slope evolution. Slopes steeper than

26| degrees can be
maintained in the Snake river plain only by a protective rim of rock.
Even in
tropical areaswhere differential erosion of bedrock is often obscured by deep
eathering mantles, the development of weathering crusts
plays a very impor-
tant role in controlling slope evolution. In fact there
is the possibility that as
they form on valley floors, at the foot of slopes and over piedmont slopes
where dissolved products concentrate, they eventually become sufficiently
indurated to produce frequent inversion of relief as, with the passage of time,
they become the caprock of residual plateaux. This, however, has not hap-
pened in central Australia since the Tertiary, as relics of low platforms
capped by weathering crusts occur close enough to adjacent hills for the
reconstruction of former piedmont profiles. It seems that the present plains
were already the lowlands on the Tertiary surface, topped by the hills of today
and that subsequent slope evolution involved mainly the dissection of the
crust, the etching away of less than 10 m of subjacent soft weathered rock,
and the concomitant shaping of the piedmont profile from broadly concave to
angular. In most sites the hill base was set back to a structural boundary for
which there appears to have been little subsequent retreat, possibly in part due
to a change to a drier climate. Farther west in Western Australia the old
plateau is reduced to smaller remnants. In other sub-humid environments a
caliche cap resisting erosion gives rise to tablelands.
A. Rapp (1960) was one of the first to observe caprock-face processes in
detail. Due to the greater kinetic energy of the greater drop, it is possible that
high widespread caprock-face fails tend to create more stable, concave talus
slopes with a fringe of large boulders at the base, while falls with smaller
vertical drops tend to form steep unstable talus. Many theorists have sug-
gested that during the growth of talus slopes, the higher bedrock walls
diminish in height until they are totally covered. Rapp’s observations extend
this view by suggesting that as falls become progressively shorter the sorting
of boulders becomes increasingly that of the unstable talus. The scars from
which the material dropped were scattered all over the steep walls suggesting
that the mode of evolution of the rock wall was by parallel retreat. The next
stage is essentially a development in plan when chutes and funnels begin to
break up the continuity of the simple wall, but it also involves the sections of
the slope affected in downwearing by linear dissection. It might be supposed
that if these notches then intersect, the isolated portions of caprock become
ruiniform in appearance and become progressively reduced.
In situations where there is no caprock or mountain wall, low domes form
on the interfluves between pediments or similar lower slopes. Average slopes
are usually noticeably less. In this way parallel retreat becomes ultimately a
self-extinguishing process by removing the upper slope which supplied the
material for transport across the lower slope and the former lower slope
ceases to be a debris-controlled slope once the supply of debris is eliminated.
S. A. Schumm
(1956) observed that even with the disappearance of a minia-
ture badland residual, the broadly concave pediments coalesce but with a
convexity developing at their junction on the divide.
A quantity of major importance needed in establishing stages of slope
evolution is the rate at which caprocks retreat. The maximum distance of cliff
recession caused by several rockfalis from Sawtooth Ridge with the
Pleistocene was probably about 400 m. P. Birot (1965) guessed that the 60-
degree flanks of limestone domes in the tropics might retreat 5 m in 100 000
years. M. A. Melton ( 965) supposed that pediments might extend laterally at
1
about 800 m/1 million years, with a lowering of the surface at about one-
tenth this rate. The ratio of these two rates, and an assumed differential
lowering due to progressive exposure of the pediment bedrock, suggests that
the latter would theoretically slant upward to the base of the retreating wall at
the angle of 5-6 degrees.
In contrast to stages in slope development accompanying the reduction and
removal of a caprock. dome shapes in certain massive rocks appear to be self-
perpetuating by favouring the development of expansion joints and their

concentric patterns. B. P. Ruxton and L. Berry (1961) describe how on the
stripping of rock waste off granite batholiths in east-central Sudan the bare
rock hills at first retain their simple slope profiles, but later large pits are
weathered out at the intersection of joints, cavities are formed beneath some
sheets and depressions are excavated just below the angles at the top and
bottom of the encircling rock fans. However, it appears that as a stripped
residual hill gets smaller the dome form extends downward, eliminating the
angular junctions bounding the dome and rock fans until a steep rockface
meets the fiat clay plain at an abrupt angle. Usually this change affects only
one side of the hill producing a half dome. If dome-shaped hills in massive
rocks tend to be self-perpetuating the same control of expansion joints in
bowl-shaped or concave slopes might operate to maintain their shapes.
Some of the most incontrovertible stages of slope evolution in solid rock
occur in areas where mass movement is dominant. Three main types might
be recognized. First, there are landslides with negligible rotation, like the
Chuska Mountains landslides where eastward from the escarpment crest there
are up to seven successively lower ridges. In the highest of three main
landslide areas there are huge caprock blocks ranging from 9 to 300 m in
length. 9 to 30 m wide and high, broken apart slightly along the vertical
planes of intersecting joints but still essentially horizontal. Troughs have flat
sand-covered floors but contain little coarse debris. In the second area,
caprock blocks are broken into a jumbled mass of individual blocks and
erosional debris nearly
fills the troughs so that the ridge and trough form
is
obscured. Along the base of the steep slope separating the highest from
the
second area of landslide debris a series of springs emerge which have helped
to erode a wide flat area. In the third and lowest area of
landslide debris, the
main continuous remnant is about 0-7 km wide, but debris is scattered
o\'er theCretaceous shale eastward for a further 8 km. Canyons of intermit-
tent streams exceeding 30 m in depth are the main
relief form, the blocks of
froeJur^
Caprock
A
Santl with
minor ~ Old landslide ridges from previous cycle
hardbeds
Cretaceous shale
.2nd froclore
Îst ridge block
B
Basal block glide
Figure V.5 Evolution of escarpment slopes with progressive block glide of ridge-
blocks on sand (from Watson and Wrighu 1 963). This example, observed on the
east flank of the Chuska mountains, north-western New Mexico, also shows lateral
spreading on shale of loose sand during block glide of the entire mass and the
progressive diminution in scale of the ridge and trough forms.
sandstone, now less than 1-5 m forming the major relic of the landslide
topography, protecting ridges up to 6 m in height. There is a reddish soil
mantle 0-6-0-9 m thick, a horizon of secondary carbonate and sandstone

blocks blackened by desert varnish. The entire area has been lowered by
erosion and downwasting of the landslide mass and the underlying
Cretaceous shale and. as in numerous similar situations, it is difficult to
decide on the degree to which detached masses have slid laterally and the
degree to which caprock remnants have been essentially lowered by the
removal of the adjacent or underlying less resistant material.
A second type of movement in which stages of slope evolution can be

clearly deciphered is where rotational landslipping occurs. Stages may
involve the pre-slip surface, its abrupt disorganization by slippage then either
its gradual regularizing by slopewash or the enlargement of the scar. The
relatively undissected nature of the slide material of the many postglacial
landslides in the Central Pennines gives some indication of how long stages of
IE 21
Landforms and time
12 3Z:
349
Figure V.6 A downslope senes of rotational slips (from G. Reichelt, 1967,

Erdkunde, Vol. 21). on the western slope of the Eichberg, north of Achdorf.
slope evolution in such circumstances might be. There may also be stages
within the slipping phase. In the Eichberg slip near Achdorf, it seems that the
slumped material of an initial rotational slip near the slope summit loaded a
downslope section of the hillside, inducing a second rotational movement and
that this initial disturbance perpetuated a sequence of five rotational slips in
all, with the amount of rotation decreasing downslope. Similarly, the over-
steepening of the toe of a slope could, by removing the lateral support from
segments of a hillside, propagate upslope a chain reaction of slips analogous
to that produced from above by overloading. A third type of mass-movement
involves the free fall of individual rock fragments which may form definable
stages of depositional forms in cool environments where there are seasonal
snow banks beneath rock walls. There are two clearly distinguishable stages
of slope platforms with protalus ramparts. Younger platforms have well-
defined ramparts and distinctive ditches made up of little weathered material
with a scant vegetation cover. Older talus platforms have only moderately
clear ramparts and poorly defined ditches, made up of moderately to slightly
weathered material largely buried beneath a thin soil veneer.
This would not apply to features as small as many dolines. the funnel-
NORTH SOUTH
Figiiie V.7 A slope on the south Rank of Navajo mountain showing platforms of
two ages each with a sequence of protalus ramparts. R, and associated ditches,
D
(after Blagbrough and Breed. 1967).
shaped hollows which are one of the most widespread characteristics of

limestone relief. In this case close interdependence with an underground
drainage system is an important factor in slope development, and a complicat-
ing factor is the different modes of initiation of depressions. Nonetheless, at
least some stages in the evolution of slopes on the flanks of enclosed depres-
sions are observable. These slopes therefore hold general implications for
landform study as well as being of particular interest to karst specialists. In a
bare karst, where an appreciable area of limestone is exposed at the surface,
there is often no reason why a depression should develop at a particular point,
and solution tends to lower the limestone surface as a whole. One of the most
effective ways in which solution concentrates at a particular point in lime-
stone depends on an initial stage when the limestone, as a covered karst, is
still by a largely non-soluble stratum; thin or more permeable points in

overlain
the cover rock then localize solutional activity at particular points on the under-
lying limestone, perhaps for substantial periods of geomorphological time. A
pocket of deeply weathered limestone may therefore already be in existence
before the first stage in the surface slope development appears as the cover
rock founders. In part the initial hollow may be the result of removal of
some of the cover rock by piping. The Mendip area, therefore, is typical by
having a very large number of its enclosed depressions on the Mesozoic
rocks which flank the outcrop of massive Carboniferous limestone. If a connec-
tion with an underground drainage system is maintained, the depression en-
larges. In some situations the greater biological activity in the deeper soils of
the depression favours more rapid solutional activity, but it seems common
for rillwash to develop on the slopes of depressions enlarged beyond a certain
size. This induces progressive infilling of the floors which reduces perme-
ability. arrests deepening at the base of the slopes and favours relatively
accelerated solution on the upper slope. As in the Indiana karst, ponds may
form, sometimes as rapidly as collapse features appear. Their infilling with silt
and organic debris is particularly rapid. Similar developmental stages have

been envisaged for the slopes of depressions created in calcareous rocks by
water draining off an adjacent insoluble or impermeable rock. An example is
the pockets developed in the chalk of northern France at the contact with the
overlying Tertiary strata. Occasionally surface depressions appear above an
underground stream which may evacuate much of the debris, leaving the
walls of a cylindrical shaft as the initial form. Some collapse features, like the
shaft doline of Modro Jezero. near Imotski in Yugoslavia, have scree banked
up on their lower slopes. From this stage the infilling of finer material would
then lead to slope forms similar to those of covered karst origin, which
become increasingly concave as the debris fill grows higher. In the initiation
or reactivation of slope development in enclosed depressions seismic shocks

are sometimes significant.
Figures V.8 and V.IO Observed changes in slope form, (/I iôve) Evolution of sides
of a doline {H. Alojzij, 1953, Kraska Ilovica). {Below) Stages of coastal cliff
development in north-east Yorkshire {Agar, 1960).

Zone of solution of carbonates
and refatfye accumulation
of sesqufoxides
CT’-v^-
Zone of chemical precipitation
of limestone
Water level during

Biostasis Chemical deposits
’
I I
'
of friable limestone
Friable limestone
Zone of detntai and

chemical sedimentation
RHEXISTASIS Deposit of powdery

Due to organic movement nodules of sill
(lowering of base level)
Landfonns and time 353
One of the main limitations on the study of slope evolution is the typical
absence of sequences of sediments from which the phases of erosion on the

upper slopes might be inferred. However, in many areas the scope of the
study of lake sediments in this context awaits investigation and in many semi-
arid or arid environments such sedimentary sequences may remain on lower
slopes in basins of internal drainage. J. H. Durand (1959) describes such a
situation in Algeria and sketches some typical changes in soil and slope
development (fig. V.9). Here terra rossa formed on upper slopes under a damp
and warm climate due to the relative enrichment in sesquioxides of the soil in
relation to the parent material of limestone or dolomite, and also due to the
addition of soil material from an external source. Water draining from the
slopes, rich in dissolved calcium carbonate, led to the precipitation of friable
limestones on to lake floors at the foot of the slopes. The absence of clastic
material in these limestones indicates a period of biostatis with no erosion at
this stage (fig. V.9. A). In a subsequent phase lowering of local base-levels and
the drying-up of lakes encouraged erosional activity which extended some
way up the slopes. At this stage the rhexistasis. indicated by clayey silts and
by the powdery limestone nodules formed in contraction fissures by the segre-
gation of calcareous material from other sediments, was due to tectonic
movements rather than to climatic change as the moisture regime conditions
had not changed (fig. V.9.B). As
new equilibrium was established erosion
a
ceased and the drainage water again became sediment-free. The calcium
carbonate was then deposited on the lower slope as a layered crust due to its
precipitation from sheetwash. each layer hardening between successive sheet-

wash episodes (fig. V.9.C). In a fourth phase, characterized by desiccation, the
reduced effectiveness of the sparser vegetation in filtering sediment out of
surface water led to some erosional downwearing of the upper slope and to
the transportation of red silts and reddish soils to add to the depositional
layers on the lower slope (fig. V.9.D). The red silts are the product of the
more intense erosion.
Of rapid changes, forms in unconsolidated materials provide the clearest
examples. In Rhodesia, sand ridges up to 120 m apart are now only 3-4 m
high, sheet erosion probably being the main agent in the destruction of
inferred former dunes. It is likely that erosion and trough-filling would be
most rapid initially when steep slopes still prevailed. Similar in the degree to
which unconsolidated material undergoes rapid change is the instructive
sequence of forms observed, in the moliisol banks in the Mackenzie
delta
area, by J. R. MacKay (1966). In early summer, the lower scarp face and
Figure V.9 Phases in slope development in semi-arid

Algeria linked with stability
of the vegetation cover, erosion on the upper
slope and the accumulation of
stratified slope deposits at the base
{after Durand. 1959).
354 Introduction to Geotnorphology
frequently also the middle is covered with slumped debris which accumulated
in the previous autumn. In general, the scarp face is cleaned olT by mid-July.
When first exposed, the surface is usually smooth from top to bottom, but it
may roughen as the season progresses. As thawing proceeds, ‘caprock’ hum-

mocks. often bordered by roots are undermined, overlying and frequently
falling to the base of the scarp. A ridge and gully system encroaches on the
scarp face until it becomes partially or totally ridged. Towards the end of
summer the ribbing tends to disappear, perhaps reflecting the decreasing
effect of stream erosion towards the end of the summer. As it appears to be a
universal characteristic of slope development, the transportation of material
from the toe of the slump as quickly as it accumulates favours active slumping
and the maintenance of a steep profile, and other features of this small scale
example might be relevant to the study of slope development on larger scales

in space and time. S. A. Schumm (1956) observed seasonal changes on micro-
pediments showing comparable contrasts in degree. During the spring much
of the runoff from these slopes flows beneath the layer of aggregates to
reappear on the pediment at the base of the slope. However, the proportion of
surface runoff increases during the summer as the slopes are compacted by
rainbeat and become less permeable.
The erosion of marine cliffs, where rapid, may also point to some more
broadly applicable generalizations about slope development. For instance, in
1956, January-March storm-wave erosion of the clifT base at Highland, on

Cape Cod. left a basal scarp 1 • 5-4- 2 m high. Although this scarp was slowly
buried by colluvial material from the clifT face and was gone as a cliff feature
by May 956, the line marking the top of the scarp was
1 stilt distinguishable in
September 1956. This example lends support to the belief that important
stages in the history of a slope’s evolution might be no more than mere

subtleties in the present-day form. It also points to the usefulness of the study
of coastal slopes or cliffs as models of situations where basal undercutting is
undoubtedly effective.
Equally significant is the cessation of undercutting, and particularly instruc-
tive is the contrast between abandonment due either to base-level fall or to

accretion at the foot of the coastal slope, transportation of debris tending to
be more rapid in the former case. Even the undercutting of solid-rock coastal
cliffs and slopes may lead to well-defined evolutionary morphological stages.
Examples occur along the north-east Yorkshire coast, as described by R.

Agar (1960). due to the undercutting of cliffs made up of Jurassic shales,
sandstones, and shales with thin hard ironstone bands. In the last inter-glacial
the cliff-line receded from about the present low- water mark to within 20 m
of the present clifT. Large fallen blocks remained on the foreshore (fig.
V. 10. A). With the onset of the Wiirm glaciation, undercutting ceased with the
lowering in sea-level, and the abandoned foreshore underwent weathering.
colonization by vegetation and soils formed. The shale under the sandstone
boulders may have been protected from these changes. The cliff-top
weathered back, and debris cones and landslide debris filled the cliff-foot
notch (fig. V.IO.B). In late-glacial times narrow stream gorges developed in

the drift was rapid postglacial sea-level rise reaching to within 5 m of
but it
present sea-level which led to the clearing away of much of the drift by
undercutting; there was, however, no appreciable erosion of solid rock (fig.
V. lO.E). Renewed rise of sea-level from the pause at —5 to —3 m OD level,
passing the present level and rising to a maximum of 9 m OD, led to under-
cutting and removal of debris cones from most of the old interglacial cliff
foot. Undercutting of the solid but was apparently insuf-

cliff also started
ficient to undermine the The slow undercutting of the solid cliff and
cliff-top.
slow downwearing of the foreshore started about a.d. 1000 and continues
today (fig. V. lO.G). The result of undercutting in postglacial times is a cliff on
average about 30 m high. At points along the coast where the lower zone of
active undercutting has not yet affected the full height of the cliff, recession of
the cliff-top is negligible (fig. V.IO.E-G). Conversely, with the full height of
the cliff affected, parallel retreat takes place with measured rates of recession
being similar for the basal notch and for the cliff-top. At Huntcliff, the
portion of a Roman Signal Station that has fallen away indicates a cliff-top
recession of 3-6 m/ 100 years at a point where undercutting affects the full
height of the cliff.
Fig. V.IO.B illustrates a common pattern in the interglacial evolution of

coastal slopes and cliffs with the lowest sections of an old solid-rock cliff
being present early in the phase of sea-level fall and retaining its original
steepness, but with the cliff above the level of protection tending to wear back.
In southDevon such upper slopes are irregular and have declined to angles of
20-35 degrees (Orme, 1962). This formation of an upper convexity contrasts
with the essentially static nature of the north-east Yorkshire cliff-tops during
the shorter time span of postglacial times.
D. Stages in Drainage System Evolution

‘The principle of first importance in the writer’s view of geomorphology is the
concept of essentially universal superposition (or superimposition) of drainage in

fiatland regions.’ F. A. Melton, 1965
. superimposition of drainage on the basement may not be a necessary con-

‘. .
sequence of the removal of coastal-plain sediments.’ R. F. Flint, 1963
One of the basic considerations in studying the history of drainage pattern

evolution is the consequences of an actively downcutting stream
encountering
different subjacent rocks. If a series of strata are essentially
conformable the
Figure V.l J Buried land surfaces showing the contrast between the hypothetical
peneplain and actual reconstructions based on the boring logs from intensively
drilled oil-fields.
A. The buried peneplain, assumed to have been bevelled by erosion, on which

superimposed drainage could develop with the downcutting of drainage estab-
lished on the unconformable cover of younger strata {from Sparks. 1960).
B. Cross-section of the south Alamyshik oil-field, Uzbek S.S.R., Turkestan {from

Martin, 1966), showing an escarpment on the flank of an anticline and its econ-
omic significance as an oil trap. The position of wells is also indicated.
lowering of a stream-bed from one stratum on to the next is accepted simply
as valley downcutting in which any attendant changes might be limited to
those of slope form, with the pattern of drainage remaining unchanged.

However, since many lowlands are made up of unconsolidated sediments
lying unconformably on much older and harder rock, the possibility is fre-
quently considered of a stream actively downcutting through a cover rock
and ingraining its pattern on to subjacent rocks with disregard for their
different structural attitude. The likelihood of this development was enhanced

in W. M. Davis’s scheme of drainage superimposition by the assumption that
the unconformity had the virtual flatness of the hypothetical peneplain with
the differing resistances of underlying structures planed off.
Although the lowering of a drainage system from a high rock on to one
beneath is inevitable, one of the difficulties in accepting the idealized scheme
as a starting-point in drainage development is that the assumption of a rela-
tively subdued erosional surface at the unconformity beneath the cover rock
is increasingly found to be mistaken. Borings now indicate that there were
waterfalls at least 25 m high on the Paleozoic bedrock relief now buried in the
Lower Mississippi valley, and that all the way from Cairo. Illinois, to the Gulf
of Mexico, the buried pre-Recent surface is broadly rolling and markedly
incised by an intricate valley system.
Another difficulty, posed by the hypothetical initial conditions, is that of

visualizing how the initial surface, unless it be a volcanic cone produced
overnight, comes into existence without being subjected to the existing pattern
of drainage in the process. For instance, the emergent marine areas frequently
supposed to provide an ideal gently inclined smooth surface on which
‘consequent’ streams develop are in reality more likely to be mere seaward
strips in the path of outlets from existing drainage systems with an existing
pattern perhaps closely tied to structures continuing seaward beneath the
covering fringe of marine sediments.
A problem of applying this concept in the field is the absence in many areas
of any geological evidence to demonstrate that a cover rock once spanned the
area. In such instances the base of a possible cover rock is projected geometric-
ally from its outcrop in an adjacent area. However, this could be a misleading
reconstruction as in many areas Tertiary deformation was substantial and
widespread. G. H. Dury (1959) considered that a Carboniferous cover as an
C. Cross-section of Athabasca ‘Bituminous Sands’ area, north-eastern Alberta

{from Martin, 1966). The basal Cretaceous sandstone units of the McMurray
formation were laid down in a connected valley system where erosion-resistant
layers in the underlying Devonian-Mississippian sequence stood out as
north-west
to south-east trending escarpments. Note the thinning north-eastward
of the
underlying Devonian strata, due to interstratal solution in salt-bearing
formations,
and the resultant subsidence in post-Devonian strata.
initial stage in superimposition of drainage in Donegal could be ruled out,

‘especially since the projected base of the Carboniferous passes well above
summits in west Donegal’. This hypothetical geometry, however, is mistaken
as there are small Carboniferous outliers actually in existence below the
present summit levels. Conversely, some reconstructions which are geometric-
ally possible for one area are irrelevant or geologically impossible in others.
The extension north and west of the Cretaceous outcrops appears feasible in
lowland England; yet it is pointless to extricate the same strata from beneath
the Tertiary lavas in Northern Ireland as the latter have the discordances
between structure and drainage which the Chalk cover might be invoked to
explain elsewhere, and impossible to extrapolate from its small outlier en-
closed at the base of exhumed late-Cretaceous valleys in Co. Kerry in south-
west Ireland. In other instances the circular argument, that observed dis-
cordances between drainage and structure point to the existence of a former

cover rock, is sometimes used where independent evidence of the existence of
a former cover is lacking.
A second basic consideration is the mode in which a river enlarges its
valley. After W. M. Davis’s imaginary, initial stage of drainage development,

the ‘consequent’ streams became incised across a structural grain of the
buried peneplain assumed to strike at right-angles to their course. The
remainder of the cover rock was gradually removed as ‘subsequent’ streams
started to hollow out strike vales by a mechanism termed headward erosion.
This is assumed to be a self-accentuating process which gets more powerful
the larger a catchment area it cuts. Thus the most powerful stream in an area
with no self-regulating mechanism envisaged to check it, breaks through
divides with Davisian zeal and cuts off the line of descent of less powerful
neighbours by decapitating them. Amid all this excitement and the large
number of interpretations tacitly depending on certain assumptions about the
efficacy of headward erosion it is difficult to realize that the mechanism itself
has scarcely ever been studied or described. Almost invariably it is the form
of the ground that provides the evidence.
A search for the source of the boundless energy of headward erosion is
hampered by features due to artificially accelerated erosion in many head-

stream gullies. However, it is clear that in many instances such gullies are not
cut into an unmodified surface but in some cases represent the transformation
of lines of concentrated flow of sub-surface seepage into surface gullies. This

change seemed to explain the headward extension of gullies in the New
Forest. Hampshire, where longitudinal growth appeared to result from the
coalescence of discontinuous hollows rather than by upslope migration of a
channel head. In fact. R. E. Horton, defined the summit of a divide as a ‘belt
of no erosion’. M. A. Melton (1958) considers that the maintenance of a

length of stream channel depends on a certain catchment area needed to
supply runoff sufficient to remove the products of mass wasting tending to

obliterate the channel. As B. T. Bunting (1961) also argued, the extension of
incipient drainage or seepage lines into crestal areas must gradually cease
when the crestal area, reaching some minimum width, can no longer yield
sufficient moisture to the adjacent seepage hollows or cusps for the processes
of corrosion to continue.
An appreciable exaggeration of the erosional potential of first order head-
streams results if headward erosion is applied loosely to include knickpoint or
waterfall recession and spring-sapping in this term. The last two are funda-
mentally distinct due to the organized drainage system that exists above the
rapids, fall or spring. From a study of laboratory streams L. M. Brush and
M. G. Wolman ( 1 960) concluded that in actual streams a knickpoint prob-
ably would not be recognizable more than several miles upstream from its
initial position, unless the original fall were extremely high, and that it would
be most unlikely to travel the entire length of a natural stream. Added to this
is the field evidence that most falls of great size are not situated in high
mountains but on plateaux often of great age and are sometimes relatively
close to the sea.
It seems that if a history of drainage evolution emphasizes the formation
and headward growth of new channels this view is largely a logical deduction
from the basic assumption of an initial, ideal undissected surface, following
inevitably as an artifact of the initial premise rather than as the result of
observation of actual landforms.
In some areas where structurally controlled drainage appears to have
flowed for geological spans of time, the concept of superimposition is super-
fluous. R. F. Flint (1963) considered that superimposition of drainage on a
basement may not be a necessary consequence of the removal of coastal plain
Zone seems to be
sediments, as the general drainage pattern north of The Fall
much the same as it was before Cretaceous sedimentation began.
J. W. Ambrose (1964) concluded that for a 5 million-square-km tract of the
Canadian Shield a drainage pattern, once adjusted to structure, will persist

indefinitely, and if exhumed would be reactivated virtually intact.
buried and
The hills and valleys, of the order of 2000 million years old. are closely
related with bedrock strike and type. The inter-stream areas are more re-
sistant granites and the streams and lakes

lie in softer gneisses and schists. In
northern Scotland, the valleys of the River Lossie and the Black Burn, south-
west of Elgin, correspond in part with pre-Devonian valleys.
A factor similar to that of the persistence of rigid geological controls is that
many contemporary major rivers had precursors following a similar approxi-
mate trend to their own at times in the geological past. The St Lawrence
drainage has probably emptied near where it now does since well back in the
Tertiary. Historically the area that is now the Amazon basin existed
as a
trough throughout most of the Mesozoic and Paleozoic eras and possibly into
the pre-Cambrian The course of the Lower Senegal river has probably
era.
not changed significantly since Miocene time. The Congo has been evacuating
its cuvette along the same course since Miocene or even Oligocene times. F.
Dixey (1938) considers that the upper parts of the main rivers in Madagascar
and East Africa are of ancient origin and might date back as far as Jurassic
times. In the more immediate past and affecting innumerable rivers of much
smaller size is the deepening of channels by the order of 30 m or more
beneath their present levels by downcutting during Pleistocene low stands of
sea-level.The buried channel beneath the Quinnipac river as it enters Long
Island sound is at least 50 m below sea-level and possibly more than 60 m. In
northern Connecticut, pre-Wisconsin entrenchment is more than 30 m below
present sea-level as is that of the Pleistocene Elizabeth river; the entrench-
ment beneath the Pleistocene James river is 50 m. Drowned valleys filled to

varying degrees by alluvial material are common around Australia like the
River Hunter where depths of 75 m have long been known. The preglacial
Teifi is typical of many streams in Wales by having a maximum depth at its
mouth of 22 m below present sea-level, although a depth of 40 m is known in
the River Tawe valley 10 km from its mouth. Depths of at least 20 m occur in
the Avon, 35 m in the Severn, and 25-30 m in the Solent. The broad implica-
tion of the existence of these buried valleys for stages in the development of
the lower courses of rivers seems clear. A relative downward movement of
sea-level by several dozen metres would not lead to any substantially new
river valley development but merely to the re-excavation of existing buried
channels. For the Bristol Avon, as with the River Tawe. it has been suggested
that the buried channel was already eroded by the First or Second Phase of
the Last Glaciation. With ice recession and subsequent marine transgressions
the channel was probably infilled, subsequently to be re-excavated during
Figure V.12 Some examples of stages in drainage system development.

A. Development of a dissected alluvial fan in south-west United States (from
Beaty, 1961): (i) initial conditions, (ii) immediately after faulting, and (iii) present
situation.
B. General changes in the drainage of southern Uganda due to tilting (from J. C.

Doornkamp and P. H. Temple, 1966, Geog. Jour., Vol. 132): (i) before the middle
Pleistocene reversal, and (ii) the present lakes and rivers.
C. Effects of back tilting on tributaries in part of southern Uganda (from
Doornkamp and Temple, 1966): (i) lacustrine deposits of a former lake west of
Mbarara, and (ii) the system of lakes and marshes (stippled) to the east and their
overflow into Lake Victoria.
D. Drainage systemin the Lower Sind before and after the Indus diversion in
1758-9 (from H. Wilhelmy, 1966, Erdkunde, Vol. 20), based on information
from M. R. Haig, H. T. Lambrick, and others.
marine recessions of the Third and possibly the Second Phase of the Last
Glaciation and finally to be refilled with post-glacial deposits before and
during the Flandrian Transgression.
Perhaps because of the artificial separation of uplift from erosion in
W. M. Davis’s imaginative models, the degree to which the history of many
drainage networks reflects tectonic changes has perhaps been underestimated.
Although identifiable alterations due to strike-slip faulting are very localized,
modifications due to other crustal disturbances are more widespread. In
northern Utah, when the Bear Lake was 2 m above its present height an outlet
developed at the west side of the valley. The most logical explanation for a
subsequent easterly migration of the outlet is movement along the Bear
Lake fault. Similar movements might explain why all fans on the western
flank of the Black Mountains in Death valley have deeply incised channels
cutting far into their apex region, leading down to a newer part of the fan (fig.
V. 12.A). In another tectonically active area, westerly uplift resulted in an

eventual reversal of the rivers Kagera. Katonga. and Kafu from their former
courses to the Albert and Edward into the present Kyoga-L. Victoria system
(fig. V. 1 2.D). If the rivers like the Kagera. Katonga. and Kafu were originally
Congo headstreams flowing to the Atlantic, this connection was severed
during or before early Miocene times and subsequently were controlled by
successive rejuvenations due to tectonic movements in the rift valley. In the
Murray basin in Australia, a displacement of 15 m on the Cadell Fault near

Mathowa diverted the ancestral Murray to the north. The abandoned section
of its former course. Green Gully, is now a linear depression on the elevated
tilt-block. Multiple branching indicates that more than one phase of ancestral
river activity was responsible for the pattern of ancestral rivers. The example
in the best traditions of classical Uniformitarianism is the Casiquiare Channel
curiosity which connects the Rio Negro, a principal tributary of the Amazon,
with the Orinoco. The direction of flow in this channel depends on the flood
stages of the two river systems and reverses seasonally. Here only a minor
regional tilting would cause the diversion of a sizeable portion of the Orinoco
drainage into the Rio Negro.
Although tectonic movements may initiate diversions or even reversals of
drainage, some rivers maintain their courses during such movements to the
extent that this applies to most drainage systems with the term antecedent
describing the case where the maintained course becomes a valley discordant
to the new structure. In the Lower Truckee Canyon area the surface of
Pleistocene volcanic lavas was initially near-horizontal and the river cut down
concurrently with a westward tilting so that the present canyon is antecedent.
Because no sediments have been found upstream to indicate ponding during

of downcutting in the Lower Truckee Canyon appears to have
uplift, the rate
equalled or exceeded the rate of uplift. The nearbv upper Missouri river is
Landforins and time 363
also to some extent antecedent. In India, slow or periodic tilting in the

Pleistocene caused the old Brahmaputra river to become antecedent in places
by necessitating river scour into comparatively resistant Pleistocene sedi-
ments.
In addition to active tectonic movements initiating diversions this signifi-
cant stage in the history of a drainage system may be due to a combination of

the erosional and depositional activities of a river itself, in some situations
operating in conjunction with static geological factors like lithology. At the

downstream end of rivers the very low gradients on some lowland plains
mean that either heavy sedimentation in a channel, or only a slight regional
tilt or subsidence, may favour diversions of river courses involving huge
volumes of water. For instance, continued subsidence on the western flank of

the Mississippi modern sub-delta is causing diversion of the river to the coast
1 60 km west of the present bird’s-foot delta. A feature of such situations is
the possibility of a chain reaction as the excess flow in turn causes diversions
in a receiving river. The sudden change of course of the Tista river with
resulting addition of its waters to the Brahmaputra river may well have been a
contributing factor towards the diversion of the latter (fig. V. 1 2.D).
In middle courses of streams, stages in drainage pattern development may
accompany and follow lateral stream cutting. In areas of horizontal strata or
homogeneous rock, a river deepens its course with little lateral erosion, but
where there is even a slight regional dip or the systematic development of less
resistant strata on one valley wall compared with the other there is a greater
or lesser element of lateral shift in the position of a downcutting stream.
F. L. Stricklin (1961) describes slopes of the Seymour valley, Texas, which
suggest that the valley floor was cut primarily by two major streams. The
bedrock floor beneath the Seymour has a dominant northerly slope on the
south side of the Brazos and a dominant easterly slope on the north side
ranging from 1-4 m/km. The bedrock surfaces are smoothly bevelled and
overlain by coarse, cross-bedded alluvium, indicating their origin by lateral
stream planation. and appear to reflect a northerly shift of the Brazos and an
easterly shift of one of its large tributaries. The position of the Pearlette ash
indicates that the shifting in channel position has probably been in operation
for a few million years. In an area farther north, in the Three Forks basin,
Montana, the Bridger Range began to rise in late Pliocene or early
Pleistocene times and because of the northerly tilt the main streams like the
ancestral Madison. Jefferson, and Gallatin tended to shift northward with
time. On the eastern side of North America in Tertiary times, the attitude of
the consolidated rocks in the Teays river basin apparently exercised some
regional control on the river and its tributaries, causing the former River
Teays to migrate laterally by down-dip This 1500-km-long river flowed
shift.
from its headwater region in the Piedmont of Virginia and North Carolina.
north-westward and west and finally into the present Illinois. A pre-
dismemberment of this former major drainage feature
Pleistocene stage in the
was its diversion at Scary. E. C. Rhodehamel and C. W. Carlston (1963)
believe that down-dip shift of the stream was involved in this diversion, that
an uncommon thickness of sandstone formed a local base-level and greatly
slowed the rate of downcutting immediately downstream from the point
where the diversion eventually occurred, although the final mechanism that
accomplished the diversion at the ‘elbow of capture’ is not known. However,
Figure V.13 Further examples of stages in drainage system development. (Above)

Structural movements causing the reversal of the Lemki river, east central Idaho,
in late Pliocene to Middle Pleistocene times, a change typical of much of an area

where drainage probably has been structurally controlled since the advent of
range-front faulting in Miocene times (from Ruppel, 1967). (Below) Sequence of
alluvial terraces (from P.F. IVilliams and B. R. Rust, 1969, Jour. Sed. Petrol.,
Vol. 39) in the Donjek valley, Alaska. Downcutting may be due to a change in
hydrological regime, prolonged glacial wastage in this area increasing melt-water
volume and raising total discharge.
as this and some other diversions like that illustrated in fig. V.14 resemble in
many details diversions on inclined strata in the southern Pennines. an
explanation suggested for the diversion stage in river drainage history in that
area might have some broader application. Fig. V. 1 5 illustrates the suggested
mechanism, termed down-dip breaching (Pitty, 1965). Stages in drainage
diversion may start with indentation along a weaker scarp section by dip-
controlled stream erosion, followed by lateral erosion within the indentation,
combined with an eventual overtopping of the resultant lowered divide which

could involve an accumulation of alluvium on the erosional surface within the
embayment. A remnant of this surface may persist as a knoll in the funnel-
shaped entrance to the top through the divide cut by the diversion. Alterna-
tively in some parts of the southern Pennine scarplands with a low dip or
where the bench formed between two resistant strata is narrow, alluviation in
the channel may make diversion possible over the lower stratum forming the
up-dip margin of the drainage basin. This diversion mechanism, termed up-
dip breaching (Pitty. 1966) leads to a diversion channel cut obliquely to the
strike (fig. V.16).
A feature of river diversions is that they may result from a rare or even
unique event. One of the clearest examples followed the overflow of Lake
Bonneville into the Snake river, perhaps in the early part of the last glacial
figure V.14 Successive positions of laterally shifting streams in Texas

{from
Stricklin. 1961). as determined by outcrops of the late Kansan or
early
Yarmouthian Pearlette Ash.
about 20 000 years ago. The overspill itself was perhaps the result of drain-
age changes in the Bear river. The flood was of such great volume that at
bends and restrictions in the Snake river canyon it was some hundreds of
metres deep and it overflowed the canyon walls, took short cuts across the
plain and created spectacular bars and plunge pools where the overflow
waters returned to the canyon. This diversion followed earlier diversion
stages in the history of this river which occurred when the extrusion of the
Snake river basalt in the Cedar Butte locality dammed the river, with the
water impounded in the ponded lake extending about 60 km upstream.
When a river system exposes limestone on its floor distinctive, although
not universally similar, stages in drainage evolution usually proceed until
Figure V.15 {Above) Stages in the development of a down-dip breach {Pitt\\
1965).
Figure V.16 (Below) Stages in the development of a counter-dip breach.

much of the limestone is removed. The final stage is the re-establishment of
normal drainage on the impermeable basement beneath the limestone strata.

The gradual diversion of drainage underground depends on more factors than
the solubility and jointed nature of limestone. Tectonic movements are par-
ticularly significant in the earlier stages. In Yugoslavia, tectonic movements
along major lines of weakness have been fundamental in the deranging of
stream patterns and in the draining of lakes. Some lakes have actually been
observed to drain temporarily after earth tremors. In some areas normal

surface drainage persisted before orogenic movements took place as fissures
were then few. In the Cracow upland, dry valleys and cave systems developed
only after Miocene orogenic movements had occurred. A second factor in the
diversion of drainage underground is the necessity for there to be a low point
in the limestone outcrop at some distance below the general level of the
limestone land surface to serve as a point of hydrographic concentration. In a

ponded karst, where impermeable rocks surrounding the limestone form a
relatively high rim, development of underground drainage is restricted. An
important consideration in the contrasting type of karst, the perched karst,
should the limestone extend to the sea, is the degree to which efficient points
of hydrographic concentration developed during low stands of sea-level in

Pleistocene times. On the Adriatic coast, submarine springs are numerous
and water-levels encountered 200 m below the bed of the Rijecine seem to
have been affected by a base-level below the Adriatic itself. In general, the
lower the point of hydrographic concentration the more the karst evolution
resembles that of the dinaric type; its periodic lowering, particularly where it
is the level of an impermeable rim which controls the unwatering of the

limestone mass, may favour the development of a sequence of levels in the
abandoned drainage system (fig. V. 1 7. A). The present valley occupied by the
Plitvice lakes is an old cave-level, a higher level having been formed by
the floor of dry caves which enter the gorge above the present lakes.
M. M. Sweeting (1950) has demonstrated how the levels of caves in north-
west Yorkshire tend to occur close to certain altitudes. A third factor con-
cerns the degree to which water flowing over a limestone surface is capable of
enlarging initial tectonic or structural lines of weakness, and of exploiting the
potential depth for cave development where the point of hydrographic con-
centration is low. In some limestone streams, particularly in chalk areas, the
water at its source has as high a concentration of calcium carbonate in
solution as at
any point on its lower course (fig. V. 1 7.B). In consequence, the
most spectacular features associated with stages in the diversion of drainage
underground occur where catchments, made up largely of non-calcareous
rocks, channel water low in dissolved carbonates on to a limestone outcrop.
Quite often such an allogenic stream in fact maintains a surface course over
the limestone. However, seems that a zone of bedrock
it is sometimes
Figure V.17 Aspects of the development of drainage systems in calcareous rocks.
A. Levels in the Sainte Anne cave at Tilff linked with downcutting of the Ourthe
river, Belgium {from C. Ek, 1962, Rassegna, Spel. Italiana, Vol. 14).
B. Changes in chemical quality of streams at a geological contact, as observed in

the Hampshire basin. Where Eocene sands still cover the chalk the calcium
carbonate content of surface streams is well below the consistently higher level of
those where chalk is exposed at the surface (the values are means of five observa-
tions. 1 96 1-2).
dissolved beneath such a stream. The zone of solutional penetration, with
clastic debris filling the cavities, was 70 m deep at the Kentucky dam site in
the Tennessee valley. downward penetration of

Impermeable beds limit the
solutional activity. The larger sub-river openings were 30-300 m long, and
joint controlled, but occupied a zone only 20-30 m wide at the top and
tapered downwards. From the information provided by the unusually intensive

borings in the river-bed of the Tennessee valley it is readily envisaged how, as
a general pattern, a fall of local or regional base- level could rapidly transform a
limestone valley into the steep-sided gorges which typify many limestone
areas. Alternatively an allogenic stream may sink on meeting a limestone
outcrop and a cave system develops. If the insoluble material in the headwater
area is relatively easily eroded and evacuated through the cave system, the
valley upstream from the sinkhole may be deepened by vertical erosion more
rapidly than the general level of the limestone surface is lowered by solution.
In this case a blind-valley develops near the geological contact, with a horse-
shoe-shaped cliff or steep slope facing upvalley and the sinkhole near the base
of the blind-end. A striking example is the Pazin river at the edge of the shale
outcrop in Istria. This drainage system also illustrates another unusual stage
in drainage evolution in limestone areas as this river feeds several outlet
points (fig. V. 18.A). At some distance on to the outcrop of limestone roofing
an underground stream the possibility of shaft dolines developing by collapse
and subsequently enlarging into karst windows appears to be a distinctive if
perhaps rather unusual stage in drainage system evolution. The Rak river in
Slovenia is a classic example (fig. V.18.B). The river to which it is tributary,
the LJubljanica. is truncated into six parts with the surface reaches across
basin or polje floors each separated by sills. Integration of the surface drain-
age appears to be progressing towards the headwaters. Like the present Pazin
stream the LJubljanica headwaters are thought formerly to have flowed in
several directions.
A final consideration in the study of stages in drainage development is that

of paleohydrology. as one of the major uncertainties in drainage histories is
how climatic changes, as well as influencing terrace formation, affected the

density of drainage. The density of valleys dry and drained during the
Pleistocene in Belgium and many other lowland regions were approximately
the same for clay, sand, chalk, and shales, yet today’s density of drained
valleys varied with permeability of rock and a lot of chalk and sand valleys
are dry. One possible explanation
is that the average groundwater-table level
averaged over spans of thousands of years might have been substantially
higher perhaps due to greater effective precipitation resulting from
lower
temperatures and possibly higher precipitation totals. Also, the observed rise
of watertables on the banks of reservoirs when first impounded might suggest
that some influence of sea-level rise on watertables on a regional scale could
1
be considered. A second speculation that might be worth quantifying is the

degree to which the greater viscosity of the colder water would reduce
infiltration rates and thereby favour greater surface runoff. The largest
changes in drainage density occur in most present-day sub-humid to arid
environments. In the Texas High Plains, permanent streams are now virtually
non-existent and valleys and associated depressions exist now only as faint
but obvious surface features filled with eolian and fluviatile debris, yet are cut
in bedrock and may be used during extended periods of wet weather.

Similarly remnants of old drainage patterns of once permanent streams exist
in the Sahara and the Kalahari. Similar elongate basins appear to be the
remnants of dismembered river systems in western Australia. In eastern

Australia, ancestral rivers in the Murray basin were of much greater magni-
tude than those of the present system.

To some extent inevitable was the association of the disintegration of
drainage networks due to reduced runoff, with the blocking of channels with
dunes of wind-blown sand. Even over short spans of time, with a few seasons
of exceptional drought such as is liable to occur in parts of South Africa,
wind-blown sand may cause a temporary choking of the watercourse and even
reversal of drainage. In some instances the river system of the succeeding
phase of greater runoff became re-established in accordance with the dune
pattern. Wind-aligned drainage systems have been noted in the northern
Great Plains, like the north-west to south-east alignment of tributary streams
of east-flowing rivers in South Dakota, in Navajo country in Arizona, and
over a large area in the Colorado Plateau where numerous south-west-flowing
tributaries incised by 60 m and considerably more are parallel with present
wind direction. Most areas of aligned drainage are downwind from large
intermittent streams, the beds of which receive periodic replenishments of
sand available for wind transport during dry periods. It seems that any
deviation away from the direction of prevailing winds by smaller tributaries is
counteracted by the accumulation of eolian material in this segment of the

channel.
In more humid areas in mid-latitudes there are also indications of environ-
mental changes causing substantial changes in drainage features. One of the
Figure V.18 Critical stages in drainage development in limestone areas. {Above)

A large number of resurgences supplied from a small number of sinkholes
(from
Male^, ]968, Geografski Glasnik, Fol. 30) showing the significance
A/.
of the
Pazin sink in the underground drainage of central and southern
Istria. (Below)The,
Skocjan valley, east-north-east of Postojna, Slovenia
(after A. Serko, 1949,
Geografski Vestnik. Vols 20-1) where collapse has revealed
the underground
drainage system of the Rak River in one large
‘karst window’ and three much
smaller ones and left two natural bridges.
more widespread examples, investigated and discussed in detail by

G. H. Dury (1960; 1964) is the systematically developed large valley chan-
nels curving in wide-radius meanders which are now abandoned, partly filled
with sediment, and occupied by much smaller active meander belts.

Changes in load-discharge relationships caused by climatic changes and
associated changes in erodibility of the ground surface were responsible for
the creation of ‘climatic terraces’. These terraces do not necessarily give a
very clear history of stages in the vertical changes of a stream because, first,
there may be little correspondence between the gradients of the present and
may converge or diverge
of the terrace-building stream, and terrace profiles
downstream. Climatic terraces are most clearly displayed upstream from a
relatively unchanging level, such as might be imposed by a resistant bar
across the stream course. This is the case in the Tapitallee Creek. New South
Wales, where terraces converge downstream on a sandstone barrier.
Secondly, the exact nature of inter-relationships between terraces and
climatic change is complicated by non-linear and even non-monotonic relation-
ships between changes in precipitation amounts, extent and amount of vegeta-
tion cover, periodicity of rainfall, and dilTerences in their effects on erosion,
transport, and deposition of sediments. However, in middle latitudes,
aggradation and terrace building seems to be associated with decreases in
precipitation. In many sub-humid areas in Australia it seems that stable
ground surfaces and associated soil formation occurred in relatively humid

conditions when there was greater vegetation cover on the hillslopes,
whereas instability and the removal of hillslope sediment into stream

courses was associated with a relatively dry climate.
Climatic changes also induced significant changes in areas covered by
lakes. For instance, the New Black Sea level 2-3 m above its present height
(2500-500 was preceded by the Neoeuxine era when the Black Sea was
B.c.)
merely a small freshwater lake 100 m below its present level.
E. Stages in Glaciated Areas
The discussion and study of stages in glacial activity is of broader interest

than other erosional and depositional agents. However, there are three
main reasons why stages in landform development associated with the strati-
graphical reconstructions of Quaternary events are not described in detail.
First, the results of successive phases of glacial erosion appears to be cumula-

tive with a presumed progressive deepening, widening and straightening along
preglacial valleys. In the west-central Lake District many of the best-
developed cirques are located in structurally weak zones where preglacial
erosion had already hollowed out depressions as headwaters to significant
minor streams. Too little is known about the mechanisms of cirque erosion to
support the long-established speculations about stages of cirque expansion

progressively encroaching on uplands to form serrated ridges and peaks. It is
just conceivable that the absence of divide-lowering processes in glaciated
highlands might offer an equally specific explanation for the development of
sharp outlines in the relief. The tendency for cirques to occur in one quarter
of the compass will probably attract closer interest. Recently, A. P. Crary
(1966) suggested tentatively that if glacial erosion in fiords is limited to an
area at the junction between grounded ice and floating ice, multiple basins
might result from changes in glacial conditions, such as in ice thickness,
causing a shift in the junction position from one phase to the next. In general,
however, geomorphological studies of the forms of glacial erosion do not

often suggest that the recognition of successive stages in development is a
significant consideration in interpreting the details of the forms created by
progressive erosion of successive phases of glacial activity. Also there are
suggestions from many areas that valleys occupied briefly by ice during the
Quaternary had attained present-day proportions in preglacial times.
A second consideration relates to a somewhat similar conclusion in rela-
tion to deposition forms, for successively younger ice advances most effec-
tively destroy the depositional features of earlier glaciations should they over-
ride them. In consequence, if older glacial deposits survive they are isolated
fragments high on valley-sides or in the most distal parts of an ice-invaded
area, and are usually without enough surface expression to be of direct
relevance to landform studies. For instance, Irish Sea ice, perhaps as an early
Wiirm glacier, probably advanced south across the Cheshire Plain to reach its
greatest extent at the Wolverhampton line. Here there is a great concentration

of large erratics from the Lake District, south-west Scotland, and the Irish
Sea. However, since there is no end-moraine feature, this line is a clear
instance of a clue crucial to Pleistocene geology but with no direct relevance
to the study of landforms as these are largely absent. Similarly in the north-
west United States till of the Donner Lake commencing 400 000-
glaciation,
600 000 years ago has lost all original morainic forms. However, geo-
morphologists interested in studying Pleistocene stratigraphy investigate
phenomena like these, and others benefit indirectly as the chronology of
climatic change in the Pleistocene becomes more clearly established.
A third factor is that those areas which have been glaciated longest will
tend still to retain an ice cover burying the underlying land surface, as in
Antarctica where there may have been intermittent if not continuous polar ice
for the last 1 1 million years.
There are two aspects to stages in the evolution of depositional landforms
in glaciated areas. The first is the results directly attributable to advance and
retreat of ice. The second is the succession of events during a glacial phase
involving the activities of running water, wind, and frost interwoven
with the
Figure VJ9 Stages in glaciated areas.
A. Stages from Sweden {from J. Lundqvist in

in the retreat of continental ice
Rankama. 1965). The ornamented lines provide a geochronological time scale,

based on G. de Geer’s pioneer work on varves rather than a geomorphological
map delineating landforms.
B. Possible relations between successive advances of a valley glacier {from R. S.

Sigafoos and E. L. Hendricks, 1961, U.S, Geol. Surv, Prof. Paper 387- A). Solid
lines show position of moraines that remain after readvance: (i) Similar in down
valley extent but with axes of flow displaced laterally, (ii) closely similar in
downvalley extent but differing in longitudinal gradient (alternatively the younger
advance might have the steeper gradient), and (iii) different downvalley extent but
with same axis of flow.
deposition of morainic features, the predominant direct results of stages in ice

activity. Although every area of major end-moraine development is a highly
distinctive area where more than one moraine remains, it may reflect any of
the basic conditions illustrated in fig. V.19. End moraines might epitomize the
changing times of landform development, were it not that the identity, age,
and continuations of many end-moraines are often problematic. It is easy to
underestimate how actual moraines represent prolonged and complicated
histories and often may bear little resemblance to textbook models.
J.K. Charlesworth’s Lammermuir-Stranraer moraine, identified in 1926
along the northern edge of the Southern Uplands of Scotland and interpreted
as a major readvance of Highland ice, has subsequently been reinterpreted as
essentially stagnant-ice fluvioglacial deposits away from its western end. Also
it is probably only in the Lammermuir area that the last ice advance was from
the Highlands.The last ice to reach much of the northern edge from the
Southern Uplands came from the higher ground to the south. Similarly in the
South Wales end-moraine J. K. Charlesworth (1929) saw a feature of the last
glaciation, whereas more recently G. F. Mitchell ( 1960) suggested that it could

be Riss in age. with the Wiirm glacial advance confined to North Wales.
It is on forms produced by phases .of ice-retreat and readvance in post-
glacial times that stages in ice-deposited landforms are more clearly displayed
due to the comparative recency of the events and the clearer documentation,
including intensive pollen analysis. Again, however, much of the information
available reflects the stratigrapher’s interest in problems of chronology and
correlations in time rather than in the differences in detail of the landforms.
Selected regional examples of the sorts of pattern of advances encountered
are probably more useful illustrations of present knowledge than possible
generalizations. The definition of two climatic phases of postglacial time,
however, provides a basic starting-point of broad significance. First the
Hypsithermal interval was the time 5000-8000 years ago when many
glaciers shrank most in postglacial times. The Neoglacial describes the
climatic episode covering the last 5 millennia during which, at various times,
glaciers grew or were re-created. In arctic Canada, moraines were built about
2400 years ago, with similar features created in western Greenland 3000-
4000 years ago. Many glaciers after reaching their Neoglacial maximum,
commonly within the latter part of the nineteenth century, began a fluctuating
recession until cooler and wetter conditions began in the 1940s. In Alaska the
Sherman and Sheridan glaciers, although they advanced several kilometres
beyond their present position in Wisconsin times, readvanced in the
eighteenth and early nineteenth centuries. There was a readvance again in the
nineteenth century until 1900. when the
late
m be-
Sheridan glacier was 700
yond its Sherman 1175m forward. Recession followed
present position and the
but in 1930 both readvanced to approximately 500 m beyond their present
positions. On the opposite side of the continent the Barnes ice-cap re-advanced
to the outer moraine in c. a.d. 1250. followed by slow retreat until a.d. 1550
when readvance created a major moraine. Subsequent retreat was interrupted
by a readvance to another distinct moraine formed about a.d. 1700. Con-
ditions since have been more stable but with a readvance forming a moraine
about A.D. 1 840. Distinct from the phases quoted so far is the Washington
and Oregon area in north-west United States, where the greatest Neoglacial
advance was between 1800 and 500 years BP. In Argentina the first post-
glacial readvance took place 4600 BP and may have reached a maximum
about 2000 BP. This is essentially the Sub-Atlantic period when advances
occurred in Alaska, Canadian Arctic. Greenland, Iceland, and Europe. Also
like many other parts of the world Argentinian glaciers advanced during the
last three centuries, although, unlike many areas, this was the greatest posi-
tion in postglacial times.
Taking a broader view than the minor retreats and re-advances during a
major glaciation, certain stages in landform evolution might be linked very
tentatively with general intensification to the last maximum glaciation fol-
lowed by more clearly defined stages in its subsequent decline. The initial
phase is an interglacial period with long-continued biochemical weathering of

the land surface. It might be supposed that in upland areas, if the onset of
glaciation were slow enough, there would be intensive frost spalling on valley
walls and the development of screes and rock glaciers downvalley from the
glacier front. This would provide significant quantities of debris to the glacial
moraines and might be a factor in valley widening. However, in some areas

like Central Scotland, material weathered either by biochemical agencies or by
intense periglacial action immediately prior to the ice advance was not neces-
sarily picked up by the ice. For stages in deglaciation there is much more
tangible evidence, although it has been used to develop a range of contrast-
ing interpretations. It is established, however, that many forms associated with
deglaciation do not appear simultaneously but often occur in sequence.
Just as the importance of freezing conditions preceding a glacial advance
might have varied widely according to local circumstances, no set of
generalizations apply to air and soil temperatures in deglaciated areas. In the
area vacated by the retreat of ice of the Late Weichsel stadial in lowland
Britain, fossil permafrost features indicate that permafrost conditions de-
veloped after this particular retreat. In contrast, in coastal areas, like the
fiords of southern Norway, deglaciation probably took place fairly rapidly
sudden break-up. Another variable factor is

with floating of ice leading to its
the amount of melt water released during deglaciation. On the Labrador

coast, where the absence of melt-water forms may reflect glacier reduction
due to diminished precipitation rather than by temperature change, sharp-

crested morainic ridges are in consequence well preserved. Similarly the
Landforms arid time 377
relatively dry climate in the western Canadian plains meant that features were
mainly composed of till rather than of fluvioglacial materials and that they
remain well preserved due to only slight postglacial erosion.
The most significant aspect of an understanding of stages in deglaciation
has been the fundamental change in basic postulates in the last two decades.
The classical concept of deglaciation, elaborated about the turn of the century
for upland areas in many countries, was the retreat of a continuous ice margin
towards lower ground ponding elongated glacial lakes against the flanks of
upland nunatak areas or mountains like those in central Norway and Sweden
where the snow-line rose so high during deglaciation that no accumulation
area was left. Had more time been available for field surveys these postulated
lakes might have been reconstructed on the evidence of shorelines or shore-
line deposits, but frequently the lakes were merely inferred to have existed at
an altitude determined by the level of the channel floor assumed to have
served as a marginal lake outlet. While field workers gradually showed that
such shorelines frequently had neither form nor deposits at the postulated
altitudes, C. M. Mannerfelt (1945) put forward his ideas on the downwasting
of ice-sheets. His interpretations, worked out in southern Sweden, have sub-
sequently been successfully tested in other areas, particularly in Britain.
Similarly, in the Susquehanna valley, stages of deglaciation are now thought
to have been predominantly those of downwasting and stagnation rather than of
frontal retreat. Also stagnation of ice characterized the final stages of the
earlier short Illinoian glaciation. At this stage in deglaciation lowland areas
are occupied by large stagnant ice bodies with melt water slipping readily
between and beneath the decaying ice. In higher areas which were centres of
ice dispersal and which were high enough to remain as zones of accumulation
while lowland ice melted, the ice-sheet contracted upwards and penultimate
stages of deglaciation resembled initial stages of its onset as the ice-sheet
separated into progressively smaller units. As the mountain ridges emerged
from the ice. melt-water streams formed overflow channels at low points in
the ridges. Many glaciers were fed from more than one direction and with
some more active than others, reversals of melt-water
glacier tongues being
flow occurred and cols followed by tongues of diffluent ice as well as by melt
water. At this stage in the deglaciation of the Canadian Cordilleran region
differentiation of sedimentary products of reworked till took place, with small
deposits of lag gravel forming in upland areas and silt and finer material
remaining suspension and being sluiced from the uplands into silt-filled
in
valley bottoms. A study of stages in deglaciation depends on a careful

deter-
mination of loess and drift relationships. The Farmdale loess, blown from
valley trains in the Mississippi valley, for instance,
surrounds a drift lobe but
does not lie on it, indicating that it was blown chiefly while the
glacier was at
its maximum stand and subsequently while the till remained ice-covered by
downwasting ice. By contrast the Iowan loess was blown from the Iowan drift
during ice-retreat, and the maximum thickness is at the border of the drift.
Loess deposits form an extensive belt along the north European plain, but is
only patchily distributed in south-east England and is almost completely
absent in Scotland.
Apart from the case of well-defined end-moraines the association of glacial
is scarcely practicable nor
depositional landforms with stages of deglaciation
realistic in most areas. The complex variability of modes and rates of ice-
retreat and the associated depositional forms is shown clearly in a study by
K. Virkkala ( 1 96 1 ) of the Hameenlinna area in southern Finland. In this area,
there are several different zones in the glacial forms, related to stages in
deglaciation. Ice retreated first from the south-east part of the area, dis-
tinguished as zone A in fig. V.20. It is now occupied by numerous small
morainic ridges and kames aligned at right-angles to the direction of ice-
retreat. These features probably reflect halts or even slight readvances punc-
tuating the ice-retreat. In zone B large kames, much pitted in the Turenki
locality, indicate standstill of the ice margin for considerable periods, an ice con-
dition also favouring delta formation. However, in some parts of this zone
the ice behaved differently and washboard moraines are numerous. Zone C is a
2-3-km broad belt in which no ice-marginal formations occur. In the succeed-
ing zone D large eskers reflect a rapid rate of ice-melting and ice-retreat along
an even front. In contrast, the fragmentary ablation moraines dominating the

2-4-km-wide zone E indicate that ice disappeared from this zone largely by
downwasting. Delta-like marginal formations occupy much of belt F, but
between Alajarvi and Renko the behaviour of the ice sheet contrasts with that
in the remainder of the Hameenlinna region; a conspicuous feature is the
extensive drumlin field signifying the work of thick moving ice. In zone G
there is yet another example of substantial changes in glacial deposition
patterns within very short distances. On the northern side of Renkajarvi the
ice retreated along an even front, creating imposing eskers which in places
spread out on to extensive plateaux. At the southern end of Renkajarvi broad
areas of ablation moraine indicate downwasting; the ice here had retreated,
for the most part, across dry ground.
Within their span C ,4 dates have challenged so many correlations that,
apart from the fact that the problem of chronology is primarily the geologist’s
speciality anyway, it would be unwise to examine the broader stages of the
main Pleistocene glaciations too closely here. However, since the glacial
stages are the main support of the skeleton of temporal stages into which
all
the decipherable stages of landform evolution might be fitted, some brief
review is relevant. Three important points might be made. First, the degree to
which glacial stages or shorter phases are synchronous over a regional as well
as a global range depends entirely on the degree of general correspondence
which each expert considers constitutes a correlation. The classification of
Wisconsin glacial deposits, for instance, is much criticised or strongly

defended by Pleistocene stratigraphers. One reason for difficulties is that an
advance in one place need not correspond to an advance in another place,
whether these be on separate continents or a mere mountain divide apart. In
Scandinavia, the last major ice sheet began to form on the divide of the fjell
range, but later moved eastwards, partly due to cyclone tracks moving south-
east and southward. This might have caused reduced snow supply to
northerly parts and even recession, which might explain why ice began to re-
treat at quite an early stage from the northern mountain areas in Norway.
This reconstruction shows that, due to the significance of local conditions, par-
ticularly relief and aspect, the progress of glaciation does not necessarily imply
a parallel development with the glaciation of an area as a whole. Local ice
culmination may possibly have occurred at dilTerent times in different places.

