Tnau 2528

Download as pdf or txt
Download as pdf or txt
You are on page 1of 325

CLIMATOLOGY

METHUEN'S ADVANCED GEOGRAPHIES

NORTH AMERICA rl
An Historical, Economic and Regional
Geography
By LL. RODWELL JONES, Ph~D., B.Sc., and P. W.
BRYAN, Ph.D. With 104 Maps and Diagrams.
Third Edition, Revised. Demy 8vo. 21S. net.
SOUTH AMERICA
An Economic and Regional Geography
By E. W. SHANAHAN, M.A., D.Sc. With 50 Maps
and Diagrams. Second Edition, Revised. Demy 8vo.
0145. net.

THE POLAR REGIONS


A Physical and Economic Geography of the
Arctic and Antarctic
By R. N. RUDMOSE BROWN, D.Se. With 23 Maps.
Demy 8vo. I2S. 6d. net.
ASIA
An Economic and Regional Geography
By L. DUDLEY STAMP, D.Sc., B.A., M.I.P.T. With
324 Maps and Diagrams. Second Edition, Revised.
Demy Bvo. 27s. 6d. net.
CLIMATOLOGY
By A. AUSTIN MILLER, M.Sc. With 72 Maps and
Diagr"!Fs, Demy Bvo. 125. 6d. net.
FRANCE
A'Regional and Economic Geography
- By HILDA ORMSBY, B.Sc. With Maps. Demy 8vo.
In the Press.

METHUEN & CO. LTD. LONDON


CLIMATOLOGY
BY

A. AUSTIN MILLER, M.Sc.


LECTURER IN GEOGRAPHY IN THE UNIVERSITY OF' READI~G

WITH 72 MAPS AND DIAGRAMS

METHUEN & CO. LTD.


36 ESSEX STREET w.e.
LONDON
First ,Published in I9JI

PRINTED IN GREAT BRITAIN


PREFACE

T
HE obje.ct of this book is to provide advanced students, '
and especially students of geography, with a.. re.asoned
account of the world's climatic types. It is not intended
to exempt the student from the reading of original works, but
to give a foundation on to which he or she may build. Without
entering into the physics of meteorological processes, with which
the student is expected to be already familiar, effect is as far
as possible related to cause, and, since the book ~s intended
primarily for geographers, prominence is given to the human
aspect and the practical application. In short, the book attempts
to be .reasoned and not merely descriptive, hence the arrange-
ment adopted is not regional but is based on climatic types,
with a view to emphasizing the essential similarity of environ-
°ment in regions similarly situated and climatically allied. Par-
ticular attention is paid to the normal type, the regional peculi-
arities of the more important areas being dealt with separately.
after the general description. The classification followed departs
only in matters of detail from those in general use, but the bound-
ary lines adopted are not always coincident with those genera!ly
recognized; the reasons for these departures are set out in some
detail in Chapter III.
Except in matters of treatment and presentation the book
makes no other claim to originality, and the author acknowledges
his indebtedness to a mass of climatological literature much too
large and varied to be listed in. detail. Generally this is
acknowledged in the text or in foot-notes, but since the book
is intended only as a text-book for students, chapter and verse
are not always given in the references. The object in view in
quoting references is rather to allow the student to follow up
a particular line 'of inquiry should he wish to do so, and this
purpose is effected by means of a. short guide to further reading
at the end of each chapter; but it sometimes happens that an
interesting side-track is exposed in the tex,t which it is impossible
to pursue further and in such ca~es the way is pointed out in
a foot-note. In general only those books and journals are
quoted which would be readily accessible to a student in the
library of the university or college.
vi CLIMATOLOGY
For the sake of uniformity th~ same 'general order has been
followed in the description of each climatic type, but different
types lend themselves to most satisfactory treatment in different
ways; thus the arrangement is sometimes by seasons and
sometimes by elements.
When controversial topics arise it is not always feasible to
give due consideration to all views held, and generally one ex-
planation only, the one that appears most satisfactory to the
author, is given. .:The last chapter, in particular, dealing with
the evolution of climate, covers such an enormous subject in
such a short space that it is impossible to do more than provide
a bald outline of the sequence of events. In the present in-
adequate state of our knowledge there are inevitably numerous
apparent contradictions of fact and divergences of opinion among
authors as to the explanations. Limitation of space has forbidden
full treatment of these and it has been necessary to ignore the
difficulties in attempting to present a consecutive account.
The Fahrenheit scale of temperature is used throughout and
rainfall is expressed in inches, sihce students are generally more
familiar with these units. Often the decimal points are ignored,
temperature and rainfall figures being rounded off to the nearest
degree or inch, for except when dealing with very small rainfall
amounts or very small temperature variations these are of little
'significance; moreover, they give a misleading impression of
exactitude which, in point of fact, does not exist, especially
w}.len dealing with mean values.
It is assumed that the student is equipped with a good modern
atlas containing the usual maps of annual and seasonal tempera-
ture, pressure and rainfall, vegetation maps, etc., which are not,
therefore, reproduced in the book.
The author wishes to express his gratitude to Mr. L. C. W.
Bonacina for kindly reading Chapter IV and for offering valuable
suggestions and criticisms, and to Dr. H. A. Matthews for gener-
ous assistance and for undertaking the reading of the proofs.
A. A. M.
READING, 1931
CONTENTS
CHAP • . PAGE

I THE MEANING AND SCOPE OF CLIMATOLOGY I

II THE ELEMENTS OF CLIMATE 8


III THE FACTORS OF CLIMATE 33
IV THE CLASSIFICATION OF CLIMATES 54
V EQUATORIAL CLIMATES 71
VI TROPICAL CLIMATES. 90
VII TROPICAL MONSOON CLIMATES. 108
VIII WESTERN MARGIN WARM TEMPERATE CLIMATES 13 8
IX EASTERN MARGIN WARM TEMPERATE CLIMATES 15 8
X COOL TEMPERATE CLIMATES 17 6
XI COLD CLIMATES 206
XII ARCTIC CLIMATES

XIII DESERT CLIMATES 23 2

XIV MOUNTAIN CLIMATES 250


XV CHANGES OF CLIMATE 266
INDEX 29 1
LIST OF MAPS AND DIAGRAMS
FIG. PAGE
I JANUARY ISANOMALOUS LINES 12
2 JULY ISANOMALOUS LINES • 12
3 CLIMATIC CONTROL OF SETTLEMENT 14
4 CLIMOGRAPHS 15
5 EQUIPLUVES AND ISOHYETS. 20
6 MEAN ANNUAL CLOUDINESS 22
7 WIND ROSE FOR MENDOZA (ARGENTINE) 24
8 GENERALIZED CIRCULATION OF THE ATMOSPHERE 25
9 STRUCTURE OF A CYCLONE 29
10 THE POLAR FRONT 30
II ANNUAL VARIATION OF INSOLATION 34
12 INSOLATION AT DIFFERENT LATITUDES ON 21ST JUNE 34
13 ALTITUDE OF THE SNOW LINE 38
14 YEARLY MARCH OF TEMPERATURE IN CONTINENTAL AND MARINE
CLIMATES 43
15 MEAN ANNUAL RANGE OF TEMPERATURE 44
16 JANUARY ISOTHERMS AND MEAN ANNUAL MINIMA OF TEMPERA-
TURE OVER THE GREAT LAKES 50
17 SUPAN'S TEMPERATURE ZONES 55
18 KOPPEN'S TEMPERATURE ZONES (SIMPLIFIED) 56
19 SOUTHERN LIMIT OF THE TAIGA • 60
20 FOREST TYPE AND DURATION OF GROWING SEASON 62
21 SEASONAL DISTRIBUTION OF RAIN ·64
22 TEMPERATURE ZONES • 65
23 DISTRIBUTION OF CLIMATIC TYPES 69
24 SCHEMATIC DISTRIBUTION OF CLIMATIC TYPES 69
25 ANNUAL MARCH OF TEMPERATURE AND RAINFALL IN EQUA-
TORIAL CLIMATES. 73
26 MARCH OF RAINFALL IN EQUATORIAL AND TROPICAL CLIMATES 76
27 PRESSURE, WINDS AND RAINFALL IN THE MALAY ARCHIPELAGO 83
28 TEMPERATURE AND RAINFALL REGIMES IN THE EAST INDIES 86
29 RAINFALL REGIMES ON TRADE WIND COASTS 91
30 DISTRIBUTION OF TROPICAL CYCLONES 93
31 ANNUAL MARCH OF TEMPERATURE AND RAINFALL IN TROPICAL
CONTINENTAL CLIMATES 95
32 DVRATION OF RAINY SEASON IN AFRICA 99
33 VEGETATION OF AFRICA 99
34 MEAN PRESSURE AND WINDS, JANUARY 113

ix
x CLIMATOLOGY
FIG. PAGE
35 COLD WEATHER RAINFALL AND STORM TRACKS II3
36 DIURNAL RANGE OF TEMPERATURE, JANUARY II4
37 MEAN DAILY RELATIVE HUMIDITY, JANUARY II4
38 MEAN TEMPERATURE, MAy • II5
39 MEAN MAXIMUM TEMPERATURE, MAy • II5
40 HOT WEATHER RAINFALL (MARCH-MAy) IIi
41 WEATHER ON 14TH MAy, 1897 II8
42 WEATHER ON 3RD JULY, 1889 II8
43 ANNUAL MARCH OF TEMPERATURE AND RAINFALL IN INDIA 120
44 ST~AMLlNES DURING MONSOON • 125
45 MONSOON RAINFALL 125
46 RAINFALL AT HUE AND MADRAS. 130
47 ADVANCE OF THE MONSOON RAINS INTO AUSTRALIA 133
48 TRANSITION OF CLIMATES IN SOUTH AUSTRALIA AND VICTORIA 140
49 YEARLY MARCH OF TEMPERATURE IN MEDITERRANEAN CLIMATES 141
50 RAINFALL CONTROL OF VEGETATION AND CROPS IN SWANLAND 154
51 RAINFALL REGIMES IN EASTERN MARGIN WARM TEMPERATE
CLIMATES 160
52 ANTARCTIC DEPRESSION UNFAVOURABLE TO INLAND RAINS. 162
53 ANTARCTIC DEPRESSION FAVOURABLE TO GOOD INLAND RAINS 162
54 RAINFALL REGIMES IN THE GULF ATLANTIC STATES 167
55 RAINFALL REGIMES IN COOL TEMPERATE CLIMATES 182
56 MAXIMUM, MINIMUM AND MEAN TEMPERATURES AT CAMBRIDGE 184
57 PRINCIPAL TRACKS OF DEPRESSIONS IN WESTERN EUROPE 188
58 CYCLONE TRACKS, NORTH AMERICA 191
59 TRACKS OF CONTINENTAL DEPRESSIONS AND OF TYPHOONS 198
60 CHIEF STORM TRACKS, JAPAN * 201
61 ApPROXIMATE POSITIONS OF SUMMER ISOTHERMS IN POLAR
REGIONS 225
62 NORTHERN LIMIT OF PRODUCTIVE DATE-PALM 246
63. MECHANISM OF UP-VALLEY WINDS IN MOUNTAINS 251
64 MEAN HEIGHT OF SNOW LEVEL IN THE INN VALLEY IN N ORTRERN
TYROL 257
65 ALTITUDE AND GROWING SEASON IN THE EASTERN ALPS 259
66 EVIDENCE FOR CLIMATIC ZONES IN PERMO-CARBONIFEROUS
TIMES. 272
67 LAND AND SEA DURING THE EOCENE 278
68 LAND AND SEA AT MAXIMUM GLACIATION 278
69 CLIMATIC CONDITIONS DURING THE CONTINENTAL PHASE 284
70 'CLIMATIC CONDITIONS DURING THE MARITIME PHASE 284
71 CLIMATIC CONDITIONS DURING THE FOREST PERIOD 285
72 CLIMATIC CONDITIONS DURING THE PEAT-BOG PHASE 285
CLIMATOLOGY
CLIMATOLOGY
CHAPTER I
THE MEANING AND SCOPE OF CLIMATOLOGY