From this it follows that broader correlations may be difficult. For instance,
although a sequence of three glacial drifts in the Argentine Andes is fairly
well established, correlation with other regions remains to be attempted.

Similarly, it is only a possibility that four glaciations separated in the Hupei
region of China might correspond to the classical four glaciations in the
European Alps. Even in Scandinavia it is uncertain whether the Old Baltic
ice-stream, which preceded an equivalent of the main flow of Wiirm ice,
developed during final phases of Riss glaciation or during the initial phases of
the Wurm glaciation.
A second point is that within the Tee Age’ ideas on the number of major
glaciations are increasingly qualified due to growing evidence indicating
climatic oscillations which preceded them and which make a fixed notion
about four glaciations too rigid a frame.
A third, substantial problem is that of the broad time zone that now defines
the beginning of the Pleistocene. A faunal date from East Africa puts the
Villafranchian back to 1-75 million years, and ages for an approximate time
equivalent in North America, the Late Blancan, are between 1-5 and 3-3
million years. These ages are far older than the period, probably starting
approximately 600 000 years ago, covering the classic four glaciations of the
Alps. It is now recognized that the Pliocene-Pleistocene boundary cannot be
definedby the first incoming of glaciation, but that the earliest evidence of a
marked climatic deterioration be taken as the start of the first major phase of
glaciation for a given region.
F. Stages in Coastline Evolution
Some studies dealing with changes in coasts and coastlines record some of the
most exactly known stages in the evolution of landforms. Coastal stretches
where cliff retreat, even in consolidated rocks, is measurable within spans of

decades or centuries, provide some of most precisely definable stages in
the
slope evolution. Therefore, in terms of geomorphology as a whole, and as
illustrated in the earlier section on slope development, the significance of such

sites is not confined to their own limited strips of coastline. Within the context
of coastal development, however, such sites made up of consolidated rocks

are unusual. As on rapidly eroding coasts, the stages in the development of
depositional landforms in coastal environments may be defined with unusual
accuracy, particularly where the records of detailed maps go back several
decades or more. However, along all but a few very short stretches of coast
information is insufficient and interpretations uncertain. In fact, set against
the detail obtainable at one or two sites there is, in current interpretation of
stages of coastal development, considerable uncertainty and major recent
reconsiderations of basic concepts. Many factors make generalization about
stages of coastal evolution difficult if not misleading. In addition to the
complications of considering both section and plan, the juxtaposition of
erosional and depositional stretches of coastline is difficult to interpret and
their separate consideration may be unrealistic. Other sources of variation
include changes in the relative level of land and sea due to isostatic adjust-
ments of land levels, postglacial rises in sea-level and the superimposition of
both effects on local tectonic movements. In consequence, it is substantially
more difficult to support the idealized sequence of forms that might occur in
the plan of a coastline than was permissible two decades ago. D. Johnson’s
(1919) elaboration of early notions explaining a contrast between emergent
and submergent coasts is now seen not merely as over-idealized but also as
totally unrealistic. Also the stratigraphical evidence and ideas of use in study-
ing stages of heights in land-sea relations in coasts even a decade ago are
now obsolete. Although stratigraphical field study is permitting some pro-
gress, it is now appreciated that undeformed strand-lines are at present very
difficult to identify, and that the eventual complete correlation between rocky
and depositional coasts is perhaps intrinsically unobtainable.
Classically sea-level was believed to have been at about + 60 m during the
Giinz/Mindel interglacial (325 000 BP), at -(-30 m during the Mindel/Riss
interglacial (195 000 BP), and at -f- 18 m during the Riss/Wiirm interglacial
(105 000 BP). These three levels were named, respectively. Milazzian,
Tyrrhenian, and Monastirian, but their interpretation is now being extensively
revised on increasingly stratigraphical evidence. One reason for the revisions
is that the Mediterranean region including type localities of the initial
scheme
was tectonically unstable prior to the last interglacial. There is also inde-
pendent evidence that sea-level itself might have been at about the same
elevations during the three interglacials. Substantial revisions even appear
necessary for changes within postglacial times. For instance, in Scotland for
the last 80 years or so the raised beaches have been grouped at levels of 30,
15. and 12-5 m, yet after recent careful measurements in the Forth Valley
area, J. B. Sissons (1963) is uncertain of the total number of shorelines and
how fragments of these might be correlated with other fragments within even
this area. One of the weaknesses in D. Johnson’s deductive scheme is that,
apart from areas like northern Canada and Scandinavia where postglacial
isostatic rebound outstripped postglacial sea-level rise, or tectonically active
areas where eustatic and crustal movements are scarcely distinguishable, all
coastlines have been, in geological terms, very rapidly and emphatically

drowned in the 1 5 000-5 000-BP interval.
Although the difficulties of pointing to stages in the history of erosional

planation along coasts are now widely appreciated, there remain a few
favoured and restricted localities, particularly on small islands with rock
benches for which a wave-cut origin is highly likely. For instance, on the San
Benito Islands, a group of three off the Pacific coast of Baja California, the
most pronounced features are three wave-cut marine terraces. The uppermost,
occurring only on West Island, dips north-west at approximately 5 degrees
and ranges in elevation from about 90 to 150 m. The intermediate level, best
developed between 30 and 75 m on East Island may also be represented on

West Island at the same approximate range in elevations. All three islands
have terraces near present sea-level between 6 and 30 m. Some terraces dip
symmetrically outward from the centre of the island. Others are tilted, due to
differential tectonic uplift subsequent to wave erosion. Similarly, the most

striking feature of Middleton Island in the Gulf of Alaska. 1 10 km from the
mainland, is the step-like terraces at 3, 14. 20, 25, and 32 m, cut in Pliocene
or Pleistocene bedrock.
The clearest examples of staircases of relict erosional benches notched in
higher areas inland from coasts are found in mobile, tectonically unstable
areas where abrupt uplift and periods of stillstand occurred during
Pleistocene times. It is in these very areas, however, that vertical movements
vary widely in amount and even in direction of movement and where, in
consequence, present altitudes of benches and terraces have little significance.
Their interpretation requires the additional information provided by sea-

cliffs, buried channels, alluviated river mouths, and the stratigraphy of off-
shore sediments as well as well-authenticated and independently established

data on eustatic movements of sea-level during the Pleistocene. The work of
E. C. Buffington and D. G. Moore (1963) relating to the San Clemente area,
southern California, exemplifies the intricacy of reconstructions in the tec-
tonically active areas where the morphological evidence is best preserved. At

least six terrace levels were developed during the Pleistocene in the upper
Miocene-lower Pliocene Capistrano sedimentary rocks. These terraced areas
are separated by relatively depressed and predominantly depositional areas
such as stream mouths graded to a level about 90 below the present sea- m
level. Buffington and Moore adopt a figure of 60 m for the greater stand of
sea-level above the present during pregiacial and interglacial times. Combined
with a tectonic depression of land surface 80 m below its present elevation in
the San Clemente locality, this early high sea- level would have created condi-
tions for the formation of the fifth terrace in post-Capistrano times. This
could have been in the late Pliocene or during one of the early Pleistocene
interglacial phases (fig. V. 18. A). The formation of the lower four terraces
may have occurred subsequently during relative stillstands of the sea during
the intermittent re-elevation of the coast (fig. V. 2 1 .B). With a continued elevation
of land of perhaps 195 m and a eustatic lowering of sea-level by approxi-
mately 100 m during the last stage of the Pleistocene, land that is now 250 m
below present sea- level would have been exposed to sub-aerial erosion, a
new base-level formed (fig. V.21.C), and vigorous erosion cut gullies down
the coastal slope. These were then rapidly drowned by a rise of sea-level
approximately to its present position (fig. V.21.D), a trend continued by a
slower eustatic rise of sea level (fig. V.21.E). The latter phase of sea- level rise
was slow enough for the planation of a rock platform and the deposition of
1 2 m of sediment on it. Since drowning, deposition appears to have been the
dominant process.
Generally speaking, evolution of forms on rocky coasts is too slow for
changes to be discernible. A. Guilcher (1949) has suggested that some high
cliffs in hard rocks might be fossilised features formed a hundred thousand
years ago. In consequence stages in coastline changes are usually best seen
in glacial or alluvial deposits. However in the case of erosion of soft rocks
former stages are only reconstructions of a past that has disappeared beneath
the waves and are not part of the present-day forms like those representing
stages in coastal deposition. In the latter case several stages are sometimes
discernible. This is particularly so in shallow shelf areas where the post-
glacial rise in sea-level involved substantial lateral displacements of shorelines
and associated reworked shelf deposits at rates of tens of metres per annum.
For instance, much of the 600-700-km-wide shallow platform between
Timor and the north-west Australian coast, the Sahul Shelf,was exposed
1 8 000 years ago, as was the entire As the initial rise was rapid
Orinoco shelf.
even delta shorelines, like the Mississippi, were moved landward. As long as
the last major rise of sea-level was taking place, new areas of coastal plain
were being inundated and fresh supplies of sediment were encountered. The
clays were dispersed but the coarser eonstituents were pushed forward to
become beaches. On all tropical coasts, the amounts of sand amassed in the
coastal belts was especially huge. Beach volume increased as long as the rise
of sea-level continued and the surplus sand was blown downwind to form
coastal dunes. The beach-dune systems reached their greatest volume with
the
POST - CAPISTRANO TIME
Late Pliocene or early Pleistocene Interglaciol
Sea level up 60 m 8. land depressed

highest terrace formed
MIDDLE -LATE PLEISTOCENE
level
ôur more erosionol marine terraces formed

level
•'during stiUstands which punctuate re-elevation of land
sea
onset of continenlol gloctalion 8. eustatic fall of sea level
sea
0 PLEISTOCENE 4*'^
STAGE MAXIMUM 0 present
fi
present
— ^ Gullies developed to new base
»n exposed Miocene bedrock
above
Sea drops lOOm 8. land

level
continues to rise o minimum of 170m
below
metres
PLEISTOCENE 4^ STAGE
2—3
metres
tjPttt?
^ Rapid depression of land drowns

lower portions of sea gullies
PRESENT CONDITIONS
shelf break
Sediment apron ;«;^^^^^^Hleislocene recent

'sediment tense on shelf
>• ".^^yT^Seo guUies cut
'///into continental slope
Sea level, rising from Pliestocene low, has eroded

present continental shelf, truncating sea gullies
Figure V.21 Development of terraces and sea gullies during Pleistocene eustatic
changes in sea-level on a tectonically active coast [after Buffington and Moore,
1963). The vertical arrows indicate the relative movements of sea or land. The
diagram is based on conditions in the San Clemente locality, southern California,
approach of stillstand. Around 10 000 BP the rate of rise decreased markedly
and in areas of rapid supply of river sediments, delta progradation became

possible. As sea-level steadied about 5 000 BP new sediment supplies
for the
beach-dune systems were consequently no longer encountered, and marine

processes brought about a net loss to the system due to reworking by storm
waves and burrowing organisms. However, in areas where local conditions
ensure continued and abundant supply of sand as on the east coast of Malaya,
a succession of parallel ridges may result. If a berm survives a storm in such
an environment a new one will form on its seaward side due to accretion
during the following calmer conditions. At present beach ridges or their

remnants are found as far as 6 km inland along the Endau river. Inland they
reaeh heights of 8-9 m above present sea-level and diminish in height sea-
wards. In Mexico, beach ridges adjoining the Grigalua river extend 40 km

inland.
Fig V.22 illustrates one of the many points around the coasts of the British
Figure V.22 Postglacial episodes on the sandy shore of south Lancashire (from
R. K. Gresswell. 1953. Sandy shores in south Lancashire). At Stage B the dunes
are made up of the Shirdtey Hill Sand and the beach of Downholland silt.
Isles where there are records of postglacial changes in the forms of uncon-
solidated materials. R. K. Gresswell ( 1 964) considered that, in early postglacial
times, the land-level on the east Lancashire coast, in relation to sea-level, was
at least 45 m higher than today. Marine transgression, presumably due to
ecstatic sea-level rise exceeding isostatic land-level rise, culminated in the
formation of the Hillhouse beach, possibly 5000 years ago (fig. V.22.A). The
cliff foot associated with this feature is at 5 m OD suggesting a correlation
with similar features found between 0 and 1 0 m OD in many other coastal
areas of Britain. Eustatic rise then either diminished considerably or ceased
altogether, while either isostatic rise continued or local crustal warping
occurred. This slight relative elevation of the land favoured the deposition of
the Shirdley Hill Sand and the Downholland silt on the prograding shoreline
(fig. V.22.B), and forest and peat bog grew on the upper parts of the beach as
the coastline continued to retreat westward (fig. V.22.C). Finally land-level
fell in relation to sea-level and the erosion of the peat and other beds started
V.22.D). M. Schwartz’s (1968) work, extending ideas of P. Bruun, has
(fig.
shown how with other factors equal, within the last few millennia where some
approach to a balance between forces may be possible on a coastline of
deposition, the effect of a sea-level rise would be the landward displacement of
the beach profile, nearshore deposition compensating for the erosion of the
upper beach in such a way as to maintain water depth adjacent to the shore.
Factors not likely to remain equal include the amount of sediment supply
alongshore which is often the most important factor in determining whether
short-term stages of coastline evolution will be erosional or depositional.
Although the facts of late Quaternary times make the classical Johnsonian
theory linking the origin of barrier islands and related features to shorelines
of emergence unrealistic, there are some situations where accumulation and
emergence The emergent beach at Gisborne, Poverty Bay in New
co-exist.
Zealand, comprises a belt of nearly 100 low. closely spaced and continuous
ridges about 5 km wide, rising from 4-5 m at the coast to 12 m inland and is
considered to be largely the result of tectonics. From the evidence of volcanic

ash falls warping appears to have taken place shortly before 1500 b.c. and
between a.d. 1000 and ad. 1650. The effect on the shoreline plan of the
throwing-up of barrier beaches is to straighten its outline at the mouths of
outlets and low-lying indentations. If the sea has little tidal range and if the
climate is relatively dry. lagoons may develop. With marine oozes introduced
by incoming tides and the addition of river alluvium the filling up of lagoons
may proceed in clearly definable stages. Along the Pacific coast of Baja
California, Mexico, between 24° and 30° N, wind-blown sand is gradually

filling up many tidal lagoons. In this area the stages of lagoon and barrier
development have been studied in detail. At stage A, fig. V.23, the deepest
part of the lagoon was directly behind the barrier, so that the tidal currents
and resulting turbulence were strongest adjacent to the barrier. Deposition

was confined to shallow marginal areas. The tidal flats expanded towards the
barrier, but as strong and turbulent currents kept the channel clear of sedi-
ment it remained deep and stationary. Eventually, because of the distinctive

condition of little addition of new sediment, a balance was approached as
increased narrowing of the channels must have increased velocity and turbu-
lence of the oscillatingwedge of tidal water. Since the stage B barrier was
deposited 1800 years ago, the barrier has prograded seaward approxi-
c.
mately 1 6 km, perhaps due to changing climatic conditions leading to

greater supply of sediment from adjacent land margins.
Another coastal environment where stages in coastline evolution are pro-
minent is in river mouths where the term ‘drowned valley’ is a misnomer

since nearly all coasts have been inevitably drowned in postglacial times. R. J.
Russell (1967) suggests that a comparison between the north and south coasts
of the Baltic well illustrates the major fallacy of the subdivision between
coastlines of emergence and submergence. Instead, stages of alluviation in
river mouths are recognizable, with the effects in the ria or classical ‘drowned
valley’, either due to their depth or to the dearth of sediment supply or to
diminished river discharge being least apparent. Many estuaries being pro-
gressively filled, contract in volume, depth and surface area until adepositional
plain extends right to the coast often with an associated fringe of deposits
strung out by marine processes (fig. V.24). Deltaic coasts are the other
extreme where the scale of alluviation may be colossal; the volume of Recent
Figure y.2S Stages in the development of a coastal lagoon barrier

(from
F. B. Phleger and G. C. Ewing. 1962. Bull. geol. Soc. Am., Vol. 73), the Laguna
Ojode Liebre. Baja California.
Figure V.24 Stages in the evolution of some estuaries and their transformation by
valley infilling {from F. Ottman, 1965, Introduction a la geologic marine et
littorale).
Landfarms and time 389
alluvial fill in the Lower Mississippi valley between Cape Givardeau,

Missouri, and the end of the exposed valley walls in Louisiana is about
4200 kml
In environments which favour their continued growth the rate of advance
of some small features falls into recognizable stages, like Orford Ness which
J. A. Steers (1939) described as extending 6 km southward since A.D. 1530.
On the other hand, over spans of time, perhaps ranging from decades to
millennia, certain self-arresting features may prevent unlimited extension of a

depositional feature. Thus, as a spit narrows a river outlet, the current velocity
and turbulence in the channel increases until at a critical point they in-
hibit deposition as a theoretical equilibrium state is approached. Frequently,

however, outlets through a depositional barrier may move erratically with
time. Breaches in spits open many times as on the Lake of Barberie which
closes the Senegal delta, but are soon replaced if average conditions remain
essentially the same. Spurn Head, which curves 5-6 km across

The history of
the mouth of the Humber has been examined in detail. G. de Boer (1964)
suggests that episodes of destruction at intervals of about 240-250 years
have punctuated considerable periods of persistence and growth. Rapid ero-
sion of a wedge-shaped area of boulder clay at the root of the spit appears to
be a factor leading to the periodic breaching near the mainland end of the spit.
The breach operates as a flood-tide channel until deposition on the landward
side leads to the growth of a new spit inside the remnants of the dismembered
former spit. This mode of evolution may be particularly closely linked with
the rapid retreat of the boulder clay cliffs at the root of the spit. In some
situations the breaching of a spit may form a barrier island; in many instances
the rate and persistence of sediment supply alongshore determines whether a
depositional form is vulnerable to reworking. Where longshore drift is active
on the fringe of an expanding delta, the ridges of previous spit-like develop-
ments may remain. These conspicuous forms are termed cherniers, as ever-
green oaks flourished along those in the Mississippi delta. In south-west
Louisiana, the system of cherniers is 100 km long, the regularity of spacing
appearing to indicate that periods of erosion and retreat have alternated
regularly with periods of accumulation and advance of shorelines.
The upward growth of coral reefs, in keeping pace with postglacial sea-
level rise, is a significant enough stage in coastal evolution in itself. In
addition, the sand cays on reefs might fall into an evolutionary scheme,
although D. R. Stoddart’s (1965) study describes cay types in terms of equi-
libria with the energy level of waves. R. F. Folk (1967) prefers
to recognize
an evolutionary sequence, beginning with the development of a sandbank
or
bar that hardly breaks the surface at high tide. Secondly, as the
island grows
to a certain size
and permanence, a large area of dry sand remains beyond the
reach of the highest tides, and during a third stage the
island grows by
Angeirs high
'
/////
1766-^—
Smeaton's lighthouse
high
0674-1776)
lighlhouse
0776) ^^^1684 1
V— g'-^'îves
^y^l786
Low lighthouses
i
0776-1851)
accretion of beach or dune ridges, with the colonization by vegetation of the

inner and older ridges. Both these steps are relatively ephemeral compared
with the fourth stage when the island reaches sufficient width to allow vegeta-
tion to cover a central area out of reach of wave erosion. Building out of the
lee-side, perhaps to several times the size of the island, in its third stage
makes this possible and the growth of vegetation, the burrowing activities of
crabs and nesting operations of birds gradually level out the older dunes in
the central area.
Dune systems on coastal margins are usually intimately related to stages in
the evolution of the adjacent depositional coastline. On many of the world’s
coasts. Recent dunes include two contrasting types. The older dunes are
leached, oxidized, exhibit at least incipient soil profiles and are usually fixed
by vegetation and are believed to have increased in volume as long as sea-level
was rising. The newer system lying seaward has smaller volumes, continues to
shift and appears inter-related with modern beaches and shore characteristics
reflecting sand supply since sea-level steadied some millennia ago. The older
dunes are subdued forms compared with the bolder outline of the more
continuous coastal fringe of newer dunes.
G. Stages involving more than one Process and in the Evolution of

the Land Surface as a whole
One of the reasons why stages in landform evolution are not often clear-cut is
because more than one factor may be involved simultaneously. In some other
situations the stages in landform evolution become those of one process
succeeding a different formerly dominant one. Fluvioglacial phenomena are
perhaps the most frequently recognized examples of such a combination. One
of the most common developments is the ice-ponding of lakes and the effects
on melt-water discharge of its periodic release. There may be one of two
effects. First, ice-dammed lakes periodically, perhaps annually, empty rapidly

due to water escaping beneath the ice. This self-emptying termed Jokulklatip
in Iceland, results when the height of the dammed lake, approaching nine-
tenths that of the ice-dam, is sufficient to float the ice. The abruptness and
volume of the escaping flow causes catastrophic floods downvalley from the
glacier snout. A significant result is the progressive destruction of ice-
deposited landforms as one of the stages of deglaciation. R. J. Price (1966)
Figure V.25 Development of spits.
A. Spurn Point, east Yorkshire, 1684-1852 {from de Boer, 1964).

B. Dawlish Warren. Devon, 1851-1962 (from Kidson. 1964).
C. Progressive extension across the mouth of the Senegal river
(from J. Larras,
1 95 7, Plages et cotes de sable).
describes how between 1948 and 1963 the Casement glacier, Alaska,
retreated nearly 2 km north-east of a lake-covered esker system which melt-
water streams began to destroy after the draining of the lake. A particularly
common change is that the greater or more abrupt forms of melt water may
trench the valley train of fluvioglacial sediments, thereby creating a terrace.
However, after the abundant melt water of the early stage in deglaciation, the
net longer term effect during glacier retreat is to decrease the competency of
streams and to cause valley alluviation. Most significant are the spillways cut
when melt water spills over a low point in a watershed boundary either as a
lake spillway or as a sub-glacial chute. An example of the effects of such a
catastrophic overflow is in the Hellemofiord area, at 67°N in Norway where
a narrow canyon cuts 150-200 m down into the floor of a broad flat valley.
Drainage patterns may also be modified substantially by diversions due to ice.
In North Wales, streams Alyn system were diverted south by the

like the
barriers of the Welsh and Irish Sea ice. The spectacular diversion channel
some 16 km west of Clinton. Iowa, in which Goose Lake lies, might have
been first cut when the Illinoian glacier blocked the former Mississippi at
Clinton, creating a lake in its valley which eroded the diversion channel as it
spilled over the divide between the Maquoketa river and the Wapsipinicon
river.The former course of the Allegheny river to the Erie basin was perhaps
similarly diverted to the headstreams of the Ohio river. Eroded remnants of
sands, silts, and clays once deposited in these glacial lakes provides evidence
of their former occurrence, as in the north-trending valleys of the Illinois

Ozarks. In north-west Europe, the northerly drainage was dammed by the
Riss ice-sheet. As a result, a waterlogged area existed in front of the advancing
ice whereas the southerly drainage in North America tended to keep the
area in front of the ice relatively dry. The effect was less marked in the Wiirm
glaciation as drainage was possible away from the ice towards the North Sea.
However, it is possible, as in any situation where more than one agent is
involved, that glacial diversions as an important stage in the development of a

drainage system might be exaggerated. Abundance of melt water is only a
brief phase in glaciations which in turn occupied only a fraction of
Quaternary time. For instance, the diversion of the ancient River Teays to
form the Ohio river, originally interpreted by some as one of the largest
drainage diversions wrought by glaciation in North America, is according to
analyses of the ancient channel sediments probably a preglacial event. On the
other hand, the effect of glaciation on drainage patterns that might be caused
by isostatic depressions of part of a land area might be considered. Diversion
by this means may have been possible, for instance, where the lowest pass-
point in the direction of a deep canyon between the south-west part of the
mountain plateau area and Rombaksbotn, near Narvik, is only 7- 5 m higher
than that in the direction of the valley zone to the north-west.
In some areas glaciation, without introducing a diversion stage into the
development of a drainage system has left a pattern of a river re-establishing

its course. Drumlins, for instance, have significantly controlled the headward
growth of tributaries to the Rock river in Wisconsin and limited the size of
their watersheds, and have determined the winding course of the incised River
Kelvin in Glasgow. The flooding by rising sea-levels of glacier-deposited

landforms is a distinctive stage in the evolution of many lowlands. The
Uppsala esker continues northward as an underwater ridge some 50 m high.
Dogger Bank, rising to within 20 m of the sea surface, is a relic of a moraine
which from near the Yorkshire coast east towards the northern tip of
strikes
Denmark. Westport Bay in the west of Ireland provides a classic example of

the drowning of a drumlin field and many less extensive areas, as near Galway
or at Berwick, on the mouth of the Tweed, show the stages of drumlin
destruction by the sea, ending in a low mound of boulders.
A more unusual example of a glacial action inter-related with another
process occurs in volcanically active areas in high latitudes, like Alaska.

More than one-eighth of the surface area in Iceland is made up of palagonite
breccias and tuffs, believed to be formed from basalt magma under consider-
able hydrostatic pressure, escaping upwards and sideways through the ice.
The long rather flat-topped ridge of Dalsheidi is such a flow, occupying the
axis of an ancient glaciated valley. Although the possible stages of evolution
outlined in fig. V.26 describe an unusual example, similar stages in valley
development are also observed in fiuvially eroded valleys invaded by a lava
stream.
Inevitably, the interdependence of stream and shore processes in their
contribution to landform development has already been mentioned in several
contexts, but it is perhaps worth recalling this point here. Just as it would be
difficult to isolate stages in the evolution of streams from the changes at the
coast affecting regional base-level, so are many stages in coastline evolution
profoundly influenced by river activity. The significance of wind removal or
accumulation of material in introducing distinctive stages into landform
evolution in certain coastal, fluvial, glacial, and even slope environments is
worth reiterating.
Stages in slope and stream evolution may also be interlinked, phases of

accelerated slope weathering confining by colluvial infillings a channel in
which the evacuation of debris becomes more difficult. Conversely with
reduced colluvial supply, and consequently more rapid removal of bedload,
the channel floors are incised and slope foots undercut. Of significance for
broader contexts are theoretical schemes in which the adjustment between
stream and slope, or lack of it, is of less importance than the evolution of
their
combined effects. Some of the reconstructions of the land surface as a whole
are by necessity only very tentative suggestions. Studies which attempt the
Figure V.26 The landform evolution of a sub-glacier lava flow (from Walker and
Slake, 1966). Basalt may have flowed 35 km or more beneath the ancient
Dalsheidi glacier in Iceland.
daunting task of reconstructing stages in the evolution of the land surface as a

whole are often directed towards one of two objectives. The purpose may be
largely geological, aimed at reconstructing the landform of a given period or
periods in the past. Such studies of stratigraphical geomorphology often
include the description of unconformities which have no direct significance in
the present-day landscape. Alternatively the objective may be, largely by the
use of relict fragments in the present land surface, the reconstruction of stages
in the evolution of the present relief form.
An example of a study largely orientated towards reconstructions of strati-
graphical geomorphology is J. C. Frye and A. B. Leonard’s (1957) study of

the Great Plains, at present a semi-arid near-treeless zone of west-central
United States. In the Miocene, an erosional surface of gentle slopes and broad
valleys developed on Mesozoic and Permian rocks. Locally minor tributaries
were sharply incised but maximum relief probably did not exceed 75 m with
deposition in lower parts (fig. V.27.A). Progressive reduction of relief pro-
ceeded throughout the Neogene by gradual engulfment of the erosional relief
by the spreading of alluviation on to valley slopes. Bedrock areas were
reduced to circum-alluviated hills and small discontinuous areas along former
major divides (fig. Neogene deposition culminated in plains of low
V.27.B).
relief (fig. V.27,C) marked only by natural levees, channel scours, and small
low swells where bedrock was not buried. A thick and widespread mantle of
Earliest Pliocene
Figure V.27 Pliocene and Pleistocene evolution of

the land surface in the central
and southern High Plains of the United States
(from Frye and Leonard, 1957).
Figures V.28 and V.29 Schemes of land-surface evolution in Tunisia (from
Coque, 1962). (Above) a hamada in the extreme south. (Below) A slope and river
channel in unresistant rocks.

dense calcium carbonate, locally encrusted on top developed over most of the
region. At the beginning of the Pleistocene relief contrasts developed due to
phases of entrenchment and associated alluvial terraces produced by sharp
climatic changes from late Cenozoic times onward. The fall of the Pearlette
ash in late Kansan times provides a precise datum for the next stage (fig.
V.27.D). A feature of the present relief (fig. V.27.E) is that the incision of the
valleys is sufficiently deep for the maintenance of perennial streams.
The most decisive stages in the evolution of relief near the desert in
A
T
T IMUT nTfronn]]]
iQtentic soil profile
rn
[P
rp
]]
•i-
llllllll III mill 1
Cretace DUS rocks n
_j _j
Figure V.30 Tertiary and earlier Quaternary pedological and

geomorphological
history in the south-east of the Australian and zone {from Jessup, 1961).
1000 2 000 3 000 metres
Figure F3] Stages in landform evolution involving a sequence of effects due to

different, consecutive processes.
A. A meandering river course replaced by a lower, shorter incised course follow-

ing diversion due to deposition of boulder clay plugs (from Jones, 1965). The
Tunisia were pre-Quaternary. The same Quaternary climatic changes as led
to stages in relief development farther north (fig. V.29) operated in the south
but with effectiveness reduced to the extent that changes in pre-Quaternary
landforms were negligible. The Mio-Pliocene deposits in the valleys cut down
between the residual hills of the hamada surfaces puts the main period of
dissection before the end of the Miocene, possibly as a response to large
radius folding related to late- Atlas orogenic movements. Valley incision con-
tinued during the Oligocene and for most of the Miocene to depths of 55-
60 m below the surrounding tableland, with slope-retreat broadening the
valleys (fig. V.28.A). Subsequent deposition of several dozen metres of sedi-
ment has reduced the relief contrast somewhat and in many cases made the
direction of fiow of the formerly well-developed drainage net uncertain. In

part the sediments came from the erosion of slopes adjacent to the rising level
of the sedimentary fills (fig. V.28.B). The development of a conglomeratic
crust cemented and indurated with calcium during the Villafranchian pro-
vides a detailed record of relief at the beginning of the Quaternary
(fig. V.28.C). One feature is that the maximum depth of accumulation of
debris throughout Quaternary time was only 5 m. Some slope development
occurred in relation to local base-levels, but this was essentially a Villa-
franchian feature and there is no indication of exhumation of the buried
valleys (fig. V.28.D).
On the northern fringe of the Sahara in Tunisia R. Coque (1962) en-
visages that, according to the amount of precipitation in a drainage basin, a
main channel tends to develop sufficiently to evacuate the large quantities of
incision of the pre-existing meandering course of the River Teifi into the plateau
can itself be recognized as an earlier stage, perhaps related to uplift in preglacial
times.
B. A sequence of former shorelines notching a moraine spur at Elvegard in the

Narvik district (from R. Dahl, 1968, Norsk geog. Tidsskr, Vol. 22), the result of
isostatic rebound.
C. A sequence of former shorelines notching a glacial outwash fan, Gurreholms

Elv, Kjove Land, East Greenland (from D. E. Sitgden and B. S. John, 1956, Geog.
Jour., Vol. 131), again the result of isostatic rebound.
D. Landslide dams providing instances of distinctive stages in the development of

the long profile of a river in a mountainous region (after A. Buxtorf, 1922,
Eclogae Geol. Helvetiae, Vol. 16). These conspicuous steps are in the longitudinal
profile of the Doubs river in the French Alps.
E. An example of a land surface interpreted as the product

of successive cycles of
erosion (from King, 1967). The section shows the relation envisaged
for many
parts of Africa between skyline tors, interpreted as relics
of an older denudation
cycle, and sub-skyline tors, seen as youthful features of a new cycle.
water and debris occasionally carried on to lower slopes by rillwash and

sheetwash during torrential rains (fig. V.29.A). The pattern of stages in the
dissection of these surfaces, shown in fig. V.29 is similar regardless of
regional or local base-levels; climatic changes appear to be the controlling
factor. A phase of higher temperatures, reduced precipitation, and associated
thinning of vegetation cover led to alluviation replacing lateral planation.
Debris accumulations blocking channels encouraged lateral diversions and
the consequent building-up of alluvial fans. This mantle of sediments thinned
downslope, the steeper gradient facilitating, to some extent, the evacuation of
small-sized debris (fig. V.29.B). The relatively thin depth of the alluvial
mantles, being only a few metres thick at the most, introduced only a very
slightly more pronounced concavity into the outline of the slopes and in-
dicates that the periods of accumulation were short lived. However the
porosity of the mantle accentuated the effect of the drier climate and gypsum
crusts developed (fig. V.29.B). The reduction in sediment supply to channels,
however, favoured some incision of the main drainage channels by the debris-
free water of the occasional floods, a process accentuated at the beginning of
the ensuing pluvial phase (fig. V.29.C). Subsequently the effect of greater and
more frequent floods of the more humid phase was a reversion to lateral
planation leading to the development of a second, lower pediment surface (fig.
V.29.D).
According to R. W. Jessup’s (1961) schematic interpretation of the evolu-
tion of relief in the south-eastern part of the Australian arid zone, the alterna-
tion of pluvial and non-pluvial periods during the Tertiary and Quaternary
created a series of relief patterns unlike those postulated for Tunisia. Unlike
many parts of Australia there are, in the south-east portion of the arid area
no remnants of the actual land surface on which a deep lateritic soil developed
(fig. V.30.A). Continental uplift and down-warping of the Lake Eyre basin
in the Miocene led to the dissection of the post-laterite erosion surface
(fig. V.30.B). Subsequent cementation of the upper part of the detrital deposits
formed a silicrete capping (fig. V.30.C). The extensive younger erosional land
surface developed into the lower part of the kaolinized materials when earth
movements ceased (fig. V.30.D). As much of the area was a vast internal
drainage basin, sedimentation continued from late Tertiary into the
Quaternary. Intensified desiccation led to the deflation of lacustrine sediments

which were deposited over a wide area and even on the higher tablelands. As
in Tunisia, gypsum (fig. V.30.E), wind-blown from dried-up lakes, was laid
down. Wind transportation ceased as the climate became more humid, stony
tableland soils developed (fig. V.30.F) together with renewed but limited
drainage incision and surface stripping.
Among workers in some humid mid-latitude areas, particularly in Britain,
there was. until a decade ago, intense interest in the search for stages in the
evolution of relief as a whole, studied by mapping erosion surface remnants
and the interpretation of altitudinal ranges covering any discrete groupings
that emerged. This pattern of investigation followed that of H. Baulig’s
pioneer work on the denudation chronology of the Massif Central (Baulig,
1928). The altitudinal ranges were regarded as an expression of phases
of partial peneplanation before a relative fall in sea-level led to similar level-
'•
lings at lower altitudes, related to the new base-level. However, weathering

crusts, distinctively weathered soils, dust falls, and other reference levels
are usually absent, the evidence is largely morphological, was prepared before
former sea-levels were indicated and dated by non-morphological evidence,
and is based on theoretical concepts regarded by some geomorphologists as
oversimplified. Reappraisals of the existing evidence of denudation chrono-
logies, detailed replies to criticisms or to alternative interpretations and the
accumulation of additional facts are few. It is therefore difficult to evaluate the
present significance of these interpretations of denudation chronology in
contemporary geomorphology; nonetheless the records of relatively flat seg-
ments of land surface at certain discrete altitudinal ranges provide a most

useful introductory generalization in the description of the relief of an area.
Appendix
Some simple methods of

field measurement
Traditionally the younger student interested in landforms has been trained to

evaluate arguments, to acquire patiently over years of field experience an eye
for the country, to practise the arts of field sketching and photography, and to
be content in the meantime with the interpretations of more experienced eyes.
On reflection, it seems possible that the intellectual importance of debate in
training the mind has, in geomorphology, obscured the importance of the

basic step of science, that of collecting incontrovertible basic measurements.
Equally obscured is the fact that a basic step in learning is the finding out of
facts for oneself.
In the belief that some readers may wish to replace many of the examples
scattered through this book with those of their own observation, it is con-
sidered that this book would be incomplete without some suggestions on how
those with little experience and with limited means and time at their disposal
either to do fieldwork or to search for the appropriate advanced manual or
on techniques, might go out into
article the field and return with some
measurements or samples to be analysed.
A. Slope Measurement
During the last 20 years geomorphologists have come to regard hillslopes as
the most important relief form, and to agree that the measurement of profiles
in the field is a desirable stage in slope study. Slopes are usually measured
with an Abney level, a conveniently carried pocket-sized instrument used in
conjunction with ranging poles and a tape measure. With little expense and
requiring no great skill an alternative is the construction and use of a slope
pantometer, a simple device reflecting in concept the theme of the present
appendix. Made from well-seasoned wood or with right-angle girders of light-
weight alloy, the device consists of two uprights, each with a bolt near top and
bottom (fig. App.l.A). One of the uprights has a large protractor scale at-
tached, centred on the upper bolt. Two cross-pieces with holes exactly 1-5 m
Appendix 403
apart complete the parallelogram. One of the spirit bubbles in a builder’s
level makes it possible to set the uprights vertically for measuring the
slope declivity. To facilitate the drawing of the results, tables of horizontal and
vertical equivalents for slope angles are easily prepared, and the value for
each angle in the profile sequence is read off and accumulated to give the
points with which to trace the profileon graph paper. Representation of the
results in the form of a subdivided histogram provides a compact summary
of the relative frequency of measurements of a given angle, and by summing

separately the frequencies for each of three equal sub-lengths of the pro-
file also conveys simultaneously some impression of the slope shape (fig.
App.LB).
The procedure for measuring a series of slope angles down a profile is to
place the protractor upright on the position occupied by the leading upright
during the previous measurement. However, before taking measurements,
three decisions on the position and length of the profile line are required to
ensure some comparability between profiles. First, in dome-shaped hills with
curved contours the two dimensions of the slope-profile will vary with down-
slope changes in the plan shape and are perhaps best avoided in preliminary
studies. Secondly, the actual alignment of the profile, always at right angles to
the contours, might be drawn through a randomly located point designated rn
in fig. App.l.C, or perhaps the lines might be located to answer a specific
problem, such as the variation in slope form in relation to caprock thickness

or proximity to stream channel. Thirdly, to avoid exaggerating the extent as
well as the flatness that exists on a slope-profile summit such as 5 in
fig. App.l.C, the inclination of the ridge crest, or the steepest gradient that
exists in any direction away from the vicinity of the profile summit, should be
measured. The summit area shown in fig. App.l.C, enlarged in frame D, is
traversed by an inclined ridge crest lettered MNO, which an extension of the

slope-profile line intersects at N. As the hypothetical contours are 200 yards
apart with a 100-foot contour interval, the inclination of the ridge is about 9
degrees. In consequence declivities less than 9 degrees do not exist in this
area. Therefore in the zone XY where the inclination along the slope profile
line is between 6 and 12 degrees, trial measurements from different starting-
points will eventually establish a slope length averaging 9 degrees which
defines at its upper end the summit of the slope where there is
profile. In cases
a valley-floor gradient, the base of a slope. B, would be usefully defined by
criteria similar to those which limit the summit end.
B. Soil Moisture Content
Innumerable aspects of the behaviour of erosional processes are

closely
related to a single factor. Dependent on this neglected factor, the water
Figure App.J Some procedures in the field study of hillslopes.
A. The slope pantometer, a simple device for the measurement of large numbers of
unit-length slope angles {from A. F. Fitly. 1968, Jour. Geol., Vol. 76).
B. A subdivided histogram for representing slope-angle frequencies according to

their position on the slope-profile (from Fitly, 1969). The example shows which
angles are represented on a gritstone-shale slope, typical of many escarpments in
the southern Pennines, and how the steepest angles are largely confined to the
summit third of the profile.
C and D. The laying out of the slope profile orthogonal to contours and its
delimitation (from A. F. Fitly, 1966, Zeit. Geomorph.,

f. Vol. 10). (C) The loca-
tion is confined to an area where contours are straight and parallel in order to hold
variations' in a third-dimension constant. The line may be drawn through a grid

reference point described by a randomly selected number, rn. S = summit,
B = base. (D) The definition of the summit point on a slope profile where the
summit is not flat.

Appendix 405
content of the soil, is the mechanical behaviour of soil, such as elastic, plastic,
or liquid states. These in turn influence the shrinkage, contraction, and shear-
ing properties. The simple addition of water may, by altering the mode of
failure, explain differences between landslide, mudflow, and sheetwash. Soil
water will influence the proportion of precipitation which infiltrates into the
soil and the amount which runs off on the surface; it will influence the
effectiveness of freezing temperatures. In addition to physical conditions,

most of the chemical and biological reactions of the soil are affected directly
or indirectly by the moisture content of the soil, as is the vegetation cover.
Although many attempts have been made to express indirectly the total
water content of the soil in terms of climatic data, and although few direct
observations appear in the geomorphological literature, the measurement of
soil-water content is straightforward.
Procedure
Although field collection of samples is involved, the drying and weighing is
usually done in the laboratory. The main consideration is to avoid moisture

loss from the soil sample between field and laboratory. Circular tins, approxi-
mately 6 cm in diameter and 1-2 cm deep, provide suitable sample containers
provided that the lid is reasonably airtight. The collection of several samples
from the same site at the same time, and their subsequent weighing one at a
time at progressively longer intervals after the sampling time gives a measure
of the efficiency of the containers. If the loss is measurable a correction factor
can be worked out for use in subsequent calculations. Alternatively, a plastic
bag packed with about 250 gm of soil and sealed tightly with a rubber band
will, on return to the laboratory, provide a sub-sample which, if extracted
from the centre of the bulk sample, resembles field moisture conditions closely.
In the laboratory a tin and wet soil are weighed, Wm. The lid is then
removed from the sample tin, and tin and lid are placed in an oven. A
temperature of 105°-1 10° C is maintained for 24 hours. The lid is replaced
and the container placed in a desiccator where silica gel maintains a dry
atmosphere, thus avoiding hygroscopic absorption of atmospheric humidity

during cooling. When cool, the weight of the dried soil and tin is obtained,
Wd. The tin, cleaned and dried, is then weighed, Wt. The moisture content, m,
of the soil, expressed as a percentage of the dry soil, isgiven by
Wm-Wd °° P""
" Wd~-Wt ""
It should be noted that as the result

is expressed as a percentage of the dry
soil, soil moisture contents from waterlogged sites may be as high as 200-
300 per cent. Also the calculation does not give a direct measure of the
volume of water in the soil. This can be calculated only if the volume of the
field sample is measured when collected. However, difficulties in extracting an
undisturbed core of soil make this extra measurement unnecessary for many
practical purposes.
C. Estimation of Stream-flow
The arguments for measuring stream-flow are compelling. Many authorities
maintain that erosion by running water is still to be regarded as the dominant
agent in landform sculpture, and detailed hydrological investigations have
enabled geomorphologists, like G. H. Dury, to suggest new hypotheses con-
cerning drainage evolution. Further, the volume of stream-flow is an essential
item for the calculations of denudational loss which provide a vital aid in
interpreting rates of change in the evolution of landforms. Not least, as water
is one of man’s most vital resources, no geomorphologist can equip himself
with techniques to turn more readily to practical ends than those offered by
some knowledge of stream-flow observation.
Fluctuations in stream-flow at a point are readily and accurately recorded
by measuring the water height, or stage, in relation to a fixed datum. It can
hardly be emphasized too strongly that one needs only a ruler to learn at first
hand a great deal about the actual behaviour of streams. As an additional

observation, the timing of float velocities provides a useful indication of the
volume of stream-flow involved. While it is evident that floats necessarily
provide only an approximation to the flow rate, specially constructed flumes
with continuous recording equipment provide detail beyond the needs of most
geomorphological investigations. In addition flumes are too expensive to form
the basis of a dense network which could provide information on geo-
graphical contrasts in a small area. As an alternative, the current meter, being
portable, has an advantage for a study of variations in the intensity of fluvial

processes within a study area, but this instrument is expensive and is costly to
maintain and to calibrate. The float method has, therefore, some advantages.
L. B. Leopold, M. G. Wolman, and J. P. Miller (1964) point out that floats
are particularly valuable during floods. In addition the lack of sophistication
in the method can be offset by careful planning because uniformity of practice
gives the highest degree of comparability between stream gaugings and also
because careful selection of the reach for measurement is the most important
factor if gauging of stream-flow is to be reliable.
Procedure
The channel should be open and straight, preferably with high, vertical banks
to contain high water-flows, and with a natural or artificial constriction
B
Appendix 407
downstream which, by maintaining a smooth water surface upstream, estab-
lishes a clear relation between stage and discharge. There should be some
permanent reference point to use as a datum for determining stage height. A

stick, weighted with a nail at one end to make it float upright, provides a
serviceable float.
Fig. App.2.A illustrates the measurement of the positions of 7 floats, in a
hypothetical stream reach, as they cross the upstream line UV, and then again
as they cross the downstream line DE. The distance between these two lines.
Figure App.2 Estimation of stream flow by floats (for explanation, see text).
in is divided by the float velocity measured in seconds. The result

metres, is
then multiplied by a correction factor, 0-85 to give a velocity representative
of total depth of water flowing beneath the float. For a bed of
canal-like
smoothness a correction factor of 0-90 would be appropriate as would a
factor of 0-80 for a very rough boulder-strewn channel. MN
is a line across
the stream, half-way between the lines UV and DE. The
points at which the
paths of the floats cross this line in VI.2.A are shown in fig. App.2.
fig.
together with the corrected velocities which are joined to

provide a velocity
distribution curve.
The measured reach is now divided into 3 or more panels of equal width.
In App.2.C there are 5. indicated by Roman numerals I
fig.
to V and
separated by panel boundaries labelled a to/ The velocity,
V, at the line MN.
for the midpoint of each panel, is read
off from the appropriate point on the
velocity distribution curve (fig. App.2.D).
Stream depths at the panel boundaries are measured

along both lines UV
and DE. The average of a pair of depth readings provides the values for the
depth at the panel boundaries a to/ along the mid-reach line, MN. The width,
u’, between panel boundaries along MN is also measured. The discharge, q,
for the panels is then obtained,
,
=w X —
depth a depth 6
1-
X VI
2
„ ,
depth b + depth c ,,
qV = u’X —
depth c -r
E- —- X
depth/
F
2
The total discharge, 0, in cubic metres per second, is the sum of the indi-
vidual panel discharges.
Ten measurements of discharge, if they cover a wide range of runoff
conditions, may be sufficient to define a graph between stage height and
discharge. With the aid of this graph, known as a rating curve, it is then
possible to estimate the discharge volume from a reading of the stage alone. It
is worth noting that the scatter of points about the rating curve is a helpful
indication of the consistency of the discharge measurements, and that
the shape of rating curves may vary from one station to another due to
differences in channel configuration.
D. Particle-size Analysis of Sands and Pebbles
Particle-size analysis is a fundamental stage in the description of sediment

transportation and deposition, and of depositional landforms. In areas where
mechanical weathering is important particle-size analysis again provides basic
data. As a lengthy pre-treatment is required to remove organic matter from
soils and as the size analysis of clay and fine silts involves several stages, some
initial familiarity with the usefulness of particle-size analysis is most readily
obtained by studying problems involving only sandy or coarse silt sediments.
The sieving of sands and coarse silt particles, free from organic matter, is
straightforward.
Procedure
Take about 100 gm of sediment which have been thoroughly dried in an oven
at a temperature not greater than 105°— 1 10°C. Pass the sediment through a
2-mm sieve and weigh the coarser fraction before discarding it. When the
fraction finer than 2 mm has been weighed, the weight of the coarser fraction
can be recorded as a percentage of the total sample.
Arrange a nest of about five sieves with the mesh aperture size increasing
Appendix 409
upwards and with a collecting pan at the base. Pour the sub-sample of
sediment finer than 2 mm on to the uppermost sieve and cover with the lid
and shake the nest of sieves. After about 8 minutes further effort will not
produce an appreciable increase in the degree of sorting obtained. In shaking,
pressure should be applied to lid and base of the nest to maintain a tight fit
between sieves. The basic motion in shaking should be rotary, to keep the
particles moving round the sieves, combined with vertical jolts to facilitate the
sorting process. The contents of each sieve are emptied on to a sheet of paper
and then funnelled into a container for weighing, together with any grains
wedged in the mesh of the sieve which can be dislodged by careful brushing.
The percentage of the total weight retained at each size division is obtained.
Ws
X 100
Wd
where Ws is the weight of material retained at a given size grade and Wd is
the original dry weight of sub-sample. The weight of material on the collect-
ing pan is mesh size of finest
the percentage of the total weight finer than the
sieve. The percentage weight of material on the pan, when added to that of
material retained on the finest sieve, gives the percentage of the total weight
finer than the mesh size of the next-to-the-finest sieve size. In this way a
cumulative percentage worked out for material finer than each sieve size,
is
the cumulative curve drawn up, and indices for describing median size and
sorting interpolated, as shown in fig. III.4.

The approach for analysing the sizes of gravels and pebbles is similar in
principle to sieving. The three main axes of a pebble - length, breadth, and
depth - are defined at right-angles to each other and measured using either a
ruler or preferably calipers. The pebble is weighed. For a sample of about
fifty pebbles it is possible to calculate the cumulative percentage weight of

pebbles smaller than a series of given length or breadth dimensions, or of a
mean value of the three axes, and then to plot a cumulative curve. Ratios
I ~h b
between the axes such as Cailleux’s flatness index can be calculated.
2a
These results are plotted as a histogram, pebble analysis involves little
expense compared with the cost of sieves and the principles of particle-size
analysis and its results can be appreciated from a study of pebbles alone if the
cost of sieves is prohibitive. However, it should not be beyond the ingenuity
of some to find cheap, mass-produced items to provide a range of aperture
sizes that would subdivide sand-sized particles. The actual sizes might
be
difficult to estimate accurately but a series of samples would be effectively
ranked. The most important point in analysing sediments is to avoid wherever
possible the collection, in one sample, of material which can be seen to be of

more than one type. For instance, in stratified materials, sample parallel
to the bedding; in rivers sample upstream from confluences where unlike

materials become mixed.
E. Heavy Minerals
A complete heavy mineral analysis requires great expertise, specialized equip-
ment, and a lot of time. However, in certain localities it is possible to study
how one heavy mineral moves in relation to lighter particles by separating the
magnetic component of a sand, if present, with a simple horse-shoe magnet.
For example, using this method on barchan dunes in southern Peru,
Hastenrath (1967) demonstrated a deficiency of the heavy mineral on the slip
face of the dunes.
Table App.l Deficiency of heavy minerals on the slip face of barchan dunes in
southern Peru, demonstrated by the use of a horse shoe magnet (Hastenrath, 1967)
PERCCNTAGE MAGNETITE OF TOTAL SAMPLE IN

THREE DUNES
East horn 0 8 1 1 11 3
West horn 3 5 1 8 7 6
Windward face 1 5 3 1 3 4
Crest I 8 2 4 1 8
Slipface 0 8 — 1 2
F. Soil pH
The measurement of pH is useful in weathering and in vegetation studies since
it forms one of the most comprehensive single summaries of chemical condi-

tions in the soil. For instance, on a granite catena in Rhodesia the pH of
twenty-nine samples was highly correlated with base saturation (r = 0- 73). In
southern Scotland, soils developed from greywackes and shales on Broad

Law have low pH values of 4- 0-4- 9 which correlate with the low exchange-
able base content and degree of base saturation. In most situations where
values higher than 8-5 occur, these usually indicate the presence of sodium
carbonate. Values of pH which are low when compared with soils in different
geomorphic situations or formed closer to bedrock, or from land surfaces of

different ages, indicate the severity of leaching. For example, a surface
horizon of a soil developed on granodiorite at Elberton, Georgia, had a pH of
4 -8-5 -2 compared with 8-7 at the surface of the unweathered bedrock.
Changes downslope may reflect local changes in leaching intensity, or, as is
particularly the case on chalk scarp slopes, the rapidity of renewal of un-
weathered fragments from a base-rich bedrock. Changes of pH in relation to
Appendix 411
time of exposure of depositional landforms are particularly noteworthy. Table

App.2 illustrates the relationship between exposure time of glacial moraines
and the pH of the surface soil. Although must be stressed that pH readings
it
can have no precise significance, the examples show that estimates of pH are
well worth recording.
Table App.2 The relationship between pH and age of moraines (from Stork, 1963,
and from R. F. Chandler, 1942, Soil Sci. Soc. Am. Proc., Vol. 7).
LOCALITY
Storglaciar, Kebnekajse Mendenhall Glacier, near

massif, N. Sweden Juneau, Alaska
Exposure in Exposure in
years pH years pH
0 81 15 5-4
5 7-7 90 51
10 7-8 250 4-3
15 6-5 approx. 1000 3-7
30 61 — —
50 5-4 — —
Procedure
To estimate soil pH by the Barium Sulphate method, stoppered test tubes 20

cm long and 1 cm internal diameter are used. The amount of soil placed in the
test tube depends on the texture of soil, ranging from a depth of 1-5 cm for a
clayey soil to 4 cm for a gravelly soil. The depth is then made up to 5 cm with
an appropriate amount of barium sulphate. Distilled water is added to bring
the height of the column up to 10 cm. This is thenmade up to 1 1 5 cm by the-
addition of Soil Indicator, a soil-testing reagent. The top end of the test tube is
then stoppered with a second rubber bung and the suspension shaken vigor-
ously. The test tube is then placed vertically in a rack, and when 2-5 cm of the
supernatant liquid is clear, the colour of the liquid is compared with a range
of standard colours on a special pH chart. Interpolation between the chart
colour chips is not easy, and improved results are obtained if several samples
are analysed at the same time and then ranked in order of increasing intensity
of colour.
Among which complicate the interpretation of pH observations
the factors
moisture. Waterlogged soils are commonly close to neutral
is soil
but may, on
drying, fall from 7 to 4 as a result of oxidation. High
carbon dioxide concen-
trations may lower pH by as much as a unit compared with similar soils
where the carbon dioxide concentration is low. At a given point variations of

0-5 pH units may occur during the course of a year.
The complexity of factors which might produce a given pH value makes
most interpretations tentative, and very rarely can full use be made of really
accurate measurements of pH of a field soil. In consequence little information
is lost if measurements are made to the nearest O-l of a pH unit. Therefore
more sophisticated methods than the Barium Sulphate method are unneces-
sary for most field study purposes.
G. Dissolved Solids
A comprehensive study of chemical weathering requires a great deal of

expertise and equipment. However, the average rate of chemical denudation
is very nearly that of any of the major constituents and calcium carbonate,
which is quantitatively the most important in many drainage waters, is neither
difficult nor expensive to determine accurately. Its study, in addition to
providing first-hand insight into chemical weathering also can improve one’s
appreciation of the practical problems of maintaining water quality and of
controlling pollution.
Procedure
Pipette 100 ml of a drainage water sample into a flask or dish, and add 1 ml of
4N sodium hydroxide solution. Add 1 calcium hardness indicator tablet
which, when dissolved, colours the solution pink. Fill a burette with N/50
E.D.T.A. and add this dropwise to the sample. The end-point of this titration
is reached when the sample solution becomes violet, and the addition of a
further drop of N/50 E.D.T.A. produces no further colour change. If possible
the sample should be agitated during the titration and the flask or dish should
stand on a white background. Calcium carbonate content, or calcium hard-
ness in non-limestone areas, in parts per million is given by
^ E.D.T.A. used, in ml x 1000

Too
The reagents are standard for the determination of water hardness and are
therefore readily obtained.
H. Suspended Sediment
Although the measurement of suspended sediment is one of the most
important techniques in measuring contemporary denudation rates, the
technique when compared with those already described, requires greater
Appendix 413
skill and much more time. Also an expensive, highly sensitive balance is
essential.
Procedure
The bung of two openings; a nozzle about 1 cm in

the sampling bottle has
diameter to point upstream and a tube through which air can be expelled as
the bottle fills up. The bottle is raised and lowered at points across a stream.
By maintaining a constant rate of lowering and raising the amounts of water
entering the bottle in shallower or in slower moving parts of the stream will
be proportional to the depth and velocity of flow. Accordingly, the bulk

sample is representative of the amount of sediment moving through the cross-
section.
In the laboratory the volume of the sample is carefully measured. It is then

heated to evaporate some of the water. A two-piece funnel is assembled with a
filter paper, of known weight and capable of retaining particles as small as 0- 2
microns, placed on the filter disc of the lower part of the funnel. The upper
part of the funnel is then clamped securely on top. A pump,
water-filter
attached to a tap, maintains a vacuum in the flask which supports the funnel.
When all the water has passed through the funnel, the filter paper is removed,
oven-dried, placed in a desiccator to cool, then weighed. The computed
weight of sediment is divided by the weight of the sample. The result, multi-
plied by one million, gives the concentration of suspended sediment in parts
per million.
Any small area with some relief offers scope for landform study. In fact
with the contrast between north- and south-facing slopes differences equal to
several degrees of latitude are on the doorstep. With contrasts between
seasons, an indication of past climates is to some extent compressed into the
present If the student of geomorphology sets up a simple hypothesis, sees
some advantage in making measurements to test it, and carefully collects field
data by using any one of the above methods, or others not described here, his
achievement will be substantial. Even if there is no apparent or comprehen-
sible pattern, his accomplishment is still to have performed one of the most
difficult tasks in landform study, that of collecting field measurements.
Selected Bibliography
General
BAKKER. J1960 Some observations iti connection With recent Dutch investiga
B
tions about granite weathering and slope development in different climates and
climate changes. Zett / Geomorpit Suppbd l.pp 69-92 ,
BECkiNSALE, R p 1966 Land, ati and oceaii. 4th Kv ed (Butterworth. London).

448 pp
BIRD. E c F 1968 Coasts (Aust Nat Univ Press. Canberra). 246 pp
BIRD, j B The pinsiogiaphv oj Aictic Canada (John Hopkins Press.
1967
Baltimore). 336 pp
BIROT. p 1960 Le cycle d'eiosion sous les dijjfeients climats (Centro de Pesquisas
de geografia do Brasil. Univ of Brazil. Rio de Janeiro) 137 pp trans by C I ,
Jackson and K M Clayton. 1968. Thu c\cle of erosion in different climates

(Batsford. London). 144 pp
BIROT. P 1959 Precis de geograplue physique generaleiArmand Cohn, Pans) 403

pp trans by M Ledesert 1966 General pin sical geograpin (Harrap, London)
.
360 pp
BIROT. p 1965 Geographic physique geneiale de la zone intertropicale (Centre
Doc Univ Pans). 290 pp
.
BLACK. R F 1969 Geology, especially geomorphology, of northern Alaska. A rctic,

Vol 22, pp 283-99
BLOOM A. L 1969 r/ie si(i/ace o/'t/ie eart/i (Prentice Hall, New York) 152 pp
BUTZER. K w 1964 Enviionment and aichaeo!og\ (Methuen. London) 524 pp
BUTZER.K w.HANSEN.c L . and G L AD FE LTE R, B G 1968 Desert and ! ner in
Nubia Geomorphologt and pre historic environments at the Aswan reset von
(Univ Wisconsin Press, Milwaukee), 562 pp
CH ARLESWORTH, j K 1957 T'/ie 2«a/er«a/ p eta (Arnold. London), Vol 1.592
pp.Vol 2. pp 593-1700
CLAYTON, k M (Ed) 1964 A bibliography of British geomorphologv. j9m Geo
morph Reseat ch Group Occ Paper, No 1. 21 1 pp
Selected bibliography 415
COQUE, R. 1962. La Tunisie presaharienne (Armand Colin, Paris), 476 pp.
COTTON, c. A. 1 960. Geomorphology, an introduction to the study of landforms. 7th

rev. ed. (Whitcombe and Tombs, Christchurch). 505 pp.
DAVIES, J. L. 1969. Landforms of cold climates (Aust. Nat. Univ. Press, Canberra),
200 pp.
DAVIS, w. M. 1909. Geographical essays (Ginn, Boston). 777 pp.; republished 1954
(Dover, New York).
DERRUAU, M. 1963. Precis de geomorphologie (Masson, Paris), 393 pp.
DURY. G. H. 1959. The face of the earth (Penguin, Harmondsworth), 223 pp.
EASTERBROOK, D. j. 1969. Principles of geomorphology (McGraw-Hill, New

York). 462 pp.
EMBLETON, c. and KING, c. A. M. 1968. Glacial and periglacial geomorphology

(Arnold, London), 608 pp.
FAiRBRiDGE. R. w. (Ed.). 1968. The encyclopedia of geomorphology (Rheinhold.

New York), 1 295 pp.
FLINT, R. F. 1957. Glacial and Pleistocene geology (Wi\ty, New York), 553 pp.
GiLLULY, A., WATERS, A, c.. and WOODWARD, A. o. 1968. Principles of geology

(Freeman, San Francisco), 687 pp.
GODARD, A. 1965. Recherches de geomorphologie en £cosse dit nord-ouest (Cen.

Nat. Rech. Sci., Paris), 701 pp.
GUiLCHER, A. 1954. Morphologk littorale et sous-marine (Presses Univ. France.

Paris).215 pp.; trans. and rev. by B. W. Sparks and R. H. W. Kneese, 1958.
Coastal and submarine morphology (Mtlinien. London), 274 pp.
HOLMES. A. 1965. Principles of physical geology. 5th rev. ed. (Nelson, London),
1288 pp.
HORTON. R. E. 1945, Erosional development of streams and their drainage basins:
hydrophysical approach to quantitative morphology. Bull. geol. Soc. Am., Vol. 56,
pp. 275-370.
INSTITUTE OF GEOGRAPHY, U.S.S.R. ACADEMY OF SCIENCES. 1969. The
physical geography of China (Praeger, New York), Vol. 1 . 448 pp.; Vol. 2. 337 pp.
JENNINGS, j, N. and MABBUTT. J. A. (Eds). 1967. Landform studies from Australia
and New Guinea (Cambridge Univ. Press, Cambridge). 434 pp.
KEEN, M. J. 1968. An introduction to marine geology (Pergamon. Oxford), 218 pp.
KING, c. A. M. 1959. Beaches and coasts (Arnold, London), 403 pp.
king, l. c. 1967. The morphology of the earth, 2nd ed. (Oliver and Boyd.
Edinburgh). 726 pp.
lake, p. 1958. Physical geography, 4th ed. (Ed. J. A. Steers) (Cambridge Univ.
Press. Cambridge), 483 pp.
LINTON. D. L. 1951. Problems of Scottish scenery. Scot. geog. Mag., Vol. 61.
pp. 65-85.
lliboutry, l. 1964. Traite de glaciologie (Masson. Paris), Vol. 1. 427 pp.; Vol. 2,
1040 pp.
LONG WELL, c. R., FLINT, R. F., and s AND E Rs. J. E. 1969. Physical geology (Wi\ty.
New York), 685 pp.
MACAR. p. (Ed.). 1966. L’evoliition des versants {XJniv. Liege et Acad. Roy. Belg.,
Liege), 384 pp.
MACHATSCHEK, F. 1968. Geomorphologie. 9th ed. (Teubner, Stuttgart), 209 pp.;

trans. by D. J. Davies, 1969. Geomorphology (0\Wer and Boyd, Edinburgh), 212
pp.
MARK. J. E. 1900. The scientific study of scenery (Methuen, London), 368 pp.; 4th
rev. ed.. 1912, 372 pp.
MATTSON. A. 1962. Morphologische Studien in Siidschweden und auf Bornholm
iiber nicht glaziale Formenwelt der Felsskulptur. Lund Studies in geog., Ser. A,
No. 20. 357 pp.
M AULL. o. 1958. Handbuch der Geomorphologie (Deuticke. Vienna). 600 pp.
MCGILL. J. T. 1960. Selected bibliography of coastal geomorphology of the world

(Los Angeles). 50 pp.
MILLER. T. G. 1 953. Geology and scenery in Britain (Batsford. London). 224 pp.
NICHOLS, R. L. 1960. Geomorphology of Marguerite Bay area. Palmer peninsula.

Antarctica. Bull. geol. Soc. Am., Vol. 71, pp. 1421-50.
NICHOLS. R, L. 1966. Geomorphology of Antarctica, Am. geophys. Union,

Antarctica Res. Ser.. Vol. 8. pp. 1-46.
PENCK, w. 1924. Die morphologische Analyse; ein Kapitel der physikalischen

Geologic (Engelhorns, Stuttgart). 283 pp.; trans. by H. Czech and K. C. Boswell.
1 953, Morphological analysis of land forms (Macmillan, London). 429 pp.
PINCHEMEL, p. 1954. Les plaines de craie du nord-ouest dtt Bassin Parisien et du

sud-est du Bassin de Londres et leiirs bordures (Armand Colin. Paris), 502 pp.
RAPP, A. 1960. Development of mountain slopes in Karkevagge and surroundings,

northern Scandinavia. Geografiska Annaler. Vol. 42, pp. 65-187.
RAN KAMA. K. (Ed.). 1965. The Quaternary (Interscience, New York), Vol. 1,
300 pp.; Vol. 2, 477 pp.
ROGNON. p. 1967. Le massif de I'Atakor et ses bordures {Sahara centrale). etude
geomorphologique (Cent. Nat. Rech. Sci., Paris), 559 pp.
RUSSELL, R. J. 1967. River plains and sea coasts (Univ. of California Press,
Berkeley). 1 73 pp.
SHARP. R. P. 1960. Glaciers (Oregon state system of higher education, Eugene),

78 pp.
SISSONS, J. B. 1967. The evolution of Scotland’s scenery (Oliver and Boyd,

Edinburgh), 259 pp.
SMALL, R. J. 1970. The study of landforms (Cambridge Univ. Press, Cambridge),

486 pp.
SPARKS. B. \v. I960. Geomorphology {Longmans, London). 371 pp.
STEERS, J. A. 1964. The coastline of England and Wales. 2nd ed. (Cambridge Univ.
Press. Cambridge), 750 pp.
thornbury. w. d. 1969. Principles of geomorphology, 2nd ed. (Wiley, New York),
594 pp,
TRICART, j. 1963. Geomorphologie des regions froides (Presses Univ. France,

Paris), 289 pp,; trans. by E. Watson, 1970, Geomorphology of cold environments
(Macmillan, London), 320 pp.
TRICART, J. 1965. Principes et methodes de la geomorphologie (Masson, Paris),

496 pp.
TRICART, J. 1965. Le modele des regions chaudes forets er savanes (Societe d’edition
d’enseignement superieur, Paris), 322 pp.
TRICART, J. and CAiLLEUX. A. 1960-61. Le modele des regions seches (Centre
Doc. Univ., Paris), Vol. 1, 129 pp.: Vol. 2, 179 pp.
TRICART, J. and CAILLEUX, A. 1967. Le modele des regions periglaciaires (Socictc
d’edition d’enseignement superieur. Paris). 512 pp.
WEBER, H. 1967. Die Oberflachenformen des festen Lands (Teubner, Leipzig),
367 pp.
WEST, R. G. 1963. Problems of the British Quaternary, Proc. geol. Ass., Vol. 74.
pp. 147-86.
WEST, R. G. 1968. Pleistocene geology and biology, with especial reference to the
British /s/es (Longmans, London). 377 pp.
WILHELM Y, H. 1958. KUmamorphologie der Massengesteine (Westermann,
Braunschweig), 238 pp.
woLDSTEDT, p. 1954-65. Das Eiszeitalter (Enke, Stuttgart). Vol. 1, 374 pp.;
Vol. 2. 438 pp.; Vol, 3, 328 pp.
WOOLDRIDGE, s. w. and MORGAN, R. s. 1959. An outline of geomorphology: the
physical basis of geography (Longmans. London). 409 pp.
WRIGHT. H. E. and FREY, D. G. (Eds). 1965a. The Quaternary oj the United States
(Princeton Univ. Press, Princeton). 922 pp.
WRIGHT. H, E. and FREY, D. G. (Eds). 19656. International studies on the
Quaternary (Geol. Soc, Am.. Special Paper No. 84), 565 pp.
WRIGHT. H. E. and OSBURN, w. H. (Eds). 1968. Arctic and alpine environments
(Indiana Univ. Press. Bloomington), 308 pp.
ZEUNER, F. E. 1959. The Pleistocene period. 2nd ed. (Hutchinson, London). 447 pp.
Chapter 1. Definitions, nature and basic postulates
A. Definitions
fenneman. n. m. 1936. Cyclic and non-cyclic aspects of erosion. Bull. geol. Soc.
^m.. Vol. 47, pp. 173-86.
hjulstrom. f. 1935. Studies of the morphological activity of rivers as illustrated
by the river Fyris, Bull. geol. Inst. Uppsala. No. 25, pp. 221-527.
LEOPOLD, L. B., woLMAN. M. G.. and MILLER, j. p. 1964. Fliivial processes in

geomorphology (Freeman, San Francisco). 522 pp.
TRICART, j, 1965. Principes et methodes de la geomorphologie (Masson, Paris),
496 pp.
WOOLDRIDGE, s. vv. 1951. The role and relations of geomorphology, in L. D.

Stamp and S. W. Wooldridge (Eds), London essays in geography (Longmans.
London), pp. 1 9-31.
B. Nature of Geomorphology
1. Description and interpretation
ALEXANDER, c. s. 1966. A method of descriptive shore classification and mapping

as applied to the northeast coast of Tanganyika, Ann. Ass. Ainer. Geogrs. Vol. 56,
pp. 1 28-40.
ARBER, M. A. 1949. Cliff profiles of Devon and Cornwall, Geog. Jour., Vol. 114,
pp. 191-6.
CAiLLEUX. A. and TRICART, J. 1956. Le probleme de la classification des faits
geomorphologiques, Ann. Geogr.. Vol. 65. pp. 162-86.
COTTON, c. A 1 954. Deductive morphology and the genetic classification of coasts.

Sci. Monthly. Vol. 78. pp. 1 63-81.
CURTIS. L. F.. DOORNKAMP. J. c.. and GREGORY, K. J. 1965. The description of
relief in field studies of soils, yoiir. Soil Sci., Vol. 16, pp. 16-30.
DARBY. H. c. 1962. The problem of geographical description. Trans. Inst. Brit.

Geogrs. No. 30, pp. 1-14.
DURY, G. H. 1963. Geographical description; an essay in criticism, j4nsf. Geog.. Vol.
9, pp. 67-78.
FOURNEAU. R. 1963. Essai de cartographic geomorphologique. Rev. Beige Geog..

Vol. 87, No. 3,pp. 1-7.
HAMILTON, R. A. et al. 1957. Surveying aneroids; their uses and limitations. Geog.
Jour.. Vol. 123. pp, 481-98.
HORTON. R. E. 1932. Drainage basin characteristics, Trans. Am. geophys. Union.

Vol. 1 3, pp. 350-61.
MccoY, R. M. 1969. Drainage network analysis with K-band radar imagery. Geog.
Rev.. Vol. 59. pp. 493-512.
MILLER. V. c. and miller, c. f. 1961. /’/loiogeo/ogy (McGraw-Hill. New York),

248 pp.
MORiSAWA M. 1958. Measurements of drainage-basin outline form. Jour. Geoi,

Vol. 66, pp. 587-90.
MONKHOUSE, F. J. and WILKINSON, H. R. 1971, Maps and diagrams. 3rd rev. ed.
(Methuen, London). 522 pp.
OLLIER, c. D. 1967. Landform description without stage names, Aust. Geog. Studies.
Vol. 5, pp. 73-80.
piTTY. A F. 1967. Some problems in selecting a ground-surface length for slope-
angle measurement. Rev. geomorph. dyn.. Vol. 18. pp. 66-71.
REED. B., GALVIN, c. J.. and MILLER, J. p. 1962. Some aspects of drumlin
geometry. Am. Jour. Sci.^ Vol. 260, pp. 200-10.
SAViGE AR, R. A. G. 1965. A technique of morphological mapping, Ann. /4ss. Amer.
Geogrs. Vol. 55, pp. 514-38.
SPIRIDONOV, A. I. 1952. Geomorfologicheskoe kartografirovanie (Moscow)-, trans.
by S. Kommling, 1956, Geomorphologische Kartographie (VEB Deutscher Verlag
der Wiss., Berlin), 160 pp.
TRICART, J., HiRSCH, A. R., and LE BOURDiEC, F. 1965, Presentation d’un extrait
de carte geomorphologique detaillee, Zeil.f. Geomorph., Vol. 9, pp. 133-65.
ZAKRZEWSKA, B. 1967. Trends and methods in land form geography, Ann. Ass.
Amer. Geogrs. Vol. 57, pp. 1 28-65.
2. Process and form
ALLEN, L. R. L. 1968. Current ripples: their relation to patterns of water and

sediment motion (North-Holland, Amsterdam), 433 pp.
BAGNOLD. R. A. 1941. The physics of blown sand and desert dunes. 3rd ed.. I960
(Methuen, London). 265 pp.
BRADLEY, w. H. 1966. Tropical lakes, copropel, and oil shale. Bull. geol. Soc. Am.,
Vol. 77, pp. 1333-8.
WOOLDRIDGE, s. w. 1958. The trend of geomorphology. Trans. Inst. Brit. Geogrs,
Vol. 25, pp. 29-35.
3. Artistic and scientific elements
AMERICAN GEOLOGICAL INSTITUTE. 1957. Glossary of geolog}’ and related

sciences (Amer. geol. Inst.. Washington), 325 pp.
BAULiG, H. 1956. Vocabulaire Franco-Anglo-Alleinand de geomorphologie, (Les
Belles Lettres. Paris), 229 pp.
de BOER. G. 1964. Spurn Head: its history and evolution. Trans. Inst. Brit. Geogrs.
No. 34. pp, 71-89.
BRYAN, K. 1940.The retreat of slopes, /45s. i4/77er. Geogrs, Vol. 30. pp. 254-67.
CHORLEY, R. J. 1958. The shape of drumlins. Jour. GlacioL. Vol. 3, pp. 339-44.
CHORLEY, R. J., DUNN, A. J., and BECKINSALE, R. p. 1964. The history of the
study of landforms (Methuen. London). Vol. 678
1, pp.
CRAIG. G. Y. 1969. Communication in geology, Scot. Jour. Geol.. Vol. 5, pp. 305-
21 .
DAVIES, G. L. 969. The earth in decay.

1
A history of British geomorphology 1578-
1878 (Macdonald. London), 390
pp.
Dt likowa. a. 1962. Notion et terme ‘periglaciaire’, Biul. Pervglacjalny. Vol. 1 1.
pp. 149-64.
HAMELiN, L. E. and CLiBBON, p. 1962. Vocabulaire periglaciaire bilingue, Cahier
de geogr. de Quebec. Vol. 6, pp. 201-26.
HUTCHINGS. G. 1960. Landscape drawing (Methuen. London). 102 pp.
INTERNATIONAL FEDERATION FOR INFORMATION PROCESSING AND

INTERNATIONAL COMPUTATION CENTRE. 1968. IFIP-ICC vocabulary of
information processing (North-Holland, Amsterdam). 208 pp.
LCGRAND, H. E. 1960. Metaphor in geomorphic expression. Jour. Geol., Vol. 68.

pp. 576-9.
LEOPOLD, L. B., and SKiBiTZKE. H. E. 1967. Observations on unmeasured rivers,
Geografiska Annaler. Vol. 49A, pp. 247—55.
TERS, M. 1960. Methodes et techniques modernes en geomorphologie. Inf geog..

Vol. 24. pp. 156-65.
4. Qualitative and quantitative aspects
AHNERT, F. 1966. Zur Rolle der elektronischen Rechnenmaschine unddes mathema-

tischen Modells in der Geomorphologie, Geog. Zeit.. Vol. 54. pp. 1 18-33.
BAULiG. H. 1959. Morphometrie, Ann. Geogr.. Vol. 68. pp. 385—408.

CHORLEY, R. J. 1966. The application of statistical methods in geomorphology, in
G. H. Dury (Ed.). Essays in geomorphology (Hememann, London), pp. 275-387.
CREAGER. J. s.. MCMANUS, D. A., and COLLIAS. E. E. 1962. Electronic data
processing in sedimentary size analysis. Jcwr. Sed. Petrol. Vol. 32. pp. 833-9.
DEVDARiANi. A. s, 1967. Matematicheskiy analtz v geomorfohgii (hieita Press,

Moscow), 155 pp.
DO OR NK AMP, J. c. (Ed.), 1969. The use of computers in geomorphological

research, Brit. Geomorph. Research Group. Occasional Paper Mo. 6. 90 pp.
FERRAR, A. M. 1961. The depth of some lakes in Snowdonia, Geog. Jour.. Vol. 127,
pp. 205-8.
K RUM BE IN. w. c. and MILLER, R. L. 1953. Design of experiments for statistical
analysis of geological data. Jour. Geol. Vol. 6 1. pp. 510-32.

LCGRAND, H. E. 1962. Perspective on problems of hydrogeology. Bull. geol. Soc.
Am.. Voi. 73. pp. 1 147-52.
MACK IN. J. H. 1963. Rational and empirical methods of investigation in geology, in
C. C. Albritton (Ed.). The fabric of geology (Freeman. Cooper, Stanford), pp. 135-
63.
MELTON. M. A. 1958. Correlation structure of morphometric properties of drainage

systems and their controlling agents. Jour. Geol. Vol. 66, pp. 442-60.
MORIS AWA, M. 1964, Development of drainage systems on an upraised lake floor.

Am. Jour. Scl. Vol. 262. pp. 340-54,
PEGUY. c. p. 1948. Introduction a I’emploi des methodes statistiques en geographic

physique. Rev. geog. Alpine. Vol. 36, pp. 1—103.
PINCHEMEL. p. 1950. L’etude des reseaux hydrographiques. Bull .455. geog. Fr..
Nos 208-9. pp. 72-80.

PITTY, A, F. 1968. Some comments on the scope of slope analysis based on fre-
quency distributions, Zeit.f. Geomorph., Vol. 12. pp. 350-55.
PITTY, A. F. 1969. A scheme of hillslope analysis. 1. Initial considerations and

calculations. Univ. Hull Occ. Papers in Geog., No. 9, 76 pp.
SCHEIDEGGER. A. E. 1960. Mathematical methods in geology, Am. Jour. ScL,
Vol. 258, pp. 218-21.
SCHUMM, s. A. 1960. The effect of sediment type on the shape and stratification of
some modern fluvial deposits. Am. Jour. ScL, Vol. 258, pp. 1 77-84.
SLAYMAKER. H. o. (Ed.). 1968. Morphometric analysis of maps, Brit. Geomorph.

Res. Group Occ. Papers. No. 4, 67 pp.
SMITH, F. G. 1966. Geological data processing: using fortran iv (Harper and Row,
New York), 284 pp.
SPATE, o. H. K. 1960. Quantity and quality in geography, Ann. ,4ss. Ainer. Geogrs.
Vol. 50, pp. 377-94.
STRAHLER, A. N. 1954. Statistical analysis in geomorphic research. Jour. Geol..

Vol. 62, pp. 1-25.
SVENSSON. H. 1959. Is the cross-section of a glacial valley a parabola?. Jour.

GlacioL. Vol. 3, pp. 362-3.
TRICART, j. 1947. Sur quelques indices morphometriques, C. r. Acad. ScL. Vol.

225. pp. 747-9.
TROEH, F. R. 1965. Landform equations fitted to contour maps. Am. Jour. Sci..
Vol. 263. pp. 616-27.
WOOD, w. F. and swell, j. b, 1957. The dispersion of geomorphic data around
measures of central tendency, Ann. Ass. Amer. Geogrs. Vol. 47. pp. 184-5.
5. Laboratory and field work

BROWN. E. H. 1969. The teaching of fieldwork and the integration of physical
geography, in R. U. Cooke and J. H. Johnson (Eds). Trends in geography: an
introductory survey (Pergamon. Oxford), pp. 70-8.
bruun. p. 1966. Model geology: prototype and laboratory streams. Bull. geol. Soc.
Am.. Vol. 77, pp, 959-74.
BROUN, p. 1968, Model geology: prototype and laboratory streams: reply. Bull,
geol. Soc. Am., Vol. 79, pp. 395—8.
butzer, k. w. and cuerda, j. 1962, Coastal stratigraphy of southern Mallorca
and its implications for the Pleistocene chronology of the Mediterranean sea. Jour.
Geol.. Vol. 70, pp. 398-416.
CAZALis, p. 1961. Geomorphologie et processus experimental. Cahiers geog.
Quebec. No. 9, pp. 33-50.
DERBYSHIRE, E. 1958. The identification and classification of glacial drainage
channels from aerial photographs. Geografiska Annaler. Vol. 40,

pp, 88-95, 1
HOoke. r. Le
1968. Model geology: prototype and laboratory streams: discus-
B.
sion. Bull. geol. Soc. Am.. Vol. 19.
pp. 391-4.
KELLER, w. D. 1963. Field work —our scientific birthright. Jo;/r. Geol. Ed., Vol. ! I,
pp. 119-23.
K INC. c. A. M. 1 966. Techniques in geomorphology (Arnold, London), 342 pp.
MCGEE, w. j. 1897. Sheetflood erosion. Bull. geol. Soc. Am., Vol. 8, pp. 87-1 12.
MEIER, M. F. 1967. Why study glaciers, in the context of water resources?, Trans.
Am. Geophys. Union. Vol. 48, pp. 798-802.
OGILVIE. A. G. 1936. The earth sculpture laboratory. Geog. Jour.. Vol. 87. pp.
145-9.
(isTREM. G. 1965. Problems of dating ice-cored moraines, Geografiska Annaler,

Vol. 47A. pp. 1-38.
POUQUET, J. 1969. Geomorphologie et ere spatiale, Zeit.f. Geomorph., Vol. 13.

pp. 414-71.
POWELL. J. w. 1875. Exploration of the Colorado River of the West and its tribu-
taries (Govt, printing office. Washington), 285 pp.

scHACHORi. A. and sEGtNER. I. 1962. Sprinkling assembly for simulation of
design storms as a means for erosion and runoff studies. Bull. /Iss. Internl Hydrol.
Soc.. Vol. 7. pp. 57-72.
STEWART, A. B. 1965. Soil in the field and in the laboratory, 7oi(P. Soil Sci., Vol. 16.
pp. 171-82.
vox ALL, w. H. 1969. The relationship between falling base level and lateral erosion
in experimental streams. Bull. geol. Soc. Am., Vol. 80, pp. 1379-84.
wuRM. A. 1935. Morphologische Analysen und Experiment. Zeit.f. Geomorph.,

Vol. 10. pp. 1-24.
6. The role of geomorphology
BARKER. J. P. 1959. Recherches neerlandaises de geomorphologie appliquee. Rev.
geomorph. dyn., Vol. 10. pp. 67-84.

HOW.ARD, A. D. 1967. Drainage analysis in geologic interpretation: a summation.
Bull. Amer. Ass. Petrol. Geols, Vol. 5 I. pp. 2246-59.
KENNEDY, w. Q, 1962, Theoretical factors in geomorphological analysis. Geol.

Mag., Vol. 99. pp. 304-12.
KLiMASZEWSKi, M. 1960. ProWemes concernant la carte geomorphologique
detaillee, son importance et pratique, Przeglad Geogr., Vol. 32, pp. 459-85.
LCGRAND, H. E. 1962. Perspective on problems of hydrogeology. Bull. geol. Soc.

Am.. Vol. 73. pp. 1147-52.
LINTON, D, L. 1968. The assessment of scenery as a natural resource. Scot. geog.
Mag.. Vol. 84. pp. 219-38.
TRICART, J. 1962. Z. 'epiderme de la Terre. Esquisse d’une geomorphologie appliquee

(Masson, Paris), 167 pp.
TRICART, J. 1966. La place de la geomorphologie dans I’etude de la mise en valeur

des deltas tropicaux. in UNESCO. Scientific problems of the humid tropical zone
deltas and their implications (UNESCO. Paris), pp. 15-22.
WALTON. K. et al. 1966. The vertical displacement of shorelines in Highland Britain,

Trans. Inst. Brit. Geogrs. No. 39, 145 pp.
ZVEDER, L. N. 1967, Reconstructing Early Jurassic plains of accumulation from

occurrences of diamonds in present drainage network. Internl Geol. Rev., Vol. 10
(1968). pp. 1362-6.
C. Basic Postulates
1. Catastrophism and uniformitarianism
DOBBiE, c. H. and wolf, p. o. 1953. The Lynmouth flood of August 1952, Proc.
Inst. Civil Eng., Vol. 2, pp. 522-88.
DOEGLAS, D. J. 1959. Sedimentology of recent and old sediments; a comparison,

Geol. en Mijnbouw, Vol. 9, pp. 228-30.
DOUGLAS, I. 1964. Intensity and periodicity in denudation processes with special

reference to the removal of material in solution by rivers, Zeit. f. Geomorph.,
Vol. 8. pp. 453-73.
GOULD, s. J. 1965. Is Uniformitarianism necessary?. Am. Jour. Sci., Vol. 263,

pp. 223-8.
GRETENER, p. E. 1967. Significance of the rare event in geology, Bull. Amer. Ass.
Petrol. Geols, Vol. 51,pp. 2197-206.
HARRINGTON. J. w. 1 967. The first, first principles of geology. Am. yoi/r. Sc/.. Vol.
265, pp. 449-61.
HUME. J. D. and SCHALK. M. 1967. Shoreline processes near Barrow. Alaska: a

comparison of the normal and catastrophic. Arctic, Vol. 20. pp. 86-103.
KiDSON. c. 1953. The Exmoor storm and the Lynmouth floods. Geography, Vol. 38,
pp- 1-9.
malde, h. e. 1968. The catastrophic Late Pleistocene Bonneville flood in the Snake
River plain. Idaho. U.S. Geol. Siirv. Prof. Paper 596. 52 pp.
MILLER. D. J. 1960. Giant waves in Lituya Bay. Alaska, U.S. Geol. Surv. Prof. Paper
354-C. pp. 51-86.
NAIRN. A. E. M. 1965. Uniformitarianism and environment. Palaeogeog..
Palaeoclim.. PalaeoecoL, Vol. 1, pp. 5-11.
PLAYFAIR, J. 1802. Illustrations of the Huttonian theory of the earth (Creech,

Edinburgh). 528 pp. (Facsimile reproduction by Univ. of Illinois Press,
Urbana.)
SCHUMM. s. A. and chorley, r. j. 1964. The fall of Threatening Rock, Am. Jour.
Sci., Vol. 262. pp. 1041-54.
STEARNS, H. T. 1962. Evidence of Lake Bonneville flood along Snake river below
King Hill. Idaho, Bull. geol. Soc. Am.. Vol. 73. pp. 385-88.
WOLMAN. M. G., and MILLER. J. p. 1960. Magnitude and frequency of forces in

geomorphic processes. Jour. Geol.. Vol. 68. pp. 54-74.
2. Stillstands and the mobility of earth structures
AXELROD, D. I. 1962. Post-Pliocene uplift of the Sierra Nevada, California. Bull,

geol. Soc. Am., Vol. 73, pp. 183-98.
BADEN-POVVELL, D. F. w. 1963. Isostatic recovery in Scotland. Nature, Vol. 199,
pp. 546-7.
BLOOM, A. L. 1963. Late-Pleistocene fluctuations of sea-level and postglacial crustal
rebound in coastal Maine. Am. Jour. Sci.. Vol. 26 1. pp. 862-79.
BLOOM, A. L. 1967. Pleistocene shorelines: a new test of isostasy. Bull. geol. Soc.
Am., Vol. 78. pp. 1477-94.
BOURDIER, F. 1959. Origines et succes d’une theore geologique illusoire: I’eustat-
isme applique au terrasses alluviales. Rev. geomorph. dyn., Vol. 10, pp. 16-29.
CHAMBERLIN. T. c. 1909. Diastrophism as the ultimate basis of correlation. Jour.
Geol., Vol. 1 7, pp. 685-93.
CHORLEY, R. J. 1963. Diastrophic background to twentieth-century geomorpholo-
gical thought. Bull. geol. Soc. Am.. Vol. 74. pp. 953-70.
CHRISTENSEN. M. N. 1965. Late Cenozoic deformation in the central Coast Ranges
of California, Bull. geol. Soc. Am.. Vol. 76, pp. 1 105-24.
CHURCHILL, D. M. 1965. The displacement of deposits formed at sea-level. 6500
years ago in southern Britain. Quaternaria. Vol. 7. pp. 239-49.
DiXEY. F. 1938. Some observations on the physiographica! development of central
and southern Africa. Trans, geol. Soc. S.Afr.. Vol. 41. pp. 1 13-71.
FAiRBRiDGE. R. tv.. 1960. The changing level of the sea. Scl. American, Vol. 204,
pp. 70-9.
FAIRBRIDGE. R. w. 1961. Eustatic changes in sea level, in L. H. Ahrens et ai.

Physics and chemistry of the earth. Vol. 4 (Pergamon. New York), pp. 99-185.
FARRAND, vv. R. 1962. Postglacial uplift in North America. Am. Jour. Sci..
Vol. 260. pp. 181-99.
DONN. w. L.. FARRAND, VV. R., and EWING, M. 1962. Pleistocenc ice volumes and
sea-level lowering. Jour. Geol.. Vol. 70, pp. 206-14.
FLINT. R. F. 1966. Comparison of interglacial marine stratigraphy in Virginia.
Alaska, and Mediterranean areas. Am. Jour. Sci.. Vol. 264. pp. 673-84.
GILLULY. J. 1966. Orogeny and geochronology. Am. Jour. Sci.. Vol. 264, pp. 97-
111 .
GUiLCHER, A. 1969. Pleistocene and Holocene sea level changes. Earth Sci. Rev.,
Vol. 5, pp. 69-97.
HAILS, J. R. 1965. A critical review of sea-level changes in eastern Australia since the
Last Glacial. Aust. geog. Studies. Vol. 3. pp. 63-78.
KING, L. c. 1956. A geomorphological comparison between eastern Brazil and

Africa (central and southern). Quart. Jour. geol. Soc.. Vol. 1 12. pp, 445-74,
KUENEN. p. H. 1950. Marine geo/ogy (Wiley, New York). 568 pp.
KUENEN. p. H. 1 955. Sea level and crustal warping, Geol. Soc. Am. Spec. Paper 62.
pp. 193-204.
kukkam AKi, T. J. (Ed.). 1964. Symposium on Recent crustal movements in Finland

with bibliography, Fennia, Vol. 89, No. 2, 89 pp.
LIU KUANG-YEH. 1964. Neotectonic movement of the North China platform,

InternlGeol. Rev., Vol. 10(1968), pp. 857-69.
McGinnis, ld. 1968. Glacial crustal bending. Bull. geol. Soc. Am., Vol. 79, pp.
769-76.
MILLER, A. A. 1939. Attainable standards of accuracy in the determination of pre-

glacial sea levels by physiographic method. Jour. Geomorph., Vol. 2, pp. 95-1 15.
s HALER, N. s. 1 894. Pleistocene distortions of the Atlantic seacoast. Bull. geol. Soc.
Am., Vol. 5,pp. 199-202.
SUESS, E. 1888. Das Antlitz der Erde (Tempsky, Vienna), Vol. 2: trans. by H. B. C.
Sollas and W. J. Sollas, 1906, The face of the earth (Clarendon Press, Oxford),
Vol. 2,556 pp.
TANNER, w. F. 1968. Tertiary sea level symposium - introduction, Palaeogeog.,

palaeoclim., palaeoecoL, Vol. 5, pp. 7-14.
VERSTAPPEN, H. T. 1960. On the geomorphology of raised coral reefs and its
tectonic significance, Zeit.f. Geomorph., Vol. 4, pp. 1-28.
WEBB, s. D. and TESSM AN, N. 1968. A Pliocene vertebrate fauna from low elevation
in Manatee County, Florida. Am. Jour. Set, Vol. 266, pp. 777-81 1.
WELLMAN, H. w. 1966, Active wrench faults of Iran, Afghanistan and Pakistan,

Geol. Rundschau. Vol, 55, pp. 7 16-35.
3. The Cycle of Erosion
baulig, h. 1928. Les hauts niveaux d’erosion eustatique dans ie Bassin de Paris.
Ann. Geogr., Vol. 37. pp. 289-305 and 385-406.
BAULIG. H. 1952. Surfaces d’aplanissement. Ann. Geogr.. Vol. 61, pp. 161-83 and
245-62.
BRETZ, J. H. 1962. Dynamic equilibrium and the Ozark land forms. Am. Jour. Sci.,
Vol. 260, pp. 427-38.
BROWN, E. H. 1961. Britain and Appalachia: a study in the correlation and dating of
pianation surfaces, Trans. Inst. Brit. Geogrs, No. 29. pp. 91-100.
DAVIS, w. M. 1909. The geographical cycle, in Geographical Essays (Ginn, New
York), 777 pp.; reprinted, 1954 (Dover, Boston), pp. 249-78.
fenneman. n. m. 1936. Cyclic and non-cyclic aspects of erosion. Bull. geol. Soc.
-4m., Vol. 47. pp. 173-86.
hempel, l. 1958. Probleme der Oberfiachenformung in Grossbritannien unter
klimamorphologischer Fragestellung. Pet. geog. Mitt., Vol. 102, pp. 13-27.
hills, e. s. 1961. Morphotectonics and the geomorphological sciences, with special
reference to Australia, Quart. Jour. geol. Soc., Vol. 117, 77-89.
pp.
HODGSON, J. M., c ATT, J. A., and WEIR, A. H. 1967. The origin and development of
Clay-with-flints and associated soil horizons on the South Downs, Jour. Soil Sci.,
Vol. 18. pp. 85-102.
LOUIS, H. 1967. Reliefumkehr durch Rumpfflachenbildung in Tanganyika,

Geografiska Anmler, Vol. 49A, pp. 256-67.
M AC AR. P. 1 955. Appalachian and Ardennes levels of erosion compared. Jour. GeoL
Vol. 63, pp. 253-67.
NEEF. E. 1955. Zur genese das Formenbildes der Rumpfgebirge, Pet. geog. Mitt..
Vol. 101. pp. 183-92.
PITTY. A. F. 1968. The scale and significance of solutional loss from the limestone
tract of the southern Pennines, Proc. Geol. Ass.. Vol. 79, pp. 153-77.
RICH, J. L. 1938. Recognition and significance of multiple erosion surfaces. Bull,
geol. Soc. Am.. Vol. 49, pp. 1695-722.
SPARKS, B. vv. 1949. The denudation chronology of the dip-slope of the South
Downs. Proc. Geol. Ass.. Vol. 60, pp. 165-215.
TILLEY. P. 1968. Early challenges to Davis’ concept of the cycle of erosion. Prof.
Geogr. Vol. 20. pp. 265-9.
4. Morphoclimatic zones
bOdel, j. 1948. Die klima-morphologischen Zonen der Polarlander. Erdkunde.

Voi. 2. pp, 22-53.
bOdel. J. 1951. Klima-morphologische Beobachtungen in Siiditalien. Erdkunde.

Vol. 5. pp. 73-6.
bOdel. j. 1963. Klima-genetische Geomorphologie. Geol. Rundschau. Vol. 15.
pp. 269-85.
bOdel. j. 1969. Das System der klimagenetischen Geomorphologie. Erdkunde.
Vol. 23. pp. 165-83.
CHORLEY. R. J. 1957. Climate and morphometry. Joi/r Geol.. Vol. 65, pp. 628-38.
COTTON, c. A. 1947. Climatic accidents in landscape-making (Whitcombe

and Tombs, Wellington). 353 pp.; facsimile reproduction, 1969 (Hafner, New
York).
FRYE. J. c. 1959. Climate and Lester King’s ‘Uniformitarian nature of hillslopes’.
Jour. Geol.. Voi. 67, pp. 111-13.
GALLi'OLiviER. C. 1969. Climate: a primary control of sedimentation in the Peru-

Chile Trench. Bull. geol. Soc. Am.. Vol. 80. pp. 1849-52.
HOLMES, c. D. 1956. Geomorphic development in humid and arid regions. Am.

Jour. Sci.. Vol. 253, pp. 377-90.
KING. L. c. 1957. The uniformitarian nature of hillslopes. Trans. Edin. geol. Soc..
Vol. 17. pp. 81-102.
LEHMANN, H. 1954. Das Karstphafiomen in verschiedenen Klimazonen. Erdkunde.

Vol. 8.pp. 1 12-22.
LOUIS, H. 1957. Rumpffiachenproblem, Erosionzylkus und Klimageomorphoiogie,

Pet. geog. Mitt.. Erganzungsband 262.
PENCK. A 1914. The shifting of the climatic belts. Scot. geog. Mag.. Vol. 30,
pp. 281-93.
Selected bibliography All
SEUFFERT, o. 1969. KliiTiatische und nichtkiimatische Faktoren der Fussflachen-
entwicklung im Bereich der Gebirgsvorlander und Grabenregionen Sardinians,
Geol. Rundschau. Vol. 58, pp. 98-110.
TRICART. J. 1957. Comparison entre les conditions de fagonnement des lits
fluviaux en zone temperee et zone intertropicale. C. r. Acad. Sci., Vol. 245. pp.
555-7.
TRICART, J. 1953. Climat et geomorphologie. Inf. Geog., Vol. 17, pp. 39-51.
TRICART, J. 196 1. Les caracteristiques fondamentals du systeme morphologique des

pays tropicaux humides. Inf. Geog.. Vol. 25, pp. 155-69.
TRICART, J. and c AILLEUX, A. 1965. Introduction a la geomorphologie climatique

(S.E.D.E.S.). 307 pp.
TROLL, c. 1948. Der subnivale oder periglaziale Zyklus der Denudation. Erdkunde.
Vol. 2.pp. 1-21.
visHER. s. s. 1937. Regional contrasts in erosion in Indiana, with especial attention

to the climatic factor in causation. Bull. geol. Soc. Am.. Vol. 48, pp. 897-930.
5. Structure, process and stage
BRETZ, J. H. 1962. Dynamic equilibrium and the Ozark land forms. Am. Jour. Sci..
Vol. 260, pp. 427-38.

CHOLLEY, A. 1950. Morphologie structurale et morphologie climatique. Ann.
Geogr.. Vol. 59, pp. 321-35.
DYLiK, J. 1957. Dynamical geomorphology. Its nature and methods. Bull. Soc. Sci.
Lettres Lodz Cl. Ill, No. 8. pp. 1-42.
HACK, J. T. I960. Interpretation of erosional topography in humid temperate regions.

Am. Jour. Sci., Vol. 258A (Bradley volume), pp. 80-97.
hack, j. t. 1966. Circular patterns and exfoliation in crystalline terrane.
Grandfather Mountain area. North Carolina. Bull. geol. Soc. Am., Vol. 77,
pp. 975-86.
HEMPEL. L. 1959. Rezente und fossile Zertalungsformen im mediterranean Spanien.
Die Erde. Vol. 90, pp. 38-59.
hills, e. s. 1961. Morphotectonics and the geomorphological sciences, with special

reference to Australia, Quart. Jour. geol. Soc.. Vol. 1 1 7. pp. 77-89.
hollingworth, s. e. 1962. The climatic factor in the geological record. Quart.
Jour. geol. Soc.. Vol. 1 1 8. pp. 1-21.
king, l. c. 1953. Canons of landscape evolution. Bull. geol. Soc. Am., Vol. 64.
pp. 721-62.
KITTS. D. B. 1963. Historical explanation in geology. Jour. Geol.. Vol. 7 297-
1, pp.
313.
melton, f. a. 1959. Aerial photography and structural geomorphology. Jour.

Geol.. Vol. 67, pp. 351-71.
K'iini, h. 1967. The dependence of relief on the structure and composition of the
bedrock in Western Inari, Finnish Lapland, Feimia, Vol.
97, No. 2, 28 pp.
428 Intioduction to Geomorphologv
SYLVESTER BRADLEY, p c 1967 Evolution versus entropy, Proc Geol Ass,

Vol 78. pp 137-47
TUAN, Yi FU 1962 Structure, climate and basin land forms in Arizona and New
Mexico, /I h/i Ass Amei Geogis, Vol 52, pp 51-68
6 The necessity for simplification of geomorphological complexity
ALLEN, p 1964 Sedimentological models. 70 (1/ Sect Petio/. Vol 34. pp 289-93
CHORLE\, 1962 Geomorphology and General Systems Theory.
R J US Geol
Stirv Prof Papei 500 B. 0 pp 1
CHORLEY, R J and HAGGETT. P (Eds ) 1967 Models in geogiaptn (Methuen,

London), 8 6 pp 1
CONACHFR, A J 1969 Open systems and dynamic equilibrium in geomorphology

a comment, /fust geog Studies. Vol 7, pp 153-8
HOLMES, c D 1964 Equilibrium in humid climate physiographic processes Am

Join Sa. Vol 262. pp 436—45
HOOkE. R LCB 1968 Steady State relationships on and region alluvial fans in
closed basins. Am Jour Sci . Vol 266. pp 609-29
HOWARD. A. D 1965 Geomorphological systems -equilibrium and dynamics Am
Jour S’ci. Vol 263. pp 302-12
k RUM BE IN, u c 1968 Statistical models in sedimentoloaj. Sedimentologi

Vol 10. pp 7-23
MILLER J p 1961 Solutes in small streams draining Single rock types. Sangre de
Cristo Range. New Mexico. U S Geo! Sun IVatei siipph Papei 1535 F. 23 pp
OLLIER C D 1968 Open systems and dynamic equilibrium in geomorphology.
Aust geog Studies. Vol 6 pp 167-70
SMALLEt. I J and VITA FiNZi, c 1969 The concept of system in the earth
sciences. particularU geomorphology. Sk// geol Soc /Im.Vol 80. pp 1591-4
WRIGHT, c w 1958 Order and disorder in nature. Proc Geol zlsi . Vol 69.
pp 77-82
D. Some Conclusions about Geomorphology
ALBRITTON. C c (Ed) 1963 The fabitc of geologi (Freeman and Cooper.

Stanford). 372 pp
BAULIG. H 1957 Les methodes de la geomorphologie d'apres M Pierre Birot. i4;i«
Geog. Vol 66, pp 97-124 and 221-36
BAULIG H 1958 La legon de Grove Karl Gilbert, Aim Geog. Vol 67. pp 289-
307
BIROT. p 1958 Les tendances actuelles de la geomorphologie en France. Zeit f
Geomoiph .Vol 2. pp 123-34
DORALL R D 1968 Geomorphology - the emergence of a science. Geogi flp/uca
Vol 4.pp 24-8
DYLiK, J. 1953. Caracteres du developpement de la geomorphologie moderne. Bull.

Soc. Sci. Leitres Lodz Cl. IIL No. 4, 40 pp.
FOLK. R. L. and PERM, J. c. 1 966. A portrait of Paul D. Krynine, Jour. Sed. Petrol..
Vol. 36. pp. 85 1-63.
GELLERT. J. 1957. Systcmatik und Problemstellung der physischen Geographic.

Geog. Berichte, Vol. 2. pp. 89—1 02.
HARRISON. J. M. 1968. Geological sciences in the world scientific community.

Quart. Jour. geol. Soc.. Vol. 124. pp. 1—8.
MARTIN. L. 1950. William Morris Davis; investigator, teacher and leader in geomor-
phology. Ann. Ass. Amer. Geogrs. Vol. 40. pp. 1 72-80.
MILLER. J. p. 1959. Geomorphology in North America. Przeglad Geogr.. Vol. 31,

pp, 567-86.
PEEL, R, F. 1967. Geomorphology: trends and problems. The Advancement of

Science. Vol. 24, pp. 205-16.
RUSSELL, R. J. 1949. Geographical geomorphology. Ann. Ass. Amer. Geogrs.
Vol. 39, pp. 1-11.
SCHUMM, S. A. and lichty. r. w. 1965. Time space, and causality in geomor-

phology, Am. Jour. Sci.. Vol. 263. pp. 1 10-19.
STEERS. J, A. 1 960. Physiography: some reflections and trends. Geography. Vol. 45.
pp. 1-15.
WOOLDRIDGE, s. w. 1958. The trend of geomorphology. Trans. Inst. Brit. Geogrs,

No. 25. pp. 29-35.
Although most geomorphologists would feel out of their depth in deep space, the
following articles dealing with the surface features of the Moon and Mars are ex-
amples of sources that will be of general interest to many students of terrestrial land-
forms:
gilbert, g. k. 1893. The Moon’s face — a study of the origin of its features. Bull.
Phil. Soc. Washington. No\. 12 pp. 241-92. ,
mackin. j. h. 1969. Origin of lunar maria. Bull. geol. Soc. Am.. Vol. 80. pp. 735-
48.
MARCUS. A. H. 1968. Martian craters: number density. Science, Vol. 160, pp.
1333-4.
Quaide. w. l. and oberbeck, v. r. 1969. Geology of the Apollo landing sites.

Earth Sci. Rev.. Vol. 5, pp. 255-78.
Rennilson, t. t., dragg. d. l., morris, e. c., SHOEMAKER. E. M., and

turkevich, a. 1966. Lunar surface topography in Surveyor I Mission report —
Part 11, Scientific data and results {Jet Propulsion Laboratory Tech. Rep. 32—1023,
Pasadena), pp. 7-44.
Ronca, l. b. 1966. An introduction to the geology of the moon, Proc. Geol. Ass..
Vol. 77, pp. 101-25.
ronca, l. b. 1969. Recent advances in lunar geology. Proc. Geol. /4s5.. Vol. 80,
pp. 365-78.
430 Inti oductwn to Geomoi phologv
RONCA, L B 1970 An introduction to the geology of Mars, Proc Geol Assoc.
Vol 81. pp 111-28
RONCA L B and green, r r 1970 Statistical geomorphology of the lunar

surface. Bull geol Soc Am , Vol 80, pp 337-52
SHTE^NBERG G s 1965 Morphologv of moon craters and Cirques, compared to

certain volcanic formations of Kamchatka, liiternl Geol Rex Vol 8 (1966) ,
pp 1440-50
Chapter II. Landforms and structure
A Geophysical Considerations
1 Earth movements and mountain building
BEMMELEN. R w VAN 1968 On the Origin and evolution of the earth’s crust. Geo/
Rundschau. Vol 51.pp 657-705
BILLINGS. M p 1960 Diastrophism and mountain building. Bull geol Soc Am.
Vol 7l.pp 363-98
ELDER.! w 1968 Convection the ke\ to dynamical geology, Piog. Vol 56.
pp 1-33
GASKELL. T F 1970 P/n SICS of f/io eort/i (Thames and Hudson. London), 2 6 pp 1
GiLLULN.J 1963 The tectonic evolution of the western United States Quart Jour
geol Soc. Vol 119. pp 133-74
HARLAND. \\ B 1969 The origin of continents and oceans, Gco/ Mag. Vol 106.
pp 1 00—4
HLCZEN B c . THORP. M and EWING. M 1959 The
. floors of the oceans 1 The
North Atlantic. Geo/ Soc Am Special Papa. t>5. 122 pp
HILLS E s 1963 Outlines of structural geology (Methuen. London) 483 pp
KUENEN. p H 1967 Geosvnciinal sedimentation. Geo/ Rundschau. Vol 56, pp 1-

19
LOVERING, J P 1958 The nature of the Mohorovicic discontinuity. Trans Am

geophxs Union. Vol 39, pp 947-55
MESCHERIKOV. j A. 1959 Contemporarv movements in the earth’s crust, /«te/H/
geol 7?ei . Vol l.pp 40-51
MiLNES, A. G 1969 On the orogenic history of the central Alps Join Geol.
Vol 77. pp 108-12
POLDFRVAART, A. 1955 The crust of the earth A svmposium. Geol Soc Am

Spec Paper 62. 767 pp
rucklin. H. 1963. Die Entstehung des Gross-reliefs der Erde, Geog. Zeit., Vol. 51.
pp. 183-238.
SITTER, u. DE. 1954. Gravitational gliding tectonics. Am. Jour. ScL, Vol. 252,
pp. 371-95.
THOMPSON, G. A. and TALWANi, M. 1964, Crustal structure from Pacific Basin to
Central Nevada, yoar. geophys. Res., Vol. 69, pp. 4813-37.
TRUMPY. R. 1960. Paleotectonic evolution of the central and western Alps. Bull,
geol. Soc. Am., Vol. 71, pp. 843—907.
WYLLiE. p. 1963. The Mohorovicic discontinuity and the orogenic cycle, Trans.
Amer. geophys. Union, Vol. 44, pp. 1064—71.
WYLLIE. p. 1965. A modification of the geosyncline and tectogene hypothesis, Geol.

Mag., Vol. 102, pp. 23 1-45.
2. Continental drift
BULLARD, E. c. 1964. Continental drift. Quart. Jour. geol. Soc., Vol. 120, pp. 1-33.
GARLAND, G. c. 1966. Continental drift, Roy. Soc. Canada Spec. Publ., No. 9,
140 pp.
HARLAND, w. B. 1967. Tectonic aspects of continental drift, Sci. Progress, Vol. 55,
pp. 1-14.
HEEZEN. B. c. 1960. The rift in the ocean floor, Sci. American. Vol. 205, pp. 99-
111 .
HURLEY, p. M. 1 968. The confirmation of continental drift. Sci. American, Vol. 218,
pp. 52-64.
M AACK. R. 1969. Kontinental drift und Geologic des sudatlantischen Ozeans (Gruy-
ter, Berlin), 1 64 pp.
RAVEN. T. 1960. Alpine folding as related to continental drift. Eclogae Geol. Helv..
No. l.pp. 161-8.
RUNCORN, s. K. (Ed.). 1962. Continental drift i Academic Press. New York), 338 pp.
SHERLOCK, R. L. 1934. Notes on the Amazon, Geol. Mag.. Vol. 71, pp. 1 12-16.
STONELEY, R. 1966. The Niger delta region in the light of the theory of continental
drift, Geol. Mag., Vol. 103. pp. 385-97.
TO IT, A. L. DU. 1937. Our wandering continents (Oliver and Boyd. Edinburgh).
366 pp. (reprinted 1 957).
WEGENER, A. 1922. Die Entstehung der Kontinente und Ozeane, 3rd ed. (Vieweg,
Braunschweig), 144 pp.; translated by J. G. A. Skerl, 1924. The origin of conti-
nents and oceans (Methuen. London), 212 pp.
3. Vulcanicity
ALSop, L, E. and Oliver, j. e. (Eds).1969. Premonitory phenomena associated with

several recent earthquakes and related problems, Trans. Am. Geophys. Union.
Vol. 50, pp. 376-410.
BULLARD, F. M. 1962. Volcanoes in history, in theory, in eruption (Univ. of Texas

Press. Austin). 441 pp.
BENioFF. H. 1951. Earthquakes and rock creep. Bull. Seismol. Soc. Am., Vol.4i,
pp. 31-62.
BOUT, p. and derruau,

m. 1966. Recherches sur les volcans explosifs du Japon,
Mem. et Doc. Centre Rech. Doc. Cartog. geog. Ed. C.N.R.S., Vol. 10, No. 4,
89 pp.
BULLEN, K. E. 1954. Seismology {Meihum. London), 132 pp.
CO ATS, R. R., HAY, R. L., and ANDERSON, c. A. 1968. Studies in volcanology. Geol.
soc. Am. Mem. 116. 678 pp.

COTTON, c. A. 1952. Volcanoes as landscape forms (Whitcombe and Tombs,
Christchurch). 416 pp.: facsimile reproduction. 1969 (Hafner. New York).
DOLLAR. A. T. J. 1957, The Midlands earthquake of February 11, 1957. Nature.

Vol. 179.pp. 507-10.
EATON, J. p. and MURAT K. J. 1960. How volcanoes grow. Science. Vol. 132.
pp. 925-38.
FiSKE. R. s. and kovanagi. r, v. 1968. The December 1965 eruption of Kilauea
volcano. Hawaii, U.S. Geol. Star. Prof. Paper 607, 2 1 pp.
GUTENBERG. B. and RICHTER, c. F. 1954. Seismicity of the earth (Princeton Univ.

Press, Princeton). 3 1
0 pp.
JONES, J. G. 1968. Intraglacial volcanoes of the Laugarvatn region, south-west
Iceland. Quart. Jour. geol. Soc.. Vol. 1 24. pp. 197-2 1 2.
KARNiK. v. 1969. Seismicitv of the European area (Reidel. Dordrecht). Parti,

364 pp.
MACDONALD, G. A. 1953. Pahochoc, Aa. and block lava. Am. Jour. ScL. Vol. 251,
pp. 169-91.
MCBIRNEY, A. R. 1959. Factors governing emplacement of volcanic necks. Am.

Jour. ScL. Vol. 257, pp. 431-48.
NEUMANN VAN PAD ANG, .M. 1955. Present position regarding the catalogue of the
active volcanoes of the world. Bull, volcanologique 17. 14 1-4.
RANDALL. B. A. o. 1959. Intrusive phenomena of the Whin Sill, east of the River
North Tyne. Geol. Mag.. Vol. 96, pp. 385-92.
RICHTER, c. F. 1958. Elementary seismology (Freeman. San Francisco), 768 pp.
RiTSEMA, A. R. 1969. Seismo-tectonic implications of a review of European earth-

quake mechanisms. Geol. Rundschau. Vol. 59, pp. 36-56.
RITTM AN, A. 1 960. Vulkatte und ihre Tatigkeit (Enke, Stuttgart). 336 pp.; trans. by
E. A. Vincent, 1962. Volcanoes and their activity (Interscience, New York).
305 pp.
SMITH. R. L. 1960. Ash flows. Bull. geol. Soc. Am.. Vol. 71, pp. 795-842.
THOR ARLINSSON. s. 1968. Some problems of volcanism in Iceland. Geol.

Rundschau. Vol. 57. pp. 1-20.
YAGi, K.. KAWANO, Y., and AOKi. K, 1963. Types of quaternary activity in north-
eastern Japan. Bull, volcanologique, Vol. 26, pp. 223-35.

WENTWORTH, c. K. 1927. Estimates of marine and fluvial erosion in Hawaii. Jour.
GeoL Vol. 35, pp. 1 1 7-33.
WENTWORTH, c. K. 1954. The physical behaviour of basaltic lava flow. Jour. GeoL.
Vol. 62, pp. 425-38.
WILLIAMS, H. 1 954. Problems and progress in volcanology. Quart. Jour. geol. Soc..
Vol. 109.PP. 31 1-32.
4. Hydrothermal activity and meteorite craters
CASSIDY, w. A. et al. 1965. Meteorites and craters of Campo del Cielo. Argentina,
Science. Wo\. 149, pp. 1055-64.
ESCHER. B. G. 1955. Three caldera-shaped accidents: volcanic calderas, meteoric
scars and lunar cirques. Bull, volcanologiqiie, Vol. 16, pp. 55-70.
FLEISCHER, R. L., PRICE, p. B., and WOODS, R. T. 1969. A second tektite fall in
Australia, Earth and Planetary Sci. Letters. Vol. 7, pp. 5 1-2.
HELLYER. B. 1969. Statistics of meteorite falls. Earth and Planetary Sci. Letters.
Vol. 7.pp. 148-50,
KRiNOV. E. L. 1961. The Kaalijarv meteorite craters on Saarema island, Estonian

SSR, Am. Jour. Sci.. Vol. 259, pp. 430—40.
SPENCER. L. j. 1933. Meteorite craters as topographical features on the earth’s

surface. Geog. Jour.. Vol. 8 1 , pp. 227—48.
RINEHART, J. s. 1969. Old Faithful geyser performance 1870 through 1966. Bull,
volcanologiqiie. Vol. 33, pp. 153-63.
WHITE. D, E. 1957. Thermal waters of volcanic origin. Bull. geol. Soc. Am.. Vol. 68.
pp. 1637-58.
WHITE, D. E. 1967. Some principles of geyser activity, mainly from Steamboat
Springs, Nevada. Am. Jour. Sci.. Vol. 265, pp. 641-84.
B. Geological Considerations
AUDEN. J. B. 1954. Drainage and fracture patterns in north-west Scotland, Geol.

Mag.. Vol. 91. pp. 337-51.
braddock, w. a. and eicher. d. l. 1962. Block-glide landslides in the Dakota

group of the Front Range foothills. Colorado, Bull. geol. Soc. Am.. Vol. 73,
pp. 3 7-24.
1
BRADLEY, w. c. 1963. Large scale exfoliation in massive sandstones of the Colorado

plateau. Bull. geol. Soc. Am.. Vol. 74, pp. 5 1 9-28.
CAINE, N. 1967. The tors of Ben Lomond, Tasmania. Zeit.f. Geomorph.. Vol. 1 1.
pp. 4 1 8-29.
CAMER.MAN. c. 1953. La gelivite des materiaux pierreux. Bull. Soc. Beige Geol..
Vol. 62. pp. 1 7-34.
CHAPMAN, c. A. 1958. Control of jointing by topography. Jour. Geol., Vol. 66, pp.
552-8.
. . .
434 Intiodiiction to Geomorphology

CORBEL, J 1957 Les Karsts du nord ouest de I’Europe. Inst Etudes Rhodaniennes,
Mem et Doc. Lyon, 12, 54! pp
EPSTEIN, j B 1967 Structural control of wind gaps and water gaps and of stream
capture in the Stroudsburg Area. Pennsylvania and New Jersey, U S Ceol Siirv
Prof Papei 5505. pp B80-B86
FLINT. R F 1963 Altitude, lithology and the Fall Zone in Connecticut. yoKi Geol
Vol 71, pp 683-97
GILBERT, G R 1895 Lake basins created by wind erosion. Jour Geol. Vol 3,
pp 47-9
HANCOCK. P L 1968 Joints and faults the morphological aspects of their origins
Proc Geol ^ss.Vol 79. pp 141-51
HARRISON. K and THACKFRAV, A D 1940 On the direction of certain valleys
Geol Mag.Vo\ 77 pp 82-8
HILL, p A. 1966 Joints their initiation and propagation with respect to bedding.
Geol Mag.VoX 103, pp 276-9
HODGSON. R A 1961 Regional study of jointing m Comb Ridge-Navaio Mountain
area, Arizona and Utah, Bull Amer y4ss Pettol Geols Vol 45, pp 1-38
HODGSON. R A 1965 Genetic and geometric relations between structures in base
ment and o\erlying sedimentary rocks, with examples from Colorado Plateau and
Wyoming, 5ii// Amei Ass Petrol Geols. Vol 49, pp 935-49
HOLLiNGWORTH. s E. TAILOR. J H and K E L L A w At G A 1944 Large
, . Scale
superficial structures in the Northamptonshire ironstone field. Quait Join geol
Soc. Vol 100. pp 1-44
K ALTERHERBERG. J VON and KUHN VELTEN. H 1967 Klufte und
Talrichtungen Turon des sudostlichen Munsterlandes Geol Rundschau.
im
Vol 56. pp 726-48
LiEBLiNG. R s and KERR, p F 1965 Observations on quickclay. 5ii// geo/ Soc

^m.Vol 76. pp 853-78
MARR. J E 1906 The influence of the geological Structure of English Lakeland upon
Its present features - a study in physiography. Quail Join geol Soc. Vol 62,
pp IXVI-CXXMII
MiLLE, G DE. SHOULDICE. J R . and NELSON, H w 1964 Collapse Structures
related to etaporites of the Prairie formation. Saskatchewan, Bull geol Soc Am .
Vol 75. pp 307-16
NicKELSEN. R p and HOUGH. V N D 1967 Jointing in the Appalachian plateau of

Pennsylvania. Bull geol Soc Am Vol 78. pp 609-30
NICHOLSON. R 1963 A note on the relation of rock fracture and fiord direction
Geogtafiska Annaler. Vol 45. pp 303-4
o'brien. c A E 1957 Salt diapirism in south Persia. (7eo/ogie Afi/ntoi/u’ (The
Hague) Vol 19. pp 357-76
PLAFKER. G 1964 Oriented lakes and lineaments in northeastern Bolivia. Bull
geol Soc Am . Vol 75. pp 503-22
PRICE. N J 1959 Mechanics of jointing in rocks. Geo/ A/ng.Vol 96. pp 149-67

RANDALL. B. A. o. the relationship of valley and fiord directions to

1961. On
fracture pattern of Lyngen. Troms. northern Norway. Geografiska Annaler,
Vol. 43. pp. 336-8.
ROBERTSON. E. c. 1955. Experimental study of the strength of rocks. Bull. geol.