T
HE subject of climatology is intimately interwoven with
the affairs of everyday life. This industrial era, in
which a large percentage of the population does its
daily work under cover from the elements, certainly feels the
controlling hand of weather' less than an earlier agricultural
age, but the climatic control of its daily life and habits is prob-
ably as great to-day as it ever was. The agriculturist is still
almost entirely at the mercy of the weather and climate, but
the industrialist's dependence, though less direct, is not less real,
and often the site-area of an industrial concentration itself de-
pends on local climatic advantages, as Lancashire well knows.
CLIMATE AND TRADE. Climate limits the choice of crops and
therefore the local production of food; and climate determines
the site for the cultivation of those other foodstuffs and raw
materials of industry which modem life demands; this climatic
control of production and requirements is one of the bases of
the world's trade. Furthermore climates, in general, change
from north to south, and the products of different climates move,
therefore, chiefly along meridional routes, although much deflected
by economic factors; climate therefore controls the direction as
well as the existence of trade routes. The great wind belts, too,
are determinants of ocean routes, for steam power has diminished
but not removed the obstacle of a head wind and current, and
the same clockwise swirl of air and water in the North Atlantic
which led Columbus to America on the trades and back on the
westerlies still aids much of the steam shipping in these waters;
round the world with the' Roaring Forties' is almost as uni-
versal for the steamship to-day as it was for the' windjammers'
of fifty and a hundred years ago. The development of air lines
is insisting again on the importance of air currents, and a depend-
ence even greater than that of the sailing ship is forced on the
aeroplane or airship.
CLIMATE AND DAILY LIFE. In every climate the life habits
1
2 CLIMATOLOGY
of the natives have been regulated, often after many bitter
experiences, in accordance with prevailing conditions. Nature
enforces obedience to certain rules of diet, clothing and behaviour
by native and visitor alike, the neglect of which is fraught with
serious consequences. The adoption of many native methods
and habits has always been a wise procedure for immigrant or
conqueror, and this has ultimately led to his absorption into the
life of the zone. Thus, in the long run, climate triumphs over
) extraneous introductions and a certain continuity of character
is assurtd ,which reflects the individuality of the zone.
CLIMATE AND RACE. Controversy centres round the discus-
sion of the extent to which the physical and psychological char-
acters of the inhabitants of different climates can be attributed
directly to the climate. Is the dark pigment of the negro's
skin the result of centuries spent in an environment of torrid
heat? Can the dim light of the equatorial forests be held re-
sponsible for the dwarfing of the pygmies? There is no doubt
that many of these characters are racial and hereditary, for the
'black man ' does not become noticeably less black in a tem-
perate environment, and the pygmy does not grow appreciably
taller when transplanted from his native forest. The' yellow'
race shows little or no climatic control and is found from the
Arctic (Esquimaux) to the Equator (Malays) and from arid climates
(Gobi) to humid (Java).
The relative importance of heredity and environment provide
a fruitful subject for inconclusive debate, but it must at least
be admitted that there is a certain selective influence at work.
For the' black' races are limited in the natural state to tropical
climates, and in general they do not thrive- beyond their limits,
whije the gradual paling of complexion and hair polewards in
Europe is an incontrovertible fact which cannot be devoid of
significance.
CLIMATE AND CHARACTER. Psychologically each climate tends
to have its own mentality, innate in its inhabitants and grafted
on its immigrants. There appears to be a direct relationship
between mental vigour and changeability of climates; all the
world's great civilizations are now in regions experiencing abrupt
and often unexpected changes of weather; the temperate zone
governs the tropical zone by virtue of its infinitely greater energy
and initiative. In U.S.A., where other factors are presumably
uniform, there is a remarka:ble difference in the output of table
and prominent men between the northern (temperate) and
southern (sub-tropical) States .. The enervating monotonous
climates of much of the tropical zone, together with the abundant
and easily obtained food-supply, produce a lazy and indolent
MEANING AND SCOPE 3
people, indisposed to labour for hire and therefore in the past
subjected to coercion culminating in slavery.
CLIMATE AND COLONIZATION. It is increasingly clear that
colonization, as distinct from occupation and exploitation, can
only succeed when the climate of the colony is somewhat similar
to that of the. parent country. North Europeans succeed best
in Canada, South Europeans in Brazil; Spaniards colonized the
Argentine but Scots and Welshmen are needed in Patagonia and
the bleak Falkland Islands. This is a general principle which
has held true throughout history; the 'Barbarian~' of the
Steppes, accustomed to a continental climate, were rapidly
converted to the ways of India on reaching the plains through
the North-Western gates and lost their individuality, just as the
, Shepherd Kings' soon became Egyptian in an Egyptian climate..
The Roman Empire exceeded its climatic limits in the coloniza-
tion of Britain and Germany, and was glad to retire, leaving
little in language or literatur~ to bear witness of its occupation.
France, especially southern France, Italy.and Spain, on the other
hand, were climatically somewhat similar to Rome, and it was
here that Roman culture left a lasting mark. Conversely the
Teutonic peoples found a climatically suitable outlet for their
colonizing activity in Britain, but were beyond their climatic
range along the Mediterranean in Italy and Spain.
This aspect of colonization is to-day one of the most interesting
and important applications of climatic study. Three well-known
examples, briefly enumerated, will serve to illustrate this : -
I. The climatic difficulties, particularly the impossibility of
sustained manual labour in the moist heat of Queensland, which
provide the most formidable obstacle to the 'White Australia'
policy (see p. 14). .
2. The moderation of an equatorial climate by altitude on
the East African plateau, which has made possible the establish-
ment of real colonies here (see p. 37).
3. India, with its excessive heat and moisture from the
tropical monsoon, which is one of the danger-points in the British
Empire, because it cannot be settled by Europeans, but must
be ruled by a transitory autocracy of British officials whose
real home is elsewhere.
CLIMATE AND HEALTH. The governing and policing of depend-
encies and the prosecution of trade and industry call for the
presence of white overseers in climates which are unsuitable for
white occupation. Of such pioneers climate has taken a heavy
toll of life and health, while even among the native population
the death-rate in certain climates is extremely high. Climate
was at one time accorded full blame for such appalling death-
4 CLIMATOLOGY
rates, but it is now clear that germ-borne and insect-borne
diseases are potent influences, and that insanitary conditions
and overcrowding of a native population, together with an ill-
balanced dietary, are factors predisposing to ill-health. Con-
sequently the tendency has been to rebound to the other extreme
and to exonerate climate from all blame. But it is clear that
by encouraging decay, by nourishing swarms of insect life, by
lowering man's resistance to disease, and in a host of other ways
climate is at the bottom of this ill-health; and it is clear that
the maint~nance of health can only be accomplished by an enor-
mous expenditure of labour and money on hygienic precautions
and the treatment of disease. There is, of course, another side
to the picture, and certain climates possess undoubted pathological
virtues. Mountain air, because of its clear, dry, rarefied nature,
is frequently recommended for diseases of the lungs; desert air,
equally clear and dry, but less rarefied and therefore not so
trying to the heart, is beneficial for heart complaints. The
tonic effect of a sea-voyage is nqt entirely imaginary, and sun-
light, which may be sought with more chance of success in some
climates than others, is credited with considerable medicinal
virtues.
THE DATA OF CLIMATOLOGY. It is a common definition of
climatology that it deals with average weather conditions; and
it is clear that averages, to be of value, must be based on careful
observations over a considerable period of time. These are the
raw materials of climatology, these observations of temperature,
rainfall and other climatic elements ",bich ha'Te been collected,
mostly during the last fifty years, in nearly all the civilized
and many other parts of the world. This miscellaneous collection
of data is of very unequal value, especially +0 the geographer;
many of them have little or no biological significance, while others,
such as rainfall, temperature, humidity and sunshine, are geo-
graphical influences of the first magnitude.
The climatologist demands accuracy and reliability in climatic
figures, a reliability which is only achieved by records over a
long period of time; for this purpose thirty-five years is con-
sidered the minimum, this figure being based on a fairly well-
established cycle of climatic variation which runs its course in
about this period. In thirty-five years, therefore, a station may
be considered to have experienced all types of weather which
are likely to occur there, and the mean of the observations may
be taken to reflect average conditions. But owing to the in-
completeness, as yet, of meteorological organization, there are
huge areas on the face of the earth where records are not avail-
able for anything approaching this length of time, and in fact
MEANING AND SCOPE 5
there remain wide expanses for which no data exist at all. In-
complete data, extending over "a few years only, may serve a
useful purpose as affording samples of the climate, though they
should always be treated with caution, but where no data exist
at all the geographer must be able, by applying, the general
principles of the science, to deduce the climate of the place from
a variety of observed responses. The nature of the relief, the
vegetation, the habits of the people, their architecture and their
occupations will be attuned to the prevailing rainfall and tem-
perature conditions, and it is the response rather than.,th~ cause
which the geographer wishes to know. .
CLIMATOLOGY AND GEOGRAPHY. The attitude of the geo-
grapher herein differs somewhat from that of the meteorologist,
for the latter will try to convert these observations into average
figures for temperature, precipitation, etc., while the geographer
is more usually concerned with the reverse process, the trans-
lation of average figures into certain biological responses.
He will develop the habit of thinking in such terms as these,
and a wet-bulb temperature of 80°, for example, will not be a
figure merely, but will convey a picture of stifling heat with all
work in the open air at a standstill.
Figures may be looked upon as a convenient shorthand
method of describing climatic environment, and are, therefore,
of incalculable value to meteorologist and geographer alike; but
valuable as such average figures are, to r<?ly too fully on them
is to miss much that is important in climatic study, especially
when climate is being considered as a geographical influence.
Accurate and detailed figures frequently have a real value,
because they represent conditions which are critical, as, for
example, the Is-inch isohyet in U.S.A. which in general defin.es
the limit of cereal cultivation. But such values are not always
of universal application; to attempt cereal cultivation with
IS inches of rain on the South African plateau would be to court
disaster.
CLIMATOLOGY AND METEOROLOGY. Mean values, such as
mean monthly temperatures and even mean daily maxima and
minima, tend to obscure those extremes of heat and cold which
may have such disastrous sequels for plants, animals and even
man himself, especially when these are existing near the limits
of their natural range. Hot and cold waves, thunderstorms
and tornadoes, droughts and floods are events which may occur
rarely in the annals of any area, but their effects are so far-reaching
that they deserve careful mention among the more prosaic facts
of mean temperatures, rainfalls and humidities.
A very significant difference between the climatic belts which
1
6 CLIMATOLOGY
lie equatorwards and polewards of the mid-latitude high-pressure
belt is to be sought in the striking differences in the nature of
those two great air currents; the steady uneventful progression
of the trades provides a striking contrast with the swirling,
turbulent, eddying rush of the westerlies; climate is the control
in the former, weather in the latter.
There are, then, certain climates in which a proper apprecia-
tion of life conditions can only be attained by a study of weather
types; not only the normal succession of everyday weather,
but aJ.so .those rarer, but mo,re moving types which may be
infrequent, but which may have consequences of vital import-
ance. Moreover, in some climates certain weather types occur
with such regularity as to be valuable elements of climate, whose
frequency and distribution are worthy to be recorded in the
same way as rainfall and sunshine amount. Yet in the de-
scription of the climate of a country, weather types are usually
split up into· their component parts and recorded as isolated data
of each element. Weather types are the integrals which go to
make up the climatic whole and there is a danger of their losing
their individuality unless climate is carefully examined, as it
were microscopically, to appreciate its texture. Average figures
create an illusion of steadiness and uniformity which is seldom
justified by the facts; the study ·of. weather types provides the
corrective.
NEED FOR SCIENTIFIC TREATMENT. While it is true that
climatology is primarily descriptive and lacks, or rather dispenses
with, the careful analysis of causes which is essential in meteor-
ology, there is much to be said for an analytical method of
approaching the subject. Such a method has two practical
r~commendations: in the first place the understanding of a
phenomenon is a valuable aid to memory, and in the second
it enables intelligent anticipation of similar results where similar
causes are at work. The structure of the science being revealed,
its outward form is the easier to remember, and the parts unseen
may the more accurately be supplied from one's knowledge of
the general plan.
FACTORS AND ELEMENTS OF CLIMATE. The climate of a place
is defined by a number of elements, or,component parts, such as
the temperature and humidity of the air, the rainfall, the wind
velocity, the duration of sunshine and a host of others of less
importance and less significance to man. These elements are
the results of the interaction of a number of factors, or deter-
mining causes, such as latitude, altitude, wind direction, distance
from the sea, relief, soil type, vegetation, etc. A clear distinction
sh'6uld be made between elements and factors, as defined above,
MEANING AND SCOPE 7
and the use of the tetIns sholild be carefully restricted. The
distinction is not always easy; for example, wind direction is a
factor of great importanr:e in determining climate, yet in some

respects it is an element climate; wind velocity is undoubtedly
an element, y~t maya "'V act as a factor, for it may control
precipitation by the rate at which it brings up supplies of moisture
from the sea (see p. II9)' The !ength of day, i.e. the duration
for which the sun is above the horizon, is a factor since it helps
to determine temperature; but the d~ration of actual sunshine
is an element with far-reaching effects on plant and aflimal life.
Factors and elements of climate. are clearly of two kinds,
the first mathematically determined and therefore cqpstant, the
second variable and unreliable. Among the former is latitude,
a. factor which determines the length of day throughout the
year, a matliematically controlled and ther:efore reliable element.
Latitude also determines the intens~ty of insolation, and this,
together with the length of day, influences .the duration of sun-
shine and the temperature, elements which, however, being also
influenced by other and variable factors, such·as.prevailing wind,
marine influence, etc., are not constant. Ocean curcents provide
an example of the other group of variable factors which exert
fluctuating influences on the temperature, rainfall and .sunshine
of adjacent coasts.
SUGGESTIONS FOR FURTHER .READING
For the relation of climate to health and human activities reference
should be made to R. de C. Ward's Climate, considered especially in Rela-
tion to Man, 2nd ed., 1917; E. Huntington's Civilisation and Climate,
3rd ed., 1924; and The Human Habitat, 1928, by the same author.. .see
also G. Taylor, • The Frontiers of Settlement in Australia', Geog ..Rev.,
1926; G. T. Trewartha, • Recent Thought on White Acclimatisation in
the Tropics " Geog. Rev., 1926. .
CHAPTER II
THE ELEMENTS OF CLIMATE
OLAR RADIATION. Radiation from the sun consists of rays

S of three different natures according to their wavelength,


heat rays, light rays and actinic rays, each of which, inter-
cepted by solid bodies, produces its peculiar effects in varying
degrees according to the nature of the surface on which it falls.
The light rays are, of course, responsible for the phenomenon of
daylight, light rays and actinic rays are necessary for the life
processes of plants, but from a climatological point of view the
heat rays are the most important of the three, and temperature
is the most important manifestation of solar energy.
TEMPERATURES IN THE SUN. T.emperature figures quoted in
climatology are generally' shade' temperatures, i.e. the temper-
ature of the air measured with due precautions taken to exclude
the influence of the direct rays of the sun, but it is a common
experience that it is much hotter in, the sunshine, wherefore it
is frequently of interest and importance to know the value of
, temperatures in the sun'. These are generally measured by
means of a thermometer with the bulb coated with lamp-black
mounted iir a glass tube exhausted of air (the black bulb in
vacuo), and the element which they measure is the intensity of
the sun's radiant heat.
Many mountain resorts frequented by invalids have air
temperatures often in the neighbourhood of zero in winter, yet
the bright sunshine produces a ,feeling of warmth and comfort
and allows light clothing to be worn without any sensation of
cold. On days such as this the black bulb in vacuo may be
recording over 100°, which indicates the power of the sun's rays.
Apart from their warming capacity these rays have beneficial
tonic effects, stimulating the system and imparting a general
feeling of well-being.
Radiant energy, like all forms of vibration, can be reflected
from solid surfaces and intensified, or rather waste can be dimin-
ished, by the reflection. Reflection from walls is a frequent
device for the ripening of peaches and pears, reflection from
bare ground also assists in the ripening of melons and other
creeping plants, while reflection from water surfaces enhances
8
ELEMENTS 9
the climatic reputation of water-side resorts and is a powerful
agent in producing sunburn. . .
SHADE TEMPERATURES. Of all the meteorological elements
there is none of such vital importance to living things as tem-
perature, and there is none which exercises so profound a 'control
over human distributions; Waterless deserts may be made
habitable and productive by irrigation, but clothing and artificial
heat are inadequate to overcome the handicap of the low tempera-
tures of the polar caps and high altitude, since these remain
unproductive of vegetation. The temperature requit.ements of
plants cannot be artificially supplied on a large scale as can their
moisture requirements, man can convert tropical desert into
tropical cultivation but he cannot change tropical cultivation
to temperate, or vice versa. Temperature is therefore recognized'
as the element of chief classificatory value and its accurate
measurement and statement is a matter requiring careful atten-
tion. By temperature at any given time is meant the tempera-
ture of the air measured under standardized conditions and
with certain recognized precautions against errors introduced by
radiation from the sun or other heated body.1
The mean daily temperature is, strictly speaking, the mean of
twenty-four readings taken at hourly interva~s throughout the
day and night; 'but except where self-recording instruments are
used twenty-four hourly observations are not usually available
and the mean daily temperature is calculated from readings
taken at certain times of the day, usually morning, afternoon
and evening. For this purpose various combinati2ns are recom- .
mended,2 such as 7 a.m. + 2 p.m. + 9 p.m. -;- 3,' 7 a.m. + 2
p.m. + 9 p.m. + 9 p.m. -;- 4'i 6 a.m. + 2 p.m. + 10 p;m. -;- 3,
etc., all of which give a fairly satisfactory mean differing little
from the mean of twenty-four 'hourly observations. Often the
mean is calculated by dividing the sum of the maximum and
minimum temperatures by 2, a 'Pr:actice which saves trouble as
these thermometers need only be inspected once a day; This,
however, gives a result which is'usually rather high, but is·goQd
enough for generalizations. , . (
The mean monthly temperature is the total of the daily. means
for the month. divided by the number of' days in the month::
The mean annual temperature should strictly be calculated
from the daily means added together and divided by 365, but
the usual practice is to divide the total of the monthly means
1 See Observer's Handbook, published by the Meteorological Office.
2See A. McAdie, Mean Temperatures and their Corrections in the United
States, Washington, 1891; and W. Ellis, ip' Quart. Journ. Roy. Met.
Soc., 18go.
10 CLIMATOLOGY
by IZ, which gives practically the same result. The mean
annual can be considered reliable if based on thirty-five years'
observations, but this is not always necessary; in fact. it is
found that in equatorial climates the mean annual temperature
based on two or three years' observations is reliable within a
small margin of error. But in any case the figure is of little
value outside equatorial climates, since the monthly tempera-
tures all depart from it by considerable amounts. Extremes
mutually cancel out to give a mean which may be the same
for c1itnates of very different type, e.g. :-
Hottest Coldest Mean
Month. Month. Annual. Range.
Peking 78.8 23·5 53. 1 55·3
Scilly Isles 60·8 45·3 52 •2 15·5