Soc. Am.. Vol. 66. pp. 1275-3 14.
scHMtTTHENNER. H. 1954. Die Regeln der morphologischen Gestaltung im

Schichtstufenland. Pet. geog. Mitt., Vol. 98. pp. 3—10.
SELWOOD. E. B. and COE. K. 1963. Large-scale terminal curvature affecting the
cliffs west of Castletown Berehaven, west Cork, Proc. Geol. Ass., Vol. 74, pp.
461-5.
SWEETING, M. M. 1955. The landforms of north-west Co. Clare, Ireland, Trans.
Inst. Brit. Geogrs, No. 2 1, pp. 33—49.
WILSON, 1952. The influence of rock structures on coastline and

G. cliff de-
velopment around Tintagel, north Cornwall, Proc. Geol. Ass.. Vol. 63, pp.
20-48.
WISE, D. u. 1964. Microjointing in basement. Middle Rocky Mountains of Montana
and Wyoming, Bull. geol. Soc. Am.. Vol. 75. pp. 287-306.
WRIGHT, M. D. 1964. Cementation and compaction of the Millstone Grit of the

central Pennines. England. Jour. Sed. Petrol. Vol. 34. pp. 756-60.
ziscHiNSKY. u. 1969. Ober Bergzerreissung und Taizusschub. Geol. Rundschau.

Vol. 58. pp. 974-83.
Chapter III
A and B. Selected References on Climate, Weather, and Hydrology
BARRY, R. G. and CHORLEY. R. j. 1968. Atmosphere, weather and climate. 2nd ed.,
1 97 1 (Methuen, London), 3 1
9 pp.
BECKINSALE, R. p. 1957. The nature of tropical rainfall. Tropical Agriculture.
Vol. 34. pp. 76-98.
BRUCE. J. p. and Clark, r. h. 1966. Introduction to hydrometeorology {Pergamon.

Oxford), 3 1
9 pp.
FOURNIER, F. 1949. Les facteurs climaiiques de I’erosion du sol. Bull. Ass. geog.
Fr.. No. 202-3, pp. 97-103.
GEIGER. Das Klima der bodennehen Luftschict (Vieweg. Brunswick).
R. 1967.
646 by Scripta Technica Inc.. 1965. The climate near the ground
pp.; trans.
(Harvard Univ. Press. Cambridge. Mass.). 611 pp.
HUTCHINSON. G. E. 1957. A treatise on limnology (Wiley, New York). Vol. 1.
1015 pp.
JOHNSTON, G. H., and BROWN, R. j. E. 1961. Effect of a lake on distribution of

permafrost in the Mackenzie River delta. Nature. Vol. 92, pp. 25 1-2.
LANGBEIN, w. B. and SCHUMM, s. A. 1958. Yield of sediment in relation to mean

annual precipitation. Trans. Am. geophys. Union. Vol. 39. pp. 1076-84.
LINTON. D. L. 1959. River flow in Great Britain 1955-56, Nature. Vol. 183,
pp. 714-16.
ME INZER, o. E., et at. 1942. W>’rfro/ogi’ (McGraw-Hill, New York), 712 pp.
MUNN, R. E. 1966. Descriptive microineteorologv (Academic Press, New York),

245 pp.
SUTTON, o. G. 1 960. Understanding weather (Penguin, Harmondsworth), 21 5 pp.
WILLIAMS, p. j. 1961. Climatic factors controlling the distribution of certain frozen

ground phenomena. Geografiska Annaler. Vol. 43, pp. 339-47.
WARD, R. c, 1967. Principles of Avrfro/og.v (McGraw-Hill, London). 402 pp.
B. Amounts and Motions of Water and Ice
BASCOM. w. N. 1959. Ocean waves. Sci. American. Vol. 201, pp. 74-84.
BAS COM, w. N. 1964. fVaves and beaches. The dynamics of the ocean surface
(Doubleday. New York). 267 pp.
DARBYSHIRE, J. 1955. An investigation of storm waves in the north Atlantic

Ocean. Proc. Roy. Soc. A. Vol. 230, pp. 560-9.
DEFANT. A, 1953. Ebbe und Flut des Meeres dcr Atmosphdre und der Erdfeste
(Springer-Verlag, Berlin); trans. by A. J. Pomerans, \95&. Ebb and Flow (Univ. of
Michigan. Ann Arbor), 121 pp.
DR APER, L. and DOBSON, p. J. 1965. Rip currents on a Cornish beach. Nature. Vol.
206, p. 1249.
HOYT. w. G. and langbein. w. b. 1955. f/oods (Princeton Univ. Press, Princeton).
469 pp.
KAMB, B. 1964. Glacier geophysics. Science. Vol. 146. pp. 353-65.
KAMB. B. and LA CHAPELLE, E. 1964. Direct observation of the mechanism of

glacier sliding over bedrock. Jour. GlacioL Vol. 5. pp. 159-72.
KiDSON. c. 1953. The Exmoor storm and the Lynmouth floods. Geography. Vol. 38,
pp. 1-9.
LLiBOUTRY, L. 1965. How glaciers move. New Scientist. Vol. 28. pp. 734-6.
LLiBOUTRY. L. 1968. General theory of subglacial cavitation and sliding of temper-

ate glaciers. Jour. GlacioL. Voi. 7. pp. 21-58.
MATHEWS, w. H. 1964. Water pressure under a glacier. Jour. GlacioL. Vol. 5.
pp. 235-40.
MEIER, .M. F. and POST. A. 1969. What are glacier surges?, Canad. Jour. Earth ScL.
Vol. 6. pp. 807-17.
MELLOR. M. 1959. Ice-flow in Antarctica, dowr. GlacioL. Vol. 3, pp. 377-84.
MORIS AWA, M. 1968. Streams, their dynamics and morp/io/ogp (McGraw-Hill, New
York). 1 75 pp.
NYE, J. F. 1952. The mechanics of glacier flow,yoim. GlacioL Vol. 2, pp. 81-93.
OSTENSO, N. A.. SELLMANN, p. V., and PEVVE, T. L. 1965. The bottom topography
of Gulkana glacier, Alaska Range. Alaska. Jour. GlacioL, Vol. 5, pp. 65 1-60.
PATERS ON, vv. s. p. 1969. The physics of glaciers (Pergamon, Oxford), 250 pp.
SH ARP, R. P. 1954. Glacier flow; a review. Bull. geol. Soc. Am., Vol. 65. pp. 82 1—38.
STENBORG. T. 1969. Studies of the internal drainage of glaciers. Geografiska

Annaler, Vol, 51 A, pp. 1 3-4 .
TRiCKER, R. A. R. 1964. Borcs, breakers, waves and wakes (Mills and Boon,
London), 250 pp.
WEERTMAN, J. 1957. On the sliding of glaciers. Jour, GlacioL. Vol. 3. pp. 33-8.
WEERTM AN. J, 1964. The theory of glacier sliding. Jour. GlacioL, Vol. 5, pp. 287-
303.
C. Mechanical and Frictional Forces
B AGNOLD, R. A. 1941. The physics of blown sand and desert dunes, reprinted. 1 960
(Methuen. London), 265 pp.
CHORLEY, R. J. 1959. The geomorphic significance of some Oxford soils. Am. Jour.
ScL. Vol. 257. pp. 503-15.
CULLING, VV. E. H. 1963. Soil creep and the development of hillside slopes. Jour.
GeoL, Vol. 7 1
. pp. 1 27-6 1
KING. L. c. 1957. The uniformitarian nature of hillslopes. Trans. Edin. geol. Soc.,
Vol. 1
7, pp. 81-102,
KIRK BY. M. J. 1967. Measurement and theory of soil creep. Jour. Geol., Vol. 75,
pp. 359-78.
SCHEiDEGGER, A. E. 1961. Theoretical geomorpliologv {Springer, Berlin), 333 pp.
SCHEiDEGGER, A. E. 1964. Some implications of statistical mechanics in geo-

morphology, Bull. Intend, ,45s. Sci. Hydro!., Vol. 9, pp. 12-16.
STR ABLER. A. N. 1952. Dynamic basis of geomorphology. Bull. geol. Soc. Am., Vol.
63. pp. 923-38.
D. Sediments and Mechanical Characteristics of Soils
ANDEL. T. H. VAN. 1959. Reflections on the interpretation of heavy minerals. Jour.

Sed. Petrol., Vol. 29. pp. 153-63.
ANDERSON, H. VV. 1951. Physical characteristics of soils related to erosion, yoKc.

Soil and Water Conservation, Vol. 6, pp. 129-33.
avenard. j. m. 1962. La solifluxion ou quelques methodes de mecanique des sols
appliquees au probleme geomorphologique des versants (S.E.D.E.S.. Paris).
164 pp.
BARKER. J. p. 1957. Quelques aspects du probleme des sediments correlatifs en

climat tropical humid. Zeit.f. Geomorph., Vol. I. pp. 1-43.
baver, l. d. 1948. Soil physics {Wiley, New York). 398 pp.

BE AUJEU-GARNiER. J. 1955. Sur la presence de formation de type dit

'periglaciaire' en Algerie orientale, C. r. Acad. ScL. Vol. 240, pp. 1246-8.
BERTHOIS, L. 1 950. Methode d’etude des galets. Application a I’etude de revolution
des galets marins actuels. Rev. geomorph. dyn., Vol. 1, pp. 199-225.
BERTHOIS, L. 1959. Techniques d'analyses graniilometriques (Centre Doc. Univ.,
Paris), 64 pp.
BLACK, R. F. 1 954. Permafrost: a review. Bull. geol. Soc. Ant.. Vol. 65, pp. 839-56.
BR E WEK. R. 1964. Fabric and miiteral analysis of soils (Wiley. New York). 470 pp.
BR Y AN. R. B. 1968. The development, use and efficiency of indices of soil erodibility.
Ceodcnna. Vol. 2, pp. 5-26.
CAiLLEL'X. A. and r At LOR. G. 1954. Cryopedologie: elude des sols geles

Hermann,
( Paris), 2 1
8 pp.
CAILLEL'X. A. and tricart. j. 1959. Iniiiaiion d i’etude des sables el des galets
(Centre Doc. Univ., Paris). Vol. 1. 376 pp.: Vol. 11. 194 pp.: Vol. 111. 202 pp.
t APPFK. L. and CASSit. u. F. 1969. The mechanics o! engineering soils. 5th ed.
(Spoil. London). 309 pp.
CLARK. M. L. LI WIN. L. and SMAiL. R. J. 1967. The sarsen stones of the

Marlborough Downs and their geomorphological implications. Unit'. Soulhantp-
ton. Research Series in Oeug.. No. 4. pp. 3-40.
CZEPPL. z, 1959. Remarks on frost hease. Czasupismo Geogr.. Vol. 30. pp. 195-
202 .
DorcLAS, o. 1 . 1968. Grain size indices, classification and environment.

Sedimenioiogv. Vol. 10, pp. 83-100.
Dt LiK. I. 1963. Nouveaus problcmcs du pergelisol pleistocene. Acta Geogr. Lodz.

No. 1 7. 93 pp,
E ASTERHROOK. D. J, 1964. Void ratios and bulk densities as means of identifying
Pleistocene tills. Bull. geol. Soc. Am.. Vol. 75. pp. 745-50.
FOLK. R. L. 1962. Of skewnesses and sands, ./one. Sed. Petrol. Vol. 32. pp. 145-6.
FOLK. K. L. and WARD. w. c. 1957. Brazos river bar: a study of the significance of
grain size parameters. Jour. Sed. Petrol.. Vol. 27. pp. 3-26.
FOLK. R. L. and weaver, c. e. 1952. A study of the te.xture and composition of

chert. Am. Jour. Sci.. Vol, 250, pp. 498-5 10.
FRIEDMAN. G. M. 1961. Distinction between dune, beach and river sands from their
textural characteristics. Jour. Sed. Petrol. Vol. 3 I, pp. 5 14-29.
FR lEDM AN, G. M. 1 967. Dynamic processes and statistical parameters compared for
size frequency distribution of beach and river sands. Jour. Sed. Petrol. Vol. 37.
pp. 327-54.
GRIFFITHS. J. c. 1967. Scientific method in analysis of sediments (McGraw-Hill,
New York). 508 pp.
HAILS. J. R. 1964. A reappraisal of the nature and occurrence of heavy minerals

along parts of the east Australian coast, Aust. Jour. Scl. Vol. 27, pp. 22-3.
HAILS. .1. R. 1967. Significance of statistical parameters for distinguishing sedimen-

. .
tary environments in New South Wales, Australia. Join Seel Petiol. Vol 37.
pp 1059-69
HERDAN G 1960 Small pai tide sialtsiics, 2nd Tsv ed (Butterworths. London).
418 pp
INMAN. D 1952 Measures for descnbina the
L size distribution of sediments, yoid
Seel Petiol. Vol 22. pp 125-45
IRANI R R andcLAVTON F c 1963 Pai tide size ineasiiiemeiU, inteipt elation
and applicalion (Wiley, New York) 165 pp
JOHANSSON, c E 1963 Orientation of pebbles in running water A laboratory
study. Geog! afiska Annalei. Vol 45, pp 85-112
jUNGERius, p D 1965 Some aspects of the geomorphological significance of soil

textures, in Eastern Nigeria, Zeitschrifl fur Geomot phologte, Vol 9, pp 332-45
KiTTLEMAN. L R 1964 Application of Rosin’s distribution in size frequency
analysis of clastic rocks, doin See! Peiiol. Vol 34. pp 483-502
ROSTER E 1962 Moglichkeiten und Grenzen granulometrischer und morpho
metrischer Untersuchungsmethoden m der geographischen und ueologischen
Forschung, ft'/ geog Mitt Vol 106. pp 111-15
KRiNSLEY, D H and DONAHUE, J 1968 Environmental interpretation of sand

gram surface textures bv electron microscopy. Bull iieol Soc Am Vol 79 pp
743-8
KRiNSLES.D H andooNAHUE. J 1968 Pebble surface textures. Geo/ Mag Vol
105.PP 521-5
KRiNSLES.D H andPUNNELL. B M 1965 En\ ironmcntal history of Sand arains
from the Losver and Middle Pleistocene of Norfolk, England. Quail Join geol
Soc Vol 121. pp 435-61
RRUMBEiN. w c 1938 Size frequency distribution of sediments and tlic normal /j/i/
curve Join Sed Peiiol. Vol 8 pp 84-90

LOVEDAV.i 1962 Plateau deposits of the southern Chiltern Hills. P/oc Geol Ass
Vol 73. pp 83-102
Mcc AMMON R B 1962 Efficiencies of percentile measurements for dcscribins; the
mean size and sorting of sedimentary particles. Jour Geol Vol. 70, pp 453-65
MCCANN, s B 1962 Some supposed ‘raised beach’ deposits at Corran, Loch
Linnhc and Loch Etive, Coo/ Mag. Vol 99. pp 131-42
MELTON M A 1965 The ceomorphic and paleochmatic significance of alluvial
deposits in southern Arizona. OoKi Geol Vol 73. pp 1-38
MOSS, A J 1962 The physical nature of common sandy and pebbly deposits Part I
Am Join Set. Vol 260. pp 337-73

MOSS A J 1963 The physical nature of common sandv and pebbly deposits Part 11.
Am Join Sci. Vol 261.pp 297-343
PA AS. w 1962 Rezente und fossile Boden aul niederrheinischen Terrassen und
deren Deckschichten. £/sctT/o//e; imd Cegeiiwaii. Vol 12.
pp 165-230
penw.l r and c ATT. J a, 1967 Stone orientation and other structural features of
tills in East Yorkshire. Geo/ Mag. Vol 104, pp 344-60
PETTIJOHN, F, J. 1957. rocArs (Harper. New York). 718pp.

PFEIFFER. H. 1965. Schwermineralseifen an der siidlichen Ostsee-Kiiste. Baltica.
Vol. 2. pp. 205-13.
RAGG, J. M. and BiBBY, J. s. 1966. Frost weathering and solifluction deposits in

southern Scotland, Geografiska Aimaler, Vol. 48A, pp. 12-23.
RICHTER. K. 1932. Die Bewegungsrichtung des Inlandeises. rekonstruiert aus den

Kritzen und Langsachsen der Geschiebe. Zeii. f. Ceschiebejbrschung, Vol. 8.
pp. 62-6.
SANDFORD. K. s. 1965. Notes on the gravels of the Upper Thames floodplain

between Lechlade and Dorchester. Proc. Geol. Ass., Vol. 76. pp. 61-75.
SHEPARD. F. p. and YOUNG. R. 1961. Distinguishing between beach and dune

sands. 7oRr. Sed. Petrol.. Vol. 31. pp. 196-214.
SKEMPTON. A. w. 1953. Soil mechanics in relation to geology. Proc. Yorks, geol.

S'oc.. Vol. 29. pp. 33-62.
THAMES. J. L. and URSic. s. J. 1960. Runoff as a function of moisture storage

capacity. Jour, geophys. Res.. Vol. 2. pp. 65 1-4.
THE B AULT. J. I 968. Contribution a I'etude des galets. Rev. geomorph. dyn.. Vol. 18.
pp. 49-72.
TERZAGHI. K. 1 965. Theoretical soil mechanics (Wiley. New York). 5 10 pp,
viSHER, G. s. 1969. Grain size distributions and depositional processes. Jour. Sed.
Petrol.. Vol. 39. pp. 1074-106.
viSTELius. A. B. 1958, Struktumye diagramniy (Izd. Akad. Nauk S.S.S.R..
Moscow). 157 pp.; trans, by R. Baker. 1966. Structural diagrams (Pergamon.
Oxford). 1 78 pp.
WARD. w. H. 1945. The stability of natural slopes. Geog. Jour.. Vol. 105. pp. 170-
97.
WAUGH. B. 1965. A preliminary electron microscope study of the development of

authigenic silica in the Penrith sandstone. Proc. Yorks, geol. Soc.. Vol. 35. pp. 59-
69.
WEBSTER, R. 1966. The drifts in the Vale of White Horse of north Berkshire and
Wiltshire, Proc. Geol. Ass.. Vol. 77, pp. 255-62.
WENTWORTH, c. K. 1922. A field study of the shapes of river pebbles. U.S. Geol.
Sure. Bull. 730. pp. 103-15.
E. Geochemical Considerations
BUTLER, J. R. 1957. The geochemistry and mineralogy of rock weathering. II. the
Nordmarka area, Geochemica, Cosmochemica Acta, Vol. 6, pp. 268-8 1.
CORNWALL. I. w. 1958. Soils for the archaeologist (Dent. London). 230 pp.
FiELDES. M. and swiNDALE, L. Chemical weathering of

D. 1954. silicates in soil
formation. N.Z. Jour. Sci. Tech. B. Vol. 36, pp. 140-54.
FREDERICKSON, A. F. 1951. Mechanism of weathering. Bull. geol. Soc. Am.. Vol.

62. pp. 221-32.
1
GRIM. R. E. 1968. Clay mineralogy. 2nd ed. (McGraw-Hill, New York). 596 pp.
HARRiss. R. c. and ADAMS, j. A. s. 1966. Geochemical and mineralogical studies

on the weathering of granitic rocks. Am. Jour. ScL Vol. 264. pp. 146-73.
JAEGER. J. L. 1957. La geochimie (Presses Univ. France. Paris). 1 19 pp.
KELLER, w. D. 1957. The principles of chemical weathering (Lucas, Columbia).

1 1
1 pp.
KELLER, vv. D. and REESMANN. A, L. 1963. Dissolved products of artificially
pulverised silicate minerals and rocks: Part II, Jour. Sed. Petrol.. Vol. 33,
pp. 426-37.
LOUGHNAN. F. c. 1962. Some considerations in the weathering of the silicate
minerals. Jour. Sed. Petrol.. Vol. 32, pp. 284-90.
MASON. B. H. 1966. Principles of geochemistry. 3rd ed. (Wiley. New York). 329 pp.
PETTiJOHN. F. J. 1941. Persistence of heavy minerals and geologic age. Jour. Geol..
Vol. 49. pp. 6 10-25.
RONDEU A, A. 1 958. Geomorphologie et geochimie, C. r. Soc. geol. Fr.. pp. 288-90.
siEVER, R. 1962. Silica solubility 0-200 C., and the diagenesis of siliceous sedi-
ments, v/o«r. Geol.. Vol. 70. pp. 127-50.
SMITH, w. w. 1962. Weathering of some Scottish basic igneous rocks with reference
to soil formation, Jonr. Soil Sci.. Vol. 13, pp. 202-15.
TODD. T. w. 1968. Paleoclimatology and the relative stability of felspar minerals

under atmospheric conditions. Jour. Sed. Petrol.. Vol. 38. pp. 832-44.
WOLFF, R. G. 1967. Weathering of Woodstock granite, near Baltimore. Maryland.

Am. Jour. Sci.. Vol. 265, pp. 106-17.
F. Biological Activity
ARANDA. J. M.. and COUTTS. J. R. H. 1963. Micrometeorologica! observations in an

afforested area in Aberdeenshire: rainfall characteristics. Jour. Soil Sci.. Vol. 14,
pp. 124-33.
ATTiWEL, p. M. 1968. Loss of elements from decomposing litter. Ecology. Vol. 49.
pp. 142-5.
BAY, c. E., wuNNECHE, G. w.. and HAYS, o. E. 1952. Frost penetration into soils as
influenced by depth of snow, vegetation cover, and air temperatures, Trans. Am.
geophys. Union. Vol. 33, pp. 541-6.
BOULAiNE, J. 1961. Sur le role de la vegetation dans le formation des carapaces

calcaires mediterraneenes. C. r. Acad. ScL. Vol. 253, pp. 2568-70.
BOYER, p. 1959. De I’influence des termites de la zone intertropicale sur la con-
figuration de certains sols. Rev. geomorph. dyn., Vol. 10, pp. 41-4.
BUTUZOVA, o. V. 1 962. Role of the root system of trees in the formation of micro-
relief, Soviet Soil ScL, Vol. 4, pp. 364—72.
c AROzzi, A. V. 1967. Recent calcite-cemented sandstone generated by
the equatorial
tree Iroko (Chlorophora e.xcelsa). Daloa, Ivory Coast. Jour. Sed. Petrol.. Vol.
37,
pp. 597-600.
c HEM IN. E. 1921. Action corrosives des racines sur le marbre. C. r. Acad. ScL
Vol. 173.pp. 1014-16.
CUNNINGHAM. R. K. 1963. The effect of clearing a tropical forest soil. Jour. Soil
ScL Vol. 14. pp. 334-45.
DAHLSKOG. s. 1966. Sedimentation and vegetation in a Lapland mountain delta.
Ceografiska Annaler, Vol. 48A. pp. 86-101.
ERHART, E. H. 1956. La gcuise dcs sols eit tant qtie phenomene geologique. Esqidsse
d'une theorie geologique et geochimique. Biostasie et rhexistasie (Masson, Paris),

90 pp.
ERHART. E. H. 195 1. Sur I’importance des phenomenes biologiques dans la forma-

tion des cuirasses ferrugineuses en zone tropicale, C. r. Acad. ScL Vol. 233,
pp. 804-6.
EVANS. J. w. 1968. The role of Penitella penita (.Conrad 1837)(Family Pholadidae)

as eroders along the Pacific coast of North America, Ecology. Vol. 49. pp. 156-9.
E V E R ITT. B. L. 1 968. Use of the cottonwood in an investigation of the Recent history

of a flood plain. Ant. Jour. Sci.. Vol. 266, pp. 4 1 7-39.
HACK. J. T. and GOODLETT. J. c. I960. Geomorphology and forest ecology of a

mountain region in the Central Appalachians, U.S. Ceol. Siirv. Prof. Paper 347.
66 pp.
HEALY. T. R. 1968. Bioerosion on shore platforms in the Waitemata formation.
Auckland. Earth Sci. Jour.. Vol. 2. pp. 26-37.
HOPKINS, B. 1960. Rainfall interception by a tropical forest in Uganda, East Afr.

Agric. Jour.. Vol. 25. pp. 255-8.
JENNY. H.. GESSEL. s. P., and BINGHAM, F. T. 1949. Comparative study of

decomposition rates of organic matter in temperate and tropical regions. Soil Sci..
Vol. 68. pp. 419-32.
JONES. R. J. 1965. Aspects of the biological weathering of limestone pavements,

Proc. Geol. Ass.. Vol, 76, pp. 42 1—33.
KELLER, w. D. and FREDERiCKSON, A. 1952. Role of plants and colloidal acids in
the mechanism of weathering. Am. Jour. Sci.. Vol. 250. pp. 594-609.
KiDSON. c. 1959. The uses and limitations of vegetation in shore stabilization.
Geography. Vol. 44. pp. 241-50.
KRUM BEiN, w. E. 1969. Uber den Einfluss der Mikroflora auf die exogene Dynamik
(Verwitterung und Krustenbildung). Geol. Rundschau. Vol. 58. pp. 333-63.
LOVERING, T. s. 1959. Geological significance of accumulator plants in rock

weathering. Bull. geol. Soc. Am., Vol. 70. pp. 781-800.
LULL, H. w. and pierce, r. s. 1959. Frost and forest soil. Intend. Ass. Sci. HydroL
No. 48, pp. 40-8.
LUTZ. H. J. I960. Movement of rocks by uprooting of forest trees. Am. Jour. ScL
Vol. 258. pp. 752-6.
LUX, H. 1964. Die bioiogischen Grundlagen der Strandhaferpflanzung und

Silbergrasansaat im Diinenbau. (Ihre Untersuchung fiber die Moglichkeiten biolo-
gischer Diinenbau - und Diinenbefestigungs — massnahmen inden Sylter Diinenge-

bieten bei und fehlender
vorhaudener Sandablagerung). Angewandte
Pflanzensoziologie, Vol. 20, pp. 5-53.
lynts, g. w. 1966. Relationship of sediment-size distribution to ecologic factors in

Buttonwood Sound, Florida Bay, Jour. Sed. Petrol, Vol. 36. pp. 66-74.
MCLEAN. R. F. 1967. Measurements of beachrock erosion by some tropical marine
gastropods. Bull. Marine Sci., Vol. 17, pp. 55 1-61.
MiLLM AN, A. p. 1957. Biogeochemical investigations in areas of copper-tin mineral-
ization in south-west England, Geochim. Cosmochim. Acta, Vol. 12, pp. 85-93.
MORTENSEN, H. 1964. Eine einfache Methode der Messung der Hangabtragung

unter Wald und einige bisher damit gewonnenen Ergebnisse. Zeit.f. Geomorph.,
Vol. 8. pp. 2 1 2-22.
MORTLAND. M. M., LAWTON, K., and UEHARA, G. 1956. Alteration of biotite to

vermiculite by plant growth. Soil Sci., Vol. 82. pp. 477-81.
NESTEROFF, w. D. and MELiERES. F. 1967. L’erosion littorale du pays de Caux.
Bull Soc. geol. Fr., Ser. 7, Vol. 9. pp. 159-69.
NEUMANN, A. c. 1966. Observations on coastal erosion in Bermuda and measure-

ments of the boring sponge Cliona lamp, Limnol. Oceanog., Vol. 1 1. pp. 92-108.
piTTY, A. F. 1966. An approach to the study of karst water, illustrated by results
from Poole’s Cavern. Buxton, Univ. Hull Occ. Papers in Geog., No. 5. 70 pp.
RAHN. p. H. 1969. The relationship between forested slopes and angles of repose for
sand and gravel. Bull. geol. Soc. Am., Vol. 80. pp. 2 1 23-8.
RANSON. G. 1959. Erosion biologique des calcaires cotiers et autres calcaires d’ori-
gine animale. C. r. Acad. Sci.. Vol. 249. pp. 438-40.
RAUP. H. M. 1951. Vegetation and cryoplanation. Ohio Jour. Sci., Vol. 51. pp.
105-16.
RHOADS. D. c. 1963. Rates of sediment reworking by Yoldia limatula in Buzzards

Bay. Massachusetts, and Long Island Sound. Jour. Sed. Petrol. Vol. 33. pp. 723-7.
siEVER. R. 1962. Silica solubility. 2000°C, and the diagenesis of siliceous sedi-
ments. Jour. Geol. Vol. 70, pp. 127-50.
VANN. j. H. 1959. Landform-vegetation relationships in the Atrato delta, Ann. Ass.

Amer. Geogrs. Vol. 49. pp. 345-60.
WEBLEY. D. M.. HENDERSON, M. E. K., and TAYLOR, 1. F. 1963. The microbiology
of rocks and weathered stones. Jour. Soil Sci.. Vol. 14. pp. 102-12.
G. Human Activity
DEMEK. J. 1969. Beschleunigung der geomorphologischen Prozesse durch die

Wirkung des Menschen. Geol. Rundschau. Vol. 58, pp. 1 1 1—21.
DOUGLAS. I. 1967. Man, vegetation, and the sediment yield of rivers. Nature
Vol. 215, pp. 925-8.
JENNINGS, J. N. 1965. Man as a geological agent, Aust. Jour. Sci. Vol 28 do
• HE-
150-6. >
KiDSON, c. 196 1. The Norfolk Broads, Afa/ure. Vol. 192, pp. 314-15.
LAMBERT, J. M.,JENNINGS, J. N., SMITH. C. T., GREEN, C., and HUTCHINSON,

J. N. 1960. The making of the Broads, Roy. Geog. Soc. Research Mem., No. 3,
153 pp.
MEADE, R. H. 1969. Errors in using modern stream-load data to estimate natural
rates of denudation. Bull. geol. Soc. Am., Vol. 80. pp. 1265-74.
NELSON, J. G. 1966. Man and geomorphic process in the Chemung River valley.
New York and Pennsylvania, Ann. Ass. Amer. Geogrs, Vol. 56, pp. 24-32.
SCHULTZE, J. H. 1951-2. Ober das Verhaltnis zwischen Denudation und Boden
Erosion, Die Erde, Vol. 3, pp. 220-33.
THOMAS, vv. L. (Ed.). 1967. Man’s role in changing the face of the earth, 7th impr.
(Univ. Chicago Press, Chicago), 1 193 pp,
TiCHY, F. 1960. Die vom Menschen gestaltete Erde, Die Erde, Vol. 91, pp. 220-33.
WALLWORK, K. L. 1956. Subsidence in the mid-Cheshire industrial areas, Geog.
Jour., Vol. 1 22, pp. 40-53.
WILKINSON, H. R, 1963. Man and the natural environment, Univ. Hull Occ. Papers
in Geog., No. 1, 35 pp.
Chapter IV. Inter-relationships between processes and

landforms
Some general references
AXELSSON. V. 1967. The Laitaure delta. A study of deltaic morphology and pro-
cesses. Geografiska Annaler. Vol. 49A. pp. 1-127.
BALCHiN, w. G. v. and p YE, N, 1 955. Piedmont profiles in the arid cycle. Proc. Geol.
Ass.. Vol. 66, pp. 167-81.
BRESSAU, s. 1957. Abrasion, Transport und Sedimentation in der Beltsee, Die

Kiiste. Vol. 6, pp. 64-102.
BUTZER, K. w. 1965. Desert landforms at the Kurkur Oasis, Egypt, Ann. Ass. Amer.
Geogrs, Vol. 55. pp. 578-91.
CAVAiLLE, A. 1953. L’erosion actuelle en Quercy. Rev. geomorph. dyn., Vol. 2,
pp. 57-74.
COLBY, B. R. 1963. Fluvial sediments - a summary of source, transportation,
deposition, and measurement of discharge, U.S. Geol. Surv. Bull. 1181, pp. Al-
A47.
CORBEL, J. 1959. Erosion en terrain calcaire (vitesse d’erosion et morphologie),
Ann. Geogr., Vol. 68, pp. 97-120.

CORBEL, J. 1961. Morphologie periglaciaire dans I’Arctique, Ann. Geogr., Vol. 70,
pp. 1-24.
CORBEL, J. 1964. L’erosion terrestre, etude quantitative (methodes-techniques -

resultats), Ann. Geogn, Vol. 73, pp. 385-412.
COTTON, c. A. 1961. The theory of savanna planation. Geography, Vol. 46, pp. 89-
101 . .
COTTON, c. A. 1962. Plains and inselbergs of the humid tropics, Trans. Roy. Soc.
N.Z., Vol. 88, pp. 269-77.
DAVIDSON, J. 1963. Littoral processes and morphology on Scania flat-coasts, Lund

Studies in Geog. Ser. A, No, 23.
DENNY, c. s. 1967. Fans and pediments. Am. Jour. Sci., Vol. 265, pp. 81-105.
DRESCH, J. 1957. Pediments et glacis d’erosion, pediplains et inselbergs, Inf. geog..
Vol. 21, pp. 183-96.
DUCHAUFOUR. p. 1960. Precis de pedologie (Cen. Doc. Univ., Paris), 478 pp.
EARDLEY, A. J. 1966. Rates of denudation in the high plateaus of southwestern

Utah, Bull. geol. Soc. Am.. Vol. 77, pp. 777-80.
FITZPATRICK, E. A. 1958. An introduction to the periglacial geomorphology of

Scotland, Scot. geog. Mag.. Vol. 74, pp. 28-36.
FLINT, R. F. 1952. The ice age in the North American Arctic, Arctic. Vol. 5,
pp. 135-52.
FOX, c. 1950. Land erosion. Endeavour, Vol. 9, pp. 178-82.
GALLOWAY, R. w. 1961. Periglacial phenomena in Scotland. Geografiska Annaler,

Vol. 43,pp. 348-53.
ge'ike, a. 1868. On denudation now in progress, Geol. Mag., Vol. 5, pp. 249-
54.
GiLLULY, J. 1964. Atlantic sediments, erosion rates and the evolution of the
Continental Shelf: some speculations. Bull. geol. Soc. Am., Vol. 75, pp. 483-92.
GILLULY, J., REED, J. c., and CADY, w. M. 1970. Sedimentary volumes and their
significance. Bull. geol. Soc. Am., Vol. 81, pp. 353-76.
HADLEY, R. F. 1967. Pediments and pediment-forming processes. Jour. geol. Ed.,
Vol. 15, pp. 83-9.
HAMELiN, L. E. and COOK, F. A. 1967. Le periglaciaire par I'image îllustrated

glossary of periglacial phenomena) (Presses Univ. Laval, Quebec), 23 7 pp.
HODGSON, w. A. 1966. Coastal processes around the Otago peninsula, N.Z. Jour.
Geol. Geophys.. Vol. 9, pp. 76—90.
hOgbom, 1914. Cber die geologische Bedeutung des Frostes, Bull. geol. Inst.
b.
Univ. Uppsala, No. 12, pp. 257—389.
INMAN, D. L. and BAGNOLD, R. A. 1963. Beach and nearshore processes. II. Littoral
processes in M. N. Hill (Ed.). The sea (Interscience, New York), Vol. 3, pp. 529-
53.
KOON, D. 1955. Cliff retreat in the south west United States, Am. Jour. Sci.,
Vol. 253, pp. 44-53.
LLIBOUTRY, L. 1961. Phenomencs cryonivaux dans les Andes de Santiago (Chile),

Biul. Peryglacjalny, Vol. 10. pp. 209-24.
LOUIS, H. 1964. liber Rumpfiachen und Talbildung in den wechselfeuchten Tropen

besonders nach Studien in Tanganyika, Zeil.f. Geomorph., Vol. 8, pp. 43*-70*.
MAMMERiCKX, j. 1964. Quantitative observations on pediments in the Mojave and

Sonoran deserts (southwestern United States). Am. Jour. Set., Vol. 262, pp. 417-
35.
MENSCHiNG, H. VON. 1969. Bergfussflachen und des System der Flachenbildung in
den ariden Subtropen und Troppen, Geol. Rundschau, Vol. 58, pp. 62-82.
MOHR, E. c. j. and van baren, f. a. 1954. Tropical soils; a critical study of soil
genesis as related to climate, rock and vegetation (van Hoeve, The Hague), 498 pp.
mortensen, h. 1949. Rumpfflache - Stufenlandschaft - Alternierende Abtragung,

Pet. geog. Mitt., Vol. 93, pp. 1-14.
NOMENYi, F. 1952. Annotated and illustrated bibliographic material on the mor-

phology of rivers. Bull. geol. Soc. Am., Vol. 63. pp. 595-644.
NORRMAN, J. o. 1964. Lake Vattern. Investigations on shore and bottom mor-
phology, Geografiska Annaler, Vol. 46, pp, 1-238.
OLLIER, c. D. 1 969. Weathering (Oliver and Boyd, Edinburgh). 304 pp.
PASSARGE. s. 1929. Das Problem der Inselberglandschaften, Zeit.f. Geomorph.,

Vol.4.pp. 109-22.
PECSi, M. 1963. Periglaziale Erscheinungen in Ungern. Pet. geog. Mit., Vol. 107.
pp. 161-82.
PEEL, R. F. 1966. The landscape in aridity. Trans. Inst. Brit. Geogrs, No. 38, pp. 1-
23.
PUGH. J, c. 1956. Fringing pediments and marginal depressions in the inselberg
landscape of Nigeria. Trans. Inst. Brit. Geogrs. No. 22. pp. 15-31.
RAHN, p. H. 1966. Inselbergs and nickpoints in southwestern Arizona, Zeit.f.

Geomorph., Vol. 10. pp. 217-25.
RAHN, p. H. 1967. Sheetfloods, streamfloods and the formation of pediments. Ann.
Ass. Amer. Geogrs. Vol. 57. pp. 593-604.
RITTER, D. F. 1967. Rates of denudation. JoHr. geo/. Ft/., Vol. 15, pp. 154-9.
RUSSELL. R. J. 1967. Aspects of coastal morphology, Geografiska Annaler,

Vol. 49A. pp. 299-309.
RUXTON, B. p. and berry, l, 1961. Notes on faceted slopes, rock fans and
domes on granite in the east-central Sudan. Am. Jour. Sci.. Vol. 259, pp. 194-
206.
RUXTON, B. p. and mcdougall. i. 1967. Denudation rates in northeast Papua
from potassium-argon dating of lavas. Am. Jour. Sci.. Vol. 265, pp. 545-61.
SHARP, R. p. 1966, Kelso dunes, Mojave desert, California, Bull. geol. Soc. Am..
Vol. 77. pp. 1045-74.
sP AR K s. b. w. 1962. Rates of operation of geomorphological processes. Geography.

Vol.47,pp. 145-53.
s UND BORG, A. 1956. The river Klaralven. a study of fluvial processes, Geografiska
Annaler. Wo\. 38, pp. 127-316.
suNDBORG, ^ Some aspects on fluvial sediments and fluvial morphology.

1967.
I. General views and graphic methods, Geografiska Annaler, Vol. 49A, pp. 333-
43.
TABER, s. 1943. Perennially frozen ground in Alaska; its origin and history. Bull
geol Soc. Am., Vol. 54, pp. 1433—548.
TE PUNGA. M. 1956. Altiplanation terraces in southern England. BiuL

Peryglacjalny, Vol. 4, pp. 33 1-8.
TE PUNGA, M. 1957. Periglaciation in southern England. Tijds. Kon. Nederl Aar-

drijsh Gen., Vol. 74 , pp. 401—12.
TRICART, j., and VOGT, H. 1967. Quelques aspects du transport des alluvions
grossieres et du faponnement des lits fluviaux. Geografiska Annaler, Vol. 49A,
pp. 351-66.
TROLL, c. 1944. Strukturboden, Solifluktion und Frostklimate der Erde, Geol.

Rundschau, Vol. 34, pp. 545-694.
TWiD ALE, c. R. 1967. Origin of the piedmont angle as evidenced in South Australia,
Jour. Geol, Vol. 75, pp. 393—41 1.
vviLHELMY, H. 1958. KUmamorphologie der Massengesteine (Westermann,

YOUNG. A 1958. A record of the rate of erosion on Millstone Grit, Proc. Yorks,
geol Soc., Vol. 31. pp. 149-56.
A. Rock Breakdown
1. Erosion
BARKER, J. p. 1965. A forgotten factor in the interpretation of glacial stairways,

Zeit.f. Geomorph., Vol. 9, pp. 18-34.
BALLARD, A J. and SORENSEN, F. H. 1968. Preglacial structure of Georges Basin
and Northeast Channel, Gulf of Maine. Bull. Ainer. Ass. Petrol. Geols, Vol. 52,
pp. 494-500.
BATTLE, w. R. B., and LEWIS, w. V. 1951. Temperature observations in
bergschrunds and their relationship to cirque erosion. Jour. Geol, Vol. 59,
pp. 537-45.
BRADLEY, w. c. 1958. Submarine abrasion and wave-cut platforms. Bull geol. Soc.
Am., Vol. 69. pp. 967-74.
COTTON, c. A 1 968. Marine cliffing according to Darwin’s theory. Trans. Rov. Soc.
A.Z., Vol. 6. pp. 187-208.
CRARY, A p. 1966. Mechanism for fiord formation indicated by studies of an ice-

covered inlet. Bull, geol Soc. Am.. Vol. 77, pp. 91 1-30.
CRiCKMAY. c. H. 1960. Lateral activity in a river of northwestern Canada, Jour.
Geol. Vol. 68. pp. 377-91.
CURTIS. R, H.. and evernden. j. f. 1958. Rate of marine planation suggested by

K/A dates. Bull. geol. Soc. Am.. Vol. 69, p. 1680.
DAHL, R. 1965. Plastically sculptured detail forms on rock surfaces in northern
Nordland, Norway. Geogrqfiska Annaler, Vol. 47A, pp. 83-140.

DAVIES, G. L. 1 960, Platforms developed in the boulder clay of the coastal margins
of counties Wicklow and Wexford, Irish Geog., Vol. 4, pp. 107-16.
FAIRCHILD, H. L. 1905. Ice erosion theory a fallacy. Bull. geol. Soc. Am.. Vol. 16,
pp. 13-74.
GEiKE, j. 1878. On the glacial phenomena of the Long Island, or Outer Hebrides,
Quart. Jour. geol. Soc., Vol. 34, pp. 819-66.
GJESSiNG, j. 1 966. Some effects of ice erosion on the development of Norwegian

valleys and fiords, Norsk Geog. Tidsskr., Vol. 20. pp. 273-99.
GJESSING. J. 1967. Potholes in connection with plastic scouring forms, Geogrqfiska
Annaler, Vol. 49A, pp. 178-87.
GRAW'E, o. R. 1936. Ice as an agent of rock weathering; a discussion. Jour. Geol.,

Vol. 44. pp. 1 73-82.
HARLAND, vv. B. 1957. Exfoliation joints and ice action. Jour. Glacial., Vol. 3,
pp. 8-12.
HAYNES, V. M. 1968. The influence of glacial erosion and rock structure on corries
in Scotland, Geografiska Annaler, Vol. 50A. pp. 221-33.
HOLTEDAHL. H, 1967. Notes on the formation of fjords and fjord-valleys.
Geografiska Annaler. Vol. 49A, pp. 188-203.
KLiMASZEWSKi, M. 1964. On the effect of the preglacial relief on the course and
magnitude of glacial erosion in the Tatra Mountains, Geogr. Polonica. Vol. 2,
pp. 116-21.
LE wis, VV. V. 1 948. Valley steps and glacial valley erosion. Trans. Inst. Brit. Geogrs.
Vol. 13 (1947), pp. 19-44.
LEWIS, VV. V. 1954. Pressure release and glacial erosion. Jour. Glaciol, Vol. 2,
pp. 417-22.
LEWIS, w. V. (Ed.), 1960. Norwegian cirque glaciers, Roy. Geog. Soc. Research Ser..
No, 4. 104 pp,
LEIGHTON, M, M, and BROPHY, J. A. 1961, Illinoian glaciation in Illinois, yowr.
Geol.. Vol. 69. pp. 1-31.
LEOPOLD, L. B.. EMMETT, VV. w., and MYRiCK, R. M. 1966. Channel and hillslope
processes in a semi-arid area. New Mexico, U.S. Geol. Surv. Prof. Paper 352-G,
pp. 193-253.
LINTON. D. L. 1 963. The forms of glacial erosion. Trans. Inst. Brit. Geogrs. No. 33.
pp. 1-28.
PEEL. R. F. 195 1. A study of two Northumbrian spillways. Trans. Inst. Brit. Geogrs.
No. 15 (1949). pp. 73-89.
ROBINSON, A. H. w.. 1949. Deep clefts in the Inner Sound of Raasay. Scot. Geog.
Mag.. Vol. 65, pp. 20-5.
RUSSELL, R. J. 1963. Recent recession of tropical cliffy coasts. Science. Vol. 139.
pp. 9-15.
scHUMM, s. A. 1962. Erosion on miniature pediments in Badlands National

Monument, south Dakota, Bull geol Soc. Am., Vol. 73, pp. 719-24.
SCHWARTZ, M. L. 1968. The scale of shore erosion, Jour. Geol. Vol. 76, pp. 508-
17 .
SHARP, R. P. 1940. Geomorphology of the Ruby-East Humboldt range, Nevada,
Bull geol Soc. Am.. Vol. 5 1, pp. 221-72.
SHEPPARD, T. 1912. The lost towns of the Yorkshire coast and other chapters
bearing upon the geography of the district (Brown, London), 328 pp.
SISSONS, J. B. 1958. Sub-glacial stream erosion in southern Northumberland. Scot,

geog. Mag., Vol. 74, pp. 163—74.
SUGDEN. D. E. 1968. The selectivity of glacial erosion in the Cairngorm mountains,

Scotland, Trans. Inst. Brit. Geogrs. No. 45, pp. 79-92.
SUGDEN, D. E. 1969. The age and form of corries in the Cairngorms. Scot. geog.
Mag., Vol. 85, pp. 34-46.
THOMPSON, H. R. 1950. Some corries of north-west Sutherland, Proc. Geol /Iss.,
Vol. 87, pp. 145-55.
TiETZE, w. 196 1. Ober die Erosion von unter Eis fliessenden Wasser, Mainzer geog.
Studien Festschr. Panzer (Westermann. Braunschweig), pp. 125-42.
TS CHANG, T. L. 1957. Potholes in the river beds of north Taiwan. Erdkunde,

Vol. ll,pp. 296-303.
2. Frost action
ARNAUD. R. J. ST. and WHITESIDE. E. p. 1963. Physical breakdown in relation to

soil development. Jour. Soil Scl, Vol. 14, pp. 267-81.
BONiFAY, E. 1955. Le role du gel dans la fissuration des galets de roche calcaire, C.
r. Acad. Scl. Vol. 240, pp. 896-8.
BO YE, M. 1952. Gelivation et cryoturbation dans le massif du Mont-Perdu (Pyrenees
Centrales), Pirineos. Vol. 23. pp. 5-29.
c AiLLEUX, A 1 943. Fissuration de la craie par le gel. Bull Soc. geol. Fr., 5th ser.,
Vol. 13.pp. 51 1-20.
CZEPPE, z. 1964. Exfoliation in a periglacial climate, Geog. Polonica, Vol. 2,

pp. 5-10.
dylik, j., and klatka t. 1952. Recherches microscopiques sur la desintegration

periglaciaire. Bull Soc. Scl Lettres Lodz. Cl. III. Vol. 3. pp. 1-12.
s c HM ID. J. 1955. Der Bodenfrost als morphologischer Faktor (Hiithig, Heidelburg).
144 pp.
wiM an, s. 1963. A preliminary study of experimental frost weathering, Geografiska

Annaler, Vol. 45. pp. 113-21.
3. Other mechanical processes of rock breakdown
BEAUMONT, p. 1968. Salt weathering on the margin of the Great Kavir, Iran.
Bull
geol Soc. Am., Vol. 79. pp. 1683-4.
BIROT, p. 1954. Desagregation des roches cristallines sous Taction des sels, C. r.
ScL Vol.
Acad. 238. p. 1 145.
BLACKVVELDER, E. 1933. The insolation hypothesis of rock weathering. Am. Jour.

5c/., Vol. 226, pp. 97-113.
BLANCK, E. and PASS ARGE, s. 1925. Die chemische Verwitterung in der agyptis-
chen Wiiste (Univ. Abhandl. aus dem Gebiet der Auslandskunde, Hamburg),
Vol. 17, 1 10 pp.
KUENEN. p. 1960. Experimental abrasion 4: Eolian action. Jour. Geol., Vol. 68,
pp. 427-49.
MORTENSEN, H. 1933. Die ‘Salzsprengung’ und ihre Bedeutung fiir die regional-
klimatische Gliederung der Wiisten, Per. geog. Mill.. Vol. 79, pp. 130-5.
OLLIER, c. D. 1963. Insolation weathering: examples from central Australia, Am.

Jour. Sc/., Vol. 261, pp. 376-81.
PASSARGE, s. 1924. Das Problem der Skulptur Inselberglandschaften. Pet. geog.

Mitt., Vol. 70, pp. 66-70 and 1 17—20.
PEDRO, G. 1957. Nouvelles recherches sur Tinfluence des sels dans la desagregation
des roches. C. r. Acad. Sci.. Vol. 244, pp. 2822-4.
SHARP. R. p. 1964. Wind-driven sand in Coachella Valley, California, Bull. geol.
Soc. Am., Vol. 75. pp. 785-804.
ALEXANDER. F. E. s. 1959. Observations on tropical weathering: a study of the
movement of iron, aluminium and silicon in weathering rocks at Singapore. Quart.
Jour. geol. Soc., Vol. 1 15, pp. 123-44.
BALL, D. F, 1964. Deepweathering profile on the Island of Rhum, Inverness-shire.
Scot. geog. Mag., Vol. 80, pp. 22-7.
BARTON, D. c. 1938. The disintegration and exfoliation of granite in Egypt. Jour.

Geo/.. Vol. 46. pp. 109-11.
BIROT, p. 1947. Resultats de quelques experiences sur la desagregation des roches
cristallins, C. r. Acad. Sci.. Vol. 225. pp. 745-7.
BIROT. p. 1962. Contribution a I’etude de la desagregation des roches (Cen. Doc.
Univ., Paris), 232 pp.
BUNTING, B. T. 1961. The role of seepage moisture in soil formation, slope develop-
ment. and stream initiation. Am. Jour. ScL, Vol. 259, pp. 503-18.
DAHL, R. 1967. Post-glacial micro-weathering of bedrock surfaces in the Narvik

district of Norway. Geografiska Annaler, Vol. 49A, pp. 155-66.
DOORNK AMP, J. c. 1968. The role of inselbergs in the geomorphology of southern

Uganda, Trans. Inst. Brit. Geogrs. No. 44, pp. 151-62.
EMERY. K. o. 1960. Weathering of the Great Pyramid, Jour. Sed. Petrol., Vol. 30,
pp. 140-3.
FALCONER, J. D. \9li.The geology and geography of northern Nigeria (Macmillan.
London), 295 pp.
FITZPATRICK, E. A. 1963. Deeply weathered rock in Scotland; its occurrence, age
and contribution to the soils. Jour. Soil ScL, Vol. 14. pp. 33-43.
1
glinka, k. d. 1914. Die Typen der Bodenbildung, ihre klassifikation und geogra-
phische Verbreitttng (Gehrilder Borntraeger, Berlin), 365 pp.; trans. by C. F. Mar-
but. 1927, The great soil groups of the world and their development (Edwards
Bros.. Ann Arbor), 235 pp.
GOODCHiLD. J. G. 1890. On the weathering of limestones, Geol. Mag., Vol. 17,
pp. 463-6.
GUiLCHER, A. and PONT, p. 1957. Etude experimentale de la corrosion littorale des

calcaire. Bull. Ass. geog. Fr., No. 265-6, pp. 48-62.
HAY, R. L. 1959. Origin and weathering of late Pleistocene ash deposits on St

Vincent, B.W.I., Jour. Geol, Vol. 67, pp. 65—87.
HAY, R. L. 1960. Rate of clay formation and mineral alteration in a 4000-year-old

volcanic ash soil on St Vincent, B.W.L, Am. Jour. ScL, Vol. 258, pp. 354-68.
HENDRICKS, D. M. and WHiTTiG, L. D. 1968. Andesite weathering, your. Soil Set,
Vol. 19,pp. 135-47.
HILL, D. E, and TEDROW, J. c. F. 1961. Weathering and soil formation in the Arctic
environment. Am. Jour. Scl, Vol. 259, pp. 84-101.
KLiNGE, H. 1965. Podzol soils in the Amazon basin, your. Soil Scl, Vol. 16, pp. 95-
103.
NICHOLS, R. L. 1963. Geologic features demonstrating aridity of McMurdo Sound

area, Antarctica, Am. Jour. Scl, Vol. 261. pp. 20-31.
NOSSIN, J. J. and levelt. w. m. 1967. Igneous rock weathering on Singapore

Island, Zeit.f. Geomorph., Vol. 1 1. pp. 14-35.
NYE, p. H. 1954. Some soil-forming processes in the humid tropics. I; A field study of
a catena in West African forest. Jour. Soil Scl, Vol. 5. pp. 7-21.
OLLIER, c. D. 1965. Some features of granite weathering in Australia. Zeit. f.
Geomorph., Vol. 9. pp. 285-304.
PHEMiSTER, T. c. and SIMPSON, s. 1949. Pleistocene deep weathering in north-east
Scotland, Nature, Vol. 164, pp. 3 18-19.
RE ICHE. p. 1 950. A survey of weathering processes and products, Univ. New Mexico
Pubi. in Geol. No. 1. 87 pp.
RUXTON, B. p. 1958. Weathering and surface erosion in granite at the piedmont
angle. Balos. Sudan, Geol. Mag., Vol. 95, pp. 353—77.
SCHUNKE. E. 1969. Die Schichtstufenhange des Leine-Weser-Berglandes—
Methoden und Ergebnisse ihrer Untersuchung, Geol Rundschau, Vol. 58,
pp. 446-64.
SWEETING, M. M.. GROOM. G. E., WILLIAMS. V. H., PIGOTT. C. D., SMITH, D. I.,
and WARWICK, g. t. 1965. Denudation in limestone regions; a symposium.

Geog. Jour.. Vol. 1 3 1, pp. 24-56.
5. Weathering forms
bOgli, A. 1951. Probleme der Karrenbildung, Geog. Helvetica, Vol. 6, pp. 191—204.
bOgli. a. 1960. Kalklosung und Karrenbildung, Zeit.f. Geomorph.. Suppbd 2.
pp. 4-21.
BR AJNIKOV, B. 1953. Les ‘pains de sucre’ du Bresil sont-ils enracines, C. r. geol.

Soc. Fr., pp. 267-9.
CALKIN, p. and c AiLLEUX, A. 1962. A quantitative study of cavernous weathering

(taffonis) and its application to glacial chronology in Victoria Valley. Antarctica,
Zeit.f. Geomorph.. Vol. 6. pp. 3 1 7-24.
DAHL, R. 1966. Block fields, weathering pits and tor-like forms

in the Narvik
mountains, Nordland, Norway, Geogrqfiska Annaler, Vol. 48A, pp. 55-85.
DEMEK, j. 1964. Castle koppies and tors in the Bohemian Highlands (Czechoslo-
vakia), Biid. Peryglacjalny, Vol. 14, pp. 195-216.
DRAGOVICH, D. 1 969. The origin of cavernous surfaces (tafoni) in granitic rocks of
southern South Australia, Zeit.f. Geomorph., Vol. 13. pp. 163-81.
EAKIN, H. M. 1916. The Yukon-Koyukuk region, Alaska. U.S. Geol. Siirv. Bull.
631. 88 pp.
FEININGER. T. 1969. Pseudokarst on quartz diorite. Zeit. f. Geomorph., Vol. 13,
pp. 287-96.
FENELON, p. (Ed.). 1968. Phenomenes karstiques, Cen. Doc. Cartog. Geog. Mem. et
Doc., 392 pp.
GERBER, E. 1959. Form und Bildung alpiner Talboden, Geog. Helvetica, Vol. 14, pp.
117-238.
GERSTENHAUER, A. 1960. Der tropische Kegelkarst in Tabasco (Mexico), Zeit.f.
Geomorph., Suppbd 2, pp, 22-48.
GODARD, A. 1966. Morphologic des solces et des massifs anciens; les ‘tors’ et le
probleme de leur origine. Rev. geog. de I’Est, Vol. 6, pp. 153-70.
GUiLCHER, A. 1958. Coastal corrosion forms in limestone around the Bay of

Biscay, Scot. geog. Mag., Vol. 74, pp. 137-49.
HEDGES, j. 1969. Opferkessel, Zeit.f. Geomorph., Vol. 13, pp. 22-55.
Hsi-LiN, T. 1961. The pseudokarren and exfoliation forms of granite on Pulan Ubin,
Singapore, Zeit.f. Geomorph., Vol. 5, pp. 302-12.
Le GRAND, H. 1952. Solution depressions in diorite in North Carolina, Am. Jour.

Sci., Vol. 250, pp. 566-85.
LOUIS, H. 1963. Uber Sockelflache und Hiillflache des Reliefs, Zeit.f. Geomorph.,
Vol. 7, pp. 353-66.
LINTON, D. L. 1 955. The problem of tors, Geog. Jour., Vol. 121, pp. 470-87.
MABBUTT, J. 1952. A study of granite relief from south-west Africa, Geol. Mag.,
Vol. 89, pp. 87-96.
PALMER, H. s. 1927. Lapies in Hawaiian basalts, Geog. Rev., Vol. 17,
pp. 627-3 1.
PALMER, J. 1956. Tor formation at the Bridestones in north-east Yorkshire, and its
significance in relation to problems of valley-side development and regional glacia-
tion, Trans. Inst. Brit. Geogrs, Vol. 22, pp. 55-72.
PALMER, J. and NEiLSON, R. A. 1962. The origin of granite tors on Dartmoor,

Devonshire, Proc. Yorks, geol. Soc., Vol. 33, pp. 315-40.
SMITH, J. F. and ALBRITTON, c. c. 1941. Solutional effects on limestone as a

function of slope. Bull. geol. Soc. Am., Vol. 52, pp. 61-78.
SWEETING, M. M. 1958. The karstlands of Jamaica, Geog. Jour., Vol. 124,
pp. 184-99.
THOMAS, M. F. 1966. Some geomorphological implications of deep weathering

patterns in crystalline rocks in Nigeria, Trans. Inst. Brit. Geogrs, Vol. 40,
pp. 173-93.
THOMAS, M. F. 1967. A bomhardt dome in the plains near Oyo, western Nigeria,
Zeit.f. Geomorph., Vol. 1 1, pp. 239-61.
THORP, M. B. 1967. Closed basins in younger granite massifs, northern Nigeria, Zeit.
f. Geomorph., Vol, 1 1, pp. 459-80.
TROMBE, F. 1952. Traite de speleologie (Payot, Paris), 367 pp.
TWiDALE, c. R. 1964. A Contribution to the study of domed inselbergs, Trans. Inst.

Brit. Geogrs, No. 34, pp. 91-1 13.
WALL, j. R. D. and wiLFORD, G. E. 1966. A comparison of small-scale features on

microgranodiorite and limestone in west Sarawak, Malaysia, Zeit.f. Geomorph.,
Vol. 10, pp. 462-8.
WARWICK, G. T. 1962. In C. H. D. Cullingford (Ed.), British Caving: an introduc-
tion to speleology (Routledge, London), Chaps 1-5, pp. 11-217.
WHITE, L. s. 1 949. Processes of erosion on steep slopes of Oahu (Hawaii), Am. Jour.
5c/., Vol. 247, pp, 168-86.
WILLIAMS, p. w. 1966. Limestone pavements with special reference to western

Ireland, Trans. Inst. Brit. Geogrs, No. 40, pp. 155-72.
zwiTTKOViTS, F, 1969. Alters- und Hohengliederung der Karren in den Nordli-

chen Kalkalpen, Geol. Rundschau, Vol. 58, pp. 378-95.
B. Transportation
1. Transportation of particles in a fluid
BAGNOLD, R. A. 1937. The transport of sand by wind, Geog. Jour., Vol. 89,
pp. 409-38.
BE ADNELL, H. J. L. 1910. The Sand dunes of the Libyan desert, Geog. Jour., Vol. 35,
pp. 379-95.
behre, c. h. 1926. Sand flotation in nature. Science, Vol. 63, pp. 405-8.
CHERRY, J. A. 1966. Sand movement along equilibrium beaches north of San
Francisco, Jour. Sed. Petrol., Vol. 36, pp. 341-57.
DAVIDSON, c. 1888. Note on the movement of scree material. Quart. Jour. geol.
Soc., Vol. 44, pp. 232-8.
downing, b. h. 1968. Subsurface erosion as a geomorphological agent in Natal,
Trans, geol. Soc. S. Africa, Vol. 121, pp. 131-4.
DURAND, R. 1951. Transport hydraulique de graviers et galets en conduite. La

Houille Blanche, Spec. No. B, pp. 609-19.
EMERY, K. o. 1955. Transportation of rocks by driftwood, Jour. Sect. Petrol, Vol.

25, pp. 5 1-7.
ENGELN, o. D. VON. 1918. Transportation of debris by ice-bergs. Jour. Geol, Vol.

26, pp. 74—81.
FRECAUT, R. 1966. Les transports solides de fond des cours d’eau: techniques et
possibilites des methodes de mesure directe, Rev. geog. de I’Est, Vol. 6, pp. 321-8.
GARDNER, J. 1969. Observations on surficial talus movement, Zeit.f. Geomorph.,
Vol. 13, pp. 317-23.
GILBERT, G. K. 1914. The transportation of debris by running water, U.S. Geol

Sun’. Prof. Paper 86, 263 pp.
GIRESSE, P. 1965. Exemples de transport cotier sur le littoral Catalan espagnol.

Mecanismes de tirage. Gainers Oceanog., Vol. 1
7, pp. 99-106.
gOhren, h. 1966. Beobachtungen fiber den Einfluss des Oberwassers auf die
Sandbewegung in der Aussenweser, Die Kiiste, Vol. 14, pp. 157-69.
H ARD Y, j. R. 1 964. The movement of beach material and wave action near Blakeney
Point, Norfolk, Trans. Inst. Brit. Geogrs, Vol. 34, pp. 53-69.
JOHNSON, J. \v. 1956. Dynamics of nearshore sediment movement. Bull. Amer. Ass.
Petrol Geols, Vol. 40, pp. 2211-32.
jolliffe, p. 1964. An experimental design to compare the relative rates of

I.
movement of different sizes of beach pebbles, Proc. Geol Ass., Vol. 75, pp. 67-87.
JOPLiNG, A. V. 1966. Some principles and techniques used in reconstructing the
hydraulic parameters of a paleo-flow regime, yoi/r. Sed. Petrol, Vol. 36, pp. 5-49.
KENT, p. E. 1966. The transportation mechanism in catastrophic falls. Jour. Geol,

Vol. 74, pp. 79-82.
KiDSON, c. and carr, a. p. 1959. The movement of shingle over the sea bed close
inshore, Geog. Jour., Vol. 125, pp. 380-9.
MEADE, R. H. 1969. Landward transport of bottom sediments, /oiir. Sed. Petrol,

Vol. 39, pp. 222-34.
MCGEE, w. J. 1897. Sheetfiood erosion. Bull. geol. Soc. Am., Vol. 8, pp. 87-1 12.
MELAND, N. and NORRMAN, J. o. 1969. Transport velocities of individual size

fractions in heterogeneous bedload, Geografiska Annaler, Vol. 51 A, pp. 127—44.
PHILLIPS, A. w. 1963. Tracer experiments at Spurn Head, Yorkshire. England,

Shore and Beach, Vol. 31, pp. 30-5.
SHARP, R. p. 1964. Wind-driven sand in Coachella valley, California, Bull geol Soc.
Am., Vol. 75, pp. 785-804.
SHREVE, R. L. 1968. Leakage and fluidization in air-layer lubricated avalanches.
Bull geol. Soc. Am., Vol. 79, pp. 653-8.
STAN, and ghenovici, a 1964. River and sea transport in the Braila-Sulina
c.
sector of theDanube, Rev. Roum. Geol Geophys. Geog., Sir. Geog., Vol. 8,
pp. 45-50.
STRIDE, A. H. 1963. Current-swept sea floors near the southern half of Great
Britain, Quart. Jour, geol Soc., Vol, 1 19, pp. 175-99.
WOOD, J. 1825. Remarks on the moving of rocks by ice, Am. Jour. Sci., Vol. 9,
pp. 144—5.
2. Amounts of sediment transport
ANDERSON, H. w. 1954. Suspended sediment discharge as related to stream-

flow, topography, soil and land use, Trans. Am. geophys. Union, Vol. 35, pp.
268-81.
BERTHOis, L. and BARRIER, M. 1954. Apports sedimentaires en suspension

dans La Loire (annee 1953), Bull. Soc. Geol. Fr., Ser. 6, Vol. 20, pp. 237—42.
BOON. j. D. 1969. Quantitative analysis of beach sand movement, Sedimentology,

Vol. 13, pp. 85-103.
BiRKELAND, p. w. 1968. Mean velocity and boulder transport during Tahoe age
floods of the Truckee river, California-Nevada, Bull. geol. Soc. Am., Vol. 79, pp.
137-42.
BOGARDi, J. 1951. Mesure du debit solide des rivieres en Hongrie, La Houille

Blanche, Vol. 6, pp. 108-26.
COUTTS, J. R. H., KANDIL, M. F., NOWLAND, T. L., and TINSLEY, J. 1968. Use of
radioactive ^’Fe for tracing soil particle movement. Jour. Soil Sci., Vol. 19, pp.
311-41.
DEPETRis, p. T. and GRiFFEN, J. J. 1968. Suspended load in the Rio de la Plata

drainage basin, Sedimentology, Vol. 1 1, pp. 53-60.
DiACONU, 0. 1969. Resultats de I’etude de I’ecoulement des alluvions en suspen-

sion des rivieres de la Roumanie, Bull. Internl Ass. Sci. Hydrol, Vol. 14, pp.
51-89.
ELLISON, w.D. 1945. Some effects of raindrops and surface flow on soil erosion and
infiltration, Trans. Am. geophys. Union, Vol. 26, pp. 415—29.
FOURNIER, F. 1960. CUmat et erosion, (Presses Univ. France, Paris), 201 pp.
FOURNIER, F. 1969. Transports solides effectues par les cours d’eau. Bull. Internl
,455. Sci. Hydrol, Vol. 14, pp. 7-49.
INGLE, J. c. 1966. The movement of beach sand (Elsevier, Amsterdam), 221 pp.
kuhlman, h. 1958. Quantitative measurements of aeolian sand transport, Geog.

Tidsskr., Vol. 57, pp. 51-74.
MICHAUD, J. and cailleux, a. 1950. Vitesses des movements du sol au
Cambeyron, C. r..Acad. Sci., Vol. 230, pp. 3 14-15.
min tieh, t. 1941. Soil erosion in China, Geog. Rev., Vol. 31, pp. 570-90.
dSTREM, G. 1965. Problems of dating ice-cored moraines, Geografiska Annaler,
Vol. 47A,pp. 1-38.
parde, m. 1953. La turbiditie des rivieres et ses facteurs geographiques. Rev. Geog.
alpine, Vol. 41, pp. 399-421.
POUQUET, J. 1951. L’erosion (Presse Univ. France, Paris), 126 pp.
SCHOFIELD, J. c. 1967. Sand movement at Mangatawhiri spit and Little Omaha
bay. N.Z. Jour. Geol Geophys., Vol. 10, pp. 697-721.
scHUMM, s. A. 1956. The role of creep and rainwash on the retreat of badland
slopes, Am. Jour. ScL, Vol. 254, pp. 693-706.
SCHUMM, s. A. 1967. Rates of surficia! rock creep on hillslopes in western Colorado,

Science, Vol. 155, pp. 560-1.
SELBY, M. J. 1966. Methods of measuring soil creep. Jour. Hydrol {N.Z.), Vol. 5,
pp. 54-63.
TODD, o. 1 . and eliassen, s. 1940. The Yellow River problem, Trans. Amer. Soc.
Civ. Eng., Vol. 105, pp. 346-453.
u.s. DEPT. OF AGRICULTURE. 1965. Proceedings of the Federal Inter-Agency
sedimentation conference, 1963; U.S. Dept. Agric. Misc. Publ. No. 970, 933 pp.
VOLLBRECHT, K. 1966. The relationship between wind records, energy of longshore

drift,and energy balance off the coast of a restricted body of water, as applied to
the Baltic, Marine Geology, Vol. 4, pp. 1 19-47,
WHETTEN, J. T., KELLEY, J. c.. and HANSON. L. G. 1969. Characteristics of

Columbia River sediment and sediment transport. Jour. Sect. Petrol., Vol. 39,
pp. 1 149-66.
wiscHMEiER, w. H. and SMITH, D. D. 1958. Rainfall energy and its relationship to
soil loss, Trans. Am. geophys. Union, Vol. 39, pp. 285-91.
YOUNG, A. 1960. Soil movement by denudational processes on slopes. Nature, Vol.

187, pp. 220-2.
3. Mass movement
ANDERSSON, J. G. 1906. Solifluxion, a component of subaerial denudation. Jour.
Geol, Vol. 14, pp. 91-112.
BLACKWELDER, E. 1928. Mudflows as a geologic agent in semi-arid mountains.
Bull. geol. Soc. Am., Vol. 39, pp. 465-83.
COLLIN, A. 1846. Recherches experimentales sur les glissements spontanes des
terrains argileux, (Carilian-Goeury and Dalmont, Paris), trans. by W. R. Schriever
et al., 1956, Landslides in clay (Univ. of Toronto Press, Toronto), 160 pp.
CO OR AY, p. G. 1958. Earthslips and related phenomena in the Kandy district,
Ceylon, The Ceylon Geogr., Vol. 12, pp. 75-90.
DHONAU, T. J. and dhonau, n. b. 1963. Glacial structures on the north Norfolk

coast, Proc. Geol. Ass., Vol. 74, pp. 433-9.
DYLiK, J. 1967. Solifluxion, congelifluxion and related slope processes, Geografiska

ECKEL, E. B. (Ed.). 1958. Landslides and engineering practice. National Research

Council: Highway Research Board Spec. Rep. 29, 232 pp.
FAiRBRiDGE, R. \v. 1950. Landslide patterns on oceanic volcanoes and atolls, Geog.
Jour., Vol. 1 1 5, pp. 84-8.
gerlach, t. 1959. Needle ice and its role in the displacement of the cover of waste
material in the Tatra mountains, Przeglad Geogr., Vol. 31, pp. 603-5.
GROVE, A. T. 1953. Account of a mudflow on Bredon Hill, Worcestershire, April

1951, /’/-oc. Geol. Ass., Vol. 64, pp. 10-13.
hollingworth, s. e. 1934. Some solifluction phenomena in the northern Lake

District, Proc. Geol. Ass., Vol. 45, pp. 167-88.
JAHN, A. 1967. Some features of mass movement on Spitzbergen slopes, Geografiska
Annaler, Vol. 49A, pp. 213—25.
JOHNSON, R. H. 1965. A study of the Charlesworth landslides near Glossop, Trans.
Inst. Brit. Geogrs, No. 37, pp. 1 1 1-26.
JONES, F. o., EMBODY, D. R., and PETERSON, w. L. 1961. Landslides along the
Columbia River valley, northeastern Washington, U.S. Geol. Sitrv. Prof. Paper
367, 98 pp.
KAYSER, B. 1963. L’erosion par franes en Lucanie, Medite?ranee, Vol. 4, pp. 93—
100 .
KUPSCH, w. o. 1962. Ice-thrust ridges in western Canada, Jour. Geol., Vol. 70, pp.
582-94.
MACAR, p. 1947. Les chutes de I’lnkrisi (Congo occidental) et leurs divers modes
d’erosion, Ann. Soc. geol. Belg., Vol. 82, pp. 38-5 1.
MCLEAN, R. F. and DAVIDSON, c. F. 1968. The role of mass movement in shore

platform development along the Gisborne coastline. New Zealand, Earth Sci.
Jour., Vol. 2, pp. 15-25.
PRIOR, D. R., N., and ARCHER, D. R. 1968. Composite mudflows on

STEPHENS,
the Antrim coast of north-east Ireland, Geografiska Annaler, Vol. 50A, pp. 65-
78.
RICE, R. M., CORBETT, E. s., and BAILEY, R. G. 1969. Soil slippage related to
vegetation, topography and soil in southern California, Water Resources Research,
Vol. 5, pp. 647-59.
RUDBERG, s. 1958. Some observations concerning mass movement on slopes in
Sweden, Geol. Foren. I. Stock. Kd'rh., Vol. 80, pp. 1 14-25.
RUTTEN, M. G. 1960. Ice-pushed ridges, permafrost and drainage. Am. Jour. Sci.,
Vol. 258, pp. 293-7.
SCHMID, j. 1955. Der Bodenfrost als morphologischer Faktor (Hiithig, Heidelberg),
144 pp.
SHARPE, c. F. s. 1938. Landslides and related phenomena; a study of mass-
movements of soil and rock (Columbia Univ. Press, New York), 137 pp.
URBANEK, J. 1968. Slide classification, Geogrqficky Casopis, Vol. 20, pp. 221—
36.
WATSON, R. A. and WRIGHT, H. E. 1963. Landslides on the east flank of the

Chuska Mountains, northwestern New Mexico, Am. Jour. Sci., Vol. 261, pp. 525—
48.
WAHRHAFTiG, c. and COX, A. 1959. Rock glaciers in the Alaska Range, Bull. geol.
Soc. Am., Vol. 70, pp. 383-436.
WILLIAMS, p, J. 1957. Some investigations into solifluction features in Norway,
Geog. Jour., Vol. 123, pp. 42-58.
WILLIAMS, P. J. 1959. An investigation into processes occurring in solifluction, /Iw.

Jour. Sci., Vol. 257, pp. 48 1-90.
ZARUBA, Q. and mencl, V. 1969. Landslides and their control (Elsevier,

Amsterdam), 205 pp.
4. Transportation of dissolved material
DAVIS, s. N. 1964. Silica in streams and ground water. Am. Jour. Sci., Vol. 262,
pp. 870-91.
DOUGLAS, I. 1969. The efficiency of humid tropical denudation systems, Trans.

Inst. Brit. Geogrs, No. 46, pp. 1-19.
EK, c. 1964. Note sur les eaux de fonte des glaciers de la Haute Maurienne. Leur
action sur les carbonate, Rev. Beige Geog., Vol. 88, pp. 127-56.
EK, c. 1966. Faible agressivite des eaux de fonte des glaciers: I’exemple de la
Marmolada (Dolomites), Ann. Soc. geol. Belg,, Vol. 89, pp. 177-88.
GARRELS, R. M. 1965. Silica: role in the buffering of natural waters, Science, Vol.
148, p. 69.
GORHAM, E. 1961. Factors influencing supply of major ions to inland waters, with
special reference to the atmosphere. Bull. geol. Soc. Am., Vol. 72, pp. 795-840,
HEM, J. D. 1959. Study and interpretation of the chemical characteristics of natural
waters, U.S. Geol. Sun. Water-supply Paper 1473, 269 pp.
K R A u s K o p F, K. B. 1956. Dissolution and precipitation of silica at low temperatures,

Geochim. Cosmochim. Acta, Vol. 10, pp. 1-26.
LIVINGSTONE, D. A. 1963. The chemical composition of rivers and lakes, U.S.
Geol. Surv. Prof. Paper 440-G, 64 pp.
MACKENZIE, F. T. and GARRELS, R. M. 1966. Chemical mass balance between

rivers and oceans, Am. Jour. Sci., Vol. 264, pp. 507-25.
MOORE, E. s. and m aynard, j. e. 1929. Solution, transportation and precipitation

of iron and silica, Econ. Geol., Vol. 24, pp. 272-303, 365-402, 506-27.
ROUGERiE, G. 1961. Etude comparative de I’evacuation de la silice en milieux

cristallins tropical humide et tempere humide, premiers resultats, Ann. Geog., Vol.
70, pp. 45-50.
C. Deposition
General
ALLEN, J. R. L. 1965. A review of the origin and characteristics of recent alluvial

sediments, Sedimentology, Vol. 5, pp. 89-191.
JOHANSSON, c. E. 1965, Structural studies of sedimentary deposits, Lund Studies

in Geog., No. 42 A, 61 pp.
NORRIS, R, M. 1956. Crescentic beach cusps and barchan dunes. Bull. Amer. Ass.
Petrol. Geols, Vol. 40, pp. 1681-6.
Slope deposits
ANDREWS, J. T. 1961. The development of scree slopes in the English Lake District
and central Quebec-Labrador, Cahiers Geog. Quebec, No. 10, pp. 219-30.
BENEDICT, J. B. 1966. Radiocarbon dates from a stone-banked terrace in the

Colorado Rocky Mountains, Geografiska Annaler, Vol. 48 A, pp. 24-31.
bl'agbrough, j. w. and breed, vv. j. 1967. Protalus ramparts on Navajo

Mountain, southern Utah, Am. Jour. ScL, Vol. 265, pp. 759-72.
burkalow, a. van. 1945. Angle of repose and angle of sliding friction, an experi-
mental study, Bull. geol. Soc. Am., Vol. 56, pp. 669—708.
CAINE, N. 1967. The texture of talus in Tasmania, Jour. Sed. Petrol., Vol. 37,
pp. 796-803.
DYLiK, J 1960. Rhythmically stratified slope waste deposits, Bhil. Peryglacjalny,

Vol. 8, pp. 31-41.
MUDGE, M. R. 1965. Rockfall-avalanche and rockslide-avalanche deposits at

Sawtooth Ridge, Montana, Bull. geol. Soc. Am., Vol. 76, pp. 1003-14.
PARiZEK, E. J. and WOODRUFF, J. F. 1957. Description and origin of stone layers

in soils of the southeastern States, Jour. Geol., Vol. 65, pp. 24-34.
PECSi, M. 1967. Relationship between slope geomorphology and Quaternary slope

sedimentation, Acta Geol. Acad. Set. Hungaricae, Vol. 1 1, pp. 307-21.
PECSI, M. 1969. Genetic classification of slope sediments, Biul. Peryglacjalny, Vol.
18,pp. 15-27.
PiPPAN, T. 1969. Studies on grus and block deposits on mountain slopes in Austria,
Biul. Peryglacjalny, Vol. 18, pp. 29-42.
tinkler, k. j. 1966. Slope profiles and scree in the Eglwyseg valley. North Wales,
Geog. Jour., Vol. 132, pp. 379-85.
tivy, j. 1962. An investigation of certain slope deposits in the Lowther Hills,

Southern Uplands of Scotland, Trans. Inst. Brit. Geogrs. No. 30, pp. 59-74.
WATSON, E. 1969. The slope deposits in the Nant lago valley, near Cader Idris,
Wales, Biul. Peryglacjalny, Vol. 18, pp. 95—1 13.
Mudflows and alluvial fans
ANSTEY.R. L. 1966. A Comparison of alluvial fans in West Pakistan, Pa/:isfa/i Geog.

Rev., Vol. 21, pp. 14-20.
BEATY, c. b. 1961. Boulder deposits in Flint Creek valley, western Montana, Bull,
geol. Soc. Am., Vol. 72. pp. 1015-20.
BEATY, c. B. 1 963. Origin of alluvial fans. White Mountains, California and Nevada,
Ann. Ass. Amer. Geogrs, Vol, 53,

pp. 516-35.
BEATY, c. B. 1970. Age and estimated rate of accumulation of an alluvial fan.
White
Mountains. California, U.S.A., Am. Jour. ScL, Vol. 268, pp. 50-77.
BLissENBACH, E. 1954. Geology of alluvial fans in semiarid regions, Bull. geol.

Soc. Am., Vol. 65, pp. 175-90.
BULL, w. B. 1963. Alluvial-fan deposits in Western Fresno County, California, your.

Geol, Vol. 71, pp. 243-41.
BULL, w. B. 1968. Alluvial fans, your. êo/. jEd., Vol. 16, pp. 101-6.
BLUCK, B. j. 1964. Sedimentation of an alluvial fan in southern Nevada, your. Sed.

Petrol, Vol. 34, pp. 395-400.
Clastic deposition in rivers
BAGNOLD, R. A. 1968. Deposition in the process of hydraulic transport,

Sedimentology, Vol. 10, pp. 45-56.
FENNEMAN, N. M. 1906. Floodplains produced without floods. Bull Amer. geog.

Soc., Vol. 38, pp. 89-91.
HicKiN, E. J. 1969. A newly identified process of point bar formation in natural
streams. Am. Jour. Scl, Vol. 267, pp. 999-1010.
krigstrOm, a. 1962. Geomorphological studies of sandur plains and their braided

rivers in Iceland, Geogrqfiska Annaler, Vol. 44, pp. 328-46.
rasid, h. 1966. Morphology of the Jamuna flood plain. Oriental Geogr., Vol. 10,
pp. 57-72.
WILLIAMS, G. E. 1966. Planar cross-stratification formed by the lateral migration of
shallow streams. Jour. Sed. Petrol, Vol. 36, pp. 742-6.
WILLIAMS, p. F. and rust, b. r. 1969. The sedimentology of a braided river. Jour.

Sed. Petrol, Vol. 39, pp. 649-79.
Chemical deposition in streams
CAiLLEUX, A. 1965. Quaternary secondary chemical deposition in France, Geol

Soc. Am. Spec. Paper 84, pp. 125-39.
EK, c. and pissART, A. 1965. Depot de carbonate de calcium par congelation et
teneur en bicarbonate des eaux residuelles, C. r. Acad. Scl, Vol. 260, pp. 929-32.
GAMS, I. 1963. Nekatere znacilnosti Krke in njenih pritokov, Geografski Zbornik,

Vol. 8, pp. 92-1 10; English summary. Some characteristics of the river Krka and
its affluents, pp. 1 08-9.
GREGORY, J. w. 1 91 1 . Constructive Waterfalls, 5co/.geog.Mflg.,Vol. 27, pp. 537-46.
SENCU, v. 1967. Morphologie und Enstehung des Steinsalzkarstes bei Slanic-

Prahova, Rev. Roumanie geol geophys. geog., Vol. 1 1, pp. 49-65.
Deltas
ANDEL, T. H. VAN. 1967. The Orinoco delta. Jour. Sed. Petrol, Vol. 37, pp. 297-
310.
HAY, T. 1926. Delta formation in the English lakes, Geol Mag., Vol. 63, pp. 292-
301.
joPLiNG, A. V. 1965. Hydraulic factors controlling the shape of laminae in labora-

tory deltas, Jour. Sed. Petrol., Vol. 35, pp. 777-91.
MOORE, D. 1966. Deltaic sedimentation. Earth Sci. Rev., Vol. 1, pp. 87-104.
SHIRLEY, M. L. and RAGSDALE, j. A. (Eds). 1966. Deltas in their geologic frame-
work (Houston Geol. Soc., Houston), 25 1 pp.

STRAATEN, L. M. J. VAN. 1960. Some recent advances in the study of
u. deltaic
sedimentation, Liverpool and Manchester geol. Jour., Vol. 2, pp. 411-42.
STRAATEN, L. M. J. u. VAN. 1964. Deltaic and shallow marine deposits, (Elsevier,
Amsterdam).
Estuaries
AHNERT, F. 1960. Estuarine meanders in the Chesapeake Bay area, Geol. Rev., Vol.
50, pp. 390-401.
BERGDAHL, A. 1960. Glacifluvial estuaries on the Narke Plain, Lund Studies in

Geog., Ser. A, No. 1 3.
BERTHOis, L. 1960. Etude dynamique de la sedimenatation dans la Loire, Cahiers

Oceanogr., Vol. 12, pp. 63 1-57.
BERTHOIS, L. 1965. Techniques d'etudes estuariennes (Centre Doc. Univ., Paris),

144 pp.
EVANS, G. 1965. Intertidal flat sediments and their environments of deposition in the
Wash, Quart. Jour. Geol. Soc., Vol. 121, pp. 209-45.
GuiLCHER, A. 1953. Mesures de la vitesse de sedimentation et d’erosion dans des
estuaires breton, C. r. Acad. Sci., Vol. 237, pp. 1345-7.
LAUF, G. H. (Ed.). 1967. Estuaries, Am. Ass. Adv. Sci., Publ. 83, 757 pp.
NOTA, . J. G. and LORiNG, D. H. 1964. Recent depositional conditions in the St

Lawrence River and Gulf; a reconnaissance survey. Marine Geol., Vol. 2, pp. 198—
235.
PRICE, w. A. and kendrick, m. 1963. Field model investigation into the reasons
for siltation in the Mersey estuary, Proc. Instn. Civ. Engrs, Vol. 24, pp. 273—5 18.
ROBINSON, A. H. w. 1960. Ebb-flood channel systems in sandy bays and estuaries.
Geography, Vol. 45, pp. 1 83-99.
Coastal deposits
ALLEN, J. R. L. 1 965. Coastal geomorphology of eastern Nigeria; beach ridge barrier
islands and vegetated tidal flats, Geol. en Mijnbouw, Vol. 44, pp. 1—21.
block, b. j. 1967. Sedimentation of beach gravels: examples from South Wales,
Jour. Sed. Petrol., Vol. 37, 128-56.
pp.
ELLIOTT, E. L. 1958. Sandspits of the Otago coast, N.Z. Geogr, Vol. 14, pp. 65-74.
EVANS, o. F. 1942. The origin of spits, bars and related structures. Jour. Geol, Vol.
50. pp. 846-65.
GUILCHER, A. and KING, c. A. M. 1961. Spits, tombolos and tidal
marshes in
Connemara and west Kerry, Proc. Roy. Irish Acad., Vol. 6 IB,
pp. 283-338.
HEY, R. \v. 1967. Sections in the beach-plain deposits of Dungeness, Kent, Geol.
Mag., Vol. 104, pp. 361-70.
KiDSON, c. 1963. The growth of sand and shingle spits across estuaries, Zeit.f.
Geomorph., Vol. 7, pp. 1-22.
KING, c. A. M. and WILLIAMS, w. \v. 1949. The formation and movement of sand
bars by wave action, Geog. Jour., Vol. 1 13. pp. 70-85.
.MACNEIL, F. s. 1954. Organic reefs and banks and associated detrital sediments.
Am. Jour. Sci., Vol. 252, pp. 385-401.
NICHOLS, R. L. 1961. Characteristics of beaches formed in polar climates. Am. Jour.

Sci., Vol. 259. pp. 694-708.
PUTSY, N. p. 1965. Beach ridge development in Tabasco, Mexico. Arm. Ass. Amer.
Geogrs, Vol. 55, pp. 1 12-24.
SCHOLL, D. \v. 1963. Sedimentation in modern coastal swamps, southwestern

Florida, Bull. Amer. Ass. Petrol. Geols, Vol. 47. pp. 1581-603.
SILVESTER. R. 1965. Coral reefs, atolls and guyots. Nature, Vol. 207, pp. 681-8.
STEERS. J. A. 1937. The Coral Islands and associated features of the Great Barrier
Reefs, Geog. Jour., Vol. 89, pp. 1-28 and 1 19—46.
STOOD ART, D. R. 1965. The shape of atolls. Marine geology, Vol. 3, pp. 369-83.
TANNER, w. F. 1961. Offshore shoals in area of energy deficit. Jour. Sed. Petrol,
Vol. 31, pp. 87-95.
THOM. B. G. 1964. Origin of sand beach ridges, Aust. Jour. Sci., Voi. 26, pp. 351-2.
WIENS, H. j. 1959. Atoll development and morphology. Ann. Ass. Amer. Geogrs,
Vol. 49. pp. 35-54.
ZENKOViTCH. v. p. 1959. On the genesis of cuspate spits along lagoon shores, yoitr.
Geol, Vol. 67. pp. 269-77.
Eolian deposits
BAGNOLD, R. A. 1951, Sand formations in southern Arabia, Geog. Jour., Vol. 177,
pp. 78-87.
BEHEIR Y, s. A. 1967. Sand forms in the Coachella valley, southern California, Ann.
Ass. Amer. Geogrs, Vol. 57, pp. 25—48.
COOPER, w. s. 1958. Coastal sand dunes of Oregon and Washington, Geol Soc.
Am.. Mem. 72, 169 pp.
FiNKEL. H. J. 1959. The barchans of southern Peru, your. Geol, Vol. 67, pp. 614—47.
FLINT, R. F. and BOND, G. 1968. Pleistocene sand ridges and pans in western
Rhodesia, Bull. geol. Soc. Am., Vol. 79, pp. 299-3 14.
HASTENRATH, s. L. 1967. The barchans of the Arequipa region, southern Peru,

Zeit.f. Geomorph., Vol. 1 1, pp. 300-3 1.
INMAN, D. L.. EWING,

and CORLISS, J.
G. c., B. 1966. Coastal sand dunes of
Guerrero Negro, Baja California, Mexico. Bull geol. Soc. Am., Vol. 77, pp. 787-
802.
JENNINGS, J. N. 1967. Cliff top duHcs, Aust. geog. Studies, Vol. 5, pp. 40-9.
MADiGAN, c. T. 1936. The Australian sand-ridge deserts, Geog. Rev., Vol. 26,
pp. 205-27.
McKEE.E. D. and TiBBiTTS, G. c. 1964. Primary structures of a seif dune and

associated deposits in Libya, Jour. Sed. Petrol, Vol. 34, pp. 5-17.
PERRIN, R. M. s. 1956. Nature of ‘Chalk Heath’ soils. Nature, Vol. 178, p. 31.
SH ARP, R. P. 1966. Kelso dunes, Mojave desert, California, Bull. geol. Soc. Am., Vol.
77, pp. 1045-74.
siMONETT, D. s. 1960. Development and grading of dunes in western Kansas, Ann.

Ass. Amer. Geogrs, Vol. 50, pp. 216—41.
VERST APPEN, H. T. 1968. On the origin of longitudinal (seiO dunes, Zeit. f.
Glacial deposits
ANDREWS, J. T., and SMITHSON, B. B. 1966. Till fabrics of the cross-valley mor-
aines of north-central Baffin Island, Northwest Territories, Canada, Bull. geol. Soc.
Am., Vol. 77, pp. 271-90.
CHARLESWORTH, J. K. 1929. The South Wales end moraine. Quart. Jour. Geol.
Soc., Vol. 95, pp. 335-58.
FALCONER, G., IVES, J., L0KEN, o., and ANDREWS, J. T. 1965. Major end
moraines in eastern and central Arctic Canada, Geogr. Bull, No. 7, pp. 137-53.
GELLERT, J. F. 1966. Morphologic der Eisrandzonen der letzten skandinavischen

Vereisung in Mittel- und Osteuropa, Geog. Berichte, Vol. 1 1, pp. 99-121.
GRAVENOR, c. p. 1953. The origin of drumlins. Am. Jour. Scl, Vol. 251, pp. 674-
81.
GRAVENOR, c. p. and KUPSCH, vv. o. 1959. Ice-disintegration features in western

Canada, Jour. Geol, Vol. 67, pp. 48-64.
HARRIS, s. A. 1967. Origin of part of the Guelph drumiin field and the Galt and
Paris moraines, Ontario: a reinterpretation. Can. Geogr, Vol. 16—34.
11, pp.
HOLMSEN, G. 1963. Glacial deposits in southeastern Norway, Am. Jour. Scl, Vol.
261, pp. 880-9.
IVES. J. D. 1967. Glacial terminal and lateral features in northeast Baffin Island:
illustradons with descriptive notes, Geog. Bull, Vol. 9, pp. 62-70 and 106-14.
KIRBY, 1969. Variation in glacial deposition in a sub-glacial environment: an
R. p,
example from Midlothian, Scot. Jour. Geol, Vol. 5, pp. 49-53.
MCKENZIE, G. D. 1969. Observations on a collapsing kame terrace in Glacier

Bay National Monument, south-eastern Alaska, Jour. Glaciol,
Vol 8 on
413-25.
ORM E, A. R. 1967. Drumlins and the Weichsel glaciation of Connemara, Irish Geog.,
Vol. 5, pp. 262-74.
OSTREM, G. 1964. Ice-cored moraines in Scandinavia, Geografiska Annaler, Vol 46
pp. 282-337.
PRICE, R. J. 1966. Eskers near the Casement glacier. Alaska, Geogrqfiska Annaler,
Vol. 48A,pp. 111-25.
PRICE, R. J. 1969. Moraines, sandar, kames and eskers near Breidamerkurjokull,
Iceland, Trans. Inst. Brit. Geogrs, No. 46, pp. 17-43.
WEISSE, R. 1968. Endmoranen oder Oser ortlich von Glowen, Geog. Berichte, Vol.
13, pp. 277-91.
D. Inter-relationships between Rock Breakdown, Transport, and Deposition
1-3. Examples of studies examining time lags and the influence both of source
area and of rate of supply on further sedimentary processes and characteristics.
ANDERSON, H. w. 1957. Relating sediment yields to watershed variables, Trans.

Am. geophys. Union., Vol. 38, pp. 921—4.
CULLEN, D. J. 1966. Fluviatile run-ofT as a factor in the primary dispersal of

submarine gravels, Sedimentology, Vol. 7, pp. 191-201.
DAVIS, G. H. 1961. Geologic control of mineral composition of stream waters of the
eastern slope of the southern Coast Ranges, California, U.S. Geoi. Surv. (Vater-
supply Paper 1535-B, 30 pp.
HEIDEL, s. G. 1956. The progressive lag of sediment concentration with flood waves,
Trans. Am. geophys. Union. Vol. 47, pp. 56-66.
ICHIKAWA, M. 1958. On the debris supply from mountain slopes and its relation to
river bed deposition, Sci. Rep. Tokyo, Kyoiku Daigaku, No. 49, 24 pp.
MANNER, s. B. 1958. Factors affecting sediment delivery rates in the Red Hills
physiographic area, Trans. Am. geophys. Union. Vol. 39. pp. 669-75.
4. Covariance between phenomena
ANDRiESSE, J. p. 1969. A Study of the environment and characteristics of tropical

podzols in Sarawak (East Malaysia), Geodenna, Vol. 2, pp. 201-27.
BAS COM, w. N. 1951. The relationship between sand size and beach face slope,
Trans. Am. geophys. Union. Vol. 32, pp. 866-74.
BETTENAY, E. 1962. The salt lake systems and their associated aeolian features in
the semi-arid region of Western Australia, Jour. Soil Sci., Vol. 13, pp. 1 0-1 7.
BLUCK.B.j. 1969. Particle rounding in beach gravels. Geo/. Ma^., Vol. 106, pp. 1-14.
CAMELS, A. and cailleu.v, a. 1966. Variabilite des galets le long d’une meme
plage, C. r. Soc. geoi. Fr., pp. 84-5.
CARR, A. p. 1969. Size grading along a pebble beach: Chesil beach, England, Jour.
Sed. Petrol., Vol. 39, pp. 297-310.
e YLES, R. J. 1968. Stream net ratios in west Malaysia, Bull. geoi. Soc. Am., Vol. 79,
pp. 701-12.
FUR LEY, p. A. 1968, Soil formation and slope development. 2. The relationships
between soil formation and gradient angle in the Oxford area, Zeit.f. Geomorph.,
Vol. 12, pp. 24-42.
HOLMES, c. D. 1960. Evolution of till-stone shapes, central New York, Bull. geol.
Soc. Am., Vol. 71, pp. 1645-60.
KNEBEL, H. J., KELLEY, j. c., and WHETTEN, j. T. 1968. Clay minerals of the
Columbia river: a qualitative, quantitative and statistical evaluation. Jour. Sed.
Petrol, Vol. 38, pp. 600-1 1.
kuenen, p. 1956. Experimental abrasion of pebbles. 2: rolling by current. Jour.

Geol, Vol. 64, pp. 336-68.
MACKNEY, D. and BURNHAM, c. p. 1964. A preliminary study of some slope soils

in Wales and the Welsh borderland, Jour. Soil Scl, Vol. 15, pp. 319-30.
MELTON, M. A. 1958. Correlation structure of morphometric properties of drainage

systems and their controlling agents. Jour. Geol, Vol. 66, pp. 442-60.
MELTON, M. A. 1958. Geometric properties of mature drainage systems and their
representation in an Ej space, Jour. Geol, Vol. 66, pp. 35-56.

MELTON, M. A. 1965. Debris-covered hillsiopes of the southern Arizona desert —
consideration of their stability and sediment contribution. Jour. Geol, Vol. 73,
pp. 715-29.
PITTMAN, E. D. 1969. Destruction of plagioclase twins by stream transport. Jour.

Sed. Petrol, Vol. 39, pp. 1432-7.
PLUMLEY, w. J. 1948. Black Hills terrace gravels: a study in sediment transport,
Jour. Geol, Vol. 56, pp. 526-77.
a AGO, J. M. and bib by, j. s. 1966. Frost weathering and solifluction products in
southern Scotland, Geografiska Annaler, Vol. 48A, pp. 12-23.
ROUGERiE, G. 1951. A propos de I’etude morphoscopique des galets equatoriaux,
C. r. Soc. geol. Fr., pp. 80-2.
SCHIEDEGGER, A. E. 1966. Hortou’s law of stream numbers. Water Resources

Research, Vol. 4, pp. 65 1-5.
SMART, J. s. 1969. Topological properties of channel networks. Bull. geol. Soc. Am.,
Vol. 80, pp. 1757-74.
SNEED, E. D. and FOLK, R. L. 1958. Pebbles in the Lower Colorado river, Texas: a
study in particle morphogenesis, yonr. Geol, Vol. 66, pp. 1 14-50.
STRAW, A. 1968. Late Pleistocene glacial erosion along the Niagara escarpment of
southern Ontario, Bull. geol. Soc. Am., Vol. 79, pp. 889-910.
WAGGONER, p. E. and BINGHAM, c. 1961. Depth of loess and distance from
source. Soil Scl, Vol. 92, 396-401.
pp.
WALKER, p. H. 1964. Sedimentary properties and processes on a sandstone hillside.
Jour. Sed. Petrol, Vol. 34, 328-34.
pp.
WATSON, J. p. 1964. A soil catena on granite in Southern Rhodesia. I. Field
observations, Jour. Soil Scl, Vol. 15, pp. 238-57.
WINTERER, E. L, and VON DE R BORCH, c. c. 1968. Striated pebbles in a mudflow
deposit,South A\isiTa\ia,Palaeogeog.,Palaeoclim.,Palaeoecol,Yo\. 205-1 1.
5,pp.
WOLDENBERG, M. J. 1969. Spatial order in fluvial systems: Horton’s laws derived
from mixed hexagonal hierarchies of drainage basin areas.
Bull. geol. Soc. Am.,
Vol. 80, pp. 97-112.
5. Reworking of sediments in streams
c ARLSTON, c. w. 1965. The relation of free meander geometry to stream discharge

and its geomorphic implications. Am. Jour. ScL, Vol. 263, pp. 864-85.
HORLOCK, j. H. 1955. Erosion in meanders. Nature, Vol. 176, pp. 1034.
LANGBEiN, w. B. and LEOPOLD, L. B. 1964. Quasi-equilibrium states in channel
morphology. Am. Jour. ScL, Vol. 262, pp. 782-94.
LEOPOLD, L. B. and woLMAN, M. G. 1957. River channel patterns: braided, mean-
dering and straight, U.S. Geo/. Surv. Prof. Paper 282-B,pp. 39-85.
LEOPOLD, L. B. and WOLMAN, M. G. 1960. River meanders. Bull. geol. Soc. Am.,
Vol. 71, pp. 769-94.
mcJller, j. t. 1963. Accumulation and abrasion in a tidal area. Cartographic

methods and results, Geog. Tidsskr., Vol. 62, pp. 56-79.
sc HU MM, s. A. 1960. The effect of sediment type on the shape and stratification of
some modern fluvial deposits. Am. Jour. ScL, Vol. 258, pp. 177-84.
s c H u m m, s. A. 1963. Sinuosity of alluvial rivers on the Great Plains, Bull. geol. Soc.
Am., Vo/. 74, pp. 1089-1 100.
WERTZ, J. B. 1966. The flood cycle of ephemeral mountain streams in the south-
western United States, Aim. Ass. Amer. Geogrs, Vol. 56, pp. 598-633.
6. Dynamic depositional landforms
DIETZ, R. s. 1963. Wave-base, marine profile of equilibrium, and wave-built terraces:

a critical appraisal. Bull. geol. Soc. Am..\o\. 74. pp. 971-90.
DOLAN, R. 1966. Beach changes on the Outer Banks of North Carolina, Ann. Ass.
Amer. Geogrs, Vol. 56, pp. 699-71 1.

H AR M s, J. c. 1 969. Hydraulic significance of some sand ripples. Bull. geol. Soc. Am.,
Vol. 80, pp. 363-96.
HOOKE, R. LeB. 1967. Processes on arid-region alluvial fans, Jour. Geol., Vol. 75,
pp. 438-60.
L ANGBEiN, w. B. and LEOPOLD, L. B. 1968. River channel bars and dunes - theory
of kinematic waves, U.S. Geol. Surv. Prof. Paper 422-L, 20 pp.
LETTAU, K. and LETTAU, H. 1969. Bulk transport of sand by the barchans of

the Pampa de La Toya in southern Peru, Zeit.f. Geomorph., Vol. 13, pp. 182-
95.
MILLER, R. L. and ZEiGLER, J. M. 1958. A model relating dynamics and sediment

pattern in equilibrium in the region of shoaling waves, breaker zone, and foreshore.
Jour. Geol., Vol. 66, pp. 417—41.
PALMER, H. R. 1834, Observations on the motion of shingle beaches, Phil. Trans.,
Vol. 124, pp. 567-76.
PRICE, w. A. 1963. Patterns of flow and channeling in tidal inlets, yonr. Sed. Petrol.,
Vol. 33, pp. 279-90.
SHARP, R. P. 1963. Wind ripples, yoi/r. Geol, Vol. 71, pp. 617-36.
TRIG ART, j. and M AiNGUET, M. 1965. CaracteHstiques granulometriques de quel-

ques sables eolien du desert peruvien, aspects de la dynamique des barkanes, iJer.
geomorph. dyii., Vol. 15, pp. 1 10-21.
7. Reworking of sediments into distinctive patterns
BALL, D. F. 1967. Stone pavements in soils of Caernarvonshire, North Wales, Jour.

Soil. Set., Voi. 18, pp. 103-8.
BALL, D. F. and GOODiER, R. 1968. Large sorted stone-stripes in the Rhinog

Mountains, North Wales, Geografiska Annaler, Vol. 50A, pp. 54-9.
CAINE, T. N. 1963. The origin of sorted stripes in the Lake District, northern
England, Geografiska Annaler, Vol. 45, pp. 1 72-9.
CORTE, A. 1963. Particle sorting by repeated freezing and thawing, Science,

Vol. 142, pp. 499-501.
cosTiN, A. B., THOM, B., wiMBUSH, D. J., and STUiVER, M. 1967. Nonsortcd
steps in the Mt. Kosciusko area, Australia, Bull. geol. Soc. Am., Vol. 78, pp. 979-
92.
DZULYNSKi, s. 1963. Polygonal structures in experiments and their bearing upon

some periglacial phenomena. Bull. Acad. Polonaise Sci., Vol. 1 1, pp. 145-50.
FLEMMING, N. c. 1964. Tank experiments on the sorting of beach material during

cusp formation, Jour. Sed. Petrol., Vol. 34, pp. 112-22.
HAY, T. 1936. Stone stripes, Geog. Jour., Vol. 87, pp. 47-50.
HOPKINS, D. M. and siG afoos, r. s. 1954. Discussion: role of frost thrusting in

the formation of tussocks. Am. Jour. Sci., Vol. 252, pp. 55-9.
JOHNSSON, G. 1960. Cryoturbation at Zaragoza, northern Spain, Zeit. f. Geo-

morph., Vol. 4, pp. 74-80.
KLATKA, T. 1961. Problemes des sols stries de la partie septentrionale de la

presqu’ile de Sorkapp (Spitzbergen), Biul. Peryglacjalny, Vol. 10, pp. 291—320.
KLATK A. T. 1962. Champs de pierre de Lysogory, origine et age, Acta. geog. Lodz,
No. 12, 129 pp.
KUENEN, p. H. 1948. The formation of beach cusps, Jonr. Geol., Vol. 56, pp. 34—40.
LACHENBRUCH, A. H. 1962. Mechanics of thermal contraction cracks and ice-
wedge polygons in permafrost, Geol. Soc. Am. Spec. Paper, No. 70, 69 pp.
MAC KAY. J. R. 1962. Pingos of the Pleistocene Mackenzie delta area, Geog. Bull,
No. 18,pp. 21-63.
MULLER, F. 1959. Beobachtungen uber Pingos in Ostgronlands und in der
Kanadischen Arktis, Medd. Gr^land Danm., Vol. 153, 127 pp.
NEAL, J. T., LANGER, A. M.. and KERR, p. F. 1968. Giant
desiccation polygons of
Great Basin playas. Bull. geol. Soc. Am., Vol. 79, pp. 69-90.
philberth, k. 1964. Recherches sur les sols polygonaux et stries, Biul.

Peryglacjalny. Vol. 13, pp. 99-198.
POSER, H. 1933. Das Problem des Strukturbodens, Geol. Rundschau, Vol. 24, pp
105-21.
RUSSELL, R. j. and mcintire, w. g. 1965. Beach cusps, Bull.geol. Soc. Am.,Vo\.

76, pp. 307-20.
SCHREIBER, K. F. 1969. Beobachtungen fiber die Entstehung von ‘Buckelweiden’

auf der Hochflachen des schweizer Jura, Erdkunde, Vo!. 23, pp. 280-90.
SVENS SON. H. 1964. Traces of pingo-like frost mounds, Lund studies in Geog. series
A, No. 30, pp. 93-106.
SVENSSON, H. er a/. 1967. Poij'gonal ground and solifluction features, Lund studies
in Geog. series A, No. 40, 67 pp.
TABER, s. 1929. Frost heaving, Jour. Geo/., Vol. 37, pp. 428-61.
TALLIS, j. H. and KERSHAW, K. A. 1959. Stability of stone polygons in North

Wales, Nature, Vol. 183, pp. 485-6.
WASHBURN, A. L. 1956. Classification of patterned ground and review of suggested

origins. Bull. geol. Soc. Am., Vol. 67, pp. 823-65.
VERGER, F. 1964. Mottureaux et gilgais, Ann. Geog., Vol. 73. pp. 413-30.
8. Recombination of weathered chemical elements
BASSETT, H. 1954. Silicification of rocks by surface waters. Am. Jour. Sci.,

Vol. 252, pp. 733-5.
brOckner, w. 1955. The mantle- rock (‘laterite’) of the Gold Coast and its origin,
Geol. Rundschau, Vol. 43, pp. 301-21.
CHARLES, G. 1949. Sur la formation de la carapice zonaire en Algerie, C. r. Acad.

Set., Vol. 228, pp. 261-3.
DEMANGEOT. J. 1961. Pseudo-cuestas de la zone intenropicale. Bull. Ass. Geog.

Fr.,No. 296-7, pp. 2-17.
DROSTE, J. B. 1956. Alteration of clay minerals by weathering in Wisconsin tills,
Bull. geol. Soc. Am., Vol. 67, pp. 91 1-18.
ENGEL, c. G. and SHARP, R. p. 1958. Chemical data on desert varnish. Bull. geol.
Soc. Am., Vol. 69, pp. 487-5 1 8.
FREISE, E. w. 1936. Bodenverkrustung in Brasilien, Zeit. f. Geomorph., Vol. 9,
pp. 233-48.
GUILCHER, A. 1961. Le ‘beach-rock’ ou gres de plage, Ann. Geog., Vol. 70, pp.
113-25.
KELLER, w. D. 1964. Processes of origin and alteration of clay minerals, in Soil

clay mineralogy - a symposium (Chapel Hill, Univ. North Carolina Press), pp. 3-
76.'
LAMOTTE, M. and ROUGERiE, G. 1962. Les apports allochtones dans la genese des
cuirasses ferrugineuses. Rev. geomorph. dyn., Vol. 13, pp. 145-60.
L ANGFORD-SMiTH, T. and DURY, and attitude

G. H. 1965. Distribution, character
of the duricrust in the northwest of New
South Wales and the adjacent areas of
Queensland, Am. Jour. Sci., Vol. 263, pp. 170-90.
LARSEN, G. and CHiLiNGAR, G. V. 1967. Dlageitesis in sediments (Elsevier,

Amsterdam), 551 pp.
MAUD, R. R. 1 965. Laterite and lateritic soil in coastal Natal, South Africa, Jour. Soil
ScL, Vol. 16, pp. 60-72.
MOSELEY, F. 1965. Plateau calcrete, calcreted gravels, cemented dunes and related
deposits of the Maallegh-Bomba region of Libya, Zeit. f. Geomorph., Vol. 9,
pp. 166-85.
MOSS, R. p. 1965. Slope development and soil morphology in a part of south-west

Nigeria, Jour. Soil Sci., Vol. 16, pp. 192-209.
PRIDER, R. T. 1966. The lateritized surface of Western Australia, Aust. Jour. Sci.,
Vol. 28, pp. 443-51.
ROUGERiE, c. 1959. Lateritisation et pedogenese intertropicale. Inf. geog., Vol. 23,
pp. 199-206,
STEPHENS, c. G. 1966. Origin of silicretes of central Australia, Nature, Vol, 209,
p. 496.
STOOD ART, D. R. and c ANN, j. R. 1 965. Nature and origin of beachrock, yoifc. Sed.
Petrol., VoL 35, pp. 243-7.
TAYLOR, J. c. M., and ILLING, L. V. 1969. Holocene intertidal calcium carbonate

cementation, Qatar, Persian Gulf, Sedimentology, Vol. 12, pp. 69-107.
Y A ALON, D. H. 1 967. Factors affecting the lithification of eolianite and interpretation

of its environmental significance in the coastal plain of Israel, Jour. Sed. Petrol.,
Vol. 37, pp. 1189-99.
9. Influence of weathered material on subsequent weathering
ACKERMANN, E. 1962. Butserstein-Zeugcn vorzeitlicher Grundwasserschwankun-

gen, Zeit.f. Geomorph., Vol. 6, pp. 148-82.
bOdel, J. 1957. Die ‘Doppelten Einebnungsflachen’ in den feuchten Troppen, Zeit.f.
GiLLULY, J. 1937. Physiography of the Ajo region, Arizona, Bull. geol. Soc. Am.,
Vol. 48, pp. 323-48.
MABBUTT, J. A, 1966. Mantle-controlled planation of pediments. Am. Jour. Sci.,

Vol. 264, pp. 78-91.
MORTENSEN, H. 1921 Der Fonuenschatz der nordcliUenisclten IFfAre (Weidmann,
.
Berlin), 191 pp.
WAHRHAFTiG, c. 1965. Stepped topography of the southern Sierra Nevada, Bull,

geol. Soc. Am., Vol. 76, pp. 1165—90.
YOUNG, R. G. 1964. Fracturing of sandstone cobbles in caliche-cemented terrace
gravels. Jour. Sed. Petrol., Vol. 34, pp. 887-9.
10. The concept of quasi-equilibrium
CARLSTON, c. w. 1969. Downstream variations in the hydraulic geometry of

streams: special emphasis on mean velocity. Am. Jour. Sci., Vol. 267 pp 499-
509.
CARLSTON, c. w. 1969. Longitudinal slope characteristics of rivers of

the midcon-
tinent and the Atlantic East Gulf slopes. Bull. Inlernl /Iss. Sci. Hydrol, Vol. 14,
pp. 21-31.
DURY, G. H. 1963. Rivers in geographical teaching, Geography, Vol. 48, pp. 18-
30.
HACK, j. T. 1956. Studies of longitudinal stream profiles in Virginia and Maryland,

V.S. Geol. Sun\ Prof. Paper 294-B, pp. 45—97.
LEOPOLD, L. B. and LANGBEiN, w. B. 1962. The concept of entropy in landscape
evolution, U.S. Geol. Surv. Prof. Paper 500-A, 20 pp.
MACKiN, J. H. 1948. Concept of the graded river. Bull. geol. Soc. Am., Vol. 59,
pp. 463-512.
SCHUMM, s. A. and lichty, r. w. 1965. Time, space, and causality in geomorpho-
logy, Am. Jour. ScL, Vol. 263, pp. 1 10-19.
souCHEZ, R. 1961. Rente d’equilibre de creusement et force tractrice. Bull. Soc.

Roy. Beige geog.. Nos I-IV, pp. 145-58.
E. Inter-relationships betvveen Form and Process
ALLEN, J. R. L. 1965. Scour marks in snow. Jour. Sed. Petrol., Vol. 35, pp. 331-8.
BRUNET, p. 1956. Deux processus d’erosion en haute montagne. Rev. geomorph.

dy/!., Vol. 7,pp. 143-7.
c AiLLEUx, A. 1 948. Le ruissellement en pays tempere non-montagneux, Ann. Geog.,

Vol. 57, pp. 21-39.
CALLENDER, G. s. 1951. The effect of the altitude of the firn area of a glacier’s
response to temperature variations, GlacioL, Vol. 1, pp. 573-6.
FLINT, R. F. 1943. Growth of North American ice sheet during the Wisconsin age.
Bull. geol. Soc. Am., Vol. 54, pp. 325-62.
JENNINGS, and BiK, M. J. 1962. Karst morphology
J. N. in Australian New Guinea,
Nature,Vol. 194, pp. 1036-8.
RATHJENS, c. 1951. Der Hochkarst im System der klimatischen morphologie,
Erdkunde, Vol. 5, pp. 3 10-15.
SCHOELLER, H. 1950. Les variation de la teneur en gas carbonique des eaux

souterraines en fonction de I’altitude, C. r. Acad. Sci., Vol. 230, pp. 560-1.
s PREEN, \v. c. 1947. A determination of the elTect of topography on precipitation,

Trans. Amer. geophys. Union, Vol. 28, pp. 285-90.
F. The Significance of Changes in Weather and in Climate
1 . Local changes due to differing aspects
BALCHiN, vv. G. V. and PYE, N. 1947. A microclimatological investigation of Bath

and the surrounding districts. Quart. Jour. Roy. Mel. Soc., Vol. 73, pp. 297-319.
BEATY, c. B. 1962. Asymmetry of stream patterns and topography in the Bitterroot
Range, Montana, Jour. Geol., Vol. 70, pp. 347-54.

CAiLLEUX, A. and CALKIN, p. 1963. Orientation of hollows in cavernously

weathered boulders in Antarctica, Biul. Peryglacjalny, Vol. 12, pp. 147-50.
CARSON, c. E. and hussey, k. m. 1962. The oriented lakes of arctic Alaska, Jour.
Geo/., Vol. 70, pp. 417-39.
FLINT, R. and FiDALGo, F. 1964. Glacial geology of the east flank of the
F.
Argentine Andes between latitude 39° lO'S. and latitude 41° 20' S., Bull. geol.
Soc. Am., Vol. 75, pp. 335-52.
GARNETT, A. 1937. Insolation and relief, Trans. Inst. Brit. Geogrs, No. 5, 71 pp.
GILBERT, G. K. 1904. Systematic asymmetry of crest lines in the High Sierra of

California, Jour. Geol, Vol. 12, pp. 579—88.
HEMBREE, c. H. and RAINWATER, F. H. 1961. Chemical degradation on opposite

flanks of the Wind River Range, Wyoming, U.S. Geol. Surv. Water-supply Paper
1535-E, 9 pp.
JENNINGS, j. N. 1957. On the orientation of parabolic or U-dunes, Geog. Jour.,

Vol. 123, pp. 474-80.
LANDSBERG, s. Y. 1956. The orientation of dunes in Britain and Denmark in
relation to the wind, Geog. Jour., Vol. 122, pp. 176-89.
LEVERETT, F. 1929. Pleistocene glaciations of the northern hemisphere. Bull. geol.

Soc. Am., Vol. 40, pp. 745-60.
LINTON, D. L. 1959. Morphological contrasts of eastern and western Scotland, in R.

Miller and J. W. Watson (Eds), Geographical essays in memory of Alan G. Ogilvie,
pp. 16-45.
MAARLEVELD, G. c. 1960. Wind directions and coversands in the Netherlands,

Biul. Peryglacjalny, Vol. 8, pp. 49-58.
MABBUTT, J. A., WOODING, R. A., and JENNINGS, J. N. 1969. The asymmetry of
Australian desert sand ridges, Aust. Jour. ScL, Vol. 32, pp. 159-60.
MELTON, M. A. 1960. Intravalley variation in slope angles related to microclimate

and erosional environment, Bull. geol. Soc. Am., Vol. 71, pp. 133-44.
6strem, g., BRIDGE, c. w., RENNIE, w. F. 1967. Glacio-hydrology, discharge and
sediment transport in the Decade Glacier area, Baffin Island, N.W.T., Geograjiska
SEDDON, B. 1957. Late-glacial cwm glaciers in Wales, Jour. Glacial, Vol. 3, pp.
94-9.
SOONS, J. M. and ranier, j. n. 1968. Micro-climate and erosion processes in the

southern Alps, New Zealand, Geografiska Annaler, Vol. 50A, pp. 1—15.
SPEITZER, H. 1960. Hangformung und Asymmetrie der Bergriicken in den Alpen
und im Taurus, Zeit.f. Geomorph., Suppbd 1, pp. 211-36.
STOKES, w. L. 1964. Incised, wind-aligned stream patterns of the Colorado plateau.
Am. Jour. Scl, Vol. 262, pp. 808-16,
TEMPLE, p. H. 1965. Some aspects of cirque distribution in the west-central Lake
District, northern England, Geografiska Annaler, Vol. 47A, pp. 185-93.
THORP, J. 1934. The asymmetry of the Pepino Hills of Puerto Rico in relation to the
trade winds, 7oi//-. Geol, Vol. 42, pp. 537-45.
WHITE, E. M. 1961. Drainage alignment in western South Dakota, Am. Jour. Sci.,
Vol. 259, pp. 207-10.
WHITE, w. A. 1966. Drainage asymmetry and the Carolina capes. Bull. geol. Soc.
Am., Vol. 77, pp. 223-40.
WILLIAMS, M, A. J. 1969. Prediction of rainsplash erosion in the seasonally wet
tropics. Nature, Vol. 222, pp. 763-5.
2 and 3. Changes in the effectiveness of processes with time and examples of

past climates
ARNBORG, L., WALKER, H. J,, and PEiPPO, J. 1967. Suspended load in the Colville
River, Alaska, 1962, Geografiska Annaler, Vol. 49 A, pp. 131-44.
AXELROD, D. I. and BAILEY, H. p. 1969. Paleotemperature analysis of Tertiary

floras, Palaeogeog., palaeoclim., palaeoecoL, Vol. 6, pp. 163-95.
BACHMANN, F. 1966. Fosslle S trukturboden und Eiskeile (Kunz-Druck, Zurich),

177 pp.
BECKINS ALE, R. P. 1965. Climatic change: a critique of modern theories, in J. B.
Whitlow and P. D. Wood (Eds), Essays in geography for Austin Miller (Univ. of
Reading, Reading), pp. 1-38.
BOWEN, R. N. c. 1966. Palaeotemperature analysis (Elsevier, Amsterdam), 265 pp.
BROECKER, w. s., EWING, M., and HEEZEN, B. c. 1960. Evidence for an abrupt
change in climate close to 1 1,000 years ago. Am. Jour. Sci., Vol. 258, pp. 429-48.
B 0 D E L, J. 1951. Die Klimazonen des Eiszeitalters, Eiszeitalter und Gegenwart, Vol.