The difference between the means of the warmest and coldest


months is the .mean annual range. 1
The mean 6f the maxima recorded in a given period is the
mean maximum 1 and the mean of the minima the mean minimum. l
The difference between the mean maximum and the mean
minimum for the month is the mean diurnal range for that month.
The difference between the maxima and minima recorded in
a particular, month throughout the whole period of observation
is the absolute monthly range.
The means of the highest and lowest temperatures which
occur during any month or the year are the mean monthly and
mean annual extremes.
The difference between the mean extremes of the hottest and
coldest months is the annual extreme range. 1
. Mean temperatures are the ones usually quoted in describing
climates, but important though they arp it is more usually the
extreme temperatures which control vegetation, especially the
extreme minima which may bring killing frosts or in other ways
exceed critical values.
ACCUMULATED TEMPERATURES. Another aspect of tempera-
ture with well-marked botanical controls is the duration of
temperatures above a certain minimum amount, e.g. the minimum
for growth; zoo frostless days are required for the successful
cultivation of cotton and 150 for maize. Experiments at Rotham-
sted 2;.howed that 4zo may be considered the basal temperature
for 'Wneat; each degree excess of the mean daily temperature
over this critical value may be called a 'day degree " and if they
1 See Maps in Bartholomew's Atlas of Meteorology.
2 Summarized by R. H. Curtis in Symon's Meteorological Magazine,
190 ,5,
ELEMENTS II

are added together for the period of growth they give a measure
of the accumulated temperature. The experiments showed that
wheat required 1,960 day degrees between germination and
ripening at Rothamsted, but in Canada the requirements are
considerably less on account of the longer duration of daylighU
ISOTHERMS. The distribution of temperatures can be ex-
pressed graphically by means of isotherms, lines joining places
with the same temperature, but in the great majority of isothermal
maps it must be borne in mimI that the effect of altitude has
been eliminated by reduction of all temperatures to cQrre.spond-
ing sea-level temperatures by means of a formula. These maps
therefore show the combined effects of the other factors, latitude,
continentality, etc., and are of great value in the study'of climate;
but to obtain the approximate temperature of any place from
the map an allowance must'be made for its altitude. For bio-
logical distributions the actual temperature is of greater signifi-
cance than the sea-level figure, which has only a theoretical
value; but maps constructed on this basis are little more than
relief maps and become extremely complicated and confused
unless drawn on a large scale.
A much clearer conception of isothermal lines is obtained if
they are considered. as the lines oj. intersection of isothermal
planes with the surface of the sphere. For example, the July
isotherm of 80° passes near Mombasa and is met again at Cairo,
while between the two is Port Sudan where occurs the isotherm
of 90°. Somewhere about 3,<roo feet above this latter station
there must exist a surface with mean July temperatures of 80°,
in fact the isothermal surface whose edges meet the ground at
Cairo and Mombasa. This isothermal surface is continuous and
could be flown along by an aeroplane taking off at Momba~a,
passing over Port Sudan at a height of about 3,000 feet and
landing at Cairo. About one-third of the way along this route
the Abyssinian highlands rise into layers of air with mean July
temperatures below 80° and thus intersect the isothermal surface,
but since isotherms are reduced to sea-level values the 80° isotherm
does not appear at this point on the map.
As will be shown later the decrease of temperature with
altitude, although variable, is approximately 1° F. for each
330 feet of ascent, while the decrease with latitude is still more
variable but, except in the vicinity of the Equator, averages
about 1° F. for each degree of latitude (approximately 328,000
feet). The ver~ical decrease of temperature in mid-latitudes is
therefore about 1,000 times that of the horizontal; that is to
1 J. F. Unstead •• CllIDatic Limits of Wheat Cultivation,' Geog. ]ourn.,
1912.
I2 CLIMATOLOGY
say, the inclination of the isothermal surfaces is, on an average,
-about I in I,OOO.
ISANOMALOUS LINES. Another interesting and informative
method of representing temperature conditions is by the method
of differences. The difference between the mean temperature

I.-January Isanomalous Lines (Batchelder), differences in degrees


Fahrenheit

FIG. 2 . -July Isanomalous Lines (Batchelder)

of a place (reduced to sea-level) and the mean temperature


along its latitude is its temperature anomaly, and lines joining
places with equal and similar anomalies are known as isanomalous
lines. They indicate with great clearness those places which
are too warm for their latitude (having a positive anomaly)
and· those which are too cold (having a. negative anomaly).
ELEMENTS I3
SENSIBLE TEMPERATURE. The temperature recorded by the
thermometer does not always agree'with the sensations of heat
recorded by the human body. The sensation of heat depends
on other air conditions besides temperature, the chief of these
being air movement and humidity. For the human body is
cooled by a double process of radiation and evaporation (from
the sweat glands), and any condition which produces activity of
either gives 'an impression of cold. Temperatures of 60° below
zero are easily endured in Siberia where, under the influence of
the winter anticyclone, the air is practically still, but blizzards
with temperatures of 60° higher than this are insufferable. On
the other hand, the comfort of a cooling breeze on a hot day is
universally recognized, it accelerates both radiation and evapora-
tion from the skin. Again, 80° in the equatorial zone is distinctly
more uncomfortable than IOOo in the desert because of the
humidity, while at the other extreme the dry cold of continental
interiors is' less t$ng than the raw cold of moister climates.
Both excessive heat and excessive cold are, in fact, more trying
in wet than in dry air, for the dry air mitigates heat by accelerat-
ing evaporation from the skin, while the damp air, being a better
conductor of heat than the dry, allows escape of heat from the
body in cold weather. These factors, it is true, work in opposite
directions, but evaporation is more important than conduction
at high temperatures, while the reverse obtains at low tempera-
tures and evaporation practically ceases.
th moist climates the blood becomes diluted because evapora-
tion is checked, and conversely in dry climates (Le. deserts and
mountains) the blood is more concentrated. These physiological
conditions have interesting psycb,ological sequels; the conse-
quences of wet climates are ner~ous depression and lethargy,
while dry climates breed nervous energy, excitability and sleep-
lessness; these characters are very pronounced in visitors before
acclimatization has been attained.
WET-BULB TEMPERATURES. The combination of heat an'd
high humidity is one which the human body is least able to
endure and is a very important control of human activity. A
measure of this combination is given by the wet-bulb ther-
mometer, which, like the human body, records the value of air
temperature, decreased by evaporation from a moist surface.
Wet-bulb temperatures of 75° or 80° conduce to heat .stroke,
although dry-bulb temperatures of goO or IOOo can be safely
endured. Dr: Griffith Taylor 1 has shown that continued wet-
bulb temperatures of 70° place a limit on white colonization by
Control of Settlement by Temperature, etc.: 1916, Weather Bullenn,
l'
;14: Melbourne.
CLIMATOLOGY
rendering impossible sustained manuallabow- and by prohibiting
even sedentary and domestic occupations from being followed
with health and comfort .
. In Fig. 3 isopleths are plotted to.show the number of months
with mean wet-bulb temperatures above this amount. The
coastal regions north of the tropic are clea?ly unfavourable for
white settlement.
CLIMOGRAPHS AND HOMOCLIMES. The same author has
des.cribed an i?teresting graphical method of depicting climates
which emphaSIzes these physiological reactions. With wet-bulb
temperatures as ordinates and relative humidity as abscissa!
the twelve monthly figures are plotted, giving a twelve-sided
polygon (the c!ilpograph). According to the position of the
polygon on the' graph the nature of the climate is seen at a

'6
4 6
2 4
o

FIG. 3.-Climatic Control of Settlement (Griffith Taylor)


,.
"glance. The N.E. corner with high wet-bulb temperatures and
high relative humidity is described as 'm.lggy'; the N.W.
corner with low relative humidity but high wet-Qulb temperatures
as 'scorching', the S.W. as 'keen', and the' S.E. as 'raw.'
Climographs for four contrasted stations are shown in Fig. 4.
Stations having similar climographs are described as ' homo-
climes '. Alice Springs is the homoclime of Biskra (Algeria),
Perth of Cape Town, Brisbane of Durban, etc.
HUMIDITY. The depression of the wet-bulb temperature
below that of the dry-bulb provides one of the means employed
for estimating the humidity of the air,! i.e. the water vapour
content, a most important element both because of its biological
effects, described above, and its influence on rainfall. But its
VoJue is, in many climates, subject to rapid fluctuations under
1 For the method, see Observer's Handbook.
ELEMENTS 15
the influence of wind direction, especially in relation to moisture
supply, and this makes average figures of little significance.
Maps of the distribution of relative humidity are therefore seldom
20io 30% 40/0 50% 60% 70/0 80/0 901-0 100%

~"J
v~ 2=-f
F
~Vf:\
80° _Q.. / ~A
o M~ J'J MAOIIlI.~

~r)- itU·~
Q)..
~ N

0 ~\ A

~ k~ ID~rnA
F

A -... t:.J N
c!'s
o~ ~%' ~S\
5
50°

Q ~
30" ~
.~

20°
.+
-~
(:'
t...
~
~M
0\\
.'

~
If,
0 10:;
10 ~w,

\F
l
-to"
Relative Humidity
FIG. 4.-Climographs
(A/It' Griffith Taylor)

used, but the resultants, namely,' cloudiness and rainfall, are


shown in all atlases of meteorology.
RAINFALL. After temperature, rainfall is the most important·
of the climatic elements. The agricultural or pastoral utiliza-
tion of the land is the only real and lasting source of wealth,
and both of these are in a large measure dependent on rainfall.
16 CLIMATOLOGY
The yield of wheat, sugar, maize and' numerous other crops
varies with the rainfall of the year in such a way as to leave
no doubt that the rainfall has been the real determinant, and the
carrying power of grazing land in head of stock pe;::. square mile
emphasizes the same truth. As early as 1874, Sir W. Rawson 1
showed that the yield of sugar in Barbados was intimately related
to the rainfall of the preceding year, and was even able to fore-
cast the total production of the island, whose sugar acreage had
been constant for many years, by means of a formula, viz. that
every.inoh of rain in the year would result in a yield of 800
hogsheads of sugar from the whole island in the following year.
Sir Napier Shaw 2 devised a formula by which the wheat yield
in England could be forecast from the rainfall of the preceding
autumn as follows :-yield = 39'5 bushels per acre - t (previous
autumn rainfall in inches). From which it appears that the
autumn rainfall is excessive in the British Isles, and a reduced
autumn rainfall augurs an increased yield. In India the opposite
obtains, rainfall during germination is deficient and an increftsed
fall results in an increased yield. Here, however, such fa.ctors
as the moisture storage capacity of the soil and the evaporation
must be allowed for. Wallen 3 has carefully worked out the
relationship between rainfall and the yield of cereals (wheat,
barley, oats and rye) in Sweden.
EFFECTIVE RAINFALL. RUN-OFF AND EVAPORATION. Rain-
fall amount is stated in inches or millimetres (I mm. = 2~ inch)
for each month of the year and includes the total depth of water
resulting from all forms of condensation, whether rain, dew, fog,
frost, hail or snow. It should be borne in mind that the months
are not of equal duration; thus January, with 31 days, may
be expected to have a rainfall some IO per cent. higher than
February with 28 days, other things being equal. But in addi-
tion to the monthly and annual totals some information is
desirable on the nature of the rainfall, its persistence and its
intensity. This is provided by data concerning:-
I., The number of rainy days (a rainy day is one on which
an appreciable amount of rain, say 'OI inch, falls). The average
rainfall per rainy day can, of course, be obtained by dividing
the mean annual rainfall by the number of rainy days; it gives
a valuable measure of the intensity of the rain.
2. Maximum precipitation per day, per hour, or for shorter
1 Reports upon Rainfall of Barbados and upon its Influence on the Sugar
Crops, 1847-1871, by Governor Rawson, C.B. Barbados, 1874.
2 Proc. Roy. Soc. (Ser. A), 1905.
8 ' Sur la Correlation entre les recoltes et les variations de la tempera-
ture et de l'eau tombee en Suede,' Stockholm, K. Svenska, Vet. Ak.
Handl., 57, No.8, 1917.
ELEMENTS 17
periods. In general the rainfall of low latitudes is much more
torrential than that of high, but over a short period of time the
temperate zones can show an intensity almost equal to that of
the tropics. It inches fell in five minutes at Preston on loth
August, 1893, and nearly 4 inches in an hour at Maidenhead
on 12th July, 1901; but such' cloudbursts' rapidly exhaust the
available' water vapour and are short-lived. In the tropics,
on the other hand, where the warmer air holds infinitely greater
stores of moisture, these intensities can be maintained for hours
on end; 30 inches in 24 hours is by no means uncommon from
typhoons, 40 inches has been recorded from many stations, and
1?aguio, in the Philippines, has recorded 45 inches.
The intensity affects run-off and evaporation, thus qualifying
the,effectiveness of the rain.) For example, 30 inches of rain at
Pretoria 1 is inadequate for agriculture and the country round is
chiefly engaged in pastoral pur~uits. The explanation is supplied
by the nature of the rain, which occurs generally in the form of
he..avy downpours compacting the surface soil into a relatively
impervious layer over which the rain runs without penetrating;
and under the cloudless blue skies, which are such a typical
feature of the climate, evaporation is extremely rapid.
The rate of evaporation depends primarily on the dryness
of the air, but it is also affected by numerous other influences
such as air movement and plant cover. Steady winds, by
importing fresh supplies of air before saturation is reached,
enormously accelerate the ,process, especially if the winds are
coming from an arid area, e.g. the sirocco coming off the Sahara,
or if it is descending and therefore being adiabatically warmed,
e.g. the Foehn and Chinook winds (see pp. 214 and 252).
The nature of the ground surface, too, has an effect on the
rate of evaporation. A wet soil surface in the Transvaal, because
of its irregular surface and consequent large area exposed,2 has
been shown to evaporate 4·75 inches a week compared with
only 1·88 inches from a free water surface. Plant cover also
checks the direct loss by evaporation, and a loss of 52 per cent.
was found 3 on a summer tilled plot as compared with only
14 per cent. on land under grass sod. On the other hand, the
plant cover causes a loss by transpiration from the leaf surface
which generally exceeds the loss by evaporation; in the case of
forests this loss by transpiration is enormous.
SEASONAL DISTRIBUTION OF RAIN. It has been shown that
effective rainfall differs materially from the recorded amount
1 Rainfall and Farming in Transvaal, F. E. Plummer and H. D. Leppan.
Pretoria, 1927.
2 ap. supra cit. ap. supra cit.
18 CLIMATOLOGY
because of evaporation and run-off. Seasonal regime is a further
qualification which profoundly effects its utility and which
therefore deserves ca.reful attention in climatic descriptions.
The seasonal incidence may render much of the rainfall useless
and wasteful as in parts of Bombay which receives the unneces-
sary amount of 75 inches in four months (June to September)
and has drought throughout the rest of the year; or scanty
rain may be concentrated into a short growing season so that
optimum utilization is obtained. For example, wheat is grown
in parts of Western Australia with less than 10 inches of rain,
but only because it comes at just the right time; elsewhere
30 inches may not be enough.
RAINFALL RELIABILITY. Another feature of great signifi-
cance, especially in areas where the yearly or seasonal mean is
barely adequate, is the reliability of the rain, generally expressed
as a mean or extreme percentage departure from the normal
(see pp. IIO and 143). Regions with high variability will be
victims of drought in bad years and their crop production will
fluctuate widely. In pastoral areas the results are still more
serious, for flocks and herds are sadly depleted by the failure of
the pastures and the process of recovery may take years.
A reliable dry period at harvest time is a valuable asset,
in which respect the Mediterranean climates are especially fortu-
nate, while the unreliability of such a period in Norway reduces
the harvest to a gamble. By restricting crop rotation and by
compelling seasonal fallow, the seasonal incidence of rain enforces
a seasonal regime of labour on the farm.
TYPES OF RAINFALL AND THEIR SEASONAL DISTRIBUTION. If
the different types of rainfall are examined it will be found that
each tends to have maxima and minima at certain fairly definite
seasons.
Relief rain will occur chiefly when the supply of moisture
is greatest. This may be :-(1) when the seas are warm and
the land is growing cooler, i.e. in autumn or early winter in
inter~ediate zones. Seathwaite receives 41 inches out of 130
inches (32 per cent.) in October, November and December,
Ben Nevis 52 inches out of 171 inches (34 per cent.) in the same
three months.
Or (2) on the arrival of a strong off-sea wind as in monsoon
climates. Calicut has go inches out of IIg inches (75 per cent.)
in June, July and August; Cape York (Queensland) has 58
inches out of 82 inches (71 per cent.) in January, February and
March.
Convectional rainfall will tend to occur:-
(I) When the annual swing of the climatic belts brings the
ELEMENTS
region pnder the influence of the doldrum zone of ascending
currents of air, i.e. a summer maximum of rain, as in tropical
climates, e.g. Timbuktu has 7 inches out of 9 inches (78 per cent.)
in June, July and August.
(2) When high temperatures set up convectional currents,
often initiating thunderstorms, i.e. again a summer maximum,
as in continental interiors, e.g. Moscow has 8 inches out of
2I inches (38 per cent.) in June, July and August.
Cyclonic rain will tend to occur when cyclonic activity is
greatest, i.e. :- .
(1) When the swing of the climatic belts brings the region
into the belt of the stormy westerlies, i.e. in winter on th~e
equatorial side of this belt; e.g. Algiers has 14 inches out oti{
30 inches (47 per cent.) in November, December and January.
(2) When local superheating during periods of calm sets up
strong ascensional air movements resulting in tropical cyclones.
Calms are necessary for the generation of such storms (see p. 93)
and they occur at the maximum poleward extension of the
doldrum belt at the solstice, i.e. a summer maximum. Many
stations in the West Indies, for example, have their late summer
rain considerably augmented from this source.
ISOHYETs. The cartographic representation of rainfall amount
is by means of lines drawn through points having equal rainfall
during any given period and known as isohyets. Unlike isother-
mal and isobaric maps these require no correction for altitude,
indeed such would be impossible as the relation between rain-
fall and altitude is subject to no fixed laws (see p. 39). Higher
land generally has a heavier rainfall than lower levels, especially
where the precipitation is in the form of relief rain, and to
some extent isohyetal maps function as relief maps, but the
influence is not sufficient to obscure the other controls of
rainfall.
EQUIPLUVES. 1 The method of differences (d. p. 12) may be
conveniently applied to the study of rainfall, giving an informa-
tive impression of the raininess of a particular month, or other
period. The amount of rainfall which would be precipitated at
a given place, assuming that such rain were perfectly evenly
distributed through the year, is taken as a ' norm', and actual
mean amount of rain is stated as a percentage of this (the
pluviometric coefficient). Lines joining places with equal pluvio-
metric coefficients, and therefore with equal and similar departures
from the normal, are known as equipluves.·
Rainfall amount and seasonal regime may be instructively
. 1 For a full description of this method, see B. C. Wallis, 'Geographic:il
Aspects of Climatological Investigation: Scot. Geog. Mag., 1914.
20 CLIMATOLOGY
presented together by a combination of yearly isohyets. with
graphs of the monthly values placed on the site of the station. l
SNOW. Snowfall is included in rainfall figures, but its
measurement presents considerable difficulties. 2 It is sometimes
stated that 10 inches of snow are equivalent to I inch of rain,
but the nature of the snow, the size of the flakes (largely de-
pendent on temperature), the degree of compactness, etc., intro-
duce considerable errors, and it is to be recommended that the '
snow be either weighed directly or melted and the resultant
"