1, pp. 16-26; trans. 1959, Climatic zones of the Pleistocene, Internl Geol. Rev.,
Vol. l,pp. 72-9.
BUDEL, J. 1963. Die pliozanen und quartaren Pluvialzeiten der Sahara, Eiszeitalter
und Gegenwart, Vol. 14, pp. 161-87.
COETZEE, J. A. 1964. Evidence for a considerable depression of the vegetation belts

during the Upper Pleistocene on the East African mountains. Nature, Vol. 204,
pp. 564—6.
coo PE, G. R. 1962. A Pleistocene coeopterous fauna with arctic affinities from
Fladbury, Worcestershire, Quart. Jour. Geol. Soc., Vol. 1 18, pp. 103-23.
COTTON, c. A. 1958. Alternating Pleistocene morphogenetic systems. Geol. Mag.,

Vol. 95, pp. 125-36.
CROLL, J. 1875. Climate and time and their geological relations (E. Stanford,
London), 577 pp.
CURRY, R. R. 1966. Glaciation about 3,000,000 years ago in Sierra Nevada, Science,
Vol. 154, pp. 770-1.
CURRY, R. R. 1969. Holocene climatic and glacial history of the central Sierra
Nevada, California, Geol. Soc. Am. Spec. Paper 123, pp. 1-47.
CZEPPE, z. 1960. Annual course and the morphological effect of the vertical frost
movements of soil at Hornsund, Vestspitzbergen, Bull. Acad. Polon, Sci. Ser. Sci.
geol. geog.. No. 2, pp. 145-8.
CZEPPE, z. 1968. The annual rhythm of morphogenetic processes in Spitzbergen,

Geog. Polonica, Vol. 14, pp. 657-65.
DENTON, G. H. and ARMSTRONG, R. L. 1969. Miocene-Pliocene glaciations in

southern Alaska, Am. Jour. ScL, Vol. 267, pp. 1 121-42.
djmbleby, g. w. 1952. Pleistocene ice-wedges in north-east Yorkshire, Jour. Soil
Sci., Vol. 5,pp. 1-19.
DRESCH, J. 1954. Formes et limites climatiques et paleoclimatiques en Afrique du
Nord, Ann. Geog., Vol. 58, pp. 56-9.
DURY, G. H. 1954. Weather, climate and river erosion in the Ice Age, Science News,
Vol. 33, pp. 65-88.
DYLJK, J. 1960. Rhythmically stratified slope waste deposits, Biul. Peryglacjainy,

Vol. 8, pp. 31-41.
EMiLiANi, c. 1955. Pleistocene temperatures, yottr. Geol, Vol. 63, pp. 538-78.
EWING, M. and DONN, w. L. 1956. A theory of ice ages. Science, Vol. 123,
pp. 1061—6.
FLINT, R. F. 1959. Pleistocene climates in eastern and southern Africa, Bull. geol.
Soc. Am., Vol. 70, pp. 343-74.
FLINT, R. F. and BRANDTNER, F. 1961. Climatic changes since the Last

Interglacial, Am. Jour. Sci., Vol. 259, pp. 321-8.
FRENZEL, B. 1967. Die Klimaschwankungen des Eiszeitalters (Vieweg,

FRYE, J. c. 1962. Comparison between Pleistocene deep-sea temperatures and glacial

and interglacial episodes. Bull. geol. Soc. Am., Vol. 73, pp. 263—6.
GENTiLLi, J. 1961. Quaternary climates of the Australian region, A/in. New York
Acad. Sci., Vol. 95, pp. 465-501.
GRipp, K. 1964. Winter-Phanomena am Meeresstrand, Zeit.f. Geomorph., Vol. 8,
pp. 326-3 1.
GROVE, A. T. and warren, a. 1968. Quaternary landforms and climate on the

south side of the Sahara, Geog. Jour., Vol. 134, pp. 1 94-208.
GUiLLiEN, Y. 1955. La couverture vegetale de I’Europe pleistocene. A/m. Geog.,
Vol. 64, pp. 24 1-76.
HEY, R. w. 1963. Pleistocene screes in Cyrenaica (Libya), Eiszeitalter iind
Gegenwart, Vol. 14, pp. 77-84.
hoinkes, h. c. 1968. Glacier variation and weather. Jour. Glaciol., Vol. 7, pp. 3-
20 .
HOMMERiL, p., and LARSONNEUR, c. 1963. Quelqueseffets morphologiques du

gel intense
de I’hiver 1963 sur le littoral bas-normand, Cahiers Oceanog., Vol. 15,
pp. 638-50.
HOUTMAN, T. J. 1965. Winter hydrological conditions of coastal waters south of
Kaikoura peninsula, N.Z. Jour. Geol. Geophys., Vol. 8, pp. 807—19.

JOHNssoN, G. 1959. True and false ice-wedges in southern Sweden, Geografiska
Annaler. Vol. 41, pp. 15-33.
KAISER, K. 1960. Klimazeugen des periglazialen Dauerfrostbodens in Mittel- und

Westeuropa, Eiszeitalter und Gegenwarl, Vol. 1 1, pp. 121-41.
LADURiE, E. LE R. 1967. Histoire du climat depuis Van mil (Flammarion, Paris),

379 pp.
lOtig, g. 1965. Interglacial and interstadial periods, your. Geol., Vol. 73, pp. 579-91.
MIL ANKOViTCH, M. 1938. Astrononische Mittel zur Erforschung der erdgeschicht-
lichen Klimate, Hardbuch der Geophysik, Vol. 9, pp. 593-698.
MOBERLY, R., BAVER, L. D., and MORRISON, A. 1965. Source and variation of
Hawaiian littoral sand, Jour. Sed. Petrol, Vol. 35, pp. 589-98.
NAIRN, A. E. M. (Ed.). 1961. Descriptive palaeocUmatoIog}’ (InteTScisncs, London),

380 pp.
NAIRN, A. E. M. (Ed.). 1964. Problems in potoeoc/imaio/ogv (Interscience, London),
705 pp.
OLDFIELD, F. 1960. Late Quaternary changes in climate, vegetation and sea-level in
lowland Lonsdale, Trans. Inst. Brit. Geogrs, Vol. 28, pp. 99-1 17.
OTVos, E. G. 1965. Sedimentation-erosion cycles of single tidal periods on Long

Island Sound beaches. Jour. Sed. Petrol, Vol. 35, pp. 604-9.
PEWE. T. L., CHURCH, R. E., and ANDRESEN, M. j. 1969. Origin and paleoclimatic
significance of large-scale patterned ground in the Donnelly Dome area, Alaska,
Geol Soc. Am. Special Paper JOS, 87 pp.
PEWE, T. L. (Ed.). 1969. The periglacial environment: past and present (McGill-
Queens University Press, Montreal), 487 pp.
POSER, H. 1948. Boden- und Klimaverhaltnisse in Mittel- und Westeuropa wahrend

der Wiirmeiszeit, Erdkunde, Vol. 2, pp. 53-68.
PRENTICE, j. E. and MORRIS, p. G. 1959. Cemented screes in the Manifold valley,

north Staffordshire, East Midld Geogr., Vol. 2, pp. 16-20.
REEVES, c. c. 1 965. Pleistocene climate of the Llano Estacado, Jour. Geol, Vol. 73,
pp. 181-8.
REEVES, c. c. 1968. Introduction to Paiaeolimnology (Elsevier, Amsterdam),

228 pp.
REtD, c. 1887. On the origin of the dry chalk valleys and of the Coombe rock. Quart.
Jour, geol Soc., Vol. 43, pp. 364-73.
SCHWARZBACH, M. 1961. Das Klima der Eorzeit (Enke, Stuttgart), 275 pp.; trans.
by R. O. Muir, 1963, Climates of the past (Van Nostrand, London), 328 pp.
Segota, t. 1967. Paleotemperature changes in the Upper and Middle Pleistocene,
Eiszeitalter und Gegemvari, Vol. 18, pp. 127—41.
SHOTTON, F. w. 1960. Large-Scale patterned ground in the valley of the

Worcestershire Avon, Geol Mag., Vol. 97, pp. 404-8.
STARKEL, L. 1966. Post-glacial climate and the moulding of European relief, Roy.
Met. Soc. Symp. on World climate 8000 to 0 BC, pp. 15-33.
STRAHLER, A. N. 1966. Tidal cycles of changes in an equilibrium beach, Sandy

Hook, New Jersey, Jour. Geol, Vol. 74, pp. 247-68.
TABER, B. A. 1950. Intensive frost action along lake %\ioTes, Am. Jour. Sci., Vol. 248,
pp. 784—93.
TUCKFiELD, c. G. 1964. Gully erosion in the New Forest, Hampshire, Am. Jour.
Sci., Vol. 262, pp. 795-807.
UNESCO. 1963. Changes of climate, Arid Zone Research, Vol. 20 (UNESCO, Paris),
488 pp.
WALKER, D. and WEST, R. G. (Eds). 1970. Studies in the vegetational history of the
British Isles (Cambridge Univ. Press, Cambridge), 266 pp.
WATERS, R. s. 1961. Involutions and ice-wedges in Devon, Nature, Vol. 189,

pp. 389-90.
WATSON, E. 1965. Periglacial structures in the Aberystwyth region of central Wales,

Proc. Geol. Soc., Vol. 76, pp. 443-62.
WATT, A. s., PERRIN, R. M. s., and WEST, R. G. 1966. Patterned ground in

Breckland: structure and composition, Joj/r. Ecol, Vol. 54, pp. 239-58.
WEAVER, c. E. 1967. Variability of a river clay suite. Jour. Sed. Petrol, Vol. 37,
pp. 971—4.
WOLMAN, M. G. 1959. Factors influencing erosion of a cohesive river bank. Am.

Jour. Scl, Vol. 257, pp. 204-16.
WRIGHT, H. E. 1 961. Late Pleistocene climates of Europe; a review. Bull. geo!. Soc.
Am., Vol. 257, pp. 204-16.
YEHLE, L. A. 1954. Soil tongues and their confusion with certain indicators of
periglacial climate. Am. Jour. Sci., Vol. 252, pp. 532-46.
ZEiGLER, J. M., and TUTTLE, s. D. 1961. Bcach changes based on daily measure-
ments of four Cape Cod beaches. Jour. Geol, Vol. 69, pp. 583-99.
Chapter V. Landforms and time
A. Dating
ANDREWS, J. T. and WEBBER, p. J. 1964. A

lichenometrical study of the north-
western margin of the Barnes Ice Cap; a geomorphological technique, Geogr. Bull,
No. 22, pp. 80-104.
aver, V. 1963. Late glacial and postglacial shoreline displacements in South
America as established by tephrachronology, compared with displacements of the

Baltic shorelines, Fennia, Vol. 89, No. 1, pp. 5 1—5.
AXELROD, D. I. 1966. Potassium-argon ages of some western Tertiary floras. Am.
Jour. Sci., Vol. 264, pp. 497-506.
BESchel, r. 1958. Lichenometrical studies in West Greenland, Arctic, Vol. 11,
p. 254.
BLACKBURN, G. 1966. Radiocarbon dates relating to soil development, coastline

changes, and volcanic ash deposition in south-east South Australia, Aiist. Jour.
Set, Vol. 29, pp. 50-2.
BOWEN, D. Q. 1966. Dating Pleistocene events in south-west Wales, Nature,
Vol. 211, pp. 475-6.
cox. A., DALRYMPLE, G. B., and DOELL, R. R. 1967. Reversals of the Earth’s
magnetic field. Set. American, Vol. 216, pp. 44-54.
DAVIS, M. B, 1963. On the theory of pollen analysis. Am. Jour. Sci., Vol. 261,
pp. 897-912.
DiMBLEBY, G. w. 1961. Transported material in the soil profile. Jour. Soil Sci.,
Vol. 12, pp. 12-22.
EMiLiANi, c. 1955. Pleistocene temperatures, your. GeoL, Vol. 63, pp. 538-78.
EVERNDEN, J. F., CURTIS, G. H., and K isTLER, R. 1950. Potassium-argon dating of

Pleistocene volcanics, Quaternaria, Vol. 4, pp. 13-17.
EVERNDEN, j. F. and JAMES, G. T. 1964. Potassium-argon dates and the Tertiary

floras of North America, Am. Jour. Sci., Vol. 262, pp. 945-74.
F AEGRi, K. and IVERSEN, J. 1966. Textbook of modern pollen analysis-, 2nd rev. ed.
(Munksgaard, Copenhagen), 23 7 pp.
G E E R, G. D E. 1 9 1 2. A geochronology of the last 1 2,000 years, C. r. Xl Internl. Geol.

Congress {Stockholm), Vol. 1, pp. 241-53.
GEER, G. DE. 1951. Conclusions from C ,4 and de Geer’s chronology, Dani-

Gotiglacial, with datings, Geol. Fdren Forhandl, Vol. 73, pp, 557-71.
GODWIN, H. 1941. Pollen analysis and Quaternary geology, Proc. Geol. Ass.,
Vol. 52, pp. 328-61.
GODWIN, H. 1960. Radiocarbon dating and Quaternary history in Britain, Proc.

Roy. Soc. B, Vol. 153, pp. 287-320.
HAMILTON, E. I. 1965. Applied geochronology (Academic Press, London),

267 pp.
HEATH, J. p. 1960. Repeated avalanches at Chaos Jumbles, Lassen Volcanic

National Park, Am. Jour. Sci., Vol. 258, pp. 744-51.
HOLMES, A. 1960. A revised geological time scale, Trans. Edin. Geol. Soc., Vol. 17,
pp. 183-216.
IVES, R. L. 1962. Dating of the 1746 eruption of Tres Virgenes volcano, Baja
California del Sur, Mexico, Bull. geol. Soc. Am., Vol. 73, pp. 647-8,
KAizuKA, s. 1958, Tephrochronological studies in Japan, Erdkunde, Vol. 12,

pp. 253-70.
KAYE, c. A. 1964. The upper limit of barnacles as an index of sea-level change on

the New England coast during the past 100 years, Jour. Geol, Vo). 72, pp. 580-
600.
KELLEY, T. E. and BAKER, c. H. 1966. Color variations within glacial till, east-
central North Dakota - a preliminary investigation, Jour. Sed. Petrol, Vol. 36,
pp. 75-80.
Selected bibliography All
landim, p. m. b. and frakes, l. a. 1968. Distinction between tills and other

diamictons based on textural characteristics, Jour. Sed. Petrol, Vol. 38, pp. 1213-
23.
LiBBV, vv. F. 1965. Radiocarbon dating, 2nd ed. (Univ. Chicago Press, Chicago),
175 pp.
MERRITT, R. s. and MULLER, E. H. 1959. Depth of leaching in relation to carbonate

content of till in central New York State, Am. Jour. Sci., Vol. 257, pp. 475-80.
MOORB ATH, s. 1962. Lead isotope abundance studies on mineral occurrences in the
British Isles and their geological significance, Phil Trans. Roy. Soc., Vol. 254A,
pp. 295-360.
MUNNicH, K. o. 1960. Die C ,4 Methode, Geol Rundschau, Vol. 49, pp. 237-44.
PULLAR, w. A., 1967. Uses of volcanic ash beds in geomorphology. Earth Sci. Jour.,
Vol. l,pp. 164-77.
RONC A, L. B. and zeller, e. j. 1965. Thermoluminescence as a function of climate

and temperature. Am. Jour. Sci, Vol. 263, pp. 416-28.
ROSHOLT, J. N., EMILIANI, C., GEISS, J., KOCZY, F. F., and WANGERSKY, P. J.
1961. Absolute dating of deep-sea cores by the Pa”VTh^^'’ method. Jour. Geol,
Vol. 69, pp. 162-85.
SHACKLETON, N. 1967. Oxygen isotope analyses and Pleistocene temperatures re-

assessed, Nature, Vol. 215, pp. 15-17.
SHARP, R. p. 1960. Pleistocene glaciation in the Trinity Alps of northern California,
Am. Jour. Sci., Vol. 250, pp. 305-40.
SHARP, R. p. 1969. Semiquantitative differentiation of glacial moraines near Convict
Lake, Sierra Nevada, California, Jour. Geol, Vol. 77, pp. 68-91.
SHOTTON, F. w. 1966. The problems and contributions of methods of absolute

dating within the Pleistocene period. Quart. Jour. geol. Soc., Vol. 122, pp. 357-83.
SIMONSON, R. w. 1954. Identification and interpretation of buried soils. Am. Jour.
Sci., Vol, 252, pp. 705-32.
STORK, A. 1963. Plant immigration in front of retreating glaciers, with examples
from the Kebnekaise area, northern Sweden, Geogrqfiska Annaler, Vol. 45, pp.
1
-22 .
WAMBEKE, A. R. VAN. 1962. Criteria for classifying tropical soils by age. Jour. Soil
S’cL, Vol. 13, pp. 124-32.
ZEUNER, F. E. 1958. Dating the past, 4th ed., rev. and enl. (Methuen, London),
516 pp.
B. The Initial Form

AMBROSE, J. w. 1964. The exhumed paleoplains of the pre-Cambrian shield of
North America, Am. Jour. Sci., Vol. 262, pp. 817-57.
boye, m., moulline, m., pratviel, l., and viguier, c. 1968.
Relations entre la
forme des cours inferieurs de la Garonne et de la Dordogne et les
topographies
souterraines des terrains. Rev. geomorph. dyn., Vol.
18, pp. 83-91.
BURK, c. A. 1968. Buried ridges within continental margins, Trans. New York Acad.
ScL. Vol. 30, pp. 397-409.
CARSOLA, A. j. 1954. Microrelief on the Arctic Sea floor. Bull. Amer. .4ss. Petrol.
Geols, Vol. 28, pp. 1587-601.
DAY, A. A. 1959. The continental margin between Brittany and Ireland, Deep-Sea
Research, Vol. 5, pp. 249-65.
DURRANCE, E. M. 1969. The buried channels of the Exe, Geol. Mag., Vol. 106,
pp. 1 74-89.
FISHER, R. V. 1964. Resurrected Oligocene hills, eastern Oregon, Am. Jour. Sci.,
Vol. 262, pp. 713-25.
HAWKINS, A. B. 1962. The buried channel of the Bristol Avon, Geol. Mag., Vol. 99,
pp. 369-74.
HiNSCHBERGER, F. 1 964. La repartition des fonds sous-marins dans le vestibule du
Goulet de Brest, C. r. Acad. ScL, Vol. 258, pp. 6497-9.
HOLTED AHL, H. 1958. Some remarks on geomorphology of continental shelves off
Norway, Labrador, and southeast Alaska, Jour. Geol, Vol. 66, pp. 461-71.
KLIM ASZE wsKi, M. 1960. On the influence of preglacial relief on the extension and
development of glaciation and deglaciation of mountainous regions, Przeglad
Geogr.. Vol. 32 (Supp.), pp. 4 1-9.
M ALLORY, J. K. 1964. Topography of the seabed off the coast of southern Africa, S.
Afr. Jour. Sci., Vol. 60, pp. 105-9.
MARiNEscu, A. and SELATiu, o. 1965. An underwater valley in front of the'

Rumanian shore of the Black Sea, Rev. Roum. Geol. Geophys. Geog., Ser. Geog.,
Vol. 9, pp. 77-80.
MARTIN. H. 1969. Palaomorphologische Formelemente in den Landschaften

Sudwest-Afrikas, Geol. Rundschau,Vo\. 58, pp. 121-8.
MARTIN, R. 1966. Paleogeomorphology and its applications to exploration for oil

and gas (with examples from Western Canada), Bull. Amer. ;4ss. Petrol. Geols, Vol.
50, pp. 2277-31 1.
MENARD, H. w. and DIETZ, R. s. 1952. Mendocino submarine escarpment. Jour.

Geol.. Vol. 60, pp. 266-78.
MCM ANUS, J. 1 967. Pre-glacial diversion of the Tay drainage through the Perth gap,
Scot. Geog. Mag., Vol. 83, pp. 138-9.
s H E p AR D, F. p. 1967. The earth beneath the sea, rev. ed. (John Hopkins, Baltimore),
242 pp.
SHEPARD, F. p. and DILL, R. F. 1966. Submarine canyons and other sea valleys
(Rand McNally, Chicago). 381 pp.

sooNS, J. M. 1968. Raised submarine canyons: a discussion of some New Zealand
examples, Ann. Ass. Amer. Geogrs, Vol. 58, pp. 606-13,
WALSH, p. T. 1966. Cretaceous outliers in south-west Ireland and their implications
for Cretaceous palaeogeography. Quart. Jour. Geol. Soc., Vol. 122. pp. 63-84.
WARNER, M. M. 1965. Cementation as a clue to structure, drainage patterns,

permeability, and other factors. Jour. Sed. Petrol., Vol. 35, pp. 797-804.
WILLIAMS, G. J. 1968. The buried channel and superficial deposits of the Lower
Usk, and their correlation with similar features in the Lower Severn, Proc. Geol.
Ass., Vol. 79, pp. 325—48.
WINSLOW, j. H. 1966. Raised submarine canyons: an exploratory hypothesis, Ann.

Ass. Amer. Geogrs, Vol. 56, pp. 634-72.
C. Stages in Slope Evolution
AGAR, R. 1960. Post-glacial erosion of the north Yorkshire coast from the Tees
estuary to Ravenscar, Proc. Yorks, geol. Soc., Vol. 32, pp. 409-28.
AHNERT, F. 1960. The influence of Pleistocene climates upon the morphology of

cuesta scarps on the Colorado plateau, Ann. ,4ss. Amer. Geogrs, Vol. 50, pp. 139-
56.
BARKER, J. p. and LE HEUX, J. w. N. 1952. A remarkable new geomorphological
law, Proc. K. ned. Akad. Wet., Vol. 55B, pp. 399—410 and 554—71,
BECKETT, p. H. T. 1968. Soil formation and slope development. 1. A new look at
Walther Penck’s Aiifbereitiing concept, Zeit.f. Geomorph., Vol. 12, pp. 1-24.
BRYAN, K. 1922. Erosion and sedimentation in the Papago County, Arizona, U.S.
Geol Sum. Bull. 730, pp. 1 9-90.
CARTER, c. s. and CHORLEY, R, J. 1961. Early slope development in an expanding

stream system, Geol. Mag., Vol. 98, pp. 1 17-30.
COLEMAN, A. M. and BALCHiN, w. G. V. 1959. The origin and development of
surface depressions in the Mendip Hills, Proc. Geol. /Iss., Vol. 70, pp. 291-309.
DURAND, J. H. 1959. Les sols rouges et les croutes en Algerie, Service et Sci.,
Algiers, No. 7, 188 pp.
JAHN, A. 1968. Denudational balance of slopes, Geog. Polonica, Vol. 13, pp.
9-29.
KERNEY, M. p., and CHANDLER, T. J. 1964. The Late-glacial and

BROWN, E. H.,
Post-glacial history of the Chalk escarpment near Brook, Kent, Phil. Trans. Roy.
Soc. B, Vol. 248, pp. 135-204.
mackay, J. R. 1966. Segregated epigenetic ice and slumps in permafrost, Mackenzie

delta area, N.W.T., Geog. Bull, Vol. 8, pp. 59-80.
melton, m. a. 1965. The geomorphic and paleoclimatic significance of alluvial
deposits in southern Arizona, Jour. Geol, Vol. 73, pp. 1—38.
PENCK, w. 1924, Die morphologische Analyse; ein Kapitel der physikalischen
Geologic (Engelhorns, Stuttgart), 283 pp.; trans. by H. Czech and K. C. Boswell,
1953, Morphological analysis of land forms (Macmillan, London), 429 pp.
Ruxton, b. p. and berry, l. 1961. Notes on faceted slopes, rock fans and domes
on granite in the east-central Sudan, Am. Jour. Scl, Vol. 25 9, pp. 1 94-206.
s avigear, r. a. g. 1952. Some observations on slope development in South Wales,
Trans. Inst. Brit. Geogrs, No. 31-51.
18, pp.
scheidegger. a. e. 1961. Mathematical models of slope development, Pit//, geo/.
Soc. Am., Vol. 72, 37-50.
pp.
SCHUMM, s. A. 1956. Evolution of drainage systems and slopes in badlands at Perth

Amboy, New Jersey, Bull. geol. Soc. Am., Vol. 67, pp. 597-646.
SCHUMM, s. A. 1966. The development and evolution of hillslopes, Jour. geol. Ed.,
Vol. 14, pp. 98-104.
TRICART, j. 1957. Mise au point; revolution des versants. Inf. geog., Vol. 21,
pp. 108-16.
TWiD ALE, c. R. 1967. Origin of the piedmont angle as evidenced in South Australia,
Jour. Geol., Vol. 75, pp. 393-4 1 1.
WALKER, p. H. 1962. Soil layers on hillslopes: a study at Nowra, N.S.W., Australia,

Jour. Soil Sc/., Vol. 13, pp. 167-77.
WATSON, R. A. and weight, h. e. 1963. Landslides on the east flank of the

Chuska Mountains, northwestern New Mexico, Am. Jour. Sci., Vol. 261, pp. 525-
48.
D. Stages in Drainage System Evolution
AMBROSE, J. w. 1964. Exhumed paleoplains of the Precambrian shield of North

America, Am. Jour. Sci., Vol. 262, pp. 817-57.
BEATY, c. B. 1961. Topographic effects of faulting: Death Valley, California, Ann.
Ass. Amer. Geogrs, Vol. 5 1, pp. 234—40.
BECKINSALE, R. p. and RICHARDSON, L. 1964. Recent findings in the physical
development of the Lower Severn valley, Geog. Jour., Vol. 130, pp. 87-105.
BE ETE, J. B. 1 862. On the mode of formation of some of the river-valleys in the south
of Ireland, Quart. Jour. geol. Soc., Vol. 18, pp. 378-403.
BLANC, A. 1958. Repertoire bibliographique critique des hudes de relief karstique

en Yougoslavie depuis Jovan Cvijic, Cen. Doc. Cartog. Geog. Mem. et Doc., Vol. 6,
pp. 135-223.
BONYTHON, c. w. and MASON, B. 1953. The filling and drying of Lake Eyre, Geog.
Jour., Vol. 1 19, pp. 321-30.
BO WLER, J. M. and h arford, l. b. 1965. Quaternary tectonics and the evolution of
the riverine plain near Echuca, Victoria, Jour. geol. Soc. Aust., Vol. 13, pp. 339-
54.
BROWN, E. H. 1952. The river Ystwyth, Cardiganshire: a geomorphological analysis,

Proc. Geol. /4ss., Vol. 63, pp. 244-69.
BRUSH, L. M. and wolman, m. g. 1960. Knickpoint behaviour in noncohesive

material: a laboratory study. Bull. geol. Soc. Am., Vol. 71, pp. 59-74.
CHANG, p. 1964. Relationship of block faulting to stream development on both

flanks of Chin-Ling mountain regions Intnl Geol. Revs, Vol. 10(1968), pp. 1428-
39.
DURY, G. H. 1954. Contribution to a general theory of meandering valleys, Am.

Jour. Sci., Vol. 252, pp. 193-224.
DURY, G. H. 1959. A contribution to the geomorphology of Central Donegal, Proc.

Geol. Ass., Vol. 70, pp. 1-27.
dury, g. h. 1960. Misfit streams: problems in interpretation, discharge and distribu-

tion, Geog. Rev., Vol. 50, pp. 219-42.
dury, g. h. 1964. Principles of underfit streams, U.S. Geol. Stay. Prof. Paper 452A,
67 pp.
EK, c. 1961. Conduits souterrains en relation avec les terrasses fluviales, Ann. Soc.
geol. Belg., Vol. 84, pp. 313-40.
GAMS, 1. 1962. Slepe doline v Sloveniji; Geografski Zbornik, Vol. 7, pp. 265-306;
English summary. Blind valleys in Slovenia, pp. 301-4.
GEORGE, p. 1948. Quelques formes karstiques de la Croatie occidentale et de la
Slovanie meridionale (Yougoslavie), Ann. Geog., Vol. 57, pp. 298-307.

HOLMES, D. A. 1968. The recent history of the Indus, Geog. Jour., Vol. 134,
pp. 367-82.
HJULSTROM, F. 1949. Climatic changes and river patterns, Geografiska Annaler,

Vol.31,pp. 83-9.
JONES, 0. T. 1924. The Upper Towy drainage system. Quart. Jour. geol. Soc.,
Vol. 80, pp. 568-609.
JONES, o. T. 1965. The glacial and post-glacial history of the lower Teifi valley.
Quart. Jour. geol. Soc., Vol. 121, pp. 247—81.
LENSEN, G. j. 1968. Analysis of progressive fault displacement during downcutting
at the Branch River terraces, South Island, New Zealand, Bull. geol. Soc. Am.,
Vol. 79, pp. 545-56.
LEOPOLD, L. B. and MILLER, J. p. 1954. A post-glacial chronology for some alluvial
valleys in Wyoming, U.S. Geol. Sum. Water-supply Paper 1261, 90 pp.
LUMSDEN, G. 1. and DAVIES, A. 1965. The buried channel of the River Nith and its
marked change in level across the Southern Upland fault, Scot. Jour. Geol., Vol. 1,
pp. 134-43.
MILLER, A A 1935. The entrenched meanders of the Herefordshire Wye, Geog.

7o«r., Vol. 75, pp. 160-78.
PiTTY, A. F. 1965. A study of some escarpment gaps in the southern Pennines,
Trans. Inst. Brit. Geogrs, No. 37, pp. 127—45.
PITTY, A F. 1966. Landform studies in the Peak District of Derbyshire, unpubl.
D.Phil. thesis, Univ. of Oxford, 378 pp.
reeves, c. c. 1 966. Pluvial lake basins of west Texas, Jour. Geol., Vol. 74, pp. 269—
91.
reeves, c. c. 1968. Introduction to paleolimnology (Elsevier, Amsterdam), 228 pp.
rhodhamel, e. c. and carlston, c. w. 1963. Geologic history of the Teays

valley in West Virginia, Bull. geol. Soc. Am., Vol. 74, pp. 25 1-74.
rice, r. j. 1957. Some aspects of the glacial and postglacial history of the lower
Goyt valley, Proc. Geol. Ass., Vol. 68, pp. 217-27.
Ruppel, e. t. 1967. Late Cenozoic drainage reversal, east-central Idaho, and its
relation to possible undiscovered placer deposits, Econ. Geol., Vol. 62, pp. 648-
s ARK AR, s. K. and BASUMALLiCK, s. 1968. Morphology, structure and evolution

of a channel island in the Barakar river, Barakar, West Bengal, Jour. Sed. Petrol,
Vol. 38, pp. 747-54.
STRAW, A. 1963. The Quaternary evolution of the lower and middle Trent, East
Midld Geogr, Vol. 3, pp. 171-89.
STRICKLIN, F. L. 1961. Degradational stream deposits of the Brazos river, central
Texas, Bull, geol Soc. Am., Vol. 72, pp. 19-36.
Serko, a. and michler, l, trans. by N. Kuret, 1958, La grotte de Postojna et les
autres curiosites du Karsv, 2nd enl. ed. (Zavod Postojnska jama, Ljubljana),
191 pp.
TROLL, c. 1954. Uber Alter und Gildung von Tal Mandern, Erdkunde, Vol. 8,
pp. 286-302.
TUAN, Y. F. 1966. New Mexican gullies; a critical review and some recent observa-
tions, Aim. Ass. Amer. Geogrs, Vol. 56, pp. 573-97.
VAR ESC HI, V. 1963. Die Gabelteilung des Orinoco, Pet. geog. Mitt., Vol. 107,
pp. 241-8.
viTA-FiNZi, c. 1969. Late Quaternary alluvial chronology of Iran, Geol
Rundschau, Vol. 58, pp. 951-73.
viTA-FiNZi, c. 1969. The Mediterranean valleys. Geological changes in historical
time (Cambridge Univ. Press, Cambridge), 140 pp.
WHITE, w. B. and schmidt, v. a. 1966. Hydrology of a karst area in east-centra!

West Virginia. [Voter Res. Research, Vol. 2, pp. 549-60.
E. Stages in Glaciated Areas
BiRKELAND, p. w. 1964. Pleistocene glaciation of the northern Sierra Nevada,

north of Lake Tahoe, California, Jour. Geol, Vol. 72, pp. 810-25.
BORNS, H. w. and goldthwait, r. p, 1966. Late Pleistocene fluctuations of

Kaskawulsh glacier, southwestern Yukon Territory, Canada, Am. Jour. Scl,
Vol. 264, pp. 600-19.
CH ARLES WORTH, J. K. 1926. The readvance marginal kame-moraine of the south

of Scotland and some later stages of retreat, Trans, roy. Soc. Edinb., Vol. 55,
pp. 25-50.
CH ARLESWORTH, J. K. 1929. The South Wales end-moraine. Quart. Jour. geol.

Soc., Vol. 85, pp. 335-58.
DENTON, G. H. and STUivER, M. 1966. Neoglacial chronology, northeastern St.
Elias Mountains, Canada, Am. Jour. Scl, Vol. 264, pp. 577-99.
DOWN IE, c. 1964. Glaciations of Mount Kilimanjaro, northeast Tanganyika, Bull
geol. Soc. Am., Vol. 75, pp. 1-16.
DORY, G. H. 1953. A glacial breach in the north-western Highlands, Scot. geog.
Mag.,Vo\. 69, pp. 106-17.
DURY, G. H. 1957. A glacially breached watershed in Donegal, Irish Geog., Vol. 3,
pp. 80-171.
GROVE, J. M. 1966. The Little Ice Age in the massif of Mont Blanc, Trans. Inst. Brit.
Geogrs, Vol. 40, pp. 129-43.
jENNESS, s. E. 1960. Late Pleistocene glaciation of eastern Newfoundland, Bull,

geol. Soc. Am., Vol. 71, pp. 161-80.
JOHN, B. s. 1967. Further evidence for a Middle Wiirm interstadial and a Main
Wiirm glaciation of south-west Wales, Geol. Mag., Vol. 104, pp. 630-3.
LEIGHTON, M. M. and BRQPHY, J. A. 1966. Farmdale glaciation in northern Illinois
and southern Wisconsin, Jour. Geol, Vol. 74, pp. 478—99.
LINTON, D. L. 1951. Watershed breaching by ice in Scotland, Trans. Inst. Brit.

Geogrs,Wo\. 15 (1949), pp. 1—16.
LINTON, D. L. 1949 and 1951. Some Scottish river captures re-examined, Scot.
Geog. Mag., Vol. 65, pp. 123-31, and Vol. 67, pp. 31-44.
MCLENNAN, A. G. 1969. The last glaciation and deglaciation of central Lanarkshire,

Scot. Jour. Geol, Vol. 5, pp. 248—68.
MCKENZIE, G. D. 1969. Observations on a collapsing kame terrace in Glacier Bay
National Monument, south-eastern Alaska, Jour. Glaciol., Vol. 8, pp. 413-25.
MERCER, J. H. 1968. Variations of some Patagonian glaciers since the Late-Glacial,

Am. Jour. Sci., Vol. 266, pp. 91-109.
PENNY, L. F. 1964. A review of the Last Glaciation in Great Britain, Proc. Yorks,
geol. Soc., Vol. 34, pp. 387-411.
PORTER, s. c. 1964. Late Pleistocene glacial chronology on north-central Brooks
Range, Alaska, Am. Jour. ScL, Vol. 262, pp. 446-60.
PORTER, s. c. and denton, g. h. 1967. Chronology of neoglaciation in the North

American Cordillera, Am. Jour. Sci., Vol. 265, pp. 177—210.
SISSONS, J. B. 1967. Glacial stages and radiocarbon dates in Scotland, Scot. Jour.
Geol., Vol. 3, pp. 375-81.
sooNs, J. M. 1963. The glacial sequence in part of the Rakaia valley, Canterbury,
New Zealand, N.Z. Jour. Geol. Geophys., Vol. 6, pp. 735—56.
SYNGE, F. M. 1964. Some problems concerned with the glacial succession in south-
east Ireland, Irish Geog., Vol. 73-82.
5, pp.
szupRYCZYNSKi, J. 1968. Glaciations in the Spitzbergen area, Geog. Polonica,
Vol. 14, pp. 175-83.
TOTTEN, s. M. 1969. Overridden recessional moraines of north-central Ohio, Bull,
geol. Soc. Am., Vol. 80, pp. 1931-46.
viRKKALA, K. 1961. On the glacial geology of the Mameenlinna region, southern
Finland, Bull. Comm. geol. Finlande, No. 196, pp. 215—41.
VATEs, E. M. and moseley, f. 1967. A
contribution to the glacial geomorphology
of the Cheshire Plain,
Trans. Inst. Brit. Geogrs, Vol. 42, pp. 107—25.
F. Stages in Coastlino Evolution
ANDREWS, J. T. 1967. Problems in the analysis of the vertical displacement of

shorelines, Geog. Bull., Vol. 9, pp. 7 1-4.
BUFFINGTON, E. c. and MOORE, D. G. 1963. Geophysical evidence of the origin of

gullied submarine slopes. San Clemente, California, Jour. GeoL Vol. 7 1, pp. 356-
70.
COTTON, c. A. 1963. Levels of planation of marine benches, Zeit.f. Geomorph.,

Vol. 7, pp. 97-111.
DRISCOLL, E. M. and HOPLEY, D. 1968. Coastal development in a part of tropical

Queensland. Australia. Jo;/r. Tropical Geog.,'Vo]. 26, pp. 17-28.
FLEMMING, N. c. 1965. Form and relation to present sea level of Pleistocene marine
erosion features, Jour. GeoL, Vol. 73, pp. 799-8 1 1.
FLiNN. D. 1969. On the development of coastal profiles in the north of Scotland,

Orkney and Shetland. Scot. Jour. GeoL, Vol. 5, pp. 393-9.
FOLK, R. L. 1967. Sand cays of Alacran reef, Yucutan, Mexico; morphology. Jour.
Geo/., Vol. 75, pp. 4 12-37.
GLENiE, R. c.. SCHOFIELD, 1. c., and WARD. w. T. 1968. Tertiary sea levels in
Australia and New Zealand, Palaeogeog., Palaeoclim., PalaeoecoL, Vol. 5,
pp. 141-63.
GRES SWELL, R. K. 1964. The origin of the Mersey and Dee estuaries, GeoL Jour.,
Vol. 4,pp. 77-86.
GUILCHER, A. 1949. Aspects et problemes morphologiques du massif de Devon-

Cornwall, Rev. geog. alpine, Vol. 37, pp. 689-717.
JOHNSON, D. w. 1919. Shore processes and shoreline development (J. Wiley, New
York), 584 pp.
K ANEKO, s. 1966. Rising promontories associated with a subsidiary coast and sea-
floor in south-western Japan, Trans. Roy. Soc. N.Z., GeoL, Vol. 4, pp. 21 1-28.
KiDSON, c. 1952. Dawlish Warren: a study of the evolution of the sand spits across
the mouth of the River Exe in Devon, Trans. Inst. Brit. Geogrs., Vol. 16 (1950),
pp. 67-80.
KING, w. B. R. 1954. The geological history of the English Channel. Quart. Jour,
geol. Soc., Vol. 1 10, pp. 77-101.
KLiE WE, H. 1963. Fortschritte und Problem beider Erforschung der Kiistenentwick-
lung an der siidlichen Ostsee, Baltica, Vol. 1, pp. 116-36.
KOFOED, j. w. and gorsline, d. s. 1966. Sediments of the Choptank River,

Maryland, South eastern geology, Vol. 7. pp. 65-82.
LEWIS, w.v. 1938. Evolution of shoreline curves, Proc. Geo/. /Iss.. Vol. 49, pp. 107—
27.
MACNEiL, F. s. 1954. The shape of atolls; an inheritance from subaerial erosion

forms. Am. Jour. ScL, Voi. 252, pp. 402-27.
MITCHELL, G. F. 1960. The Pleistocene history' of the Irish Sea. Adv. Sci., Vol. 68,
pp. 313-25.
NAK AG AVVA. H. 1965, Pleistocene sea lev'els along the Pacific coast of Japan, Sci.
Rep. Toboku Univ., Ser. 2 {Geol.). Vol. 37, pp. 451-62.

ORME, A. 1962. Abandoned and composite seacliffs in Britain and Ireland, Irish
Geography, Vol. 4, pp. 279-91.

PONS, L. J., JELGERSMA, S., WIGGERS, A. J., and DE JONG, J. D. 1963. Evolution
of the Netherlands coastal areas during the Holocene; Verh. Kon. Ned. Geol.
Mijnbk. Gen. Geol., Ser. 21-2, pp. 197-208.
ROBINSON, A. H. w. 1966. Residual currents in relation to shoreline evolution of the
East Anglian coast, Marine Geology, Vol. 4, pp. 57-84.
RUSSELL, R. J. 1967. Aspects of coastal morphology, Geografiska Annaler,

Vol. 49A, pp. 299-309.
SCHLANGER, s. O. et ol. 1963. Subsurface geology of Eniwetok Atoll, U.S. Geol.

Sun'.. Prof. Paper 260-b, 75 pp.
SCHWARTZ, M. L. 1968. The scale of shore erosion. Jour. Geol., Vol. 76, pp. 508-
17 .
SHIER, D. E. 1969. Vermetid reefs and coastal development in the Ten Thousands
Islands, southwest Florida, Btdl. geol. Soc. Am., Vol. 80, pp. 485—508.
SISSONS, J. B. 1963. Scottish raised shoreline heights with particular reference to the
Forth valley, Geografiska Annaler, Vol. 45, pp. 180-5.
STEERS, J. A. et al. 1939. Recent coastal changes in south-eastern England: a

discussion, Geog. Jour., Vol. 93, pp. 399-419 and 491-512.
STEPHENS, N. 1958. The evolution of the coastline of north-east Ireland, Adv. Sci.,
Vol. 56, pp. 389-91.
STOOD ART, D. R. 1965. British Honduras cays and the low wooded island problem,
Trans. Inst. Brit. Geogrs, Vol. 36, pp. 131-47.
TANNER, w. F. 1962. Reorientation of convex shores, Am. Jour. Sci., Vol. 260,
pp. 37-43.
VARJO, u. 1964. Ober finnische Kiisten und ihre Entstehung, Fennia, Vol. 91, No. 2,
104 pp.
VERSTAPPEN. H. T. 1954. The influence of climatic changes on the formation of

coral islands, Am. Jour. Sci., Vol. 252, pp. 428-35.
WILLIAMS, w. w. 1956. An east coast survey: some recent changes in the coast of
East Anglia, Geog. Jour., Vol. 122, pp. 3 1 7—34.
WILLIAMS, w. vv. 1960. Coastal changes (Routledge and Kegan Paul, London),
220 pp.
ZENKovicH, V. p. 1962. Osnovii ucheniva o razvitsi toskich beregov, (Academy of

Sciences, Moscow), by O. G. Fry and edited by
7 10 pp.; trans. J. A. Steers, 1967,
Processes of coastal development (Oliver and Boyd), 738 pp.
6. Stages involving more than one process and in the Evolution of the
Landsurface as a whole
B ALCHiN, w. G. V. 1952. The erosion surfaces of Exmoor and adjacent areas, Geog.
Jour., Vol. 11 453-76.
8. pp.
battistini, r. 1966. Le littoral du paleokarst de la presqu’ile, de Narinda au Nord

de Majunga (Madagascar):
un example de cote contraposee. Bull. Ass. geog. Fr.,
Nos. 346-7, pp. 41-9.
BAULiG, H. 1928. Le plateau central de la France et sa bordure Mediterraneenne,

etude morphologique (Armand Colin, Paris), 589 pp.
BROWN, E. H. I960. The relief and drainage of Wales (Univ. Wales, Press), 186 pp.
BRUNSDEN, D., KiDSON, c., ORME, A. R., and vvATERs, R. s. 1964. Denudation
chronology of parts of south-western England, Field Studies, Vol. 2, pp. 1 15-32.

CLAYTON, K. M. 1953. The glacial chronology of part of the middle Trent basin,
Proc. Geol. Soc., Vol. 64, pp. 198-207.
DIKE Y, F. 1956. Erosion surfaces in Africa, Trans, geol. Soc. S. Afr., Vol. 59, pp. 1-
16.
DURY, G. H. 1955. Diversion of drainage by ice. Science News, Vol. 38, pp. 48-71.
EMBLETON, c. 1964. Sub-glacial drainage and supposed ice-dammed lakes in north-
east Wales, Proc. Geol. Ass., Vol. 75, pp. 3 1-8.
EVERARD, c. E. 1957. Erosion platforms on the borders of the Hampshire basin,

Trans. Inst. Brit. Geogrs, No. 22, pp. 33—46.
FRYE, J. c. and LEONARD, A, B. 1957. Ecologica! interpretation of Pliocene and

Pleistocene Stratigraphy in the Great Plains region. Am. Jour. ScL, Vol. 255,
pp. 1-11.
GEORGE, T. N. 1955. British Tertiary landscape evolution, Sci. Prog., Vol. 43,
pp. 291-307.
GEORGE, T. N. 1966. Geomorphic evolution in Hebridean Scotland, Scot. Jour.
Geol., Vol. 2, pp. 1-34.
GREGORY, K. J. 1965. Proglacial lake Eskdale after sixty years, Trans. Inst. Brit.
Geogrs, Vol. 36, pp. 149-62.
HOUGH, J. L. 1966. Correlation of glacial lake stages in the Huron-Erie and

Michigan basins. Jour. Geol., Vol. 74, pp. 62-77.
IVES, J. D. 1960. Former ice-dammed lakes and the deglaciation of the middle
reaches of the George River, Labrador-Ungava, Geog. Bull., No. 14, pp. 44-88.
JAMIESON, T. F. 1865. On the history of the last geological changes in Scotland,

Quart. Jour. geol. Soc., Vol. 21, pp. 161-203.
JESSUP, R. vv. 1961. A Tertiary-Quaternary pedological chronology for the south-

eastern portion of the Australia arid zone. Jour. Soil Sci., Vol. 12, pp. 199-213.
JOHNSON, R. H. and RICE, R. J. 1961. The geomorphology of the south-west Pennine

upland, Proc. Geol. Ass., Vol. 72, pp. 21-32.
JOHNSON, R. H. 1965. The glacial geomorphology of the west Pennine slopes from
Cliviger to Congleton, in J. B. Wbittow and P. D. Wood (Eds), Essays in geo-
graphy for Austin Miller (Univ. of Reading, Reading), pp. 58-93.
KIRBY, R. p. 1969. Morphometric analysis of glaciofluvial terraces in the Esk
basin, Midlothian, Trans. Inst. Brit. Geogrs, No. 48, pp. 1-18.
LEWIS, p. F. 1960. Linear topography in the southwestern Palouse. Washington-
Oregon, Ann. Ass. Amer. Geogrs, Vol. 50, pp. 98-1 1 1.
MANNERFELT, c. M. 1945. Nagra glacialmorfologiska formelement, Geografiska

Annaler, Vol. 27, pp. 3-239.
MACAR, P. 1965. Apercu synthetique sur revolution geomorphologique de

I’Ardenne, La Geographie, Vol. 17, No. 1, 1 1 pp.
MITCHELL, G. F. 1963. Morainic ridges on the floor of the Irish Sea, Irish Geog.,
Vol. 4, pp. 335-44.
NICHOLS, R. L. 1968. Coastal geomorphology, McMurdo Sound, Antarctica, Jour.

Glacial., Vol. 7, pp. 449-78.
QUENNELL, A. M. 1958. The structural and geomorphic evolution of the Dead Sea
rift. Quart. Jour. geol. Soc., Vol. 1 14, pp. 1—24.
ROBINSON, A. H. w. 1968. The submerged glacial landscape off the Lincolnshire
coast, Trans. Inst. Brit. Geogrs, No. 44, pp. 1 19-32.
SERET, G. 1965. La succession des episodes fluviatiles periglaciaires et fluvio-

glaciaires a I’aval des glaciers, Zeit.f. Geomorph., Vol. 9, pp. 305-20.
STRAW, A. 1961. Drifts, meltwater channels and ice-margins in the Lincolnshire

Wolds, Trans. Inst. Brit. Geogrs, Vol. 9, pp. 115-28.
STRIDE, A. H. 1959. On the origin of the Dogger Bank, in the North Sea. Geol. Mag.,
Vol. 96, pp. 33-44.
sweeting, m. m. 1963. Erosion cycles and limestone caverns in the Ingleborough
district, Geog. Jour., Vol. 115, pp. 63-78.
TOMLINSON, M. E. 1963. The Pleistocene chronology of the Midlands, Proc. Geol.

Vol. 74, pp. 187-202.
TODD, T. w. 1968. Dynamic diversion: influence of longshore current-tidal flow
interaction on chernier and barrier island plains. Jour. Sed. Petrol., Vol. 38,
pp. 734-46,
WALKER, G. p. L. and BLAKE, D. H. 1966. The formation of a palagonite breccia

mass beneath a valley glacier in Iceland, Quart. Jour. geol. Soc., Vol, 1 22, pp. 45-
61.
WOLFE, p. E. 1964. Late Cenozoic uplift and exhumed Rocky Mountains of central
western Montana, Bull. geol. Soc. Am., Vol. 75, pp. 493-502.
WOOLDRIDGE, s. w. 1950, The upland plains of Britain: their origin and geo-
graphical significance, Adv. Sci., Vol. 7, pp. 1 62-75.
WOOLDRIDGE, s. w. and LINTON, D. L. 1955. Structure, surface and drainage of
south-east England (Philip, London), 1 76 pp.
ZAKHAROV, Y. F. 1966. The age of relief in northern, subpolar and polar Urals,
Intnl. Geol. Rev., Vol. 9 (1967), pp. 1025-7.
1 1
Subject Index
321 Ant mounds, 151, 283

gradient, 362
Ablation, 122, 295; Antecedent drainage, 47-8,
Abney 402
level, 256
270;solut.on, Anticline, 100, 221,
Lasion,174, 176,261,264, Anticlinorium, 100
Ants, 151
246, 281, 354
AccSion, 236, 238-9, 242, Applied geomorphology,
35-7
rates, 321; estuaries, 159
Accretion Aragonite, 288
Accumulation, 168; rates,
44
Arenicola, 151, 237
rocks, 146, 190, 195
Acid crystalline Argillation, 147, 173,222
Acid environment, 290-1 Arroyos, 157
Acid reaction, 161 dacitic, 190
Ash,
Acidity, 142, 144-5, 146
Ashfalls, 72, 87
Adsorption, 142, 222
Ash tree, 332
Aerobic conditions, 145,
161 307-13
Aspect, 123, 226-7,
African cycle, 84 307-9
Asymmetry, 7,
Aggradation, 277, 297-8, 372
335 Atmospheric vapour, 153
Air photos, xiv, 246, 272, 273,
Atolls, 243—4, 313
Alders, 252
Atterberg limits. 137
Alectoria miniscula, 333
Augite, 146
Algae, 159, 164, 194, 233,290,339
Auto-assechement, 7
Alkaline environment, 29
Avalanche chute, 95 . , , i u
Alkalinity, 145 301, 311; slush,
Avalanches, 125, 216,
Allerod, 330
216
Allogenic streams, 115, 367, 369
131, 345,
Alluvial, channels, 117; deposits,
260, 261, Backwash, 120, 207
361; fans, 228, 258, 259,
Bacteria, 147, 161,
164
339,
262-3, 267, 275, 281, 299, 311,
Back-weathering, 114
360-1, 400; plains, 230, 388
400 Badlands, 27, 208, 346
Allmtiation, 229, 363, 387-9, 392,
Bahadas, 229, 236
Alluvium, 1 17, 174, 227, 260, 263, 363
Bamboo, 165
Altiplanation, 53 255, 2
Bank, erosion, 210, 231,
,
Altitude, 45, 47, 57, 91—2, 123,

125, 157,
316; recession, 115
158,253, 298-301,341, 367 277
Bankfull stage, 276,
Alumina, 145, 147, 164,222, 288-91 279-8
Barchans, 6, 8, 245-8, ,
Aluminium, 141, 146, 231

Barnacle limits, 335
Amphibole, 144 231 ; beaches, 386
Barrier, geochemical,
Amphriirile ornata, \52
coastal lagoon, 386-7, 388 ; island, 12,
Anaerobic, environment, 145; fermentation, rock, 47-8, 174, 175,
241, 303, 386, 389;
161
179,264,303
Anamorphism, 290 longshore, 238-9; olT-
Bars, 213, 240-1;
Angle of repose, 126, 136 240; river, 225,
chnre. 286; plunge-pomt,
Animals, 147
356-
234, 275, 279; spool-, 232, 281; sub- Buried, channels, 361-2, 382; relief, 343,
merged, 343 7
Basalt, 56, 87, 93, 107, 186, 191, 192, 193, Burning, 60, 78, 170, 322-3
195, 197, 268, 283, 305, 366, 393-4
Base-exchange, 142-3 Cailleux indices, 129, 132-3
Base-level, 42-3, 63, 64, 70, 367, 393; fall Calcareous Grit, 52
•"
in, 197,354 Calcareous rocks, 144, 350, 368

Basement, 342, 343, 359 Calcium, 141, 142, 146, 164, 165, 188-9,
Basic igneous rocks, 146 190, 191, 197, 222, 224, 286, 291;
Batholith, 82, 107,347 carbonate, 144, 161-2, 173, 188, 189,
Bauxite, 289, 326 255-6, 258, 287, 315, 353, 367
Bays, 267 Caldera, 88, 89
Beach, arctic, 170; cusps, 77, 201, 285; Caliche, 186, 267-9, 287, 294, 337, 346
deposits, 131, 134, 135, 269, 339; emer- Calluna vulgaris, 300
gent, 386; plan, 282; processes, 202, Caloplaca, 147
204-5, 209, 224, 254, 272-4, 280-2, Cambering, 52-3, 101
285, 293, 321; ridge, 238-41; rock, 59, Capillary movement, 137, 139
164, 286, 288, 336; sands, 129, 130, 260; Cap-rock, 101, 107, 108, 219, 268, 287,
step, 237, 240 345-8
Beaches, 70, 127, 133, 237, 239-41, 267,
357-
Carbon, 145; dioxide, 114, 143, 144, 145,
269, 272-3, 286; bay-head, 241 161-3, 233, 291, 324
Beavers, 324 Carbonates, 231. 324, 337, 348
Bedding plane, 268 Carbonation, 143, 144, 173
Bedload, 131, 174, 203-6, 217, 229, 235, Carbonic acid, 143, 144, 162
236, 359, 261, 264;amounts, 210 Carboniferous Limestone, 105, 164, 350,
Bergschrund, 181 8
Berm, 237-9, 240, 281,285 Catastrophic changes, 37-42
Bicarbonate, 233, 318 Catena, 35
Bifurcation ratio, 273, 274 Cation, exchange, 143, 145, 294; capacity,
Biological, activity, 286, 287, 292, 300, 223, 224
350; cycle, 165-6; productivity, 155, 156, Cations, 142, 160, 223,290-1
308 Cave pool, 255
Biomass, 148, 155, 159, 160, 164, 170 Caves, 67, 175, 196, 233, 367-9
Biostasis, 292, 352-3 Cavitation, 175, 184
Biotite, 144, 147 Cementation, 244, 287-8
Birch scrub, 325 Ccrrados, 151
Blind valley, 369-70 Chalk, 103, 106, 304; chalk, 28, 71, 106,
Block, field, 78, 93, 131, 209, 270, 347-8; 107, 185, 191,367,368,369
terrace, 252-3 Chalk cover, 358
Block-glide landslide, 97, 219-20, 347-8 Changes with time, 218, 313-21, 341;
Blocks, 254, 265, 318, 355; perched, 351 annual, 124-5, 168; diurnal, 123, 315;
Bornhardt, 198-9, 294 seasonal, 123, 158, 161-2, 209, 234, 277,
Borrowdale volcanic series, 181 281, 315-19,353-3
Boulder, clay, 131, 135, 173, 177, 196,228, Channel, bottom, 157, 158, 159, 183, 216;
248, 389, 398; clog, 274, 279; field, 270; cross-secdon, 277-9; fork development,
-weathering ratios, 338-9 274-6; roughness, 159, 277
Boulders, 131, 148, 157, 177, 195, 206, Channels, vegetated, 157-8
214-15, 216, 217, 228, 237, 249, 265, Chaparral, 154
267, 270, 294, 338-9, 355, 393; gliding, Chara, stonewort, 1 64
174,203 Chelators, 160-1
Braided streams, 58, 225, 228, 274-9, 364 Chemical weathering, 144, 259
Braiding, 236 Cherniers, 389
Braking effect, 157, 158-9 Chernozem, 137, 308, 312
1 1
Subject Index 49
Chert, 108, 189 Corestones, 190-1, 294
Chloride, 313, 318-19 Correlations, 9, 24-6, 254
Chlorine, 164 Cottonwoods, 167, 331

Chlorite, 107 Counter-dip breaching, 365-6
Cirque, 66, 181-2, 195, 299-300, 309-11 Covered karst, 105
Clay, 93, 106, 108, 137-8, 142, 170, 220, Cover rock, 358; reconstruction of, 357-8
223, 228, 230, 236-7, 244, 248, 258, Crabs, 152, 391
259, 267, 271, 277-9, 283, 285, 294, Crag-and-tail, 250-1, 303, 305
369; alteration, 340; degradation, 290; Crevasse, deposits, 306; fillings, 252; traces,
formation, 143, 146, 290-2; minerals, 74, 184
137, 143,222, 259; pan, 285 Crevasses, 184, 251-2, 253
Clay-with-flints, 5 Cross-strata, 224, 231, 244
Cliffs, 148, 149, 175. 196, 197, 216, 265, Cross- valley moraines, 128
318, 347, 349 Crusts, 45, 267, 345-6; calcareous, 286-7,
Climatic change, 25, 57, 63, 87, 298, 321—6, 353, 396; ferruginous (ferricrete), 286,
337, 345-6, 352-3, 369, 372, 373, 375, 288; gypseous, 286, 396, 400
399^00 Crustal, movement, 42-8; subsidence, 43;
Climatic geomorphology, 54-60, 76 uplift, 44
Climax vegetation, 166 Cryoplanation terrace, 53, 200
Clints, 196-7 Cumulative frequency curves, 128
Closed systems, 282 Current crescent, 304
Clymenella, 151 Cuspate foreland, 242
Coastal, areas, 158, 213; classification, 7; Cusps, 59
cliffs, 2, 67, 150, 176-7, 240,
296-7, 35 1, Cut and fill, 13,281
354-5, 381-3; erosion, 40, 176-7; plain, Cycle of Erosion, 7, 37, 42-3, 48-54, 60-4,
44, 61, 238-9, 241, 383; sediments, 151, 66, 68, 74, 398-9
355, 359
Coastline, evolution, 380-9 1 submerged, 343
; Dating, 327—42
Coasts, tropical, 150, 186 Debris control, 202-3, 259, 292-5, 345,
Cobbles, 131, 133, 157,339 346; mantle, 264, 296
Cockpit karst, 194 Decalcification, 190, 192
Cohesion, 138, 154 Deep-sea, cores, 324; sediments, 328
Colloidal, properties, 154; systems, 142 Deflation hollow, 296
Colloids, 222, 236 Deforestation, 169-71
Colluvial deposits, 131, 345, 354 Deglaciation, 249, 374-9, 391
Colluvium, 227, 232 Deltas, 72, 77, 86, 134, 166, 209, 233-6,
Colonizers, 333 244, 275, 280, 302, 303, 387-9
Comet-tails, 303 Dendrochronology, 331
Competence, of rocks, 53, lOl, 197, 205, Denudation, 114, 125, 172, 174, 191, 223,
219; of transporting agencies, 208, 210, 290, 299; chemical, 160, 187-8, 199,
214,224, 236, 241,297 223; chronology, 33, 401
Concretions, 164, 336 Denudational loss, 165
Cone-karst, 199
Deposition, 142, 157, 172, 224-54, 305
Conglomerate, 67 Depositional landforms, 224-54; dynamic,
Consequent streams, 357-8 225
Constrictions, 305 Desert, 143, 155, 160, 186, 187, 193, 214,
Continental drift, 81-2, 84-6, 324
244-8, 267, 269, 281, 289, 316; pave-
Continental, interiors, 209; rise,
342; shelf, ments, 193, 270; varnish, 270, 348
44, 244, 302,
342-3, inner, 238-9 Desiccation, 58, 95, 137, 139, 161, 164,
Com-ergence, 12 59 , 209, 259, 283, 285,319,323
Corals, fragments,
59; islands, 55, 164; ram- Differential, erosion, 95, 345; lowering,
parts, 321; rubble,
313; sand, 240 51-3, 347; resistance, 91, 92, 200;
Corallian, 95
weathering, 109, 197
1 1

Diorife, 107, 186 Eocene sands, 368
Dip diagram, 134 Eolianite, 287-8, 337
Dissolved load, 164, 222—4; solids, 114, Eolian processes (see wind action)
160,412 Epeirogenic movements, 43, 47, 83
Diversions, river, 298, 345, 360-7, 398, Ephemeral streams, 115, 158, 279
400; by ice, 392-3 Epidote, 339
Divides, 17, 358, 360, 365, 392 Equations, 25-9
Dolerite, 93, 107 Equilibrium, 57, 70-1, 257, 280, 282, 290,
Dolines, 305, 349-51, 369-70 292, 296, 353, 389; chart, 280; profile of,
Dolomite, 107, 167, 253 281
Domes, rock, 178, 195, 294, 346-7 Erosion, 172, 174-84; amounts, 167;
Donner Lake glaciation, 373 bedrock, 174; fluvial, 98, 101, 174, 179,
Downcutting, 47-8, 53, 356, 361, 363, 364; 354; glacial, 174, 177-84, 200,215,248,
rates, 47-8, 53, 175 372-3; lateral, 176; marine, 176-7, 286,
Downdip, breaching, 365-6; shift, 108, 11 1. 354; pre-glacial, 180, 182; rates, 87, 299,
176,363-6 329, 41-2, 60, 74,
artificial acceleration,
Downholland silt, 385-6 154, 158, 321, 322-3, 358; sub-glacial,