'7
B

July Equipluves July Isohyets


FIG. 5.-Equipluves and Isohyets

water measured. But when these precautions are taken the


result is not necessarily accurate, as a still greater difficulty is
experienced in making a correct catch, especially if the snow
is drifting. '
~ • The ways in which snow affects everyday life are so numerous
mid. so far-reaching that data of snowfall are of the utmost
importance. It interferes with the working of railway and
road transport, involving enormous yearly exp'enditure in clear-
1A good example of this may be seen in the G.]. for Nov. 1929,-P. 476.
, 2For a discussion of these (with references), see R. de C. Ward's Climate
of the United States, pp. 243-6.
ELEMENTS
ing, and capital expeRse during construction for snow-sheds
and snow-ploughs. On the other hand, the snow provides a
natural highway for sledges, on which, for example, the success
of the lumbering industry in Eastern Canada and elsewhere
largely depends. It protects underlying vegetation. from the
ill-effects of severe frost [winter wheat can be grown in Ontario
because snow comes before frost (see p. 51)] ; its late melting
may delay sowing; it holds up the winter precipitation and
then discharges it suddenly in what may be disastrous floods
when the. thaw comes.
For reasons such as these it is of interest to knQw:-
1. The average number of days on which snow falls.
2. The average duration of snow cover.
3. The average depth of snow cover.
4. The average dates of the first and last snows.
DEW, MIST AND FOG. Condensation of water vapour may
occur without necessarily r,esulting in rain; fog, mist and
cloud are such forms of condensation whose frequency and
seasonal distribution deserve mention in the climatic data10f a
station. They make very little difference to the recorded
precipitation, since they are very small in amount and much
of this is lost by evaporation, but they have quite an appreciable
effect on vegetation in certain climates. The regular dense fogs
of the Kalahari coastal strip nourish a scanty vegetation, and the
dew and mist of the cold season in the Central Provinces of
India play an important role in nourishing the wheat crop of
this great producing area.
Mist or fog will occur wherever air in contact with the ground
is cooled below the dew-point. This may happen in one of
two ways:-
1. The warm moist air JIlay .be cooled in situ (radiation fog)
either from direct loss of heat by radiation, such as commonly
occurs at night under anticyclonic conditions, or from the
sinking of cold air into valley bottoms, chilling the moist air
which lies there, especially near the stream.
2. The moist air may be cooled by drifting into contact
with a chilling surfa~e (advection fog) as when a current of air
passes from a warm sea on to cold land, or from a warm current
to a cold (e.g. along the' Cold Wall' where the Gulf Stream
meets the Labrador current), or in front of a feeble cyclone
where warm air drifts over colder.
CLOUDINESS AND SUNSHINE. Above ground-level the cooling
of air below dew-point results in clouds, which to a greater or
less extent obscure the sun. Cloudiness is stated in percentage
(or more usually tenths) of the sky covered with clouds. Statld'ns
2
22 CLIMATOLOGY
with equal cloud amount may be plotted on a map and joined
by lines called Isonephs. The map of mean annual cloudiness
(Fig. 6) show two belts of high cloud amount correspond2ng to
the equatorial and circumpolar low-pressure belts. Within the
zone of the trade winds the areas of clear sky are strikingly ex-
tended westward of the deserts as the dry air is carried seawards.
Westwards from the desert shores of Australia and California
these are shown extending through some 80° of longitude. In
fact, the only places in this belt where high cloud amounts are
recorded is where these winds blow on-shore on the eastern
margins of continents. There is, however, an interruption of
this general cloudlessness in the China Seas, the result of the.
monsoonal interruption of the trades. In addition to the me.ari

FIG. 6.-Mean Annual'Cloudiness


Figures indicate tenths of the sky covered

annual cloudiness the monthIy and daily distribution of cioud


are often important. There may be high cloudiness in the
morning, disappearing as the day advances, or clear mornings
may turn cloudy later on the arrival, for example, of a sea-
breeze.
The duration of sunshine (shown on maps by lines of equal
duration known as isohels) is not the converse of cloudiness, for
the possible hours of sunshine in each month vary according to
latitude and, to a less extent, according to altitude. Thus
within the Arctic Circle on 2ISt June the sun is above the horizon
for twenty-four hours, twice as long as at the Equa~or; the
Pole could therefore record the same sunshine hours as the
maximum at the Equator though the sky was covered with
clc1ud for half the day. The nature of the cloud (whether
ELEMENTS
cirrus, cumulus or stratus) and its position m the sky also
affect the sunshine hours.
Cloudiness is one of the characteristics of marine climates
and it is one of the means by which, in these climates, tempera-
tures are kept un~form; for clouds check solar radiation by
day and terrestrial radiation by night. They afford protection
from the sun, which is sometimes an advantage but more often
the reverse.) Coffee and certain kinds of tobacco are crops
requiring shade, which often has to be artificially supplied at
considerable expense. On the other hand, most fruits require
direct sunlight for ripening; South Africa, California and the
Mediterranean lands, the chief fruit-growing areas in the world,
are som"e of the sunniest. In t.lle British Isles Kent, Norfolk
and Devon have the highest sunshine hours and produce the
most fruit, while the fruit-growing district of Evesham benefits
from the loop of the IAoo-hours isohel which spreads up the
Severn-Avon valley.
PRESSURE ,AND WINDS. Pressure only becomes an element
of climate at high altitudes when it is decreased sufficiently to
produce physiological effects (see p. 36), but as a factor it is
directly responsible for winds and storms which are elements
of great importance. To separate those circumstances in which
wind functions an an element from those in which it is more
properly conside&d as a factor would lead to much repetition
and inconvenience, and the whole subject of pressure, wind and
atmqspheric circulation will be discussed here. The effect of
winds on sensible temperature has already been discussed, they
cool the body both by conduction Jind evaporation; but while
windy climates are generally more stimulating to animals and
man than those with persistent calms, the winds are often
injurious to plants on account of the accelerated evaporation;
a windy day can be as desi~cating as a hot day. Winds are more
'constant and their velocity is greater over the sea than on land,
since free air movement is less interfered with; even the trade
winds are not entirely reliable on land, especially on land of
considerable relief. The nearest approach to the conditions at
sea is found on fiat, featureless plains such as the prairies and the
pampas where the strength and reliability of the winds have
been turned to account in the use of numerous windmills for
pumping water and for other purposes. Windmills are familiar
features of the landscape in the fiat lands of Holland and the
Norfolk Broads.
The measurement of wind velocity has not received the
same attention as that of other elements. The anemomet6f,
by means of which wind, velocity can be measured and expressed
24 CLIMATOLOGY
(usually in miles per hour or inetres per second), is not altogether
satisfactory, since its exposure, in spite of precautions, is nearly
always subjected to eddies and gusts which introduce errors.
But at many stations wind velocity is merely estimated by its
effect on smoke, trees, etc., and stated according to Beaufort's
Scale.1
Wind direction can be represented diagrammatically by
, wind roses', in which the percentage frequency of winds from
each point of the compass is indicated by the length of a line
drawn in that direction radially from a centre. They may be
instructively combined with the frequency of other elements,
and in this way the dependence of such elements as rainfall
on wind direction may be demonstrated (see Fig. 7).

FIG. 7.-Wind Rose for Mendoza (Argentine)


Shaded inner rose shows percentage of thunderstorms according to wind direction

CIRCULATION OF THE ATMOSPHERE. The broad essentials of


the atmospheric circulation in each hemisphere are the three
belts of the trade winds, the westerlies and the polar winds;
but it should be borne in mind that these orily represent the
surface winds, often only shallow currents, and that the direction·
is generally different and the force always greater at higher
altitudes. The importance of this will doubtless be great in an
era of commercial flying, since, by careful selection of flying
level, favourable winds may often be found in both directions
along a route. The projected air route to Australia, for example,
passes over Queensland where strong and steady trade winds,
i.e. following winds on the northward journey, blow for months
at a time. But above these is a N.W. return current, i.e. a
1 See Observer's Handbook.
ELEMENTS 25
following wind on the s'buthward journey, which occurs at 12,000
feet near the tropic but is found at lower and lower levels pole-
wards until it is only 4,000 feet up at Melbourne. The existence
of a return current of air, flowing from S.W. to N.E. above the
northern hemisphere trades had long been recognized from the
drift of high clouds and the movements of dust thrown to con-
siderable heights by volcanic eruptions. It now appears that
this eastward drift is universal at least as far polewards as the
areas of permanent low pressure where the polar winds and the
westerlies meet, and probably as far as the poles themselves.
The general circulation of the troposphere therefore appears to
be as shown in Fig. 8.

Upper Circulation

FIG. S.-Generalized Circulation of the Atmosphere

THE ~ HORSE LATITUDE' FRONT OF DIVERGENCE. The belt


of high pressure known as the 'Horse Latitudes' from which
diverge, at the earth's surface, the trade winds and the westerlies,
is a front of divergence, and the two currents, being divergent,
do not interfere. I t is a zone of calms, a zone of stability, a
region of descending, and therefore dry, air currents. On land
it is a belt of deserts, widest in Africa and Asia where there is
most land, scarcely existent in North America,where the place
of the Sahara is taken by the Gulf of Mexico, i.e. where a great
desiccating agent is replaced by one of the greatest sources of
atmospheric moisture.
This zone of high pressure is not, however, continuous
throughout the year. During the summer in each hemisphere
the belt is interrupted by the low pressures which develop over
continents, but during winter it again becomes continuous, tb.e
pressure over continents being now even higher than over the
26 CLIMATOLOGY
seas. Five areas remain, however, permanent centres of high
pressure, isolated during summer into independent anticyclones;
two are in the northern, three in the southern hemisphere.
They occur over the oceans whose relatively low temperatures
in summer maintain heavy air above them. These permanent
, centres of action' are of vital concern to the climates of the'
surrounding lands, and, combined with the waxing and waning
of the continental pressure systems, are responsible for the
seasonal changes in wind direction.
THE EQUATORIAL FRONT OF CONVERGENCE. The trade
winds originating in these high-pressure areas of the northern
and southern hemispheres meet in the equatorial belt of low
pressure, the 'Doldrum' belt, which is directly due to the
intense heating which it receives. This belt of calms is a front
of convergence where two air currents meet and escape by
upward moveJ;Ilent, and this causes it to be turbulent and stormy
in great contrast with the Horse Latitude calms. But the air
currents which meet here are of equal temperature and humidity,
and thus it is less unreliable and less squally than the other
front of convergence, shortly to be described, which exists
between the westerlies and the polar winds. Since seasonal
temperature changes in these. low latitudes are very slight, there
are virtually no seasonal pressure changes; low pressure is
continuous round the globe throughout the year.
THE SWING OF THE PRESSURE BELTS. The primary cause
of the pressure belts is the unequal distribution of insolation;
the low pressure of the equatorial belt results from the excessive
insolation received and tends to coincide with the heat equator.
But the position of the heat equator is subject to a seasonal
migration, lagging somewhat behind the zenithal position of
the sun, and the low-pressure belt also swings north and south,
lagging somewhat behind the heat equator. The other pressure
belts, intimately connected with this, move north and south
also, so that certain areas lying at or about the junction of two
, zones experience a season in each; the essence of the tropical
climates, for example, is its alternation of equatorial calms and
trade wind conditions. On land the wind belts are interrupted
to such an extent and so modified by continental influences that
they are often rendered entirely unrecognizable, but except in
the Indian and West Pacific Oceans where they are displaced
by monsoons, they retain their characteristics over the sea.
Their limits here can be traced with some degree of reliability
and are stated in the table below. The lag of the wind belt~
b~hind the sun is such ~at the limits of the migrations are not
reached until two or three months after the solstices.
ELEMENTS 27
March. September.

Atlantic. Pacific. Atlantic. Pacific.

r0rth,m Bruit.
N.E. Trades Southern limit
26° N.
3° N .
25° N.
5° N .
35° N.
11° N.
30° N.
loON.
Equato'ial North,rn limit }
Calms Southern limit 0° 3° N . 7° N.
{Northern limit 3° N .
S.E. Trades Southern limit . 25° S. 28° S. 25° S. 20° S.

It will be noticed that all the belts are north of their natural
position, a fact which is explained by the greater amount of land
in the northern hemisphere, and whose consequences are far-,
reaching. It results in warm air and water (from currents)
finding their way across th~ Equator into the northern hemi-
sphere, thus increasing its temperature at the expense' of the
southern, and strengthening the Gulf Stream and the Ruro Siwo
at the expense of the Brazilian and East Australian currents
(see p. 46). Other important consequences of this will be
described later, especially in connection with the climates of
West Africa and Central America.
THE POLAR FRONT OF CONVERGENCE. By contrast with the
equatorial front there meet, along this line, air currents of very
different natures, the warm wet westerlies of equatorial origin
and the cold heavy air gavitating out from the poles. Along the
plane of contact eddies are set up, and patches of the warmer,
lighter westerly circulation penetrate into the heavier polar air
and so rise swirling upwards. Thus are born the greater number
of cyclonic storms which form such a characteristic feature of the
climates of the' temperate' zones. Since these cyclones, together
with anticyclones and their varietal forms, account for nearly all
the weather experienced in certain parts of this zone, certain
properties of these need to be fully examined before their effects
can be appreciated.
r. Anticyclones are generally much less active than cyclones i
they may persist for days or weeks on end, bringing settled weather,
whereas cyclones are hardly ever in the same place on the weather
maps for two consecutive days. They usually move at about
I8 miles per hour in Western Europe, but in N. America their
speed is greater (about 25 m.p.h.); they move faster in winter
than in summer. Their movement is nearly always easterly
(towards E., N.E. or S.E.), following the prevailing wind direction:
Well-developed anticyclones do not as a rule alternate with tMse
28 CLIMATOLOGY
.. .
rapidly moving cyclones, though they are. much better developed
in North America than in Europe.
2. Anticyclones are characterized by quieter weather condi..
tions than cyclones, with much smaller wind velocities, for the
pressure gradient in the latter is nearly always much steeper than
in the former.
3. No general distinction can be made between the tempera-
tures of the two opposite systems as wholes, each type sometimes
brings hot weather, sometimes cold, but certain temperature
conditions can be correlated with certain parts of each system.
These are explained by the direction of the winds circulating in
them. In the south and south-east sectors of a cyclone in the
northern hemisphere the winds are southerly and usually warm,
in the north and north-west they are northerly and usually cold;
while in an anticyclone the north-east is the cold sector and the
south-west the warm one. Exceptions to this general rule are,
however, introduced qy the distribution of land and sea; for
example, in the British Isles the winds in the north-east sector of a
cyclone are usually cold in winter, because off a cold continent, but
in summer these winds are hot. An important contrast should be
noted here between Western Europe and the most densely popu-
lated parts of North America, i.e. the eastern seaboard. In the
former the west winds of the south-west sector are warm in winter
(because marine) and in the latter warm in summer (because
continental). Conversely, the east winds of the north-east sector
in Europe are cold in winter and hot in summer, while New
York is cooled by an east wind in summer.
4. Similarly with regard to the rainfall there is no fixed law
as to distribution. The south-east sector is usually wettest
because here warm wet winds are travelling towards colder
latitudes, while the colder winds of the north-west sector are dry
since they are moving to warmer latitudes and increasing their
moisture capacity. But if the south winds have come over land
and the north winds over sea their rain-bearing values will be
reversed. Cyclones entering the Mississippi basin across the
Rockies bring little or no rain on their advancing edge, since the
trajectory of the wind brings it over dry land; but if the cyclone
enters from the Gulf of Mexico there will probably be heavy
falls in front of the centre from the winds coming off these warm
seas.
CYCLONES AND THE POLAR FRONT. The majority of the
cyclones of temperate latitudes originate, as has been already
described, along the' polar front' or line of discontinuity between
the cold polar winds on the one hand and on the other the warm
cii~ulation of the westerlies. Families of cyclones originate along
ELEMENTS 29
this line each characterized by fL 'warm front' or ' steering line'
on its advancing edge where the warm air rides over the cold, and
a' cold front' or' squall line ' in rear, where the eddy of cold air,
swirling round the vortex, thrusts itself under the warm air,
forcing it upwards. The cold front is much more steeply inclined
than the warm front and the disturbances associated with it are
more violent, gusts of northern air with squally showers are
characteristic. The pocket of warm air, thus included in the cold,
rises because of its lighter weight and eventually ceases to exist
as a surface current: but it is present in the upper air and can
be traced at the surface of the ground by the lower barometric
pressure with which it is associated. The plane of discontinuity

FIG. g.-Structure of a Cyclone


(Afte, V. Bjerknes)

between the two currents will be a zone of cloud, for the cold air
will ch.ill the warm and compel the condensation of its moisture.
According to the altitude of this plane of discontinuity the clouds
will change in type, on the approach of a cyclonic centre, from
cirrus to alto-stratus and finally to nimbus with copious rain as
the 'discontinuity layer approaches lower and lower.
If the observer is standing in such a position (A-B, Fig. 9)
that the tongue of warm air in passing by him is still in contact
with the ground, he may experience a brief period of fine warm
weather, , the eye of the storm', which will be followed by squally
showers as the cold front arrives. If the observer is stationed at
D (Fig. 9), the passage of the cyclone will not bring the central
area of fine warm weather, for the plane of discontinuity will be
entirely overhead; but the clouds will change from cirrus to
CLIMATOLOGY
nimbus, with rain, steady and persistent at first,' but squally later
as the cold front passes overhead. An observer at F will be
beyond the active reach of the storm centre and will only see,
away to the south, a bank of cloud passing slowly from west to
east.
POSITION OF THE POLAR FRONT. The position of the polar front
is well worth tracing on account of the important meteorological
phenomena associated with it. It should be clearly realized that,.
like the Horse Latitude front of divergence, the polar front is not
in a vertical plane, but is inclined at a low angle (probably about
a third of a degree) to the horizontal. Small variations in the
altitude of the front can therefore produce considerable differences
in the position of its line of intersection with the earth's surface.
Its usual position does not follow the parallels of latitude, .but is a

FIG. lo.-The Polar Front


.
very irregular line consisting of four stretches, each arranged from
S.W. to N.E. (see Fig. IO) and carried far polewards by the warm
air over the oceans. Along each of these stretches there originate
families of cyclones such as have already been described.
COLD WAVES AND THE POLAR FRONT. The polar circulation
is an attempt at a return current of air to the low pressure at the
Equator, but it is prevented by the intervening westerlies. It
sometimes happens, however, that the polar air spreads far to the
south of its usual limits and breaks through the westerlies to
join up, perhaps, with the trades and so to reach the Equator.
When this occurs cold waves will be experienced in temperate
latitudes along the path of the cold air. Such bulges of the polar
.front will clearly have t.heir best chance of success where mountain
barriers impede the westerly winds, i.e. east of meridional moun-
tain ranges such as the Rockies and the Urals, for these paths are
to windward of the mountains from the point of view of the polar
ELEMENTS 31
winds, but to leeward from the point of view of the westerlies.
The western slopes of these north-south barriers are to winaward
for the westerlies and to leeward for the polar winds; here, then,
the westerlies will be able to invade the polar zone and carry
warm temperatures far to the north.
T:s:E ICELANDIC AND ALEUTIAN ' Lows'. Owing to the con-
stant procession of cyclones, the polar front is normally a zone of
low pressure, but being in a zone subject to a considerable 'annual
temperature range, it is interrupted in the northern hemisphere
in winter by the growth of the continental highs of Eurasia and
North America. Thus there are left two areas of low pressure in
winter, the Icelandic and.the Aleutian, which owe their existence
to the pronounced temperature contrasts between the warm
waters of the North Atlantic and North Pacific Drifts and the
cold land surfaces of Greenland-Iceland and Alaska-Siberia. Now
the Greenland ice' cap is permanent, enduring through the
summer, consequently the temperature contrast endures and the
Icelandic low persists throughout summer as a still lower pressure
in a low-pressure belt. But Alaska and Siberia have no perman-
ent snow cover and temperatures here rise considerably in sum-
mer; the temperature contrast therefore ceases to exist and the
Aleutian low is overshadowed by the extension of the continental
lows of North America and Siberia, and ceases to have a separate
existence. . ,
In the southern hemisphere there is practically no land in
these latitudes and the low-pressure belt is permanent, continuous
and uniform.
SUGGESTIONS FOR FURTHER READING
For an account of the elements of climate the standard work is J. Hann,
Handbook of Climatology, Vol. I, translated by Ward, 1903. A. J. Her-
bertson's The Distribution of Rainfall over the Land, London, 1901, is still
valuable; also M. Jefferson, • A New Map of World Rainfall', Geog. Rev.,
1926. B. Franze's monograph on the rainfall of South America is reviewed
and summarized by H. A. Matthews in the Geog. Journ .• 1929. World
Weather Records, assembled by H. H. Clayton and published by the Smith-
sonian Institution, is a valuable collection of yearly data for upwards
of 500 stations spread over the world. See also' Climatic Data', assembled
by Cundall and Thurston in the Geog. Teacher, 1926 and 1927. Vols. II
and III of J. Hann's Handbuch dey Klimatologie also contain useful lists
of figures, as does Kendrew's Climates of the Continents, 2nd ed., 1927.
, For the relation of climate and agriculture the following articles may
be consulted. H. Mellish, • The Relations of Meteorology with Agriculture',
Q. J. Roy. Met. Soc., 1910; B. C. Wallis, • Rainfall and Agriculture in the
United States of America', Monthly Weather Review, 1915; R. deC. Ward,
, The Larger Relation of Climate and Crops in the United States', Q. J.
Roy. Met. Soc., 1919; G. Taylor, • Agricultural Climatology of Australia',
Q.J. Roy. Met. Soc., 1920; R. H. Hooker, • Forecasting Crops from the
32 CLIMATOLOGY
Weather', Q. ]. Roy. Met. Soc., 1921; H. l\f. Leake, • The Agricultural
Value of Rainfall in the Tropics " Proc. Roy. Soc., 1928.
For the circulation of the atmosphere and for modern views on the
formation of cyclones, reference may be made to V. Bjerknes, • The Struc-
ture of the Atmosphere when Rain is Falling " Q.]. Roy. Met. Soc., 1920 ;
Bjerknes, • Extra:tropical Circulation of the Atmosphere', Nature, 1920 ;
A. Stevens, • The' New Outlook in Meteorology', Scot. Geog. Mag., 1927;
V. Bjerknes et H. Solberg, • L' evolution des cyclones et la circulation atmos-
phe:ique d'apres la theorie du front polaire', Mem. Off. Nat., Met., 6,
Pans, 1923.
The standard atlas is Bartholomew's Atlas of Meteorology, being Vol.
III of Bartholomew's Physical Atlas; the Atlas of American Agriculture
is very useful, and reference may be made to the Russian Climatological
A Ilno
CHAPTER III
THE FACTORS OF CLIMATE
OLAR AND PHYSICAL CLIMATE. The latitude of a place,

S together with its altitude and relation to surrounding relief,


determines once and for all the possible duration and
intensity of the light and heat received from the sun. But this
theoretically possible amount will not be attained because of
interference by clouds and because of heat imported or exported
by air currents. These interferences are subject to no rigid laws
and cannot be evaluated with ilccuracy in estimating the climate
of a place. But the climate, in so far as it is governed by the
amount of insolation received, is capable of accurate estimation
and is known as the < solar climate '. The actual climate, result-
ing from the interference by the other factors with the solar climate,
is known as the' physical climate '.
LATITUDE is the prime factor in determining climatic zones,
for the only considerable source of warmth is the sun, and the,
heating effect is greatest where the sun's rays are most nearly
vertical. The amount of heat received by a given area depends
both on the intensity and duration of sunlight, both of which
depend on latitude. The intensity of insolation is greatest where
the sun's rays fall vertically on the surface of the earth, both
because here a bundle of rays of given width is spread over the
minimum area and because the rays are- penetrating a minimum
thickness of atmosphere and therefore ab~orption will be at a
minimum. The duration of sunlight increases in summer with
increasing latitude and decreases in winter; thus in summer the
low angle of the sun in high latitudes is partially offset by the
- greater length of day. This is a fact of great significance in cereal
cultivation and allows, for example, the ripening of wheat in
Canada in places where the growing season is less than 100 days
and where the accumulated temperatures are less than I,400 day
degrees (see p. 10). Furthermore, the long days of weak sunshine
in high latitudes seem to produce better grain than the shorter
hours of more intense sunshine of lower latitudes;' Canadian
wheat is better quality than Egyptian, Italian rice is superior to
Indian. In winter, on the other hand, the feeble sun, low in the
sky, has little power in the short hours of daylight to warm up the
33
34 CLIMATOLOGY
ground, and the heat received by day·is entirely dissipated by
radiation in the long winter night.
INSOLATION AT EACH LATITUDE. At the Equator the value
of insolation varies little throughout the year, for the day is
12 hours long, and the sun never departs far from the vertical
(see Fig. II). There will, however, be two maxima at the equi-
noxes and two minima at the solstices. Polewards the length of
day in summer increases, until at the tropic the midsummer day
is 13t hours long. It is clear that insolation at midsummer is
greater here than it can ever be at the Equator, for the duration
is longer and the intensity is just as great, the sun being directly
overhead. At midwinter, on the other hand, the day is only
IOt hours long and the altitude of the midday sun is only 43° ;
insolation at this season is clearly much less than it can ever be
at the Equator. The single maximum and minimum characteris-
tic of the tropic is, in point of fact, reached at about I2° N. and

--
',--
,~ ~ e(V' -I- ~
--
"" Or
...... :-- .«- .'j;
./ J ,V
I--""
~ 0.'1
~J
\
"
\
(,eV n
~a su 10 Soo
i""'o ~ ~
~ to~ o\~ n)

'r'!/ \ [;' ~~~


_~w~~~~~wowrom~~~w~_
J F M A M J J A SON 0
South -E-- ~ North

FIG. lI.-Annual Variation FIG. 12.-Insolation at Different Latitudes


of Insolation on 21st June

S.; for beyond this point the gain of insolation by increased


length 6f day outweighs the loss of insolation consequent on the
rather lower angle of the sun. Polewards again the length of day
increases steadily while the midsummer altitude of the sun
steadily declines. For a while the former more than compensates
for the latter and midsummer insolation increases polewards
until about 43lo, where a maximum is reached. Winter insolation
here is, of course, much less than at the tropic, for days are short
and the midday angle of the sun is only 23° (curve b, Fig II).
Further polewards still the relative values of the two factors are
reversed, the increasing length of day fails to counteract the low
angle of the sun, and the value of insolation decreases to a
minimum at 62° N. Beyond this, however, the length of day
increases rapidly until at the Arctic Circle it is 24 hours long;
the values of the factors concerned are again reversed and insola-
tion rises again. It might be thought that beyond the Arctic
Circle the lower elevation of the midday sun, without the com-
pensation of increased length of day, would cause insolation to
FACTORS 35
decline, but actually the opposite is the case and the curve
continues to rise until a finaJ. maximum is reached at the Pole,
exceeding all previous ,maxima. This is accounted for by the
increased altitude of t]:\e sun at midnight, which is 23tO at the
Pole, while at the Arctic Circle the midnight sun at midsummer
only skims the horizon. At midwinter within the polar circle
there is, of course, no insolation as the sun does not rise; insola-
tion at the Pole ceases at the autumnal equinox and begins again
at the vernal equinox (see curve c, Fig II). ..
EFFECT OF THE ATMOSPHERE. Up to this point we have con-
sidered only the theoretical value of insolation arriving at the
surface of the sphere, but·to arrive at the insolation value at the
surface of the earth a correction must be applied for the loss of
heat in the passage of the sun's rays through the· atmosphere.
The amount which penetrates will be determined by two factors :-
(I) The thickness of air to be traversed, a fixed and calculable
value, and (2) the degree of transparency of the air which varies
with conditions of cloud, dust, etc. Other things being equal, the
absorption of the sun's rays will be greatest in high latitudes
where the oblique rays have to pass through a greater depth of
atmosphere; the theoretical high values of insolation here are
therefore reduced considerably when the correction is applied.
If the coefficient of transmission is taken as 0.50, an average
amount in fairly clear weather, only 18 per cent. of the total
insolation reaches ground-level at the Pole. The lower curve in
Fig. 12 is drawn on this assumption and the upper curve shows
the value of insolation received by the outer layers of the atmos-
pher~. It should be remembered that absorption is much greater
in the cloudy belts of the equatorial calms and the stormy
westerlies than under the clear skies of the high-pressure belts
of the horse latitudes, so that the'lower curve, based on a fixed
coefficient of transmission, has little more than a theoretical
value.
The value of insolation, thus corrected, supplies an approxi-
mate measure of the sun's heating effect, but this solar control
of climate is effective in broad outline only. Certainly it de-
termines the latitudinal arrangement of the main climatic zones,
but in detail it is profoundly modified by such factors as alti-
tude, marine influence, physical features, plant cover, etc., the
resultant being the Physical Climate.
EFFECT OF ALTITUDE. Height above sea-level has a profound
influence on climate, in many respects imitating the effects .of
increased latitude. The peculiarities of climates of high altitudes
will be dealt with in detail in a later chapter, but it will be neces-;,
sary here to examine some of the more important effects. Briefly
CLIMATOLOGY
these are (r) a decreased pressure, (2) a .decreased mean tempera-
ture, (3) an increased precipitation. .
PRESSURE AND ALTITUDE. Although the decrease of pressure
with altitude is slightly more rapid in cold climates than in hot,
the difference is not significa:o.t, and as a rough approximation it
may be assumed that :-
From sea-level to 2,000 feet the decrease is 4 per cent. per 1,000 feet.
" 2,000feetto 5,000 " 3
'" 5,000 feet to 10,000 " 2'5