Downpours, 1 16, 136, 220, 230, 299 183-4; surfaces, 44-5, 51-4, 73, 344,
Drainage basin shapes and parameters, 4-6, 400
25-8, 65, 272-4 Erosional history, 286
Drainage density, 169, 369, 371 Erratics, 214-15
Drainage diversions, glacial, 175 Escarpments, 150, 194, 268, 304, 311,
Drainage networks, 169 347-8,356-7
Drainage system, evolution, 355-72; rever- Esker. 7, 252, 303, 378-9, 392, 393
sals, 364, 371; underground. 350; wind- Estuaries, 158, 160,236-9,387-9
aligned, 371 Estuarine, Clay, 111; flats, 236-7; marshes,
Drumlins, 6, 8-9, 18, 246, 253-4, 378, 393 236-7, 24
Dry valley, 7, 28, 51, 271, 367, 369 Evaporation, 56, 58, 122, 186, 287-8, 319
Dunes, 2, II, 71, 131, 134, 160, 209, 213, Evaporites, 102, 103, 105
227, 240, 244-8, 279-81. 306. 321, 338, Evapolranspiration, 56, 114-16, 153, 160,
353, 371, 383, 385, 391; gypsum, 259 169, 170, 191,231,299,302, 308
Dust, 1 1, 83, 128, 186, 214, 245, 305, 324 Evergreen oaks, 389
Dynamic, geomorphology, 15-16; equilib
rium, 13, 63-4 Fabric diagram, 134, 266
Fans, 208
Early-Pleistocene times, 363 Farmdale glaciation, 1 80; loess, 377
Earth hummocks, 283 Faults, 98-9
Earth movements (see also tectonic move- Faulting, 229, 360-1, 364
ments), 79-84 Fault-line scarp, 99
Earth’s, crust, 79-81, 88; mantle, 79-80, 88 Feedback mechanisms, 7
Earthflow (see also soilflow), 106, 138, 217, Feldspar, 92, 143, 146, 173, 188, 190,263,
226 264, 268
Earthquakes, 40-1, 83-4, 88, 101, 221, 350 Felsite, 192
Earthworms, 147, 150-1 Ferric oxide, 59, 144, 222, 288, 291, 338,
East Anglian tills, 134-5 340
Ebb-flood channels, 239 Ferricrete (see also crusts, ferruginous),
Electron, micrograph, 132; microscope, 134 202-3, 265, 270, 288
Elephants, 151 Ferrous iron, 144, 291
Emergence rates, 45-6 Ferruginization, 289
Enclosed depressions, 105, 194-7, 350-1 Festuca vivipara, 300
Enrichment, residual, 286, 289; secondary, Fetch, 119, 176, 282, 307,313
285 Field work, 29-33, 236
Enteromorpha sp., 159 Fiords, 98, 111, 179-80, 182, 305,376
1 1
Subject Index 493

Firn, 121
Geyser, 90
Firs, 326 Gibbsite, 29
Fissility, 93—4, 106 Gilgai, 58, 285

Fissures, rock, 93, 95, 96-8, 191-2, 340, Gipping Advance, 10
micro-, 184, 190; soil, 71, 106, 137, 139, Glacial, advances, 298, 378; boulders, 131;
202, 291,353 deposition and depositional landforms,

Flandrian Transgression, 362 131, 134, 248-54, 277, 373-80, 393;
Flocculation, 236 erosion, 98; retreats, 324, 377-9; trans-
Flood, deposits, 135, 232—3; plains, 167, port, 131, 263

229-31, 236, 276, 332; training, 167 Glaciated valley, 1 1
Floods, 14, 38, 116-17, 158, 197, 206, 208, Glaciations, 255, 325-6
210, 230, 233, 237, 254, 259, 276-7, Glacier {see also ice), accumulation, 310; ali-
279, 323,331-2, 391 mentation, 311; flow, 15; regime, 122;

Flotation, 207 surge, 83
Flow, coefficient, 114-16; lines, 302, 305 Glaciers, 24, 112, 121-5, 309-12, 343
Fluvioglaciai, fans, 135; deposits and deposi- Glaciomarine deposits, 224
tional landforms, 205, 252, 271, 282, 315, Gneiss, 96, 98, 101, 107, 128, 185, 191-2,
375 197, 221,264, 294,345,359
Folds, 100, 108 Goethite, 144
Foreset bed, 134, 235,238-9 Gondwana, 84, 86; surface, 49
Forest, 26, 112, 148, 150, 153, 157; coni- Gorse tussock, 304-5
ferous, 148, 164; deciduous, 150, 155, Grabens, 45, 82-3, 99
165; humid temperate, 151; litter, 155; Graded, bedding, 228; stream, 31, 297-8
oak, 153, 165; tropical, 154, 165, 170, Gradient, 119, 123, 138, 197, 222, 270-1,
189, 194, 209; zones, 116, 154 300-2; beach, 240; reverse, 194, 222,
Fossil, fauna, 331; flora, 44, 331; soils, 127, 252; shore, 176; stream, 27, 119, 157,
340 158, 236, 264, 277, 297-8, 363, 372;
Fracture, pattern, 93, 179; planes, 263 valley, 230
Fractures, 110-11 Grain-size, rocks, 92-3
Fracturing, 95, 342 Granite, 91, 96, 97, 107, 109, 143, 146, 175,
Freeze-thaw, 112-13, 125, 181, 184-6, 176, 185, 189, 190-2, 193, 197, 209,
195, 199, 218-19, 277, 307, 313, 315, 221, 257, 264, 270, 294, 301, 305, 345,
316, 318; cycles, 32, 229 347,359,410
Friction, 113, 125-6, 138, 215, 217 192,259,410
Granodiorite,
Frictional resistance, 1 18, 277 Grass, 42, 159, 300; dune, 159; mat, 157,
Frost, 115, 217; action, 106, 129, 131, 133, 283; rafts, 237; savanna, 165
139, 184-6, 197, 261, 292; creep, 218- Grassland, 26, 112, 151, 153, 160,314
19; heaving, 156; penetration, 113, 139- Gravel, 31, 107, 135, 152, 201, 208, 211,
40, 156-7, 218, 321; splitting, 270 228, 229, 230, 239, 250-1, 261, 263,
Frozen, ground, 139, 156, 282; soils, 139- 269
41, 156, 283^ Gravitational spreading, 82
Froude number, 118 Gravity, 40, 100, 101, 118, 135, 138, 151,
Fucus, 158 216, 217, 281; fall, 174, 216, 221, 227,
264-5, 346, 349; sliding, 244
Garnet, 146 Greywacke, 272
Gault Ciay, 101 Gritstone, 194, 268
Gels, 142, 290 Groundwater, 90, 188, 237, 288, 289, 290,
General Systems Theory, 70 297,319
Geographical cycle, 49 Grus, 190, 192,294, 339
Geomorphological mapping. 7-8, 35-6 Gulf Stream, 57, 67, 120
Geosynclines, 81-2 Gullies, 170, 208, 319, 354, 358
Geotectonics, 79 Gullying, 171, 301, 308,309
Geothermal heat, 124 Gypsum, 259
Hamada, 293, 396 Ice movement, 77, 123-4, 177, 270, 303;
Hanging valley, 309 intergranular, 123; plastic deformation,
Hardpan, 284, 288-9 123; velocity, 124-5
Head, 225-7, 266 Ice-pushed ridges, 59, 205, 215, 248, 306
Headlands, 28, 59, 241, 267, 274, 313 Ice-recession, 125, 252, 333, 374, 378-9
Headward erosion, 17, 358-9 Ice-sheets, 38, 42, 112, 177, 180, 181, 214,
Heavy minerals, 146, 264, 268, 339-40, 410 215, 230, 251, 304, 323-5; downwasting,
Helicoidal flow, 14, 118, 276-7 379; Keewatin, 178
HemaUte, 143, 144, 146, 193,289 Ice-wedges, 15, 139, 140, 282, 283, 285
Hercynian folding, 49 Icefield outwash, 338
Hinge fault, 99 Illinoian drift sheet, 249; glaciation, 338,
Hjulstrom effect, 201, 231, 236 377; glacier,392
Hog’s-back ridges, 103 Illite, 138, 223,290-1,315
Holocene time, 323 Illuvial horizon, 137
Homo ereclus. Early man, 168 Ilmenite, 146
Homologies, 12, 59, 202-3 Index of flatness, 131, 133; wear, 131, 133
Hornblende, 147, 339-40 Induration, 287-8; layers, 289
Horst, 99 Inferior Oolite, 264
Hortonian analyses, 26-8, 272-3 Infiltration, 228, 302, 317; capacity, 23,
Human activity, 165, 168-71,211,296,321 116, 153, 217, 295; rate, 153, 156, 308,
Human records, 72, 87, 336; ancient Indian 371
sites, 323; Norse colonies, 323; pre- Initial form, 342-4; surface, 345
historic, earthworks, 191, monument, 194; Injections, 139
Roman, camp site, 289, signal station, 355 Insects, 270
Humic acids, 166 Inselberg, 98, 109, 193, 195, 271, 290
Humification, 144, 155 Interception, 153, 156, 169
Humus, 139, 154, 155, 157, 165, 169, 170 Interglacials, 46, 47, 199, 315, 325-6, 354
Hurricanes, 39, 177, 302 Intermittent streams, 260, 263, 347
Hydration, 143, 146, 173, 188, 194,263 Intertidal, areas, 159, 177, 236-7, 275, 276,
Hydraulic lift, 1 74 277, 303; flats, 387
Hydrogen ion, 143, 144-5, 161,291 Interstralal solution, 105, 339, 357
Hydrolysis, 143, 144, 173, 176, 188,222 Involutions, 139
Hydrothermal activity, 89-91 Ionic potential, 141
Hydroxides, 145,261,264 Ionium, 328
Hydroxyl ion, 145 Ions, 222-3, 290,318
Hypersthene, 147 Iowan, drift, 379; loess, 379
Hypsithermal, 255, 324, 375 Iron, 141, 142, 144, 146, 161, 162, 164,
Hythe Beds, 108 222, 231, 267, 289, 291; hydroxides, 291,
294; oxide (see ferric oxide)
Ice, 112, 113, 172, 174,206, 215-16,218, Ironstone, 192, 354
248, 249, 251, 294-5, 303; accumulation, Isostasy, 34, 80
122, 321; buried, 253;drift, 215;floating, Isostatic recovery, 34, 45, 81, 382, 386, 399
224, 237; ground, 227; snowdrift, 306; Itabirite, 289
stagnant, 184, 249, 251-2, 303
Ice carapace, 124 Joints, 93-6, 106, 107, 109, 111, 175, 192,
Ice-contact deposits and depositional land- 196-7, 199, 263, 338,347
forms, 230, 252 Jokulklaup, 391
Ice-disintegration ridges, 25 1-2 Jurassic times, 361
Ice-drag, 205, 215
Ice-fronts, 125,251,377 Kame terraces, 250-2
Ice-jams, 1 17, 332 Karnes, 252, 378—9
Ice-marginal features, 378-9; lake, 252, 391; Kansan, sediments, 336; till, 340
ridge, 250-1 Kaolin, 90
Subject Index 495
KaoUnite, 106, 138, 143, 222-3, 258, 259, Lee-side effect, 177—8, 302—6; climatic,
283, 290-1,340 308-9; lense, 250-1

Karren, 196-7 Levees, 231-3, 234,275
Karst, 305, 350-1; bare, 350; covered, 350; Lichen, 147, 161, 333
perched, 367; ponded, 367; wells, 193—4; Lichenometry, 331, 333
windows, 369-71 Limestone, 63, 67, 97, 98, 106, 107, 128,
Katamorphism, 290 162, 173, 175, 177, 191, 193-7, 258,
Kegelkarst, 199 264, 267, 296, 347, 350, 353, 366-9;
Kettleholes, 12, 252, 253 coral, 176; pavement, 200
Kluane, drift, 303; glaciation, 214; outwash, Limoni te, 143, 144
338 Lineaments, structural, 61, 73, 103, 342
Knickpoint, 179, 359 Linear, erosion, 63, 346; dissection, 346
Kyanite, 147 Liquid limit, 106, 138, 216
Litter, accumulation, 155; fall, 156, 165;
Labradorite, 146 layer, 157, 165, 169,322
Laccolith, 108 Load, 157, 259, 260, 318, 321; saltating,
Lag-deposit, 261, 270, 293; channel, 297; 174, 187

gravel, 129, 377 Lobes, 217, 225; solifluction, 218, 220
Lagoon, 239, 241, 285-6, 305, 386-7 Local base-level, 53, 345, 353, 364
Lakes, 27, 67, 1 13, 156, 162, 163, 188, 215, Loess (see also dust), 128, 137, 190, 201,
226, 230, 251, 252, 256-7, 261, 313, 21 1, 214, 244-5, 305, 377, 379
317, 324, 326, 359-62, 366, 367, 372; Log raft, 40
enclosed, 321; floors, 282, 353; glacial, Longitudinal profile, 28, 91, 111, 178, 179,
377-9, 391-2; sediments, 353, 361 262-3, 271, 273, 297, 398-9; ‘up-and-
Laminar flow, 1 17, 125-6 down’,1 84
Landslides, 7, 41, 101, 137, 169, 176, 178, Longshore, currents, 305; drift, 211, 241,
216, 219-21, 225, 228, 302, 311, 315, 267, 281, 285,389
347-9, 355; dams, 398-9; scars, 42, 95, Lower Greensand, 108
221 Lowestoft Advance, 10
Landslips (see landslides) Lunettes, 259
Last Glaciation, 361
Late Blancan times, 380 M discontinuity, 80, 88
Late-Cenozoic surfaces, 340; upwarping, Magma, 89, 90, 393
342 Magnesian Limestone, 95, 148-9
Late-glacial times, 43 Magnesium, 141, 142, 146, 164, 188-9,
Late-Cretaceous surface, 344 224,290-1
Late-Pleistocene times, 89 Magnetite, 146
Late-Pliocene times, 122, 363-4 Manganese, 146, 161, 231; oxide veneer,
Late Tertiary relief, 180; times, 44, 56; River 290
Teays, 190 Mangrove, 160,230
Late-Weichsel stadial, 376 Maquis, 153
Lateral, erosibn, 91, 363, 365; offsetting, 99; Marine, clays, 142; transgressions, 361
planation, 400 Marls, 177
Laterite, 400 Marram grass, 168
Lateritic alterations, 290 Mass-movement, 197, 209, 215, 216-21,
Latitude, 54, 55, 56-8, 112, 133, 155, 176, 225, 255, 318, 347-9; scars, 348
255 Meanders, 2, 5, 8, 14, 67, 118, 166-7,
Latosol, 147, 340 275-9, 372, 388, 398-9
Lava, 18,53,88-9. 108, 193,305,328,333, Measurement, 22—4, 78, 21 1, 297
358.362,393 Meltwater, 124, 183^, 195, 206, 216, 230,
Lavakas, 170
248, 249, 284, 3 17, 3 1 8-1 9, 3 64, 3 76-7,
Leaf. fall. 155; 159
litter, 391-2; channels, 249, 377, 392
Leaching, 289. 291. 308 Mesozoic, rocks, 350; times, 45
1 1 1 5

Metallic-organic compounds, 160-1 Old Baltic ice-stream, 380
Meiamorphic rocks, 93, 107 Old Red Sandstone, 100, 192
Meteorite craters, 9 Oligocene times, 38, 58, 361
Methane, 145 Olivine, 107, 144, 146
Mica, 91, 107, 146 Open systems, 345
Mica-schist, 185, 188 Ordovician times, 62
Micro-organisms, 161 Organic, acids, 160-1, 166; compounds,
Mid-oceanic ridge, 66, 82, 87 160-1, 164; layer, 113; mat, 155, 156,
Middle-Pleistocene times, 364 157, drifting, 156; matter, 142, 145, 154,
Mid-Pliocene age, 47 157, 161, 164, 168, 223-4, 235, 291;
Milazzian, 381 residues, 155, 156
Millstone grit, 95, 105 Organisms, 131, 146, 160, 166-8, 176,219,
Miocene, 42, 343; times, 44, 45, 89. 361-2, 222, 231,233, 237,292
364,399 Orientation, 228
Molehills, 151; tunnels, 151 Orlhoclase, 92, 143, 190
Moles, 153 Outwash, fan, 398-9; plains, 184, 214, 215,
Mollisol, 139,202-3,284,353 230, 244, 275
Monastirian, 38 Overdeepening, 179
Monazite, 147 Overgrazing, 78, 171
Mono basin glaciation, 338 Overflow, 360-2, 365
Monsoon, 116,315; season, 3 1 366
Overspill, 40,
Montmorillonite, 106, 137,222-3,258,291, Oxford Clay, 52, 264
315 Oxidation, 144, 146, 155, 173, 190, 192,
Moraine, 227, 250-1, 259, 336, 338. 393, 231, 338, 341; -reduction potential, 145
398-9, 411; ablation, 205, 378; annual, Oxygen, 141, 144, 145, 164, 291; deficiency,
317; cross-valley, washboard, or de Geer, 160; isotopes, 324, 328-9
249-51. 3 1 7; end, 2 14, 248, 249-5 1. 270, Oxyria digyna, 160
303, 311, 315, 373-6, 378-9; ground,
178, 205, 249; ice-cored, 250-1, 253, 3 1 1, Paleoclimates, 3 16
330; lateral, 214; lee, 303; medial, 214 Paleoclimatic indicators, 286
Morainic debris, 108, 271; ridges, 378-9 Pnleoforms, 58
Morphoclimatic zones, 54-60 Paleohydrology, 369, 371
Morphological mapping, 2, 4, 198 Paisas, 295
Mosses, 153, 156, 163, 194 Pans, 194-5
Mountaintop detritus, 270, 293 Parallel retreat, 345, 346, 355
Mudbanks, 2 1 Particle, 261; disposition,
collisions, 134,
Mudcracks, 215, 283 134-6; movement, 126, 127; orientation,
Mudflats, 40, 237 134-6, 248, 266, 267; rounding, 261-3;
Mudflow, 21 1, 228-9, 248, 259 shapes, 131-3, 204, 267, 270, 409;
sphericity, 127-31, 184,262-3,
263; size,
Nardus strkta, 300 268, 408-10, distribution, 128-31; sort-
Natural bridges, 370-1 ing, 128-9, 225, 228-9, 239, 244, 264,
Nebka, 159 283,346
Needle ice (see also pipkrakes), 218 Pearlette Ash, 336, 363
Neoglacial, 190, 324, 375-6 Peat, 67, 113, 156, 164, 283, 292, 295,
Neve, 121, 122, 123 312
Nivation, 302; hollow, 8-9, 195, 200, 309 Pebbles, 127, 131-6, 174, 203-5, 210-11,
Norite, 192,270 215, 219, 229, 242, 248, 249, 252, 259,
Nunataks, 377 262-4, 287, 293, 304, 339, 409
Pediment, 50, 267-9, 292-3, 294, 296,
Ocean basins, 79 344-7, 348, 354, 396, 400
OITshore, gradient, 302; shelf, 234; topo- Pediplanation, 53, 63
graphy, xiv Peds, 136-7
1 1
Subject Index 497

356—8, 401 Pluvial, episodes, 255, 325, 400; histories,
Peneplain, 2, 38, 49—53, 64,
Penitent rocks, 1 1
324
Periglacial, conditions, 128; phenomena, xiv; Podzol, 190
processes, 36, 56, 57, 200 Podzolization, 289, 308

Permafrost, 15, 67, 113, 140, 150, 156,205, Point-bars, 230-3, 276, 277
215, 218, 255, 282, 283, 307, 313, 340,
Polje, 196-7, 199,369
376 Pollen, 324, 335
Permeability, 32, 137, 170, 294, 350 Polygons, 59, 67, 282-3
Petroleum, 74, 77 Pond, 219, 350

p-forms, 184 Ponding, 362
pH, 143, 145, 194, 301, 308, 410-12 Poplars, 331
Phacolith, 108 Pore-water pressure, 217, 219

Phi scales, 129-30 Porosity, rock, 93, 106, 185, 301; soil,
Phosphate, 334 136-8, 151, 153-4

Phosphorous, 164 Porphyry, 107
Photographs, 333, 335 Posidiction, 39
Photosynthesis, 113, 164 Postglacial, changes, 385; conditions, 255;
Phyllite, 101, 192 times, 43, 58, 177, 210, 348, 375-6
Piedmont, angle, 345; glaciers, 121, 133, Post-Wisconsin weathering, 340
214; junction, 345 Potassium, 14], 142, 143, 146, 164, 165,
Piedmonttreppe, 50 190, 290-1; -argon, 87, 327-8
Pimpled plains, 285 Potholes, 174, 183-4
Pines, 148, 153, 326, 334-5; Norwegian, Prairie, 151, 152
157; Poderosa, 148; Scots, 157 Precipitation, biochemical, 148, 160, 162,
Pingos, 12, 66, 295 164-5; chemical, 56, 141, 142, 145, 230,
Piftus aiiistata, bristlecone pine, 167 285-90, 315,353
Pinus monticola. Western White pine, 334-5 Precipitation, effective, 25-6, 1 14, 153, 209,
Piping, 195, 196, 209,350 308, 369; intensity, 116
Pipkrakes, 218-19, 283, 295, 313 Pre- African surface, 49
Pitted beaches, 59 Pre-Boreal times, 183
Pivotal fault, 99 Pre-Cambrian times, 361
Plagioclase, 92, 143, 146, 190 Pre-Devonian valleys, 359
Planation surfaces, 45, 294, 382 Preferred orientation, 10
Plant, evidence, 334; remains, 23 Pre-glacial, hollows, 181; topography, 178;
Plantations, 169; maple, 320; oak, 320 398-9; weathering, 248
valleys, 182,
Plants, 147, 148, 158-9, 160, 164, 166, Pre-Middle Tertiary times, 343
288; accumulator, 165; aquatic, 164; Pre-Oligocene relief, 344
tropical, 165 Pre-Ordovician topography, 180
Plastic, deformation, 216-17; limit, 138 Pre-Pleistocene, age, 340; drainage, 364;
Plasticity index, 138 landforms, 399; times, 47; "topography,
Plateaux, 53, 100, 108, 270, 286. 306, 346, 180
359 Pre-Recent surface, 357
Platform, abrasional, 176, 313, 383 Pre- Wisconsin, age, 175, 340; entrenchment,
Playa, 84, 169, 261,283,296 361
Pleistocene, 7, 25, 28, 39, 44, 47, 58, 70, 87. Pressure release, 175, 197, 199
107, 175, 177, 241, 244, 315, 324, 337, Proglacial deposits, 131
345, 347, 362, 363, 369, 373, 380, 382; Progressive exposure, 51-3, 347
early. 47, 175; erosion, 344; late, 328; Protactinium-231, 328
movements, 44; terraces, 340; transgres- Protalus ramparts, 200, 344
sions, 331 Pseudo-karst, 67
Pliocene, 122; -Pleistocene erosion, 344, Pseudo-pillow structures, 163
boundary, 380 Pumice, 272, 273
Plunge point. 239, 240 Pyroxene, 144, 146
Quartz, 107, 146, 257, 263, 264, 265, 288, Rosin’s Law, 128, 130
289, 294; veins, 339 Rotational slip, 181
Quartzite, 96, 107, 177, 185, 192, 194, Rubidium-strontium dating, 328
228 Rumpfflache, 50
Quasi-equilibrium, 225, 280, 295-8 Runoff. 114-17, 153, 156, 165, 170, 255,
Quaternary (see Pleistocene) 307, 317, 320, 354; coefficient, 154, 160,
Quick clay, 84; slides, 220-1, 319 173
Radiant energy, 112 Salicornia, 159

Radiation, 157 Salt, 93, 102, 103, 105, 158, 236,244,313;
Radiocarbon dating, 33, 320, 329-3 1, 333 crystallization, 176, 186, 187, 193, 194,
Rain, shadow, 298; splash, 208, 293; wash, 199, 261; dome, 104; lake, 256-7, 259,
138,209 337,357
Raised beach, 34, 133, 230, 341, 382 Saltation, 206-7, 279
Reading Beds, 51-2, 139 Sand, 139, 157, 159, 168, 174, 190, 208,
Recementation, 133 209, 211, 213, 227, 228, 229, 330, 237,
Recurrence interval, 1 17 239, 244-8, 250-1, 252, 259, 260, 271,
Reduction, 144, 145 279-81, 293, 305, 306, 317, 338, 347,
Reeds, 166 368, 369; blasting, 133, 174, 187, 257,
Reedy drift, 340-1 263, 313; cays, 313, 389; covers, 244;
Reefs, 242^, 321, 333, 389, 391 grain surfaces, 132-4, 261; megaripples,
Regolith, 192 237; ridges, 353; ripples, 213, 237, 279,
Rejuvenation, 179, 362 315; sea, 159, 244, 245; shadows, 306;
Remanent magnetism, 87 traps, 159; waves, 232, 343
Reservoirs, 169, 369 Sandstone, 93, 106, 107, 177, 185, 192, 193,
Resurgences, 370-1 202-3, 220, 257, 264, 267, 269, 348,
Reworking, 126, 135, 150-3, 174, 177, 184, 354
210,211,213, 244,274-9,304 Sandur, 230, 232
Rhacomltrium, 227 Sand veld, 171
Rhexistasis, 191,352-3 Sangamon, interglacial, 340; time, 191;
Rhizocarpon geographicum, 331 weathering, 340
Ria, 387-8 Savanna, 57, 60, 151, 155, 160, 164, 165,
Riegel, 178 170, 267, 289, 319
Riffles, 279 Scale, 2, 73; models, 29, 3 1-2
Rift-valleys, 81-2, 345,362 Scarplands, 34, 53, 100, 342-3, 356-7,
Rillwash, 64, 176, 214, 227, 283, 314, 319, 365
350,400 Scarps, 218, 219, 220, 271, 354, 365
Rip currents, 120,213,316 Schist. 96, 98, 101, 107, 185, 192, 193, 197,
Risers, 178 199, 257, 259, 264, 294, 345, 359
Riss, age, 375; glaciation, 380, 392 Schistosity, 93
Rivelin Grit, 268 Schmidt net, 34, 266
River, deposits, 130; diversions, 40; gravels, Scour, 204, 207, 231, 232, 274, 281, 303,
135; sands, 129 363; -and-fill, 235, 279; marks, 304
Rock, avalanche deposit, 334-5; barrier, Scree (see aho talus), 126, 136, 159-60,
179; breakdown, 172-9; falls, 38, 39, 227-8,259, 264-5,301,350
221, 254-5, 259, 319, 347; fans, 347; Scroll, 231
glacier, 215-16, 259, 311-12; pedestals, Sea, breaches, 177; -floor spreading, 81-2,
199; protuberance, 303; slide, 83, 101, surface, 342-3; gullies, 383-4; ice, 121,
220-1; table, 295; threshold, 303; wall, 321; -levels. Pleistocene, 361, 367; -level
349 changes, 42-8, 286, 381-2, 386, fall, 342,

Root action, 137, 252; exposure, 168 344, 354, former, 341, low stands, 361,
Roots, 157, 159-60, 287, 354 367, postglacial, 234, 241, 333, 355, 382,
Rose diagram, 134 rise, 177, 229, 240, 243, 296, 323, 355,
Subject Index 499
369, 386, 393; mounts, 342; valleys, 344; Slope deposits and depositional landforms,
water, 236, 288 225-8
Sediment, 14,22, 127-36, 147, 151;compac- Slope, deposits, stratified, 227, 353; evolu-
tion, 133; concentration (see also turbi- tion, 344-55; measurement, 402-3; pan-
dity), 133; intertidal, 152; load, 254, 297, tometer, 403-4; processes, 227, 300-2,
320; sorting, 152; source, 211, 257-60; 317, 348; profile, 403-4; wash, 209;
structures, 274-9, 340; supply, 234, 241, water, 188, 216,302,319, 353
260, 280-1, 343, 385, 386, 387, 389, Slopes, 54-5, 107, 109, 125, 156, 174, 194,
400; transport, 114, 201-15, amounts, 195, 200, 202-3, 216-21, 249, 265-7,
209-16; yield, 26, 170, 296, 308 306, 308-9, 344
Sedimentology, 12 Slump, 219, 220, 251,354
Seepage lines, 359 Snow, 112, 113, 121, 195, 206, 216, 244,
Seif dunes, 246, 303 282, 294, 300, 302, 307, 317, 318, 380;
Sensitivity, 138 bank, 349; block fall, 216; creep, 125;
Serir, 293 drift, 195, 308, ice, 306; line, 57, 121,
Sesquioxides, 146, 147,353 122, 253, 299; melt, 217, 320; pack, 125;
Settling lag, 224; velocity, 1 14, 127, 201 patch, 333
Shakwak till, 338 Sodium, 141, 142, 291,313, 318
Shale, 93, 94, 106, 347-8, 354, 369 Soil, air, 161; carbon dioxide, 144, 161-3,
Shear-planes, 124, 216; stress, 16, 125-6 255; climate, 12, 24; creep, 13, 216, 218-
Sheeting, 65, 96-8, 347 19; evidence, 334, 336—41; exposure,
Sheetflow, 159, 208 322-3; fabrics, 136; fauna, 150-3; fis-
Sheetwash, 58, 208, 294, 296, 301, 308, sures, 71; flow, 216-18; guirlands, 283;
353,400 horizons, truncated, 150; hydrology,
Shirdley Hill Sand, 385-6 136-7; layers, 136, 227, burled, 227,
Shore platform, 163, 176, 193, 194 truncated, 337; mechanics, 137-8; mois-
Shorelines, former, 398-9 ture, 152, 153, 161, 250-1, 287, 294,
Shores, frozen, 215; gradient, 14; rocky, 308, 403, 405-6; movement, 151, 159;
163; sandy, 385 porosity, 136, 153, 154; profile, 168, 190;
Shrinkage coefficient, 106 science, 2, 18, 77; shear strength, 32, 138;
Sieving, 127 structures, 340; water, 106, 136-7, 170,
Silica, 141, 143, 144, 145, 146, 147, 165, 173, 188, 267, lateral flow, 137; weather-
166, 189,222, 263,290,300 ing evidence, 338-40
Silicate minerals, 143, 144 Soils, 127-36, 170, 193, 268, 271, 273; arc-
Silicates, 173, 189, 197 tic, 184, 218; sandy, 137, 153, 171; struc-
Silicon, 141, 164, 165,231 tural, 282; tropical, 291
Silicrete, 397, 400 Solar radiation, 84, 112-14, 148
Sillimanite, 147, 340 Solifluction, 51,59, 133, 135, 167,218,219,
Sills, 90 266, 270, 307; deposits, 225; terraces,
Silt, 128, 137, 139, 156, 195,208,214,227, 225-7
228, 229, 230, 237, 245, 254-5, 259, Sols, 142, 222
271, 277-9, 294, 315, 339; -clay ratios, Solubility, 144, 145-7, 162
339; lacustrine, 230, 350; wind-blown, Solutes, 160, 257
336 Solution, 98, 141, 144, 146, 162, 164, 197,
Sinkhole,369-71 222, 285, 289, 350
Skewness, 129-30 Solutions, 164, 288, 292
Skiddaw slates, 181 Solutional, lowering, 63, 350; weathering,
Slate, 93, 101, 107, 185 113
Sliding plane, 215, 216, 217, 221 Space science, 20, 33, 67
Slip surface, 219
Spartina, 159
Slip-olT slope, 231-3 Spits, 59,238, 241—2, 260, 389-91; arrow,
Slope-angles, 26. 156,
216-21, 227-8, 267, 303; cuspate, 242, 285, 306
271,273. 283, 296,302, 345 Spray, 150, 186,313
Springs, 219, 233, 267, 300, 347; sapping, Swales, 59, 238-9, 240
202, 359; water, 255 Swamps, 161
Spruce, 148, 331 Swash, velocity, 69; zone, 120, 208, 213,
Stacks, 150 237, 240,281
Stalagmite, 233 Syenite, 192
Staurolite, 147 Syncline, 100
Steady-state, 70-1, 280 Synclinal valley, 344
Steppe, 112, 115, 155 Synclinorium, 100
Stillstand, 42-8, 64,382 Systems theory, 69-72, 76
Stone-banked terrace, 227, 311, 320
Stone, pavements, 270; stripes, 219, 283-5 Tahoe till, 294
Stones, 227, 263, 267, 270, 282, 293 Taiga, 116, 148, 156
Storm surge, 39, 40 Talus (see also scree), 202, 216, 227, 243,
Strandline, 176 267, 346, 349; platforms, 349
228-9, 237-40, 246-8, 249;
Stratification, Taxus baccata, yew, 148—9
1, 232
cross-bedding, 244, trough, 23 Taynlon stone, 1 1
Stratovolcano, 89 Tectogene, 80-1
Stream, capacity, 210-11; channels, 133, Tectonic, gravity sliding, 51, 82-3; move-
157, 217, 274-9, bouldery, 158, dunes, ments, 255-7, 345, 353, 362-3, 367,
277, gravelly, 158, sandy, 158; depth, 381
157-8, 276-8; flow estimation, 406-8, Temperature, 56, 112-16, 142, 154, 161-2,
turbulence, 174, velocity, 157, 158, 164, 186, 199, 218, 231, 255, 256, 287,
203-8, 236, 276-7, volume, 160; in- 292, 307,319,324-6
cision, 345; orders, 6-7, 26, 272 Tephra, 89
Striae, 124, 163, 177, 378; sand-blast, Tephrochronology, 336-7
257-8 Termite, 151, 152
Strike-slip fault (see also transcurrent fault), Terra rossa, 340, 353
83,84,99, 100,362 Terraces, 15, 25, 34, 43, 252, 263, 341, 343,
Strike, trench, 342; valley, 101, 359 364, 369, 372, 392; deposits, 288, valley,
Structural, control, 27, 62, 64-5, 95, 98-9, 157
220, 221, 359, 364; movements, 40, 364; Terrain unit, 5-6
surface, 344 Tertiary, deformation, 357; surface, 346;
Sub-Atlantic period, 376 times, 199, 344, 359, 363, 400; uplift,
Sub-Boreal, 324 182
Subdivided histogram, 404 Thalassia restudinum Konig, turtle-grass,
Sub-Eocene surface, 5 1-2 147
Sub-glacial, chutes, 184, 392; streams, 215 Thawing, 113, 136, 170,215,216,282,307,
Sublimation, 122 313,315,317, 354
Submarine canyons, 70, 305, 342-4 Thermoluminescence, 329-30
Submerged jet flow, 1 1 Thixotropy, 138
Sub-river cavities, 369 Thorium-230, 328
Subsequent streams, 358 Thrustfault, 43
Sub-soil, 137, 164; frozen, 218, 320: Thufur, 58, 283-5

weathering, 294 Tidal, currents, 120, 236, 241-2, 343,
Subsurface, seepage, 209; solution, 102; 386-7; range, 119, 234, 236, 237, 241,
weathering, 1 1 285,386
Sumas till, 133 Tides, ebb, 236
Superimposition, 355-6, 359 Tillite, 86
Surf zone, 120, 239, 281 Till fabric, 250-1

Surface, depressions, 102; lowering, 191 Tills, 10, 128, 133, 249-51, 255, 294, 334,
Suspended, load, 131, 207, 222, 315; sedi- 340, 377; ablation, 248; basal or lodge-
ment, 114, 115, 206-8, 211, 240, 259, ment, 248
279,412-13 Tilting, 34, 40, 99, 361-3
Subject Index 501
Time lag, 161, 224, 254-7 Veneers, 289-90

Timberline, 324 Ventifact, 133, 187
Tombolas, 59 Vermiculite, 223
Tors, 12, 95, 150, 193, 197-9, 398-9 Villafranchian, 380, 399
Tourmaline, 147 Viscosity, 114, 123, 371
Tracers, 218 Volcano, 57, 72, 342

Traction, carpet, 206-7, 279; clog, 225, 229, Volcanic ash, 146, 336, 344, 386; cones,
231,261 243, 357; eruption, 41; rocks, 41,
Transcurrent faults, 45, 99 90
Transportation, 161, 172, 199-224; clays, V-shape, 178
142; fluvioglacial, 215; glacial, 214—16; Vulcanicity, 86-9
pebbles, 203-5; solutional, 145; surfaces,
199, 295; wind, 195, 201, 205-7, 209, Wadis, 7, 28, 58, 263
210, 230, 320, rates, 213-14 Warping, 44
Travertine, 233 Washload, 207, 223
Tree-damage, 33 1-2 Water, 56, 68, 93, 95, 98, 105, 106, 114-21,
Tree-ring dating (see also dendrochronology), 136-9, 141, 156, 164, 173-5, 185, 193,
72,331 195, 216-17, 222, 282, 294, 302, 307,
Tree roots, 106, 148-50, 164, 167-8, 199; 326; depth, 158
seedlings, 147 Waterfalls, 174, 233, 357; recession, 175,
Trees, 153, 158, 166-7, 216; deciduous, 359
148; wind-thrown, 148, 150 Water flow-through time, 161
Tropical rainforest, 165; soils, 159, 339 Waterlogging, 145, 218, 235, 392
Tsunami, 83 Water-stable aggregates, 154, 155
Tufa, 233 Watertable, 171, 231, 288, 289, 296,
Tuffs, 88 369
Tundra, 91, 112, 116, 156, 159, 164, 171; Wave, action, 234, 317; height, 69, 119, 254,
arctic, 148, 165; forest, 165; shrub, 280, 281, 285; length, 119; period, 69;
148 refraction, 241, 242
Turbidity, 209-10, 212, 299, 317 Wave-cut bench, 91; cliff, 237; platform,
Turbulence, 114, 118, 201, 225, 234, 387, 158, 176, 177,297,382
389 Waves, 158
Turf, 215, 225,227,284 Weathered material, influence on subsequent
Turloughs, 194 weathering, 292-5
Weathering, 126, 141, 146, 147, 162, 173,
Unconformities, 51, 344, 356-7, 395 175, 176; biochemical, 160-4, 176; caver-
Undercutdag, 196, 308, 354-5 nous, 8, 186, 193, 199, 340-1; channels
Undersaturation, 146 195-7; chemical, 160, 172, 175,
(flutings),
Uniformitarianism, 37, 39, 362 187-92, 287, 294, 299; crust, 57, 147,
Unloading, 96 209, 285-90, 292, 326; deep-, 191, 194,
362, 382
Uplift, 342, 199, 217, 220, 289, 345; forms, 193-9;
Upper Greensand, 101 mantle, 107, 113; pits, 161, 180, 193-4,
Uranium-lead dating, 328 195; ratio, 147; reactions, 142—4; recess,
U -shape, 178 5; profile, 144, 285, 292; zone, 144, 173,
222,291
Valders glacial maximum, 330 Wet season, 164
Varved clays, 125, 250, 336, 337 Willow, 156, 166
Vashon till, 133 Wind action. 111, 119, 128, 129, 148, 159,
Vegetation, 25-6, 42, 54, 57, 58, 68, 153,
171, 174, 187, 194, 195, 211, 241, 243,
156, 157, 208, 229, 236, 245, 247, 252, 244-8, 259, 260, 269, 280-1, 296,
255, 282, 292, 303, 305, 307, 308, 319, 307, 313, 315-17, 320, 321, 371, 377,
333-4, 353, 355, 372, 391, 400; growth, 393
158 Wind-throw, 159, 313
interglacial, 340; weathering.
j -fc moraines jao
TSn- moraines, 338-9; Yarmouth
Wisconsin, deposits, 380,
^
WoSland, 1 1 2
302
^rg£:n;3U 354,373, 375,380, Zavieja,
Zircon, 146
392 Eel grass, 159
Zostera sp,
Z/r ratio, 142-3

X-ray analysis, 74, 340
Geographical Index
(excluding bibliography)
Aberdare Mountains, 57 186, 187, 188, 193, 214, 227, 245, 254,
Aberdeen, 192 283, 305, 306, 340, 373
Aberdeenshire, 178 Antrim, 193
Abidjan, 21 Apalachicola, River, 259, 280
Abyssinian Highlands, 49, 57 Appalachians, 5, 50, 53, 64-5, 109, 345
Acadia National Park, 97 Arabia, southern, 245
Achdorf, 349 Ararat, Mount, 87
Adriatic coast, 367 Aragvi, River, 301
Africa, 43, 50, 57, 85, 98, 160, 170, 223, Arctic, 67, 121; areas, 197, 218; deserts, 58,
399; central, 298, 339, 345; east, 309, 60, 112, 209
361, 380; east coast, 240; North, 158, Argentina, 299, 300, 376, 380
263, 326; West, 57, 116, 263 Argyllshire, 202-3
Aigoual, 169 Arid areas, 57, 58, 59, 60, 68, 115, 155,
Alajarvi, 379 173, 289, 298,371
Alamyshik oil-field, 356 Arizona, 91, 108, 1 15, 233, 263, 371; south-
Alaska, 34,53, 83, 101, 116, 119, 122, 124, west, 27
180, 181, 211, 214, 218, 221, 299, 303, Arkansas, River, 314
307, 313, 317-18, 324, 331, 337, 375, Arran, 214
376, 392, 393; Gulf of, 255, 300, 382; Ashop valley, 138
north, 215, 257; Range, 309, 312, central, Asia, 223; central, 211; south-east, 28
259; south-east, 255, 343 Askja, 89
Alazami, River, 301 Asov Sea, 242, 285
Albany, 107 Atakor Mountains, 202-3
Albert, Lake, 242, 260 Athabasca area, 357
Alberta, north-west, 357 Atlantic Ocean, 84-6, 234, 324, 362
Algeria, 353 Australia, 53, 62, 98, 109, 115, 160, 218,
Alice Springs, 95 223, 244, 245, 285, 313, 337, 361, 362,
Allegheny, River, 392 397, 400; central, 209, 267, 345, 346;
Alps, 82, 124, 180, 252, 338, 380; French, eastern, 311, 371; north-west coast, 383;
399 south-east coast, 241; west, 260
Alsace, 321 Avon valley, Worcestershire, 325
Alyn, River, 392 Avon, River, Bristol, 361
Amazon basin, 56, 57. 85-6, 115, 359, 361; Axe valley, 266
River, 210, 298, 362 Azores, 343
Anadyr, 171
Andarai, 151 Baffin Island,45-6, 121, 206, 249, 309-11,
Andes, 44, 56, 81, 1 15, 259; Argentine, 299, 333; north-central, 317; north-east, 193;
300, 380; Peruvian, 150; Santiago, 285 northern, 181, 282
Antarctica, xiv, 113, 121, 139, 179-80, 184, Bahia Sebastian Vizcaino, 213
1 1 5
Baja California, 213, 3 16, 382, 386-7 Bonneville, Lake, 40, 365-6; Flats, 244
Bali, 87 Bonney, Lake, 113
Balkans, 80 Boomer Beach, 317
Balkash, Lake, 235 Boston, 46
Ballangen, 192 Boulder, 193, 311; county, 320
Baltic, 48, 211, 236, 240, 387 Brahmaputra, River, 40, 363
Banbury, 192 Brandywine Creek, 223
Banchory, 178 Brazil, 57, 59, 84, 95, 107, 151, 192, 193,
Barbados, 177 263, 286, 288-9
Barberie, Lake, 389 Brazilian Shield, 103
Barcelonette, 301 Brazos, River, 91, 363
Bariloche, 300 Breaksea Sound, 179
Barind, 98 Bridger Range, 363
Barnaul, 203 Bridlington, 305
Barnes ice-cap, 121, 122,331, 376 Britain, 43, 156, 188, 377
Barnet Spring, 196 British Colombia, 122, 125, 179, 180;
Barnstable Harbour, 152 southern, 255
Barra Head, 181 British Guiana, 189,211,242
Barrington Crater, 9 British Honduras, 288
Bane, 96 British Isles, 184, 385-6
Barrow, Alaska, 242 Brittany,303
Basin and Range area, 83, 342 Broad Law, 267, 410
Bear, Lake, 362; River, 366 Brodie Island, 254, 281
Beardmore glacier, 121, 124 Buchan, 192
Beartooth Mountains, 109, 175 Bukobya, 260
Beaver Creek valley, 215 Bulawayo, 116
Belgium, 212, 368, 369 Bumbe, Lake, 242
Ben Cleuch, 108 Burma, 57
Benguela, 242 Buxton, 161-2
Ben Lomond, Tasmania, 95 Byrd Station, 214
Beni basin, 102, 103; River, 298
Bergen, 319 Caernarvonshire, 177; north, 310
Bering, -Chukchi platform, 344; Sea, 285; Cairngorns, 182
Straits, 120, 213 Cairnsmore of Fleet, 101
Berkshire, 151 Cairo, Illinois, 263, 264, 357
Bermuda, 191, 342 Caithness, 100
Berwick, 393 California, 44-5, 66, 80, 87, 124, 254, 167,
Bitterroot Range, 308-9, 3 1 187, 205, 206, 211, 213, 221, 229, 240,
Bjorneljell, 294 245, 255, 258, 274, 288, 302, 305, 311,
Black Aragvi, River, 301 317, 338-9; Gulf of, 169; northern, 233,
Black Burn, 359 299; southern, 342, 382, 384
Black Forest, 82, 178 Californian shelf, 342
Black Hills, 261,264 Cambridgeshire, 162
Black Mountains, California, 229, 362 Cameroons, 1 1
Black Sea, 285, 372 Canada, 46, 87, 97, 101, 105, 212; arctic,
Blackford Hill, 250-1 375-6; eastern, 156; northern, 382; north-
Blakeney Point, 321 west, 229
Blomesletta, 200 Canadian Cordillera, 377; Falls, 175; plains,
Blue Creek, 205 western, 377; River, 261, 264; Shield, 58,
Blue Ridge, 1 94 62, 98, 180, 326,359
Bohemian-Moravian Highlands, 200 Canary Islands, 214
Bolivia, 102, 103; north-east, 298 Cape Cod, 354
Bonavista Bay, 43 Cape Givardeau, 389
Geographical Index 505
Carmel Mountains, 153 Corran, 135
Carolina, 254, 281 Corrie an-t-Sabhail, 182; an-t-Sneachda, 182;

Carpathians, 299; Moravian, 202-3; Burn, 214; Ruadh, 182
Ukrainian, 259 Corsica, 343
Carro Cadillal, 217 Costa de Nayant, 238-9

Carson Range, 44 Cotswolds, 108, 271
Casa de Pedra, 289 Courantyne, River, 242
Cascade Range, 87 Covehithe, 40
Casement glacier, 123, 252, 303, 392 Cow Close Fell, 196
Casiquiare Channel, 362 Cracow upland, 367
Caspian Sea, 43 Crater Lake, 89

Cedar Butte, 366 Crimea, 116; south, 217
Celtic Sea, 343 Cronese Mountain, 305
Central America, 233, 299 Crooked River, 223
Central Lowlands, 108, 178 Cross Sound, 343
Cerro Bolivar, 189 Cuba, 147, 170, 199
Ceze, River, 257 Curbar Edge, 268
Chaco canyon, 314 Cyrenaica, 263
Chain Bridge, 332 Czechoslovakia, 226, 302
Chambarrau plateau, 219
Chao Phya, River, 2 1 Dakota, 298
Chaos Jumbles, 333-5 Dairy, 9
Chapel-le-Dale, 197 Dalsheidi, 393
Chemung valley, 229, 276 Danube, River, 230, 234
Chernozem regions, 170 Dartmoor, 12, 200
Chesapeake Bay, 43 Darwin, 44
Cheshire Plain, 373 Dawlish Warren, 390-1
Chile, 179, 217, 221, 300; longitudinal Death Valley, 229, 362
valley, 44 Decade glacier, 309-10
China, 57, 116, 124, 165, 211, 380; south- Deccan, 89
east, 155 Dee, River, 178
Chuska Mountains, 108, 220, 347 Dee Why Beach, 316
Clare, County, 200 Delaware, 223
Clinton, 392 Denmark, 393
Coeur d’Alene, River, 3 1 Derbyshire, 128, 194, 256,268
Coffee Creek, 233 Derwent, River, Yorkshire, 95
Colombia, 44, 206, 263 Devonshire, 200, 266, 391; south, 355
Colorado, 97, 148, 175, 192, 311, 320; pla- Disko Bay, 124
teau, 97, 371; River, Texas, 40 Djakarta Bay, 321
Columbia plateau, 305; valley, 302, upper, Dneiper, River, 161, 299; uplands, 245
219, 221,315 Dneister, River,230
Colville, River, 116-17,119,317-18 Dogger Bank, 393
Congo, 116, 339, 361; basin, 115; River, Don delta, 234
210,362 Donegal, 358
Connecticut, 26, 49, 91, 92, 180; northern, Donets, River, 320
361 Donjek Moraines, 331; valley, 364
Constance, Lake, 210, 320 Dorking, 108
Cool temperate environments and latitudes, Doubs, River, 398-9
31,59, 97, 113, 116, 155, 156, 157, 164, Dourdan, 64
195, 209, 216, 217, 237, 254, 270, 289, Dours gorge, 175
292, 300, 308,317 Driesam valley, 178
Cornwall, 90 Dungeness, 243
Corpus Christi, 240, 244 Dzerzhinsk, 276
1 1 1 1

East Anglia, 10, 134,213 Flamborough Head, 177, 213, 303-5
East River, 120 Flam valley, 183
Eastbourne, 176 Films, 39
Edinburgh, 25 Flin Flon, 180
Eglwyseg valley, 227 Florida, 39, 46, 211, 243, 342; Bay, 147
Egypt, 30, 53, 186, 245; north coast, 288; Forfar, 289
Upper, 289 Fort Churchill, 177
Egyptian sand sea, 245 Forth valley, 230, 341, 382
Eichberg, 349 France, 43, 115, 169, 213; northern, 350;
Eklunta, Lake, 83 south-west, 213
Elbe, River,276 Franklin, 5
El Dorado, 329-30 Fremont County, 263
Elberton, 410 Fresno County, western, 228, 258
6letot, 163 Front Range, 97, 175
Eleuterio, River, 212 Furuberget, 183
Elgin, 359
Elgon, Mount, 57, 309 Gallatin, River,363
Elizabeth, River, 36 Galway, 393; Bay, 108
Elizabethville, 151 Ganges delta, 234
Elkton,215 Garcia, 233
El Salvador, 190 Garpenberg, 164
Elvegard, 399 Gaule, 107
Emilia, 211 Gebel Akdhar, 158
Endau, River, 385 Georgia, 213, 241, 410
England, 108; east coast, 236; eastern, Georgia, eastern, U.S.S.R., 301
153, 171; north-west, 310; south-central, Germany, north coast, 236; northern, 244
282; south coast, 176; south-east, 177; Ghana, 154
southern, 191, 307 Gibraltar, Straits of, 120
English Channel, 177, 213, 342, 343 Glasgow, 138,254,393
Enone, Lake, 182 Glen Coe, 3 1
Equatorial areas, 57, 115, 116 Glencolurobkille, 200
Erie basin, 392 Glen Etive, 202-3
Euphrates, River, 2 12 Glen Hills, 272
Eureka valley, 339 Glen Nevis, 178
Europe, 189, 222, 223, 283, 324, 376; Glomfiord, 179
central, 325; eastern, 230; northern, 330; Gobbins Island, 193
north-west, 392 Gobi desert, 244
European plain, north, 306, 379 Godavari basin, 170; delta, 235; River,
Everglades, 147 263
Exmoor, 1 16 Goose Lake, 392
Eyre Peninsula, 109 Gorak Shep fan, 339
Gota Alv, 179, 259
Faeroes, 227 Gotland, 177
Fairbanks, 12, 30 Grand Coulee dam, 169
Falkland Islands, 270 Grand Valley, 43
Fall Zone, 180,359 Grasubreen, 251
Ferrar glacier, 124 Great Barrier Reef, 244
Findhorn, River, 264 Great Basin, 46, 83
Finland, 180; southern, 378-9; south-west, Great Glen, 311
250-1 Great Lakes, 47, 119
Finlayson Channel, 179 Great Plains (see also High Plains), 1 84, 1 95,
Fitzroy basin, 267,269 336, 395; northern, 371
Fivemile Creek, 263 Greece, 30
6 1

Green Gully, 362 Idaho, 51, 245; east-central, 364
Green River, 223 Ikroavik, Lake, 211
Greenland, 46, 58, 121, 122, 124, 249, 342, Ila delta, 235; River, 166
376; east, 179, 399; north-east, 282; nor- Illawarra, Lake, 305
thern, 315; south-west, 323; west, 375 Iller, River, 212
Grey glacier, 124 Iliinoian glacier, 214
Grigalua, River, 385 Illinois,214, 263, 264, 306, 357, 364, 392;
Gros Ventre slide, 221 northern, 180; western, 184
Guano, Lake, 283 Imotski, 350

Guerrero Negro, 213 Imperial Valley, 245, 305
Guil, River, 210 India, 81, 89, 98, 157, 165, 170, 189, 235,
Gurreholms Elv, 399 263, 363; northern, 315

Guyane, 150, 264 Indian Ocean, 243
Indochina, 57
Half Moon Bay, 274 Indonesia, 190
Hallival, 270 Indonesian 120
Strait,
Hameenlinna, 378-9 Indus, River, 360-2

Hampshire, 319, 358; basin, 368 Ingul, River, 230
Han, River, 115 Insigsuin fiord, 206, 265
Handlova, 226 Interior Plateau, 1 80
Hardangerfiord, 303 Iran, 100; northern, 261; southern, 246
Hawaii, 87-9, 189, 190, 191, 192, 195,221, Ireland, 86; south-west, 344, 358; western,
300,317 194,393
Hell Gate, 120 Irish Sea, 343, 392
Hellemofiord, 392 Irrawaddy delta, 234
Helmsdale, 133 Isle of Wight, 103,220
Herault, 210, 325 Isortoq valley, 249

Higher latitude areas, 36, 57, 114, 124, 174, Israel, 153, 176, 287,293
182, 184-5, 222, 230,317 Istria, 370-1
High Plains, 91, 105, 184, 195, 245, 287, Italy, 1 16, 193, 21 1, 212; south-east, 308
371,395 Iv5, 181
Highland, 354 Ivory Coast, Il5, 189, 190,211
Hii, River, 212
HinesHill, 257 Jamaica, 176, 194
Hohneck, 283 James, River, 361
Holderness, 303 Jamuna, River, 99
Hong Kong, 95, 170, 192, 209, 294 Japan, 192, 212,283, 302
Hoover Dam, 169 Java, 87, 299
Hopi, 88 Javorniby Mountains, 202-3
Houtzdale, 109 Jefferson, River, 363
Hudson Bay, 46, 180 Jekyll Island, 339-40
Hudson estuary, 151 Jjelle,221
Humber, estuary, 238-9; River, 18 Johnstone, 9
Huntcliff, 355 Jotunheimen, 253, 311, 330
Hunter, River, 361 Juneau, 190
Hupei region, 380 Juniata, River, 223
Hydrographers volcano, 191 Junsele, 181
Jura, 82, 283
Ibadan, 1 1
Icefield Ranges, 331 Kaap, 233

Iceland, 58, 66, 84, 87, 89, 122, 283, Kafu, River, 362
249,
337, 343, 376, 391, 393—4; north-east, Kagera, River, 362
308 Kaiso, 260
Kalagarh, 228 Lapland, northern, 311
Kalahari, 244, 288,371 Las Bela valley, 244
Kanajokk, River, 210 Lassen Volcanic National Park, 334
Kansas, 105, 329-30; central, 3 15 Leith Hill, 108
Kantz Creek valley, 279 Lekhura, River, 301
Kara-Kunt, 171 Lemki, River, 364
Karataja, River, 98 Lena, River, 1 16
Karkevagge, 16 Levantine basin, 120
Katonga, River, 362 Liberia, 195
Kaui, 190 Libya, 246-7, 263; desert, 303
Kebnekaise, 321, 334, 335 Libyan sand sea, 244
Kelso dunes, 213 Lighthouse Hollow, 26
Kelvin, River, 393 Linth,210
Kenai Peninsula, 307 Little Cataraqui Creek, 25
Kennebec, River, 252 LittleMissouri valley, 167, 230
Kentucky, 191, 223; Dam, 369 Littondale, 196
Kenya, 56, 159, 176, 240; Mount, 57 Lituya Bay, 4
Kerry, 358 LivnopoIJe, 196-7
Khadis-Khevi, River, 301 Ljubljanica, River, 369
Kharga Oasis, 213, 289 Llangollen,227
Kilia delta, 234 Llao-Lao, 300
Kilimanjaro, Mount, 57, 309 Lleyn peninsula, 177
Kilsyth Hills, 214 Long Island Sound, 361
Kimberley, 233 Long Prairie, 254
Kiruna, 38 Lossie, River, 359
Kjove Land, 399 Louisiana, 177, 233; south-west, 260, 389
Klaypeda, 213 Low Lindrick, 95
Kola peninsula, 191, 192 Lower Colorado River, 264
Koolau, 191 Lower latitude areas, 160, 164, 196, 209,
Korea, 1 15 241, 286,287,326
Kosciusko, Mount, 292 Lyn valley, 116
Krakatoa, 177, 189
Krasnopakhorskiv Forest, 152 Mackenzie, River, 66, 116, 117, 178, 218,
Krikkjokk delta, 234 224,353
Ktsiya-Khrami, River, 301 Madagascar, 170, 342, 361
Kura depression, 43 Madhupur, 40
Kurkur Oases, 53 Madison, 323
Kvikkjokk, 116 Madison canyon, 101; River, 363
Mahadanz, 189
Labrador, 214, 249, 310; coast, 249, 342; Main, River, 212
shelf, 342;-Ungava, 156 Maine, 97, 252, 293, 340, 342; Gulf of,,343
Lago Rinihue, 221 Malacca, 192
Laguna Ojode Liebre, 387 Malaspina glacier, 121, 300
Laguna Pueblo, 186 Malaya, east coast, 236, 385; southern, 195;
Laitaure delta, 118, 235; Lake, 157, 158, western, 298
209 Malham, Cove, 97; Tam, 162, 164
La Jolla, 207 Manatee County, 47
Lake District, 181, 219, 283, 298; west- Manitoba, 177, 251
central, 310 Mangami, River, 212
Laki, 89 Mangatepopo, River, 212
Lammermuir-^tranraer moraine, 375 Maquoketa, River, 3 92
Lancashire, south, 385 Marble Point, 144, 187
Languedoc, 111 Marci, 211
Marecchia, River, 212 Montrose, 108
Maren Mountains, 191 Morayshire, 264
Marguerite Bay, 253 Morganton, 194
Markland Grips, 149 Moscow, 148, 151, 152
Maryland, 115, 146, 272, 316 Munfordville, 223
Massachusetts, 152 Murray basin, 362, 371
Massif Central, 401 Mustang Island, 240, 247
Matagorda Bay, 40
Mathowa, 362 Nain-Okak area, 3 10
Maticora, River, 212 Narvik, 38, 191, 192, 194, 294,392,399

McBeth fiord, 333 Natal,340
McMurdo Sound, 186, 188 Navajo country, 88, 371; mountain, 349
Meander delta, 234 Negev, 210
Medicine Bow Mountains, 308 Neilston, 9
Mediterranean area, 57, 60, 288, 325; Sea, Nevada, 176, 206
119, 120,213,331,381 Neveri, River, 212
Meekatharra, 192 Nevy Island, 227
Melfiord, 179 New Brunswick, 342
Melito, River, 212 New England, 96, 148, 342; coast, 335
Mendenhall glacier, 190, 41 New Forest, 319, 358
Mendips, 350 Newfoundland, 43, 98; west, 342
Me Ping, River, 212 New Jersey, 211, 239
Messier Channel, 179 New Mexico, 108, 186, 195, 223, 287;
Meteor crater, 91, 95 north-central, 230; north-west, 220; south-
Meuse, River, 212 central, 3 1
Me Wang, River, 212 Newport, 223; canyon, 343