Or, stated in another way :-


The barometer stands at 30 inches at sea-level.
29 830 feet.
28 " 1,800 "
26 " 3,800 "
24 " 5,900 "
20 "10,600,,
16 ,,16,000 "

At r8,500 feet the pressure of the atmosphere is only half that at


sea-level.
The direct physiological results of diminished pressure are
breathlessness and a feeling of lethargy culminating in ' moun-
tain sickness'. The height at which this malady appears varies
with the individual and with conditions, but is usually about
r5,000 feet. It is an anremic condition, a purely physical result
of decreased oxygenation of the blood in the rarefied atmosphere.
That it is not the result of exertion is proved by the fact that
passengers on ~ountain railways suffer in the s.ame way as climbers
on foot, but the symptoms naturally become more acute in pro-
portion as the system demands more oxygen for greater exertion.
A high hremoglobin content, which increases the capacity for
absorbing oxygen, is characteristic of the blood of all native
animals and inhabitants of these high altitudes, and acclimatiza-
tion of visitors is slowly achieved by increasing the number of
red corpuscles, often by more than 50 per cent.
The decrease in density of the air brings about a marked
reduction in the absorption of insolation, an effect which is further
increased by the virtual absence, at these high altitudes, of water
vapour, which acts as a powerful absorbent at lower levels. The
result is a marked difference between sunshine and shade tem-
peratures and other peculiarities which will be dealt with later
(pp. 254 and 262). .
TEMPERATURE AND ALTITUDE. The rate of decrease of
temperature with altitude is subject to considerable variations for
focal reasons, and there may be actual inversions, but it is usually
FACTORS 37
in the neighbourhood of 1° F. for every 300 feet of ascent. It is
usually less in winter (410 feet in the British Isles) than in
summer (270 feet), less at night than during the day, less on
plateaux (290 feet) than on mountains (265 feet) and still less
on plains (365 feet). There are' numerous exceptions to these
generalizations, for example the winter lapse rate in eastern
Brazil is 1° F. in 226 feet and the summer lapse rate only 1° f.
in 500 feet. This is a result of the high humidity and aS,cending
air currents of the belt of calms centred here in summer, for in the
condensation which results heat is liberated irl great quantities
and delays the fall in temperature. The chief c;lUse of the normal
temperature lapse is the rarity of the air at higll altitudes and its
relative poverty in moisture and carbon dio}{ide, whereby its
power of absorbing heat is much diminished. The sun's rays are
allowed to pass through without warming the air, and free radia-
tion from the land surface is promoted at night or whenever the.
S~~ 'J~,:~..'i,~S tQ 'i,l\.\.~~. A..t h'N~>: k'i~ls tl\.~ <i~~<SI'J: :a.iJ:, :a.T.tQ.. 1JJD1.:~
particularly the clouds, have 'a blanketing effect which checks,
this loss by radiation. Again, in mountain districts the large
land surface exposed facilitates radiation, while much of the
surface, being in shadow, receives little or no beat from the sun.
On the other hand, although air temperatures are low, the sun's
rays, passing freely through the clear thin air, beat fiercely on
solid objects such as rock surfaces, heating them intensely and
burning the exposed skin of hands and face. Mountains and
plateaux thus enjoy climates characterized, inter alia, by lower
temperatures, and thus appear as ' islands' of, perhaps, tundra
in the rliidst of forest. The benefits which this sometimes confers
may 'be seen in Africa and South America, both of which have
enormous areas of high land in the equatorial belt; but for their
altitude these would be, sheer jungle, as impenetrable as the
Amazon and Congo forests, and as unhealthy; actually these
plateaux are among the most prosperous and promising parts of
the continents.
SNOWFALL ON MOUNTAINS. High mountains, even on the
Equator, experience regular temperatures below freezing-point and
are, subject to adequate precipitation, capped by perpetual
snow. It is important, however, to realize that precipitation, as
well as temperature, helps to define the snow-line, for the loss
by melting and evaporation must be made good by fresh falls;
the Pamirs, although rising far above the snow-line as determined
by temperature, have very little snow cover because of their
aridity, placed as they are out of the course of moisture-bearing
currents; the snow-line on the windward side of the Andes and
Himalayas is hundreds of feet lower than on the leeward side:
3
CLIMATOLOGY
Fig. I3 shows diagrammatically the position of the snow-line at
various latitudes. .
RAINFALL AND ALTITUDE. More far-reaching than the effects
on pressure and temperature is the profoul1,d modification of
rainfall distribution by high land. Mountains in general enjoy
a higher rainfall than lowlands similarly placec:l, as can be clearly
seen from almost any isohyetal map.! On the mean annual rain:
fall map of the United States, for example, the Cordilleras and
the Appalachians are clearly brought out by ridges of heavy
precipitation, and even such small relief features as the Ozarks
and the Black Hills of Dakota stand out cOnspicuously. The
latter receive 25 inches of rain in the midst of an area with only
I5 inches, and the vegetation responds in no uncertain manner.

0' Kenya 27040 (S.L.IS,OOO)

FIG. I3.-Altitude of the Snow-lh1e

Their name, in fact, refers to the dark' green forests, the basis of
a local lumber industry, which make such a striking feature in the
surrounding prairie. The higher land of the Sahara traps the
rain-bearing win4s to make altitude oases suell as Dar-Fur which
gets its water from the Jebel Maria. The in<:rease of rainfall is
explained by the cooling of the air by (r) Forced ascent up a
slope and (2) by contact with the cold surfaces 4)f higher altitudes;
consequently the effect is most pronounced Where the high land
opposes a barrier to rain-bearing winds, i.e. where the rainfall is
orographic, and least marked in regions of calms where the rainfall
is convectional. Further, the moisture capatity of air is not a
simple function of temperature, and the .change of capacity of
warm air on cooling is, degree for degree, greater than that of
cool air; hence the ascent of air in tropical latitudes results in
FACTORS 39
heavier rain than a similar ascent in temperate clim~tes.. Th~
temperature lapse, also, is steeper in low latitudes than III hIgh, so
that a gi:ren ascent brings about a greater coolirlg· . !he heavie.st
rainfalls III the world are recorded under such cc>ndItIons, Kaual 1
in the Hawaiian Islands, has an average rainfall of 47 6 inches and
Cherrapunji, in Assam, has 450 inches.
RAIN SHADOW. Hawaii lies in the belt of the N.E. trades,
wJ:ich,. coming across a warm sea, bring more tpan 1:1-0 inches to
HIlo on the N.E. coast and perhaps 180 inches to the hIgher slopes.
But this excessive rain mus1::be compensated elsewhere and there
is a compl~mentary zone of deficient rainfall in leI'! of the high land.
Here the conditions are reversed, the air is pein~ warmed by
descent and, in addition, is passing from the co}d hill-tops to the
warmer plains, its capacity for moisture is inc;reased and little
rain falls. Hilea, on the lee side of Hawaii has (mly 35 inches and
the result is the desert of Kau. The Western Ghats, opposed for
five months to the constant S.W. monsoon, h~ve 250 inches in
places on their western slope, while many stations on the eastern
slope have 25 inches or less. The effect of mountains is thus to
bring about not so much an actual increase of rainfall as a re-
distribution.
RATE OF INCREASE OF RAINFALL WITH ALTITUDE.· The
increase becomes noticeable before the mountains are actually
reached,as is shown by the figures for the following stations, all
at about 65 feet O.D. on the Gangetic plain 2:
• Dacca, 100 miles from Khasi Hills 78 'inches
Bogra, 60 92
, 110
. Mymensing,' 30
Sylhet, 20 , 15 0
This fac~ is a warning against attempts to e&timate accur~tely
the rate of Illcrease with altitude, for it is cleaf that a multItude
of oth~r factors ~re in operation. The height of mount~ins behind
a partIcular statIOn, by enforcing the ascent of tiPper aIr currents,
may give that station a rainfall out of all proportion to its height
above sea-level. Again, distance from the sea, temperature of the
rain-bearing currents, the temperature of the land surface, the
steepness of the slope and the presence or absence of gaps in the
crest all have an influen~e on the precipitation at different levels:
On the South Downs behind Bognor and Brighton ,3 where the
chalk ridge is unbroken, the rate of increase is about 2 per cent.
per 100 feet, but on the discontinuous greensand hills behind
1 Climatological data, Hawaiian Section 1922 quoted by S. S. Visher
in Climatic Laws, 1924. "
2 H. F. Blandford, The Rainfall of India. Quoted by Hann.
a M. de C. Salter, Rainfall of the British Isles, J9 2 1.
CLIMATOLOGY
"
Cranleigh it is only about I per cent.; 1:he air currents go round;
not over the hills. On the steeper slopes of the.Welsh mountains
the figure is considerably exceeded (4 or 5 per cent.) and in tropical
climates it is much higher still. . .,
ZONE OF MAXIMUM PRECIPITATION: The i~crease of precipi'-
tation is not maintained indefinitely up a slope; there comes a
level at which the rate of increase slackens and even,tually ceases,
while higher still there is an actual decrease. There is, then, a zon~
of maximum precipitation whose altitude varies slightly from
place to place, is lower in tropic than temperate zones, in humid
climates than in arid, in the cold season than the hot, in the wet
season than in the dry. In Java it appears to occur at about
3,300 feet, in the Western Ghats about 5,000 feet, in the Sierra
Nevada behind Los Angeles about 5,000 feet, and in the Alps
about 7,000 feet, while in the British Isles this zone is probably
never reached. The explanation of the decrease above the zone
of maximum precipitation is to be found in the decreased absolute-
humidity at high altitudes; temperature is lower and the mois~
ture capacity of the air is less, consequently there is less available
for condensation. All' interesting example of seasonal variation
of this zone is quoted by Hann from Swerzow. In the Tian Shan
the zone of maximum winter snowfall is about 7,000 to 9,000 feet,
and here are the coniferous forests, while the zone below is devoid
of forests because of its aridity. In summer the zone of maximum
precipitation rises above the coniferous forest and the plentiful
'summer rain nourishes a rich grass which provides pasture for the
flocks and herds which the Khirgiz maintain here in winter,
safely above the zone of heavy snow.
MOUNTAINS AS CLIMATIC DIVIDES. By interfering with the.
free flow of air currents and by influencing the distriht;ttion of
rainfall and temperature mountains tend to coincide with import-
ant climatic divides. The Dinaric Alps separate the extreme
continental climate of the Hungarian plain, with January tempera-
tures below freezing, from the warm Adriatic littoral with January
temperatures of 50°, It would never be imagined at Fiume that
such low temperatures could exist so near at hand, but unpleasant
confirmation is supplied when the passage of a depression to the
south draws off some of the cold heavy air over the lip of the
Hungarian basin down into the Adriatic as the icy , Bora'. The
Alpine-Himalayan mountain system throughout its length pro:
~ides a climatic divide of the first magnitude, excluding polar
influences from the lands to the south and tropical influences from
the lands to the north. The winters of Siberia, Turan and
China are disproportionat~ly cold, those of India are dispropor-
tionately warm; Multan (January temperature 54°) and Shangh'ai
FACTORS 41
(January temperature 38°) are on the same parallel. The
Sacramento Valley, cut off by the coast ranges from the cooling
influence of the sea, has mean July temperatures in the neighbour-
hood of goO, while San Francisco records only 57°. The contrasts
in rainfall are just as marked, roo inches and winches occur with-
in 200 miles of,each other on the west and east slopes of the.
Andes respectively in southern Chile and Argentina.
The north-and-south grain of America and the east-and-west
grain Of Eurasia offer an interesting comparison of climatic
results. In North America there is no barrier to the free meri-
dional movement of air, for the Gulf-Hudson Bay divide is below
1,000 feet and the plains offer no obstruction. The low pressure of
cyclones over the southern states can therefore draw huge
stotes of icy air from the great Canadian reservoir down to the
very shores of a tropical sea. In Asia, on the other hand, the
warm tropical air of India is kept separate from the Siberian cold
bY' a well-nigh impassable bar;rier. The contrast does not stop
there; for the direct result of the great Siberian well of cold is
to generate a huge anticyclone, the source of the winter monsoon.
In America.this cannot exist to the same degree because of mix-
ture. with warm air from the Gulf. In summer the heated south
of North America is tempered by northern influences, while in the
enclosed basin of India the unmitigated heat sets up an intense
low pressure which is the focus of the summer monsoon. Thus it is
the disposition of the relief, rather than the relative sizes of the
continents, which accounts for the feeble development of the
monsoon in America as compared with Asia.
I;NFLUENCE OF LAND AND SEA ON CLIMATE. Next to the
variation of insolation with latitude, the distribution of land and
water over the surface of the earth is the most important control
of climate. A variety of physical effects combine to make water
much more conservative of heat than land; slower to warm up,
slower to cool down, it has a moderating influence on temperature
which may be felt at considerable distances inland. The distance
to which marine influence penetrates will depend on the prevailing
wind direction coupled with the facilities which the relief offers for
free entry of ocean winds. Canada and the northern United
States, for example, lie in the belt of the prevailing westerly
winds, but marine influence is restricted by the Cordilleras to
a narrow coastal strip in British Columbia, Washington, Oregon
'and California. The East coast is a lee shore and the climate is
continental right up to the seaboard. Marine and continental
climates in this part of North America are clearly and sharply
differentiated, the coastline being simple and uniform, but iI}
Western Europe the intimate intermingling of gulf, bay and
CLIMATOLOGY,
inland sea with islands and peninsulas, t<>gether with the absence
of marked relief features athwart the path of the winds, allows
the penetration of marine influences for hundreds of miles into the
continent. The complexity of the coastline· and of continent'al
relief in Europe has its,sequel in a complexity of climates, whiCh
contrasts strangely wr1h such a compact 'continent of uniform'
relief as, for example, Africa. (The differences between Vlarine.
and' continental climates may conveniently be grouped under.
three headings: (I) Rainfall, (2) Temperature, and (3) Pressure'
and Winds. .
CONTINENTAL AND MARINE RAINFALL. The oceans are the
chief sources of atmospheric moisture, and it is to be expected
that humidity and' rainfall will be greatest where the direction
of the prevailing wind is on-shore. 'i The sudden decrease of rainfall .
inland along the w~stern mountain divide of both Americas in the
zones of the westerly winds is an indication of the narrow fringe
of oceanic climates there, just as the gradual eastward diminution
of rainfall in Europe in these latitudes shows its great extension
here. Rainfall in marine climates is chiefly orographic and usually
fairly adequately distributed throughout theyear, rainfall in con-
tinental climates is chiefly summer rain resulting from convec-
tional overturning set up by the high summer temperatures.
CONTINENTAL AND MARINE TEMPERATURES. The diurnal
variation of temperature of sea surfaces is almost a negligible
amount, and even the annual variation is extremely small;
within the tropics it seldqm exceeds 10°, in ,mid-latitudes it
reaches 20° or 25°, but these figures are only exceeded where the
oscillation of boundaries between currents brings a certain place
now under the influence of a warm current, now of a cold. Fluctu-
ations of the Gulf Stream and the Labrador current are responsible
for a yearly range of surface temperature of over 50° off the coast
of Maine, but this is local and climatically ineffective, expeci-
ally as the wind is here off-shore. The variations of air tempera-
ture over the water surface are thus kept very small and present
strong contrasts with the ranges recorded over the land. The
uniform temperatures of winds blowing off the sea, together with
the greater humidity of the air which impedes insolation by day
and radiation by night, results in:
I. A diminution of the daily range of temperature.
2. A diminution of the annual range of temperature.
3. A delay in the attainment of the daily and yearly maxima
and minima. These are features characteristic of marine climates
and are illustrated in Fig. 14, which shows temperature curves for
three stations, all situated in approximately the same latitude.
Jt will be noticed that the annual range at Valencia is less than
FACTORS 43
IS°, that the hottest month is August and the coldest February.
At Semipalatinsk the range is over 70°, the maximum is reached
early in July and the minimum early in January. .
It is interesting to compare the northern and southern hemi-
spheres in respect 'of the continentality of their climates; the
northern hemisphere contains most of the land, the southern is
essentially a sea hemisphere. In the northern 'hemisphere the
32° isotherm for January comes down to latitude 35° N. in China,
while nearly all of Asia north of 40° N. has a mean January tem-
perature below freezing. North America is slightly more favoured,
but even here the 32° isotherm extends south of 40° N. in the
Mississippi basin.
In the southern hemisphere, none of the three southern
continents reach the 32° July isotherm, although the tip of
Tierra del Fuego reaches 55° S. At the other extreme the summer

--
80 80
70 <
70
y ~
60 ,.... 60
50 ya enc a
40
~J
~
I£. ~~
~'\" - 50
40
30
20
10
r-
IN.~~

~
·s
,;;;;
W
\
'" ,I
1\
\
30
20
o
o ~ ~ o
J F MA MJ J A SON 0
FIG. ,14.-Yearly March of Temperature in Continental and Marine Climates

temperature of Tierra del Fuego is only 50°, but in Asia the 50°
July isotherm reaches 70° N. In·the absence of a great continental
mass in high latitudes there can be no great reservoir of cold air
in the southern hemisphere such as exists in Siberia and Canada to
supply the' Cold Waves' and' Northers' of the United States
and the' Buran' of Siberia. Compared with these the' Southerly
bursters' of Australia and New Zealand and the < Pampero' of
the Argentine are relatively mild.
Fig. IS shows the mean annual range of temperature and alsq
emphasizes the contrasts between the two hemispheres. Nowhere
in the southern hemisphere does the range exceed 40°, but in the
northern hemisphere huge areas in Canada have more than double
that figure and in Siberia more than treble.
CONTINENTAL AND MARINE PRESSURE AND WINDS: LAND
AND SEA BREEZES. The difference in behaviour of land and
water towards diurnal and annual temperature changes bege~s
44 CLIMATOLOGY
a, difference of pressure which results. in periodic diurnal and
seasonal winds known as land and sea breezes and monsoons.
The heatihg of the land during the·day causes an ascent of ~ir
over the land and an indraught of oceanic, air; the descent.of
air cooled by radiation over the land at night causes an expulsion
of land air out to sea. Land and sea breezes are most notice-
able and regular where tern_perature changes are most regular
and particularly in equatorial climates. In Java the nativE!
fishing industry depends on its regular occurrence, the boats
start out at night with the land breeze and return about noon
with the sea breeze. •
In the belt of the westerlies the cyclonic disturbances fre-

FIG. Is.-Mean Annual Range of Temperature

quently mask this diurnal periodicity, but in settled weather it


is usually well-marked. In the tropics the sea breeze sets in
at any time between 8 a.m. and I p.m., but its appearance is
usually punctual at any given spot. Its arrival is anxiously
awaited since it is both refreshing and healthy. Houses are
set to catch the breeze and the best residential districts are
those which are most favourably situated for its reception.
In Senegambia 1 a fall in temperature of 20° and a rise of humidity
of 42 per cent. have been noted within IS minutes of the arrival
of the sea breeze. In Senegal the inland temperatures in the
hot season reach 1I00 , but along the coast, thanks to the sea
breeze, they seldom rise above 85°.
At about sunset the sea breeze gradually dies away to a calm
1 'Comptes Rendus,' 1894, quoted by Hann.
FACTORS 45
and the land breeze springs up; frequently it is as unpleasant
and unhealthy as the sea breeze is refreshing. It is not usually
so marked as the sea breeze since the land is hotter than the
sea throughout the year in tropical latitudes and in the summer
in temperate latitudes; the diurnal temperature gradient is
therefore stronger than the nocturnal. In addition, the sea
breeze is not retarded by friction to the same extent as the land
breeze. The pressure gradients which give rise to these breezes
are not very great and are easily modified or even reversed by other
influences. In the trade wind belt they are most strongly felt when
they reinforce the prevailing wind, the opposite breeze only serving
to moderate or modify the trades. In California, for example, the
sea breeze reinforces the prevailing.west wind during the day and
is welcome as moderating the day temperatures, especially as in
summer the breeze is further strengthened by a seasonal low
pressure inland due to the high temperatures of the interior.
MONSOONS. Just as the diurnal march of temperature gives
rise to land and sea breezes, so the annual march of temperature
produces land winds in winter and sea winds in summer, known
as monsoons. Almost every large la,nd mass which is subject
to a considerable annual variation of temperature generates a
monsoon, but it is only in the case of the largest land masses
that the effect is sufficiently powerful to overcome the normal
planetary circulation, and in practice the name is restricted to
the. most conspicuous of these. They will be dealt with fully
on a regional basis in later chapters.
EFFECT OF OCEAN CURRENTS. There are many asp~cts of
continental and marine climates which are not adequately ex-
plained by the contrasts in behaviour of land and sea, and these
are usually the resrilts of ocean currents. A map of the ocean·
currents shows, as the main features of the circulation, a great
east to west movement of water in the equatorial zone com-
pensated by a west to east movement in the temperate zone,
the swirls being completed by a poleward movement along the
western coasts of the oceans (eastern margins of continents)
and an equatorward return along eastern coasts (western margins).
These currents carlY tropical temperatures towards temperate seas
on the eastern margins and temperate waters towards tropical
seas on the western margins. We expect, as a result of these move-
ments, a warming effect along eastern margins in low latitudes (e.g.
the North Equatorial Current) and a cooling on western margins
(e.g. Californian Current), while in higher latitudes it is the west-
ern margins which will be warmed (e.g. the North Pacific Drift)
and the east cooled (e.g. the Kurile Current). This effect is clearw
!>nm,g"ht out in the maps of temper,:ture anomalies (Figs. I and 2,
.CLIMATOLOGY
p. r2). These temperature anomalies are potent factors ·ip.
maintaining the areas of permanent high and low pressures,.
and the circulation of winds round these in turn provide the
driving force of the currents. The S.W. winds, for example, on
the south side of the Icelandic low drive the warm waters of the
Gulf Stream as the North Atlantic Drift to~rds the shores of
Western Europe, banking up against the continental margin a
depth of warm water such as is not equalled even in tropical
zones and whose surface temperatures are far in excess of any
other waters in similar latitudes. The analogous currents of,
the North Pacific, the Kuro Siwo and its continuation, the North
Pacific Drift, although powerful and important, only give rise.
to a positive temperature anomaly at 20° in winter (d. 40° in
the North Atlq_ntic). The greater effect of the North Atlantic
Drift is explained as" follows :-
I. The form and position of N.E. Brazil is such that much
of-the south equatorial current is deflected into the northern
hemisphere, thereby increasing the Gulf Stream at the expense'
of the Brazilian Current. The Philippines and New Guinea
which fulfil the same function in the Pacific, are less efficient
because discontinuous.
2. The greater size of the North Pacific ocean involves greater
dilution of the Kuro Siwo. ,
3. The S.E. trades, usually stronger and more constant than
the N.E., blow beyond the Equator throughout much of the year,
thus transferring much equatorial water from the southern into
the northern hemisphere. This effect is more pronounced in the
Atlantic than in the Pacific and further adds to the volume of
" the Gulf Stream.
Apart from occasional inversions due to differences in salinity,
warm water is lighter than cold, and therefore floats on the surface
where it is further warmed by insolation. It is clear, therefore,
that where the prevailing wind, and therefore the surface drift
of water, is on-shore, there the surface water is relatively warm
and the warming influence is carried inland, Where, on the
other 'hand, the prevailing drift of wind and water is off-shore,
the warm surface water is carried away and its place must be
taken by water either moving along shore or, more usually, rising
from below and therefore much colder. As examples we may
quote (r) The cold Benguela current off S.W. Africa, the result of
the S.E. trades, and (2) the cold water which rises off Cape
Guardafui while the S.W. monsoon is blowing. The latter, in
parenthesis, exerts a cold influence and produces lowest tem-
p.eratures during the season which is really summer.
THE EAST AND WEST SHORES OF THE NORTH ATLANTIC
FACTORS 47
COMPARED, If, now, we consider'-the circulation of the North
Atlantic, it is clear that there is a convergence of warm currents
accompanied by on-shore winds in·low latitudes on the American
shore, while the Eurafrican shore, apart from the relatively
small counter-current (the Guinea Current), is cooled by up-
welling of cold water along the west coasts of Spain and Africa,
The· western side of the Atlantic is therefore hotter than the
east, especially in winter, and has a smaller temperature range,
Compare the following pairs of stations in about the same latitude
and all at about sea level:
Goree (Dakar) Jan, 68,S Mean Annual 74,8 Range 16'2
{ Vera Cruz , " 7 1 '4 77'4 11'0
Banana , July 72'S 77'9 9'0
{ Pernambuco, 75'2 79'0 6'3

The mean annual isotherms of 80° diverge on crossing the


Atlantic westward; they enc~ose <;>nly about 400 miles of coast
in West Africa, but on the American side the whole coast from
15° S, to 20° N, ·has a mean annual temperature of over 80°.
Proceeding northwards the contrasts decrease until at about
30° N, there is little difference between the two sides; the mean
annual isotherm of 70° cuts both coasts in about the same latitude,
From this point northwards, however, there is a striking difference
in the opposite direction as the following figures show.
Washington , Jan, 32'9 Mean Annual 54'7 Range 43'9
{ Lisbon . 49'3 59'S 20'9

Difference 16'4 Difference 4,8 Difference 23'0


New York Jan,30'3 Mean Annual Sl,8 Range 44'2
{ Oporto , 47'0 5 8 '0 18'0

Difference 16'7 Difference 6'2 Difference 26'2


St, John'S , Jan, 24'2 Mean Annual 40'1 Range 36'0
{ Brest . 43'9 5:r 6 20'5
Difference 19'7 Difference 13'S Difference IS'S
(Nain (Labrador) , Jan, -7'1 Mean Annual 22,6 Range 54'0
(Glasgow, 3 8 '6 47'3 19'4

Difference 4S'7 Difference 24 '7 Difference 34 ,6

The American coast suffers from the double disadvantage of


continental conditions and the Labrador current, while the
European is bathed in the warm waters of the North Atlantic
Drift. The western side of the North Atlantic is thus heate.d
by hot currents in the hot zone and cooled by cold currents in
CLIMATOLOGY
colder latitudes, while the east side is cooled by cold upwellings
in the hot zone and warmed by a warm current in colder latttu9.es'.
The temperature gradient along the western margin of Eurafrica
is therefore abnormally gradual, while tha~ in eastern North
America is the steepest in the world over such a great distance.
Such a gradi~nt has important economic and climatic results;
it telescopes the temperature zones, giving a striking variety of
produce, from the almost tropical cultures of Florida to the Arctic
produce of Labrador, within a dis.tance of 2,000 miles. Meteoro-
logically it' is an arrangement predisposing to rep1arkable and
sudden changes of temperature as air may be imported from
a locality, not far distant but differing markedly in temp~rature.
In eastern Asia there is a somewhat similar telescoping of tem-
perature zones, but since the warm current (the Kuro Siwo) is
less powerful, and the cold current (the Kurile Current) has less
ready access through the constriction of the Bering Strait, the
gradient is less steep than in eastern America.
THE COLD CURRENTS OF WESTERN MARGINS IN THE TRADE
WIND BELT. In the southern hemisphere some excellent ex-
amples of the influence of currents on temperature can be traced
round the coasts of South Africa and South America. The form
of the mean annual isotherm of 70° is interesting. It is carried
up the west coast of South America within roO of the Equator
by the cold Peruvian current, and down the east coast as far
as 30° S. by the warm Brazilian current; across Africa it behaves
in a similar manner. The current, rather thaw the latitude,
is here the control of temperature at coastal stations, there is
less than 3° difference in the mean annual temperatures of Beira
(76°) and Mombasa (78.5°) although they are r6° apart in latitude.
The cold currents which wash the western shores of continents
are not merely offshoots of the Antarctic drift, they are reinforced
by cold water welling up from greater depths as the surface
water is skimmed off by the trade winds. Warmer water is
therefore found, not equatorwards, but out to sea. Their equiva-
lents in the northern hemisphere, the Californian and Canaries
currents, are not so strong nor so cold since wind circulation in
the southern (water) hemisphere is stronger and less hampered
by friction with land surfaces. Furthermore there is freer entry
for Antarctic water than for Arctic, since the South Atlantic
and South Pacific lie wide open to the south, while the North
Pacific is constricted into the narrow Bering Strait and the
North Atlantic waters are segregated by the shallow ridge of
the Scoto-Icelandic rise. A further difference is to be found in
the form of the coastal margins which are more favourable for
the free movement of water off the longitudinal coast of Chile
FACTORS 49
and the plateau edge of South-west Africa than off the broken
transverse coastline of Eurafrica. In this last respect the Cali-
fornian current resembles the Peru and Benguela currents and
differs from the Canaries current.
CURRENTS AND RAINFALL. It is quite in accordance with
expectations that rainfall is heavy on coasts washed by a warm
current and light where cold currents flow. Air passing over
warm currents will be saturated at a high temperature and
therefore a pregnant source of rain, while air which has passed
over a cold current usually has its capacity increased by warming
on coming into contact with land. Further, Warm currents
occur, with very few exceptions, where the wind is on-shore
and cold currents where it is off-shore. The currents, therefore,
only emphasize a distribution of rainfall which the prevailing
wind determines. Thus warm currents increase the rainfall in
British Columbia, the British Isles, in Japan (from the South-
East Monsoon) and in Queensland (from the North-East Monsoon),
etc., while cold currents contribute to the aridity of the Kalahari
and Atacama deserts, Patagonia, etc. The influence of the cold
Benguela current extends as far north as Banana at the mouth
of the Congo, not far north of which it meets the warm Guinea
current. Here are found some striking contrasts of rainfall.
Banana, on the coast, has only 28 inches compared with over
,fo inches in the interior, while Libreville, only 8° further north
along the coast, being under the influence of the warm Guinea
current eddying round the Bight of Biafra, has nearly roo inches.
Cape Lopez divides the spheres of influence of the two currents
and also the two rainfall types, but sometimes the warm waters
of the Bight escape south of Cape Lopez and bring heavy falls
of rain to the coast even south of Banana. A similar function
is served by the bulge of the South American Pacific coast in
Ecuador, which sharply divides the desert from the forest, the
critical point being Cape San Lorenzo.
INFLUENCE OF LAKES. To an extent roughly proportional
to the size of the body of water lakes moderate the climate of
their immediate vicinity. '. Lake-side stations on L. Constance
have a mean annual temperature '5° warmer than stations away
from the lake, and the difference in autumn and winter is as
much as rO. The Great Lakes of North America produce more
profound effects and the January isotherms are noticeably
deviated from their course (see Fig. r6). The severity of cold
waves is noticeably diminished here and the duration of the
frostless season is noticeably longer. It is particularly on the
eastern (lee) shore that these benefits are felt; the favourabl~
conditions east of L. Michigan are reflected in a fruit-growing
50 CLIMATOLOGY
area of some importance. Diurnal lake breezes are clearly
recognizable round such bodies of water as the Caspian Sea and
Victoria Nyanza, exactly similar in character and origin to land
and sea breezes. Humidity and precipitation are also increased,
especially on the lee side, as, for example, on the western side
of Victoria Nyanza from May to October while the S.E. wind
prevails.
INFLUENCE OF SNOW AND ICE. Water bodies depend for their
moderating influence on certain properties, some. of which, such
as the mixing of layers, disappear when the water freezes. The
differences in behaviour of land and sea are therefore less pro-<