Mexico, 233, 239, 385, 386; City, 169; Gulf New South Wales, 120, 267, 268, 316, 372
of, 259, 357; southern, 206 Newtondale, 95
Middleton Island, 382 New York, 120; state, 254
Mid-latitude areas, 57, 60, 107, 115, 116, New Zealand, 45, 176, 179, 199, 212, 272,
155, 174, 175, 182, 193, 196, 209, 273, 283, 386; South Island, east, 308
371-2,400-1 Ngong Hills, 159
Mikka glacier, 124 Niagara escarpment, 273, 274; gorge, 175
Milwaukee, 330 Niger valley, 288
Minais Gerais, 1 92, 1 93 Nigeria, 116, 152; south-west, 202-3;
Minnesota, 135, 184,253 western, 198
Mississippi delta, 44, 234, 363, 383, 389; Nile delta, 234; River, 24, 57, 67, 121, 229
River, 95, 120, 210, 230, 231, 259, Nisqually glacier, 206
262-3, 264, 276, 277, 392; valley, 260, Niwot Ridge, 227,311,320
298, 357,377,389 Njong, River, 1 15
Missouri, River, 262-3, 389 Norfolk, 213; Broads, 72
Mobile, River, 259 Normandy, 177
Modro Jezero, 350 Norrland, 181
Mojave desert, eastern, 245 North America, 57, 122, 185, 223, 298,
Mono basin, 338 324, 363, 392; central, 57; eastern, 303;
Monsoon areas, 116 north-west, 255
Montana, 101, 175, 195, 311, 344, 363; Northamptonshire, 15
eastern, 311; western, 309 North Atlantic, 321
Monte d'Oro Massif, 179 North Carolina, 194, 363
Monte Montarone, 193 Northcote, Mount, 218
Monterey, Mexico, 233 Northern Ireland, 358
Monterey submarine canyon, 343 North Dakota, 155, 286
Montpellier-le-Vieux, 1 1 North Pole, 58
11 1 1 0
North Sea, 120, 213, 236, 343, 392 Pennines, 173; central, 257, 348; southern,
North Uist, 254 257, 258
North West Territories, 195, 307 Pennsylvania, 223
North York Moors, 285 Periglacial areas, 14, 58, 59, 60, 133, 174,
Norway, 46, 98, 109, 111, 179, 181, 182-3, 200, 214, 218, 282-3, 295, 324
220, 221, 251, 253, 255, 270, 315, 330, Peru, 45, 269, 281, 321; southern, 213, 245,
343, 392; central, 231, 377; coast, 342; 410
northern, 191, 380; southern, 252, 324, Petsaure plain, 230
376; western, 221 Peyremale, 257
Nussbaum, Mount, 214 Plitvice lakes, 367
Po delta, 234
Oahu Island, 189, 195 Point Barrow, 242
Ob, River, 117, 118, 203,210, 277 Poland, 212, 218, 244, 271, 285; central,
Ochils, 108 227
Oder, River, 212 Polomet, River, 210
Odra valley, 27 Ponca City, 314-15
Ogmore, River, 269 Poole’s Cavern, 161-2
Ohio, 215; River, 207, 392 Post, 223
Ojakkala, 25 Postojna, 371
Oka, River, 276 Potomac, River, 332
Oklahoma, 105,314-15 Poverty Bay, 386
Olympus, Mount, 182 Presque Isle, 293
Ontario, 48; south-east, 250-1 Pribilof Islands, 34
Oregon, 89, 223,283, 321,376 Prospect Hill, 268
Orford Ness, 389 Puerto Rico, 43, 308
Orinoco delta, 234, 242; River, 298, 362; Puget Lowland, 133, 338
shelf, 383 Pyalitsa, River, 318-19
Orkneys, 181 Pyla dune, 213
Oslo, 46, 93
Oso Creek, 244 Quebec, 97, 163,221
Otago, central, 199 Queensland, 244
Otalampi, 250-1 Quinnipac, River, 361
Ottawa, 138; city, 221 Quinze, River, 97
Ourthe, River, 368
Outer Hebrides, 1 8 Rainier,Mount, 72
Oxford, 52, 264, 282 Rak, River, 369-71
Oyo, 198 Rapaalven, 2 1
Ozarks, 64,214,392 Raton Basin, 44
Osterdal, 230 Recife, 240
Osterdalsisen, 3 14-15 Red River, Louisiana, 233
Red Sea, 81-2
Pacific Ocean, 84, 85, 243 Reedy glacier, 340
Pakistan, West, 244 Regina-Hummingbird trough, 102
Palm Beach, 342 Reigate, 108
Palouse region, 302 Renko, 378-9
Pampa de la Joya, 281 Resolute, 9, 307; Bay, 195
Papua, north-east, 191 Rhine, River, 210; valley, 39
Paris basin, 157; east, 296 Rhone, delta, 234; River, 210, 234
Patagonia, 124, 309 Rhodesia, 245, 264-5, 270, 288, 353, 410
Pays de Caux, 163 Rhum, 270
Pazin, 369-7 Ridge and Trough province, 342
Peak Cavern, 256 Rijecine, River, 367
Pembina, River, 212 Rio Bueno, 176
1 1 1 1 8

Scary, 364
Rio de Janeiro, 107, 176
Scilly Isles, 305
Rio Grande, 223
Scoresby Sound, 179
Rio Grande, Brazil, 298
Scotland, 46, 100, 101, 108, 146, 192,214,
Rio Negro, 362; basin, 343
Rock River, 180, 393 266, 300, 341, 342, 343, 375, 381; cen-
Rockville, 316; Quadrangle, 272 tral, 181, 289, 376; north, 359, -central,
Mountains, 109, 298, 315; Colorado, 301,-east, 270, 301, 312,-west, 180, 181,
Rocky
148,227 249, 311, 340; southern, 169, 267, 410;
Romania, 211 west, 135, 270
Rombaksbotn, 392 Scottish Highlands, 311, 375
Rosenlund cliffs, 1 60 Seacourt Hill, 52
Ross Sea, 124, 179—80 Semi-arid areas, 58, 59, 115, 186, 210, 214,
Rothamsted, 150 260, 267, 291, 293, 353

Ruby Range, 299, 300 Sebhah, 247, 303
Ruby-East Humboldt Range, 176 Senegal, coast, 241; delta, 389; River, 361,
Rutherglen, 138 391
Ruwenzori, Mount, 57, 309 Serra do Mar, 59, 95
Severn, River, 361
Sahara, 28, 58, 68, 128, 214, 244, 326, 371, Seymour, Mount, 125; River, 363; valley,
399; central, 202-3 363

St Alban, Quebec, 22 Shakwak valley, 178
St Ambroix, 257 Shap Fell, 192
Sainte Anne cave, 368 Sheffield, 148-9
St Elias Mountains, 121, 178, 300 Sheridan glacier, 375

St Lawrence River, 59, 86, 215, 237, 359; Sherman glacier, 101, 295, 375; slide, 221
Island, 285 Shreveport, 233
St Mary’s River, 48 Siberia, 113, 171; western, 140
St Vincent, 146 Sierra de los OrgMos, 199
Sakami, Lake, 163 Sierra Nevada, 34, 83, 192
Salisbury, Rhodesia, 192, 288 Silwood Park, 151
Salmon glacier, 122 Sind, 360-2
Salpausselka I, 250-1 Singapore, 192
Salskaya steppe, 308 Siward glacier, 214
Salton dunes, 213, 245, 305 Skane, 181
San Andreas fault, 34, 45, 84, 98 Skelton inlet, 179-80
San Benito Islands, 382 Skjomen, 109, 188, 270
San Catalina Mountains, 233 Skocjan valley, 371
San Clemente, 343, 382—4 Slaetharatindur, 227
San Diego, 308 Slims, River, 215; valley, 190
San Gabriel Mountains, 154, 302 Slovenia, 369-71
San Mateo Mountains, 3 1 Smith River basin, 344
San Sanquin valley, 229 Snake, River, 365-6; basin, 38; plain, 345
Sandy Hook, 211,239 Snowshoe, 109
Sangre de Cristo Range, 230 Snowy Mountains, 2 1
Santa Barbara, 213 SogneBord, 179, 182
Santa Catalina Mountains, 263 Solent, 361
Santa Monica, 207 Somalia, 57
Sao Paulo, 192 South Africa, 171, 194, 244
Saskatchewan, 257; River, 212 South America, 85, 223, 323; east coast, 259
Sawtooth Ridge, 347 South Australia, 34, 193, 195; southern, 294
Scandinavia, 122, 254, 270, 325, 380, 382; South Dakota, 371
northern, 31 South Downs, 5 1—2
Scania, 125, 240 South Lyngen, 1 1
1 11 1 1 5 1 7

South Pole, 58 Teifi, River, 361, 398-9
South Victoria Lands, 113 Temperate latitude areas, 123, 125, 155,
Southern Bug, River, 230 164, 165
Southern Uplands, 78, 225, 266, 300, 375 Ten Thousand Islands, 243
Spain, 30, 157 Tennessee valley, 369
Spitzbergen, 46, 66, 218, 219, 326, 342 Tesuque valley, 230
Spring Cave, 329-30 Texas, 240, 244, 247, 264, 287, 365, 371
Spurn Head, 18, 177, 238-9, 389-91 Thailand, 212; Gulf of, 234
Stockholm, 125 Thames, River, Canada, 212
Stonington Island, 306 Thomaston, 340
Storfiord, 181 Threatening Rock, 3 14, 3 19
Storglaciaren, 321, 334, 41 Three Forks basin, 363
Studlands, 176 Three Pigeons plateau, 282
Styx, Lake, 256 TillT, 368
Sub-humid areas, 245, 308, 346, 371, 372 Timor, 383

Sub-tropical areas, 193, 21 Tin Mountain, 221
Sudan, 98; central, 267; south-central, 347 Tista, River, 363
Surinam, 242 Tombstone, 1 1
Surte, 221 Tone, River, 212
Susquehanna, River, 230, 276, 377 Tonga, 260
Sutherland, 133 Transcaucasia, 121
Svartisen,315 Trans-Volga region, 137
Swan Hill, 337 Trinity Alps, 124, 311, 338; Bay, 98;
Sweden, 73, 107, 108, 113, 115, 116, 117, County, 233
1 18, 124, 125, 176, 210, 215, 218, 219, Troms, 1 1
221, 230, 259, 308, 325, 374; arctic, 188; Trondheim, 46
central, 164, 250-1, 377; northern, 158, Tronto, River, 212
160, 209, 216, 218, 234, 235; south- Tropical areas. 28, 31, 57, 59, 60, 67, 95,
central, 249; southern, 377 116, 133, 150, 151, 160, 174, 189, 193,
Switzerland, 124, 283 196-9. 206, 220, 222, 237, 243-4, 259,
Sylhet Basin, 43 265, 267, 270. 289-92, 315, 347, 383;
Syria, 212 dry, 14, 57, 288; humid, 41, 57, 63, 107,
112, 113, 155, 170, 171, 174, 175, 189,
Tabasco, 239 191, 194, 217, 218, 230, 231, 263, 276
Tacoma, Mount, 83 Truckee canyon. 362; River, 206
Taijord, 220 Tucson, 229
Tagliamento basin, 116 Tunguska, 9
Tahoe, Lake. 44 Tunisia, 396, 399—400
Tama, River, 301 Turenki, 379
Tanga, 176, 240 Turkestan, 55, 356
Tanganyika, 5 Turkey, 30, 87, 212, 3 1 1; western, 234
Tapitallee Creek, 372 Turkmenia, 214
Tarawera Mountain, 9! Tweed, River, 393
Tam basin, 169
Tarraadno, River, 210 Uchau-Su valley, 2 1
Tasman glacier, 124 Uganda, southern, 360-1
Tasmania, 95, 193, 265; central, 182 Ukraine, 320; south, 321
Tatra Mountains, 218, 285 Uludal Massif, 3 1
Taunus Mountains, 219 Uraanak Bay, 124
Tawe, River, 361 Uttar Pradesh, 228
Taylor’s Falls, 184 U.S.S.R., 60, 91, 113, 115, 116, 155, 165,
Taylor valley, 214
1 13, 166, 185, 295, 301, 318; Central
Teays, River, 192, 363—4, 392 Highlands, 312; European, 154, 157, 158,
1 1 1
Wanderrie country, 260

276, 307;
319; plain, 230, Wankie National Park,
151
188, 210,
northern, 224 Wapsipinicon, River, 392
eastern 44, Wash, 236

242,’244, 342, 395; western, 338
Washington, 133, 321, 376;
^1t9’223,
;S?2toast,
north-east, 56;
46 343;
north-west, 328 313
.^ 3 ,
Watts Branch, 115, 272
Weald, 108
267, 299,
209, 259, 260,
south-west, 169, West Germany, 212
223, 285
361; western, West Indies, 190, 194
Uppsala, 393 West Kimberley 196
192, z
Western Australia, 67,
,
Ure, River, 95
362 346
Utah, 95 ; northern, 267, 269,
356 187
Uzbek S.S.R., Western Coachella valley,
Westport Bay, 393
95
Valday,307 Wharfe, River, Yorkshire,
160, 17 67
Vattern, Lake, 108, White Mountains, 1
Vecchio valley, 179 Wicken Fen, 1 62
Venezuela, 190, 212,
288
Wiluna, 192
Vermont, 96, 175, 289 Windang Island, 305
Vernon, Mount, 306 Windrush, River, 1 1
Vestspitsbergen, 200 Wingate, 287

337 323, 39
Victoria, Wisconsin, 151, 152,
Victoria, Lake, 53,
360-2 line, 373
Wolverhampton
245
Victoria Und, 139; valley, 193, Worcestershire, 325
Vikkra, River, 23 Wyoming, 109, 221,263,308
Villeneuve, 210
Virgin Islands, 43 Yakutat, 121
Virginia, 233, 363 Yangtze-Kiang delta, 234
Visingso Island, 176 Yellowstone Park, 90
Vladivostok, 342 1 17
Yenessei, River, 1 16,
Volga delta, 234 Yesilirmak, River, 212
Voronezh Province, 137, 165 Yodo, River, 212
Vosges, 82, 109,283,311
Vridi,211 Wk.h“95,»7,106,.62,eo.s.,177,393.
coast, 351,
-103-4 39 li north-east
Waal, River, 212
Wadden Sea, 238-9 valley, 1 1
Yosemite
Wadena, 135, 253, 254
Yucatan, 115
Wadi Kurf, 262-3 Yugoslavia, 350, 367
Waipaoa, River, 212 190
Yukon, 178; south-west,
Walensee, 210
375, 392;
Wales, 361; North, 227, 311, 100
Zagros Mountains, .
South, 105, 157,269,375

11 8
Author Index
(including bibliography)
Abaturov, A. B. D., 152 Bachmann, F., 472

Ackermann, E., 1 1 1, 354 Baden-Powell, D. F. W., 10, 424
Adams, J. A. S., 441 Bagnold, R. A., 11, 15, 26, 68, 126, 201,
Agar,R!, 351. 354,479 207, 419, 437, 445, 453, 460, 462
Ahnert, 420,461,479
F., Bailey, H. P.,472
Albritton, C. C., 428, 453 Bailey, R. G.,457
Alexander, C. S., 418 Baker, C. H., 476
Alexander, F. E. S., 450 Bakker, J. P„ 422, 437, 447, 479
Allen, J. R. L., 12, 304, 419, 458, 461, Balchin, W. G. V., 444, 470, 479, 485
470 Ball, D. F., 450, 467
Allen, P.. 428 Ballard, A. 447J.,
AIojzij.H., 351 Barbier, M., 455

Alsop, L. E.,431 Barker, J. B., 414
Ambrose, J. W., 359, 477, 480 Barry, R. G., 435
Andel, T. H. van, 437, 460 Barton, D. C., 450
Anderson, C. A., 432 Bascom, W. N., 273, 436, 464
Anderson. H. W., 437, 455, 464 Bassett, H., 468
Anderson,!. E., 163 Bastisse, 6., 189
Andersson, J. G., 270, 456 Basumallick, 482
S.,
Andresen, M. J., 474 Battistini, R., 485

Andrews, J. T., 128, 251, 310, 317, 331 Battle, \V. R. B., 447
459, 463, 475, 483 Baulig, H., 49, 401, 419, 420, 425, 428, 486
Andriesse, J. P., 464 Baver, L.D., 300,437,474
Anstey, R. L., 459 Baxter, F. P., 152
Aoki, K., 432 Bay,C. E., 441
Aranda, J. M., 44 Beadnell, H. J. L., 213,453
Arber. M. A., 2, 4, 4 1 Beaujeu-Garnier, J., 438
Archer, D. R.,457 Beaty, C. B., 309, 36 1, 459, 470, 480
Arkell, W. J., 52 Beaumont, P.,449
Armstrong, R. L., 473 Beckett, P. H. T., 479
Arnaud, R. J. St, 185-6, 449 Beckinsale, R. P., 18, 414, 419, 435, 472,
Arnborg, L., 472 480
Ashley, G. H., 64 Beete, J. B., 480
Attiwel, P. M., 44 Beheiry, S. A., 462
Auden, J. B., 433 Behre, C. H., 453
Avenard, J. M., 437 Behre,J. S., 207
Aver, V., 475 Bemmelen, R. W. van, 430

Axelrod, D. I., 38, 424, 472, 475 Benedict,!. B., 227,320,459
Axelsson, V., 158, 444 BeniofT, H., 432
1 1 1
Author Index 515
Bennett, R. G., 315 Bressau, S., 444

Bergdalh, A., 461 Bretz, J. H., 425, 427
Berry, L., 347, 446, 479 Brewer, R. W., 136, 268, 438
Berthois, L., 438, 455,461 Bridge, C.W., 471
Beschel, R., 333,475 Broecker, W. S., 47, 472
Bettenay, E., 257, 446 Brophy,J. A., 184, 448,483
Bibby, J. S., 78, 225, 266, 440 Brown, E. H., 421, 425, 479, 480, 486
Bik, M. J.,470 Brown, R. J. E., 436
Billings, M. P., 430 Bruce, J. P., 435
Bingham, C., 465 Bruckner, W., 121,468
Bingham, F. T., 442 Brunet, P., 470
Bird, E. C. F., 305,414 Brunsden, D., 486
Bird, J.B., 414 Brush, L,M., 359, 480
Birkeland, P. W., 455, 482 Brunn, P., 386, 421
Birman, J. H., 338 Bryan, K., 17, 345,419,479
Birot, P., 14,59,62, 63, 146, 148, 191,347, Bryan, R. B., 438
414,428,450 Budel, J., 55, 294, 324, 340, 426, 469, 472
Black, R.F.,414, 438 Buffington, E. C., 382, 384, 484
Blackburn, G., 476 Bull, W. B., 258, 460
Blackwelder, E., 338, 450, 456 Bullard, E. C., 43 1,432
Blagbrough, J. W., 349, 459 Bullen, K. E.,432
Blake, D. H., 394, 487 Bunting, B. T., 359, 450
Blanc, A,, 480 Burk, C. A., 478
Blanck, E., 186, 450 Burkalow, A. van, 228, 459
Blissenbach, E., 233, 263, 460 Burnham, C. P., 465
Bloom, A. L., 46,414,424 Butler, J. R., 440

Bluck, B. J., 205, 269, 460, 461, 464 Butuzova, 0. V., 44
Blumche,!., 185 Buteer, K., 30, 337, 414, 421, 444
Boer,G.de, 18,391,419 Buxtorf, A., 399
Bogardi, J., 455
Bogli, A., 45 Cady, W. M., 445
Bond, G., 462 Cailleux, A., 2, 8, 16, 30, 55, 59, 64, 75,
Bone, R. M., 140 129, 131, 150, 185, 299, 409, 417, 418,
Bonifay, E., 449 421, 438, 449, 452, 455, 460, 464, 470,
Boon, J. D., 455 471
Borns, H. W., 482 Caine, N., 95, 283, 433, 459, 467
Boswell, K. C., 17 Calkin, 452,471
P., 8,
Boulaine, J., 44 Callender, G. S., 470
424
Bourdier, F., Camels, A., 464
432
Bout, P., Camerman, C., 433
Bowden, K. L., 6 Cann, R. J., 469
Bowen, D. Q., 476 Capolini, J., 257
Bowen, R. N.C.,472 Capper, L., 438
Bowler, J. M., 480 Carlston, C. W., 277, 278, 364, 466, 469,
Boye, M., 449, 477 481
Boyer, P., 441 Carozzi, A. V., 441
Boynthon, C. W., 480 Carr, A. P., 454, 464
Braddock, W. A., 97, 433 Carsola, A. 478 J.,
Bradley, W. C., 97, 433, 447 Carson, C. 471 E.,
Bradley, W. 11,419
H., Carter, C. S., 26, 479
Brajnikov, B., 452 Cassidy, W. A., 433
Brandtner, F., 473 Cassie, W. F., 438
Breed, W. J.. 349,459
Catt,J. A., 304,425, 439
6 1
444 Crickmay, C. H., 447
Cavaille, A.,
Croll, J., 324, 472
Cazalis, P., 421
Cuerda, J., 30, 421
Chamberlin, T. C., 42, 424
Cullen, D. J., 464
Chandler, R. F., 41
Culling, W. 437
E. H.,
Chandler, T. J., 479
Cunningham, R. K.,154, 442
Chang, P., 480
Curray, J. R., 239
Chantladze, Z. I., 301
Currey, R. R., 472
Chapman, C. A., 97, 433
Curtis, G. H., 476
Charles, G., 468
482 Curtis, L. F., 418
Charlesworth, J. K., 374, 414, 463,
Curtis, R. H., 447
Chemin, E., 442
Czech, H., 17
Cherry, J. A., 453
Czeppe, Z., 438, 449, 472, 473
Chilingar, G. V., 468
Cholley, A., 427 452
.
Dahl, R., 188, 191, 399, 448, 450,
,
Chorley,R.J., 18,26,69-70,314,419,420,
1
Dahlskog, S., 442

423,424,426,428,435,437,479
Dalrymple, G. B., 476
Christensen, M. N., 424
Darby, H. C., 418
Church, R. E., 474 436
Darbyshire, J.,
Churchill, D. M., 424
Darwin, Charles, 147
Churchward, H. M., 337
Daubree, G. A., 95
Clark, M.J., 438
Davidson, C., 453
Clark, R- H., 435
Davidson, C. F., 457
Clayton, F. C., 439
Davidson,!., 445
Clayton, K. M., 196,414,486
Davies, A., 481
Clibbon, 420P.,
Davies, G. H., 464
Coats, R. R-i 432
Davies, J. L., 18, 419, 448
Coetzee, J. A., 472 476
Davis, M. B.,
Coe, K., 435
Davis, S. N., 458
Colby, B. R., 444 9-1
Davis, W. M., xiv, xv, 7, 1.
Coleman, A. M., 479 42-3, 48-9, 60-6,
20, 24, 30, 33, 37,
420
Collias, E. E., 357, 358, 362, 415,
68, 74-6, 191, 341,
456
Collin, A.,
425
Conacher, A. J., 428
Day, A. A., 478
Conant, L. C., 246
Defant, A., 436
Conway, V. M., 162
De Jong, J.D.. 485
Cook, F. A., 9, 445 Demek, J., 200, 202-3, 443,
452
Coope, G. R., 472 Demangeot, J., 468
Cooper, W. S., 462
Demolon, A.,189
Cooray, P. G., 456 Demorest, M., 304
Coque, R., 396, 415
Denny, C. S., 445
Corbel, J., 434, 444, 445 482, 483
Denton, G. H., 300, 473,
Corbett, E. 457
Derbyshire, E., 182, 421
S.,
Corliss, J. B., 462

Derruau, M., 415, 432
Cornwall, 1. W., 440
Devdariani, A. S., 420
Corte, A., 467 Dhonau, T. J., 456
Costin, A. B., 467
Dhonau, N. B., 456
Cotton, C. A., 4 15, 4 1 8, 426, 432, 445, 447,
Diaconu, O., 455
472,484 478
Diett, R. S., 466,
Courts, J.R. H., 441, 455
Cox, A., 3 12, 457, 476 335, 473, 476
Dimbkby’G. W., 285,
Craig, G.Y.,419 Dines, H. G., 266
Crary, A. P., 180, 305,
447
Dixey,F.,45,361,424,486
Crcager, J. S., 420
Author Index 517
Dobbie, C. H., 423 Epstein, E., 26
Dobson, P. J., 436 Epstein, 434

J. B.,
Doeglas, D. J., 423, 438 Erhart, E. H., 442

Doell, R. R., 476 Escher, B. G., 433
Dolan, R., 466 Evans, G., 152, 461
Dollar, A. T. J., 432 Evans, J. W., 442
Donahue, 132,439
J., Evans, O. F., 461
Donn, W. L., 424, 473 Everard, C. E., 486
Donner, J. J., 10, 134 Everitt, B. L., 166-7, 442
Doornkamp, J. C., 361, 418, 420, 450 Evernden, J. F., 447, 476
Dorall, R. D., 428 Ewing, G. C., 387, 462
Dorr, J. V. N., 289 Ewing, M., 424, 430, 472, 473
Douglas, I., 41, 1 15, 263, 423, 443, 458 Eyies, R. J., 464
Downie, C., 482
Downing, B. H., 453 Faegri, K., 476
Dragg, D. L., 429 Fairbridge, R. W., 47, 182, 199, 323, 325,
Draper, 436
L., 415,424, 456
Dragovich, D., 452 Fairchild, H. L., 448
Dresch, J., 445, 473 Falconer, G., 463
Driscoll, E. M., 484 Falconer, J. D., 199, 450
Droste, J. B., 468 Farrand, W. R., 48, 424
Duchaufour, P., 445 Feininger, T., 452
Dunn, A. J., 18, 419 Fenelon, P., 452
Durand, J. H., 353,479 Fenneman, N. M., 1, 51, 64, 417, 425, 460
Durand, R., 453 Fern, J. C., 429
Durrance, E. M., 478 Ferrar, A. M., 420
Dury, G. H., 357, 372, 406, 415, 418, 468, Fidalgo, F., 300, 471
470,473,480,481,482,486 Fieldes, M., 440
Dylik, J., 75, 427, 429, 438, 449, 456, 459, Finkel, H. J., 8, 462
473 Fisher, R. V., 478
Dylikowa, A., 419 Fiske, R. S., 432
Dzulynski, S., 467 Fitzpatrick, E. A., 445, 450
Fleischer, R. L., 433
Eakin, H.M., 452 Flemming, N. C., 467, 484
Easterbrook, D. 133,415,438
J., Flinn, D., 484
Eaton, I.P., 88,432 Flint, R. F., 92, 179, 180, 298, 300, 355,
Eardley, A. J.,445 359, 415, 416, 424, 434, 445, 462, 470,
Eckel, E. B., 456 471,473
Richer, D. L., 97, 433 Folk, R. L., 129, 130, 131, 132, 134, 389,
Eiler, J. P., 206 429, 438,465,484
Ek, C., 368, 458, 460, 481 Fourneau, R., 418
Eliassen, S., 456 Fournier, F., 212, 435, 455
Elliott, E. L., 461 Fox,C., 445
Ellison, W.
455 D., Frakes, L. A., 477
Embleton,C.,415,486 Frecaut, R., 454
Embody, D. R., 457 Frederickson, A. F., 440, 442
Emery, K. O., 450, 454 468
Freise, E. W.,
Emiliani. C., 324, 328, 342, 473, 476, 473
Frenzel, B,,
477 Frey, D.G., 417
Emmett, W. W., 448
Friedman, G. M., 438
Engel, C. G., 468
Frye, J. C., 55, 395, 426, 473, 486
Engeln, O. D. von, 454 Funnell, B. M., 439
Enquist, F., 335
Furley, P. A., 271, 464
1 1 1 1 1
Galileo, 28 Gretener, P. E., 423

426
Galli-Olivier, C., Griffen, G. M., 258
Galloway, R. W., 445 Grifren,J.J.,455
Galvin,C.J.,419 438
Griffiths, J. C.,
Gams, I., 150,460,481 Grim, R. 441
E.,
Gardner, J., 454 Gripp, K., 473
Garland, G. C., 43 Grob, A., 165
Garnett, A., 471 Grove, A. T., 457, 473
Garrels, R. M., 458 Grove, J. M., 483
Gaskell, T. F., 430 Guilcher, A., 239, 415, 424, 451, 452, 461,
Geer, G. de, 74, 336, 476 468, 484
Geike, A., 3 7, 445 Guillien, Y.,473
Geike.J., 181,448 Gulliver, F. P., 7
Geiger, R., 435 Gutenberg, B., 80, 432
Geiss, J., 477
Gellert, J. F., 429, 463 Hack, J. T., 64, 65, 75, 427, 442, 470
Gentilli, J., 473 Hadley, R. F., 263, 445
George, P., 481 Haggett, P., 69, 428
George, T. N., 486 Haig. M.R., 361
Gerber, 452
E., Hails. J. R., 424, 438
Gerlach, 218, 314,456 Hamelin, L. E., 420, 445
Gerstenhauer, A., 452 Hamilton, E. 1., 476
Gessel, S. P., 442 Hamilton, R. A., 418
Ghenovici, A., 454 Hancock, P. L., 434
Gilbert, G. K.. 20. 40. 96, 181, 210,241, Hansen, C. L., 414
429,434,454,471 Hanson, L. G., 456
Gilluly, A.,415,424.445 Hardy, J. R., 454
Gilluly.J., 293,345,430,469 Harford. L. B.. 480
Gipson, M., 94 Harland, W. B., 430, 43 1,448
Giresse, P., 454 Harms, 233,466
J. C.,
Gjessing.J., 178, 182,448 Harrington, J. W., 423
Gladfelter, B. G., 4 14 Harris, S. A., 463
Glenie, R. C., 484 Harrison, J. M., 429
Glentworth, R., 312 Harrison, K., 434
Glinka, K. D., 188,451 Harriss, R. C., 441
Godard, A., 133, 180,340,415,452 Hastenrath, S. L., 4 10, 462
Godwin, H., 476 Hawkins, A. B., 478
Gohren, H., 454 Hay, R. L.. 432, 45
Goldthwait, R. P., 482 Hay, T., 460, 467
Goodchild, J. G., 45 Haynes, V. M., 448
Goodier, R., 467 Hays, O. E., 44
Goodlett, J. C., 442 Healy, T. R., 442
Gorham, E., 458 Heath,:. P.,334,476
Gorsline, D. S., 484 Hedges, J., 452
Gould, S. J., 423 Heezen, B. C., 81, 430, 43 1, 472
Grant, W. J., 26 Heidel, S. G., 464
Gravenor, C. P., 463 Hellyer, B., 433
Grawe, O. R., 448 Hem,J.D.,458
Green, C., 444 Hembree, C.H., 471
Green, R, R., 430 Hempel, L., 425, 427
Gregory, J. W., 460 Henderson, M. E. K., 443
Gregory, K. J., 418, 486 Hendricks, D. M., 45
Gresswell, R. K., 385-6, 484 Hendricks, E. L., 374
1
Author Index 519

Herdan, G., 439 I.F.I.P.I.C.C., 420
Herodotus, 30 Irani, R. R., 439
Hey, R. W., 462, 473 Iversen, 476
J.,
Hickin, E. J., 460 Ives, J., 463, 486

Hill,D.E., 451 Ives, R. L., 336, 476
Hill, P. A., 434
Hills, E. S., 53, 62, 425, 427, 430 Jackson, G., 149
Hinschberger, F., 478 Jaeger, J. L., 441
Hirsch, A. R., 419 Jahn, A., 271, 285, 457, 479
Hjulstrom, F., xv, 1, 184, 201, 207, 417, James, G. T., 476
481 Jamieson, T. F., 25 1, 486
Hoadley, P. G., 6 Jelgersma, S.,485
Hobbs, W. H., 66 Jenness, S. E., 483
Hodgson, 3. M., 5 1-2, 425 Jennings, J. N., 72, 196, 269, 415, 443, 444,
Hodgson, R. A., 94, 434 463,470,471
Hodgson, W. A., 445 Jenny, H., 442
Hogbom, B., 185, 197, 445 Jessup, R. W., 397, 400, 486
473
Hoinkes, H. C., Johansson, C. E., 134, 439, 458
Hole,F.D., 152,323 John,B. S., 399,483
Hollingworth, S. E., 427, 434, 457 Johnson, Douglas W., 7, 381, 382, 386
Holmes, A., 415, 476 Johnson, J. W., 454
Holmes, C. D., 426, 428, 465 Johnson, R. H., 457, 486
Holmes, D. A., 481 Johnson, Willard D., 181
Holmsen, G., 463 Johnsson, G., 467, 473
HoUedahl, H., 179, 182, 183, 342, 448, Johnston, G. H., 436
478 JoUifTe, I. P.,454
HoUedahl, O., 220, 255 Jones, F. O., 457
Hommeril, P., 473 Jones, J. G., 432
Hooke, R. LeB., 339, 421, 428, 466 Jones, O. T., 398-9, 481
Hopkins, B., 442 Jones, R. J., 196-7,442
Hopkins, D. M., 467 Jopling, A, V., 74, 208, 454, 461
Hopley, D., 484 Jungerius, P. D., 439
Horlock, J. H., 466
Horton, R. E., xiv, 4, 6, 20, 23, 26-8, 65, 77, Kaiser, K., 474
273, 358,415,418 Kaizuka, S., 476
Hough, J. L., 486 Kaiterherberg, J. van, 434
Hough, V. N. D., 109, 434 Kamb, 436
B., 123,
Houtman, T. J., 473 Kandil, M. F., 455
Howard, A. D., 422, 428 Kaneko, S., 484
Hoyt, VV. G., 436 Karnik, V., 432
Hsi-Lin,T., 452 Karpachevskiy, L. O., 152
Hume, J. D., 423 Kaye,C. 335,476
A.,
Hurley, P. M., 43 457
Kayser, B.,
Hussey, K. M., 471
Keen, M.J., 415
Hutchings, G., 420 Kellaway, G. A., 434
Hutchinson, G. E., 435
Keller, W. D., 90, 147, 422, 441, 442, 468
Hutchinson, J. N., 444
Kelley, J. C., 456, 465
Hutton, J., 37 Kelley, T. E., 476
Kelvin, Lord, 74
Ichikawa, M., 464 Kendrick, M., 461
Illing, L. V., 469 Kennedy, W. Q., 34, 422
Ingle, J. C, 69, 281,455 Kent, P. E., 454
Inman, D. L., 129, 239, 439, 445, 462 Kerney, M. P., 479
Kerr,P. F., 434, 467 Langford-Smith, T., 468

Kershaw, K. A., 468 Larras, J., 391
Kidson, C., 391, 423, 436, 442, 443, 454, Larsen, G., 468
462,484,486 Larsonneur, C.,473
King, C. A. M., xv, 13, 31, 415, 422, 461, Lass, G. M., 227
462 Lauf,G. H.,461
King, L. C., 45, 49, 50, 54, 55, 59, 63, 73, Lawson, A. C., 180
84, 344, 399,415,424,426,427,437 Lawton, K., 443
King, W. B. R., 484 Leahy, E. J., 251
Kirby, R. P.,251,463,486 Leatherdale, J. D., 263
Kirkby, M. 437
J., Le Bourdiec, F., 419
Kistler, R., 476 LeGrand, H. E., 17, 35, 98,420,422,452
Kittleman, L. R., 439 Le Heux, J. W. N., 479
Kitts, D. B., 427 Lehmann, H., 426
Klatka, T., 449, 467 Leighton, M. M., 37, 184, 448, 483
Kliewe, H., 484 Lensen, G. J., 481
Klimaszewski, M., 422, 448, 478 Lenze, E,, 305
Klinge, H.,451 Leonard, A. B., 395, 486
Klovning, I., 183 Leopold, L. B., xv, 1, 8, 20, 54, 75, 263, 272,
Knebel, H. 465
J., 278, 406, 418, 420, 448, 466, 470, 481
Koczy, F. F., 477 Lettau, H., 466
Kofoed, J. W., 484 Lettau, K., 466
Kondiferova, E. A., 278 Levelt, W.M., 451
Koon, D., 445 Leverett,F., 122,311,471
Koster, E., 439 Lewin, J., 438
Koyanagi, R. Y., 432 Lewis, C. A., 227
Krammes, 322
J. S., Lewis, P. F., 486
Krauskopr, K. B., 458 Lewis, W. 447,448,484
V.,
Krenkel, P. A., 6 Libby, W. 477

F.,
Krigstrom, A., 232, 460 Lichty, R. W., 14,429,470

Krinov, E. L., 433 Liebling, R. S., 434
KrinsIey,D. H., 132,439 Link, W.K., 21

Krumbein, W. C., 74, 129, 420, 428,439 Linton, D. L., 197, 300, 415, 422, 436, 448
Krumbein, W. E., 442 452, 471,483,487
Kuenen, P. H., 47, 132, 187, 285, 424, 430, Liu Kuang-Yeh, 425
450,467 Livingstone, D. A., 458
Kuhlman, H., 455 Lliboutry, L., 123, 140, 285, 415, 436, 445
Kiihn-Velten, H., 434 Longuet'Higgins, M. S., 285
Kukkamaki, T. J., 425 Longwell, C. R.,416
Kupsch, W. O., 205, 457, 463 Lopatin, G. V., 223
Loring, D. H., 461
La Chapelle, E., 123, 436 Louis, H., 5 1,426,446,452
Lachenbruch, A. H., 467 Loughnan, F. C., 441
Ladurie, E. Le R., 474 Loveday, J., 439
Lake, P.,415 Lovering, T. 165, 442
S.,
La Marche, V. 168,233
C., Lovering, 430
J. F.,
Lambert, J. M., 444 Loken, O. H., 251,463

Lambrick, H. T., 361 Lull, H. W., 442
Lamotte, M., 468 Lumsden, G. L, 481

Landim, P. M. B., 477 Liitig, G., 474
Landsberg, S. Y., 471 Lutz, H. J., 148, 442
Langbein, W. B., xv, 26, 436, 466, 470 Ludqvist, J., 374
Langer, A. M., 467 Lux, H., 442

1 1
A iithor Index 521

L’Vo\ich, H. I., 170 McLennan, A. G., 483
Lyeil.C., 15,37, 62 McManus, D. A., 420
Lyants. G. W., 443 McManus, J., 478
Meade, R. H., 444
Maack, R,, 43 Meier, M. F.. 33,422,436
Maarleveld, G. C., 471 Meinzer. O. E., 436
Mabbutt. J. A., 292, 415, 452, 469, 471 Meland, N., 454
Macar, P., 271, 314, 416, 426, 457, 487 Melieres, F., 163, 443
Macdonald, G. A., 432 Mellor, M., 436
Machatschek, F., 416 Melton, F. A., 355, 427
Mackay, J. R., 353, 467, 479 Melton, M. A., 27, 63, 74, 347, 358, 420,
Mackenzie, F. T., 458 439, 465,471,479
Mackin, J. H., 21, 420, 429, 470 Menard, H. W., 64, 478
Mackney, D., 465 Mench, V.. 226,458
Macneil, F. S., 462, 484 Mensching, H. von, 446
Madigan, C. T., 463 Mercer, J. H., 483
Mainguet, M., 467 Merritt, R. S., 477
Malde, H. E., 423 Mescherikov, J. A., 430
Malez, M., 371 Michaud, J., 455
Mallory, J. K., 478 Michler, 1., 482
Maramerickx, J,, 367, 446 Milankovitch, M., 324, 474
Manner, S. B., 464 Mille, G. de,102,434
Mannerfelt, C. M„ 377, 486 Miller, A. A., 425, 481
Marcus. A. H., 429 Miller, C.F., 418
Marinescu, A.. 478 Miller, D. J., 423
Marr, J. E., 416, 434 Miller, J. P., 1, 4, 17, 54, 73, 75, 263, 272,
Marsh, C.P., 169 278, 406, 418, 419, 423, 428,429,481
Martin. H., 478 Miller, R. L., 420, 466
Martin. L., 429 Miller, T.G.. 416
Martin, R., 356-7,478 Miller, V.C., 418
Martonne, E. de, 55 Millman, A. P.,443
Mason, B., 480 Mills,!. P., 163
Mason, B. H., 441 Milne, G. M., 35
Mathews, W. H., 436 Milnes, A. G., 430
Matthes, F. E., Ill Min Tieh, T., 455
Mattson, A., 416 Mitchell, G. F., 375, 484, 486
Maud, R. R.. 469 Moberly, R., 474
Maul), 0.. 416 Mohorovicic, A., 80
Maj-nard, 458
J. E.. Mohr, E. C. 446
!.,
McBimey, A. R.,432 Moller, H., 25
McCammon, R, B., 439 Monkhouse, F. !., 418
McCann, S. B., 135, 439 Moorbath, S., 477
McCoy, R.M., 418 Moore, D., 461
McCracken, R. J., 69 Moore, D. G., 239, 382, 384, 484
McDougall, L, 446
Moore, E. S., 458
McGee, W. J., 454
30, 422, Morgan, R. S., 417
McGill,!.. 416 Morisawa, M., 5, 27. 418, 420, 436
McGinnis, L. D.. 425
Morris, E, C., 429
Mclntire, W. G., 468 Morris. P. G., 474
McKee, E D., 246, 247, 463 Morrison, A., 474
McKenzie. G. D.. 463, 483
Mortensen, H., 443, 446, 450. 469
McKenzie, P., 3 16
Mortland, M. M., 443
McLean, R. F., 443, 457 Moseley, F., 469, 483
Moss, A, J., 439 Ostrem, G., 3 1, 214, 3 10, 422, 455, 463, 47
Moss, R. P., 202-3, 469
Moulline, M., 477 Pans, N., 439
Moller, J.T.,466 Palmer, H. R., 466
Mudge, M. R.,459 Palmer, H. 195,452
S.,
Muir, J.W., 312 Palmer, 197,452

J.,
Muir, R. O., 104 Parde, M.,455

Muller, E. H., 477 Parizek, E. J., 459
Muller, F.,467 Parkin, D. W., 285
Munn, R. E., 436 Passarge, S., 186, 446, 450
Munnich, K. O., 477 Paterson, W. S. P., 437
Murata,K.J., 88,432 Pecsi, M., 446, 459
Myrick, R. M., 448 Pedro, G., 450
448
Peel, R. F., 429, 446,
Naidu, A. S., 263 Peguy, C. 420 P.,
Nairn, A. E. M., 423, 474 Peippo, J., 472

Nakagawa, H., 484 Penck, A., 55,338,426
Neal, J. T., 467 Penck, W.. 17, 34, 37, 75, 77, 300, 345, 416,
Neef, E., 426 479
Neilson, R. A., 452 Penny, L. F., 304, 439, 483
Nelson, J. G., 444 Percival, J. G., 40
Nelson, H. W., 434 Perrin, R. M. S., 463, 475
Nesteroff, W. D„ 163 Peierson, W. L., 457
Neumann, A. C., 443 Pettijohn, F. J., 440, 441
Neumann van Padang, M., 432 Pewe, T. L., 437, 474
Nichols, R. L., 227, 306, 4 1 6, 45 1, 462, 487 PfeifTer, H., 440
Nickelsen,R. 109,434
P., Phemister, T. C., 45
Nicholson, R., 434 Philberth, K., 467
Nielsen, G. A., 323 Phillips, A. W., 454
Niini, H., 427 387
Phleger, F. B.,
Nomenyi, F., 446 442
Pierce, R. S.,
Norris, R. M., 458 Pigott,C.D., 128,451
Norrman, J. O., 446, 454 Pinchemel, P., xiv, 71,416, 420
Nossin, J.J., 451 Pippan,T., 459
Nota,D.J.D.,461 Pissart, A., 460
Nowland, T. L., 455 Pillman, E. D., 464
Nye,J. F.,437 Pitty, A. F., 52, 128, 162, 202-3, 256, 257,
Nye, P. H., 152,451 258, 263, 365, 366, 404, 419, 421, 426,
443,481
Oberbeck, V. R., 429 Plafker, G., 102, 103,434
O’Brien, C. A. E., 434 Plashchev, A. V., 320
Ogilvie, A. G., 422 Plato, 37
Oldfield, F., 474 Playfair, J., 423
Olivier, J. E., 431 Plumley, W. J., 261,465
Ollier, C. D., 6, 7, 4 18, 428, 446, 450, 45 Poldervaart, A., 430
Olson, J. S., 168 Pons, L. J., 485
Orlyankin, V. A., 278 Pont, P., 451
Orme, A. R., 355, 463, 484, 486 Popov, I. V., 278
Osburn, W. H., 417 Porter, S. C., 483
Ostenso, N. A., 123, 437 Poser, H., 340, 467, 474
Oilman, F., 132, 388 Post, A., 436
Ottman, L., 30! Post, L. von, 335
Otvos, E. G., 474 Pouquet, J., 422, 455
11 1 11 6 111 1
Author Index 523

Powell, J. W., 299, 422 Robertson, E. C., 435
Pratviel, L., 477

Robinson, A. H. W., 238-9, 448, 461, 485,
Prentice, J. E., 474
487
Price, N. 434
J.,
Rognon, P., 202-3, 416
Price, P. B., 433 Romans, J. C. C., 301
Price, R.J., 252,391,464 Ronca, L. B., 330, 429, 430, 477
Price, W. A., 461, 466 Rondeua, A., 44
Prider, R. T., 469 Rosholt, J. N., 477
Prior, D. R., 457 Rougerie, G., 458, 468, 469
Pugh, J. C., 446 Riickiin, H., 43
Pullar, W. A., 477 Rudberg, S., 218, 457

Putsy, N. P., 239, 462 Runcorn, S. K., 431
Pye, N., 444, 470 Ruppel, E. J., 364, 48
Russell, I. C., 19, 168
Quaide,W. L., 429 Russell, R. J., 176, 387, 416, 429, 446, 448,
Quennell, A. M., 487 468,485
Rust, B. R., 364, 460
Ragg, J. M., 78, 225, 266, 440 Rutten, M. G., 457
Ragsdale, J. A., 235, 263, 461 Ruxton, B. P., 347, 446, 451, 479
Rahn, P. H., 271, 443, 446
Raiche, V. G., 9 Sanders, J. E.,416
Rainwater, F. H., 47 Sandford, K. 440
S.,
Randall, B. A. O., 1 1 1, 432, 435 Sarkar, S. K., 482

Ranier, J. N., 471 Savigear, R. A. G., 2, 4, 9, 419, 479
Rankama, K,, 3 74, 4 1 Schachori, A., 422
Rapp, A., 16,78, 188,218, 255, 346,416 Schaefer, W., 152
Ranson, G., 443 Schalk, M., 423
Rasid, H., 460 Scheidegger, A. E., 421, 437, 464, 479
Rathjens, C., 470 Schlanger, S. 0., 485
Raup, H. M., 443 Schmidt, J., 219, 449,457
Raven, T., 43 Schmidt, V. A., 482
Reed, B.,419 Schmitthenner, H., 435
Reed,J. C., 445 Schoeller, H., 470
Reesmann, A. L., 44 Schofield,!. C., 455, 484
Reeves, C. C., 325,474,481 Scholl, D. W., 462
Reiche, P., 45 Schreiber, K. F., 283, 468
Reichelt, G., 349 Schultze, !. H., 444
Reid, C., 474 Schumm, S. A., 4, 14, 22, 26, 63, 75, 277,
Rennie, W. F., 47 278, 314, 317, 345, 354, 421, 423, 429,
Rennilson, T. T., 429 436, 449, 456, 466,470,480
Rhoads, D. C., 152, 443 Schunke, E., 45
Rhodehamel, E. C., 364, 481 Schwartz, M. L., 386, 449, 485
Rice, R. J., 481, 486 Schwarzbach, M., 474
Rice, R. M., 457 Seddon, B.,310,471
Rich, J. L., 51, 426 Seginer, 422
I.,
Richardson, L., 480

1 1 1, Segota, T., 474
Richter, C. F., 80, 432 Selatui, O., 478
Richter, K., 135,440 Selby, M.!.,456
Rinehart,!. S.,433 Sellmann, P. V., 437
Rioux, R. L., 97 Selwood, E. B., 435
Ritsema, A. R., 432 Sencu, V., 460
Rittman, A., 432 Sengupta, S., 205, 304
Ritter, D. F., 446 Seret, G., 487
1 1 19

Serko, A., 371,482 Speitzer, H., 471
Seuffert, O., 427 Spiridonov, A. L., 4 1
Shackleton, N., 477 Spreen, W. C., 470

Shaler, N. S., 425 Starkel, L., 474
Sharp, R. P., 18, 176, 187, 207, 245, 313, Stan, C., 454
320, 338, 416, 437, 446, 449, 450, 454, Stearns, H. T., 423
463, 466, 468, 477 Steers, !. A., 220, 313, 389, 416, 429, 462,
Sharpe, C. F. S., 13,457 485
Sheldon,!., 149 Stenborg, T., 437
Shepard, F. P., 440, 478 Stephens, C. G., 469
Sheppard, T., 449 Stephens, N., 457, 485
Sherlock, R. L., 85,431 Stewart, A. B., 422
Shier, D. E., 243, 485 Stewart, !. H., 233
Shirley, M. L., 235,263,461 Stoddart, D. R., 389, 462, 469, 485
Shoemaker, E. M., 429 Stokes, W.L., 471
Shotton, F. W., xiv, 474, 477 Stonely, R., 43
Shouldice, J. R., 434 Stork, A., 334,411,477
Shreve, R. L., 454 Straaten, L. M. !. V. van, 239, 461
Shteynberg, G. S., 430 Strahler, A. N., xiv, 6, 16, 17, 26, 54, 421,
Siever, R., 166,441,443 437,474
Sigafoos, R. S., 332, 374,467 Strakhov, N. M., 41, 223
Silvester, R., 462 Straw, A., 273, 274, 465, 482, 487
Simonett, D. S., 463 482
Stricklin, F. L., 363, 365,
Simonson, R. W., 477 Stride, A. H., 454, 487
Simpson, S., 451 Stuiver, M., 300, 467, 482
Sissons,!. B., 108, 183, 230,382,416,449, Suess, E., 42, 425
483,485 Sugden,D. E., 182,399,449
Sitter, U. de, 431 Sundborg, 204, 207, 446, 447
Skakalskiy, B. G., 188,319 Sutton, O. G., 436
Skempton, A. W., 440 Svensson, H., 42 1, 468
Skibitzke, H. E., 20, 420 Sweeting, M. M., 196, 197, 269, 367, 435,
Slaymaker, H. Q., 421 451,453,487
Small, R.!., 416, 438 Swindale, L. D., 440
Smalley, L. !., 428 Sylvester-Bradley, P. C., 428
Smart, !. S., 465 Synge, F. M., 483
Smith, C. T., 444 Szupryczynski , !., 483
Smith, D.D., 456
Smith, D. I., 45 Taber, B. A., 188,475
Smith, F. G., 421 Taber, S., 447, 468
Smith, !.F., 453 Tallis, !. H., 468
Smith, R. L., 432 Talwani, M., 431
Smith, William, 76, 106 Tanner, W. F., 280, 425, 462, 485
Smith, W. W., 146 Tarr, R. 50
S.,
Smithson, B. B., 3 1 7, 463 Taylor, G.,438

Sneed, E. D., 131, 132,465 Taylor, F. B., 84
Snell, !. B.,421 Taylor, I. F., 443
Soileau, !. M., 69 Taylor, !. C. M., 469
Soons,!. M., 471,478,483 Taylor, !. H., 434
Sorensen, F. H., 447 Tedrow, F., 45
!. C.
Souchez, R., 470 Temple,P.H., 310, 361,471
Sparks, B. W., 51, 356, 416, 426, 446 Ters, M., 420
Spate, O. H. K., xix, 21, 421 Terzaghi, K., 138, 440
Spencer, L. !., 433 Tessman, N., 425
Author Index 525
TePunga, M., 447 Vareschi, V., 482
Thackeray, A. D., 95, 434 Varjo, U., 485
Thames, 440
J. L., Verger, F., 468
ThebaulhJ-, 440 Verstappen, H. T., 463, 485
Thom, B. G., 462, 467 Verstappen, W. F., 425
Thomas, M. F., 198—9, 453 Viguier, G., 477
Thomas, W. L., 444 Virkkala, K., 251, 378-9, 483
Thompson, G. A., 43 Visher, G. 440 S.,
Thompson, H. R., 449 Visher, S. S., 427

Thorarlinsson, S., 432 Vistelius, A. B., 440
Thombury, W. D., 4 1 Vita-Finzi, C., 428, 482

Thorp, J., 471 Vogt, H., 447
Thorp, M.,430 Vollbrecht, K., 456
Thorpe, M. B., 453 Von der Borch, C. C., 465
Tibbitts, G. C., 246, 247, 463 Von Richthofen, F., 2
Tichy, F., 444
Tietze, W., 449 Waggoner, P. E., 465
Tilley, P., 426 Wahrhaftig, C., 3 12, 457, 469
Tinkler, K.J., 227,459 Walker, D., 415
Tinsley, 455J,, Walker, G. P. L., 394, 487
Tivy, J., 459 Walker, H. J., 472
Todd, 0. J., 456 Walker, P. H., 465, 480
Todd, T.W., 441, 487 Wall, J. R. D., 453
Toit,A.L.du, 84-5,431 Wallis, J. R., 6
Tomlinson, M. E., 487 Wallwork, K. L., 444
Topp, G. C., 71 Walsh, P.T.,478
Totten, S. M., 483 Walton, K., 34, 423
Tricart, J., xv, 2, 3, 16, 53, 55, 59, 63, 75, Wanibeke, A. R. van, 339, 477
131, 134, 157, 185, 210, 257, 263, 315, Wangersky, P. J., 477
417, 418, 419, 421, 422, 427, 438, 447, Ward, R.C.,436
467,480 Ward, W. C., 129,438
Tricker,R. A. R.,437 Ward, W. H., 220
Troeh, F. R., 28, 421 Ward, W. T., 484
Troll, C., 55, 184, 427, 447, 482 Warner, M. M.,478
Trombe,F., 453 Warren, A., 473
Trumpy, R., 431 Warwick, G. T., 451, 453
Tschang, T. L.,449 Washburn, A. L., 468
Tuan, Yi-Fu, 428, 482 Waters, A. C., 415
Tuckfield, C. G., 319-20, 475 Waters, R. S., 200,475,486
Turkevich, A., 429 Watson, E., 459, 475
Tuttle, S. D., 475 Watson, J. P., 465
Twidale, C. R., 345, 447, 453, 480 Watson, R. A., 265, 348, 457, 480
Watt, A. S., 475
Udden, J. A., 127 Waugh, B., 134
Uehara, G., 443 Weaver, C. E., 134, 3 14, 438, 475
UNESCO,475 Webb, S. D., 425
Urbanek, J., 457 Webber,?. J., 331
J., 440
Ursic, S.
Weber, H.,417
U.S.Dept of Agriculture, 457 Webley, D. M., 443
Uvarov, L. A., 320
Webster, R., 440
Weertman, J., 123, 437
Van Baren, F. A., 446 Wegener, A., 84, 43
Vann, J, H., 443 Wegmann, E., 122
1 1 9
1

Weisse, R., 464 Wolfe, P. E., 487
Weir, A. H., 425 Wolff, R. G., 441
Welder, F. A., 275 Wolman, M. G., 1, 54, 75, 206, 263, 272,
Wellman, H. W., 425 278, 359, 406, 418, 423, 466, 475, 480
Wentworth, C. K., 87, 433, 440 Wood, J.,455
Wertz, J. B., 466 Wood, W.F.,421
West, R.G., 10, 134,417,475 Wooding, R. A., 471
Whetten, J, T., 456, 465 Woodruff, 459
J. F.,
White, D. E., 433 Woods, R. J., 433

White, E. M., 472 Woodward, A. O., 4 15
White, L. 453
S., Wooldridge, S. W., 1, 11, 33, 37, 49, 417,
White, W. A., 472 418, 419, 429,487
White, W. B., 482 Wordsworth, William, 298
Whiteside, E. P., 185, 449 Wright, C. W., 428
Whittig, L D., 45 Wright, H. E., 348, 417, 457, 475, 480
Wiens, H. 462
J., Wright, M. D., 435
Wiggens, A. J., 485 Wunneche, G. W., 44
Wilford, G. E., 453 Wurm, A., 422
Wilhelmy, H., 59, 193, 361, 417, 447 Wyllie, P.,431
Wilkinson, H. R., 418, 444
Williams, G. E., 460 Yaalon, D. H., 287, 337, 469
Williams, G. 479
J., Yagi, K., 432
Williams, H., 88, 433 Yates, E. M., 483
Williams, M. A., 472 Yehle, L. A., 475
Williams, P. F., 364, 460 Young, A., 447, 456
Williams, P. J., 436, 457, 458 Young, R. G., 440, 469
Williams, P. W., 200, 453 Yoxall, W. H., 422
Williams, V. H., 45
Williams, W. W., 40, 462, 485 Zakharov, Y. F.,487
Wiman, S., 185,449 Zakrzewska, A. B., 4 1
Wimbush, D. J., 467 Zaruba, Q., 226, 458
Wilson, G., 435 Zeigler,J. M., 466, 475
Winslow,!. H., 181,479 Zeller, E. J., 330, 477

Winterer, E. L., 465 Zemtsov, A. A., 140
Wischmeier, W. H., 456 Zenkovitch, V. P., 242, 286, 462, 485
Wise, D. V., 109,435 Zeuner, F. E.,417,477
Woldenberg, M. J., 465 Zischinsky, U., 435
Woldstedt, P.,417 Zveder, L. W., 423
Wolf, P. O., 423 Zwittkovits, F., 453

2015.131996.introduction To Geomorphology Text

Uploaded by

Copyright:

Available Formats

2015.131996.introduction To Geomorphology Text

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2015.131996.introduction To Geomorphology Text

Uploaded by

Copyright:

Available Formats

Alistair F.

METHUEN & CO LTD

SBN 416 11730 9 hardback

This book is available in both hardbound and paperback

-ti similar condition being imposed on the subsequent purchaser.

Distributed in the USA by Barnes & Noble Inc

I. Definitions, nature and basic postulates i

B. The Nature of Geomorphology 1

1. Description and interpretation 1

2. Process and form 1

3. Artistic and scientific elements 16

D. Some Conclusions about Geomorphology 72

II. Landforms and Structure 79

'ONIVmSlTY ©F JOD-SPUR LIBRAS^

III. Physical, chemical and biological basis of 112

B. Amounts and Motions of Water and Ice 1 14

3. Waves and offshore currents 1 1

4. Ice and snow 1 2

C. Mechanical and Frictional Forces 125

D. Sediments and Mechanical Characteristics of Soils 127

IV. Inter-relationships between processes and

E. Inter-relationships between Form and Process 298

F. The Significance of Changes in Weather and in Climate 307

V. Landforms and time 327

Appendix; Some simple methods of field

H. Suspended Sediment 412

Selected Bibliography 414

A book is that of the

Mr J. B. Fisher and Mr S. Moran prepared photographic prints of the

assistance in compiling information; when demands on the Map Library were

the book include Mr I. Reid (practicals), Mr and Mr

landforms in particular; my parents for encouraging me to admire and

learning for oneself, and the Department of Geology in the University of

Archiv fiir Wissenschaftliche Geographie for figs. IV.45.E, V.6, V.12.D;

Olson), IV.l l.C (and W. O. Kupsch), IV.15.D (and J. C. Harms, et al.),

IV. 20 and IV.40 (and V. P. Zenkovitch), IV.45.C (and M. Demorest), IV.50

American Association of Petroleum Geologists for figs. II.8.B-E, IV.23.A,

Journal of Glaciology by permission of the Glaciological

logists Association for figs. I.IO, IV.6.D, IV.27.A, V.25.B; Houston

Geological Society for figs. IV. 17 (and E. H. Rainwater) IV.28.G (and A. S.

D. McKee, G. C. Tibbits), IV.23.C (and E. D. McKee and G. C. Tibbits),

Bonn University Institute of Geography for figs. IV.6.A, IV.3LD; Florida

IV.7.D; University of Strasbourg for figs. III.7.B, IV.44.A.

Figs. II.15.C. Based on Crown copyright Geological Survey diagram. Repro-

branches of science. On a modest scale the part-time science of geomorpho-

down with a thermodynamic engine. In 1959 C. A. M. King’s book on coasts

and beaches appeared, emphasizing the engineers’ use of mathematical theory,

erosional forms increasingly includes the study of mass movements and

cial support from a variety of sources, has benefited geomorphology. For

problems of water supply and water-quality management. In the last two

huge store of information on hydrology, soils, and geography compiled by

absorb the attention of Pleistocene, stratigraphical, or historical geomorpho-

differing field experiences and schools of thought, geomorphology may be

hoped that sufficient detail is become familiar with

actual, observed facts of landform study in preference to summarizing the

Definitions, nature and basic

Geomorphology is the study of landforms. This obvious statement is difficult

‘much of geomorphology is stratigraphic geology’, have made the study

B. The Nature of Geomorphology