FIG. 16.-January Isotherms (Full) and Mean Annual Minima of Temperature


(dotted) over the Great Lakes

nounced in high latitudes, and large lakes, when frozen, exert


a diminished influence. The Gulf of Finland, before it freezes,
brings to bear a beneficial influence on the climate of its shores,
but in late winter and spring its influence declines. At Haparanda,
for example, where the ice closes in in November and lasts until
the end of May, it is 6° warmer in October than in April and in
September than in May, but June, when the ice has disappeared,
is only 2° cooler than August. In a similar manner a cover of
snow delays the advent of spring by preventing the land from
exerting its usual effects on warming up and by using up spring
warmth in melting and evaporating the snow; a late spring is
therefore characteristic of all snowy climates. Snow also in-
creases the annual range of air temperatu,re by increasing radiation
during winter, but, on the other hand, being a bad conductor
FACTORS 5I
of heat, it preserves the warmth of the soil by preventing the
escape of heat from the ground. For this last reason an early
fall of snow may be the means of preserving autumn-sown cereals
from winter frosts, as in Ontario where the snow comes before
the severe frosts, and supplies, in addition, moisture, on melting,
for the growing plant.
INFLUENCE OF PHYSICAL FEATURES. In shaping the smaller
details of climate the minor relief features play an important
role. The .run of the feature lines has a not inconsiderable
influence on the direction of winds, which tends to blow along
rather than across them. Valley bottoms and low-lying land
often suffer from inversions of temperature which may give rise
to severe frosts, from which the valley sides, better drained of
cold air, are exempt (see p. z59). The northern slopes of east-
and-west mountain valleys in the northern hemisphere enjoy a
more genial climate because of the greater intensity and duration
of sunlight which they en1oy,. Grain can be grown I,OOO feet
higher on the southern slopes of the Alps than on the northern,
while pasture, which extends to about 7,000 feet on the north
side, occurs up to about 8,000 feet on the southern slopes. 'South
aspect' is the main natural asset of many popular resorts such
as Nice and Torquay. A backing of high ground to keep out
northern influences is also of value, a function which the maritime
Alps effectively perform for Nice, and the highlands of Devon
less effectively for Torquay. Eastbourne is indebted to Beachy
Head and the chalk downs for shelter from rough west winds
and Bournemouth similarly shelters behind the Isle of Purbeck.
An interesting case of temperature control by relief is afforded
by Siberia and the Amur basin, where cold stagnant air accumu-
lates in winter in the basins ringed round by mountains. This
intensely .cold heavy air escapes over the lowest and easiest
notches in the basin rim and brings intense cold to the towns
situated in the gaps. Thus at Vladivostok, behind which is a
gap at 600 feet, the January temperature is nearly zero, although
only in latitude 44° N.; but further north, where the mountain
wall is more formiq.able, the temperature along the coast is
milder. At Nicolaievsk, in the wider and lower gap of the Amur,
the January temperature reaches IZo below zero, but still further:
north the crescent of the Stanovoi shelters the shores of the Sea
of Okhotsk and here is Ayan, zoo miles north of Nicolaievsk,
enjoying a temperature of only 4° below zero.
SOILS AND CLIMATE. Geological formation and the resultant
soil type are also minor' factors in determining climate. Dry
soils, such as sands, have a low specific heat responding rapidl¥
to temperature changes, while wet soils, such as clays, retain
52 CLIMATOLOGY
moisture and are therefore conservative of heat and cold; the
latter also predispose, by bad drainage, to fogs and mist. Further,
by the influence which soils exert on the nature of the vegetation,
they can powerfully modify the effects· of climate, producing
grassland (e.g. on chalk) where the climate is suited to forest,
or desert (e.g. on loose sands) where there is enough rain to nourish
grassland on any other soil, while limestone, 'in certain'cases,
gives rise to a type of scenery which is virtually desert, in spite
of an adequate or even abundant rain.
VEGETATION AND CLIMATE. Climate is the chief control of
vegetation type (see p. 58) and; the choice between forest and'
grassland is generally made by the rainfall, but there is no doubt
that the vegetation in turn reacts strongly on the climate. The
dense vegetation of th,e Selvas, for example, by its enormous
transpiration, increases the humidity of the air and gives greater
potential rainfall. This transpiration process has been utilized
for the drainage of swamps, and there are many places in Italy
and France (e.g. the Landes) which have been reclaimed by
afforestation; the original grass and swamp vegetation, in these
cases, had insufficient transpiration surface. Forests affect tem-
perature, too, particularly the maximum temperatures, which
they moderate, by casting shade, by offering a larger surface for
radiation, by absorbing heat in the process of evaporation froin
the foliage and by the production of fog, mist and cloud which:
ward off the direct rays of the sun. They act as wind-breaks,
affording shelter for crops and settlements, and by decreasing
wind velocity they decrease evaporation and thus alleviate
aridity; rows of trees are often planted for this purpose.
SUGGESTIONS FOR FURTHER READING
Han'n's Handbook of 'Climatology, Vol. I, is again the standard work
on the subject-matter of this chapter, and contains copious references.
Further information may be obtained from S. S. Visher, Climatic Laws,
1924; A. McAdie, 'Monsoons and Trades as Rain and Desert Makers,'
Geog. Rev. 1922; R. C. Murphy, 'Oceanic and Climatic Phenomena along
the West Coast of South America in 1925', Geog. Rev., 1926; C. E. P. Brooks,
The Influence of Forests on Rainfall and Run-off,' Q. J. Roy. Met. Soc., 1928.
CHAPTER IV
THE CLASSIFICATION OF CLIMATES
EED FOR CLASSIFICATION. The almost infinite combina-

N tior:s of climatic factors acting on a~ aln;-ost infinite


vanety of topography produce a bewlldenng number
of geographical climates, and it is clear that any system. of
classification adopted can recognize only the broadest types
unless it is to become unwieldy. But in spite of the seeming
complexity it becomes clear on closer examination that certain
combinations of climatic elements repeat themselves with some
degree of regularity in differ~nt parts of the world, and it is
convenient to recognize each type and to give it a name.
CLIMATIC PROVINCES. Supan,! in I896, suggested a divi-
sion of the world into 35 provinces, each characterized by a cer-
tain combination of climatic elements. These ,provinces are
reproduced in Bartholomew's Atlas oj Meteorology (the 35th, not
mentioned there, is the Antarctic province). This is simply a
list, not a classification, but these provinces might be arranged
in groups together (e.g. the desert provinces, I2 Saharan, I4
Kalahari, I7 Inner Australian, and 3I Peruvian) to form a classifi-
cation. The question now arises as to ",hat principles should
be adopted in sorting and co-ordinating geographical climates,
and clearly this will depend on the purposes for which the
classification is required.
THE OBJECTS OF CLASSIFICATION. Classifications are of two
kinds, on the one hand those intended to show genetic relation-
ship and on the other' Classifications of Convenience' based on
certain observed similarities, not necessarily related, but pro-
ducing similar effects by the co-ordination of which easier
memorization is effected. One might, for example, group to-
gether all the desert climates irrespective of their cause of aridity;
that would be a classification of convenience without genetic
basis. Or one might group together all the monsoon climates
irrespective of their temperatures and rainfalls; that would be
a clas<;ification for which there would be a true genetic basis,
namely, the fundamental control of the seasonal wind reversal,
springing from the same ultimate cause, but it would bring,
1 Grundzuge der Physichen Erdkunde. Leipzig. 1896.
4 53
54 CLIMATOLOGY
together climatic types inducing such enormously different bio-
logical responses as those in the Ganges delta and in Manchuria.
The ideal classification is one which combines the advantages of
both, and thus groups together provinces with a similarity of
elements resulting from a similarity of causes.
THE IMPORTANCE OF BIOLOGICAL RESPONSES. As a geo-
graphical science climatology i~ vitally concerned with the causal
relationships between certain distributions, and regards as essen-
tial criteria the influences which climatic conditions exert on the
sum total of human activities. Perhaps the truest indi.cator
of these is afforded by vegetation type which, in turh, invokes
certain clearly defined physical and economic responses. It is
unfortunately true that vegetation reflects not only the sum
total of climatic influences, but also of soil conditions, ground-
water supply, human interference and a host of other controls,
yet in broad outline it cannot be doubted that climate is the
predominant control of major vegetational regions. Other factors
may considerably modify the margins of the type, but climatic
provinces, on a large scale, may fairly safely be judged by their 1
characteristic plant assemblages. Any classification which de-
velops and stresses such relationships is of value to the geographer,
but to choose vegetational environment as a basis of classifica-
tion is only a first step. The data of climatology are not the
facts of vegetational distribution, but the statistical values of
the climatic elements, and our next step is to discover those
elements which most potently influence vegetation types, for we
may then accept these as the true criteria. It may be said at
once that temperature and rainfall are the elements required;
either by actual amount or seasonal distribution their control
is paramount.
CLASSIFICATIONS BASED ON TEMPERATURE. The temperature
control of climatic zones was r:ecognized by the Greeks, and
although the classical division of the earth into torrid, temperate
and frigid zones was delimited by the mathematically and
astronomically defined lines of the tropics and the polar circles,
there was, as the name implies, a recognition that temperature
was all-important. But such a scheme fell short in ignoring
the effects of land and sea. These zones are, however, still
recognized as the fundamental divisions, although under new
names and delimited in a different manner, namely, by actual
temperature values which reflect marine influence as well as
insolation values.
Supanl suggested the mean annual isotherm of 68 0 (ZOO C.)
<\s the limit of hot climates'. It conveniently encloses the trade
1 Supan, loco cit.
CLASSIFICAnON 55
wind belt and practically coincides with the limit of palms and
is in most respects, a very convenient boundary, although its
advantages are coincidental rather than causal. The polar zones
he proposed to define by the isotherm of 50° (10° C.) for the
warmest month and in this we have a line of real significance,
for it is summer warmth that matters in cold climates. Given
one month with a temperature above 50° and adequate rainfall,
trees can grow irrespective of winter conditions, which they
avoid by a resting period. The line therefore approximately
marks the limit of tree-growth, the boundary between the coni-
ferous forest and the tundra, a really critical line for living
~onditions. In Supan's zones we have a convenient framework
with a real biological significance into which a classification might

FIG. 17.-Supan's Temperature Zones

be fitted. On closer inspection, however, the mean annual


isotherm of 70° is an even better limit for the hot climates in
many places and will be adopted below in preference to the 68°
isotherm.
KOPPEN'S 1 CLASSIFICATION. These significant figures (68°
and 50°) were adopted by Koppen in a more complete classifica-
tion, based on the duration of each station above, between or
below these .critical values as follows :-
1. Tropical belt. 12 months above 68°.
2. Sub-tropical belts. 4-II months above 68°; 1-8 months
between 50° and 68°.
3. Temperate belts. 4-12 months between Soc and 68~.
w. Koppen, Geogr. Zeitschr., 1900.
CLIMATOLOGY
4. Cold belts. I-4 months between 50° and 68°; 8--'II
months below 50°.
5. Polar belts. I2 months below 50°.
A much simplified map showing the distribution of these
zones is given in Fig. I8, but it should be realized that the
temperature values used are actual values not reduced to sea-
level; the' isotherms' are therefore not isotherms in the usual
meaning of the term. On a,map of such a small scale there is
inevitably much confusion in mountain districts whe~e the zones
are crowded together and some zones have been omitted for the
sake of clearness.
This classification at lea3t has the merit of symmetry and
simplicity, but it is hardly to be expected that climatic types

FIG. I8.-Kappen's Temperature Zones (simplified)

will conform to so rigid a scheme. For example, the isotherm


of 50° for the coldest month becomes the poleward boundary
of the sub-tropical climates (at sea-level). But this line runs
through Sardinia, the toe of Italy and Morea, thus excluding
the whole of the northern Mediterranean and the greater part of
the peninsulas of Italy and Greece from the sub-tropical climates.
The Californian Mediterranean climate also becomes' temperate',
being grouped with the cool temperate climate of British
Columbia. Cape Town and Swanland, too, are excluded from the
sub-tropical climates. In many respects the isotherm of 6° C.
(43° F.) is a better value to use as the limit of the sub-tropical
climates (see p. 60), but this also fails to delimit the climatic type
with accuracy, for considerable areas of the British Isles have
January temperatures above 43°, but could not, of course, be
CLASSIFICATION 57
called sub-tropical, their summers being much too cool and the
insolation being much too feeble.
Here we are brought face to face with one of the great difficul-
ties of classification, the difficulty of delimiting a climatic
region.
CLIMATIC BOUNDARIES. These are, of course, never sharply
defined; they are transition zones across which one type merges
imperceptibly into its neighbours in all directions. But when
this is a<;lmitted it is still useful to be able to draw dividing lines
on a map, and for this purpose it is convenient to select a given
value of a given climatic element. We have already seen that
the isotherm of 50° for the warmest month is a line having real
significance; other lines may be selected for a similar purpose.
This method has been elaborated by Koppen 1 in recent classifica-
tions. Among others the isotherm of 0° C. (32° F.) for the warmest
Jllonth is used for obvious reasons to divide perpetual frost
C

climates' from tundra climates' and the isotherm of - 3° C.


C

for the coldest month. (26·6° F.) to mark the equatorward


boundary of Boreal climates' and their downward limit in
C

mountain climates, since it guarantees, in regions of sufficient


humidity, a snow blanket for several weeks in winter, a con-
dition which has important results on the vegetation.
By others the IO-inch isohyet has commonly been taken as
the boundary of aridity, and the I-inch isohyet as the definition
of a dry month. But aridity is a matter of evaporation as well
as precipitation, and such hard and fast statements as these
would be more real if qualified by estimates of evaporating
power. A slight improvement is effected by taking the 2-inch
isohyet as the definition of a dry month in hot climates and
the I-inch isohyet in cold climates, but further precision is given
by Koppen in his 1918 classification as follows :-

Temperature. 25° C. 25-20. 20-15. 15-10. 10-5· Below 5.

Desert Boundary 32 cm. 29cm. 26cm. 23 cm. 20cm. 16 cm.


Steppe 64 cm . 58 cm. 52 cm. 46 cm. 40 cm. 32 cm.
"

This is further qualified by a correction for seasonal incidence,


1 W. Koppen, Klasszjikation der Klimate nach Temperatur, Nieder-
schlag und Jahreslauf. Petermanns Mitteilungen, 1918 .. See also
review of above by R. de C. Ward in Geog. Rev., 1919 (with map). And
a discussion by James in the Monthly Weather Review, 1922.
Also Die Klimate der Erde, 1923.
58 CLIMATOLOGY
requiring 30 per cent. to be added to the above figurlis when
there is a pronounced summer maximum of rainfall, and 30 pe):;.
cent. to be subtracted when there is a pronounced winter mag-
mum. De Martonne defines a dry climate as one in which the
mean annual rainfall (measured in ems.) is less than double the
mean annual temperature (measured in degrees centigrade).
From the discrepancies between the two definitions given above
it is clear that there is no unanimity as to the limit of aridity,
or at least that the values of climatic ~lements are incompetent
to define it. Refinements of accuracy, for what they are worth,
are only attained at the sacrifice of simplicity and the boundary,
arrived at by either of the above formulre, is a value difficult
and tedious to work out and set down. .
From the examples quoted above the conclusion is reached
that a single climatic element can seldom satisfactorily define
any climatic region, and where such a limit is used it must
be interpreted liberally and regarded only as the best avail-
able.
CLIMATE AND VEGETATION. It has already been laid down·
as a general principle that a satisfactory classification of climates
, must reflect the climatic control of vegetation; that climatic
provinces must coincide as closely as possible with the major
vegetational regions of the globe. It seems likely, therefore,
that the means of classification may best be discovered by a brief
examination of these controls.
Broadly speaking, vegetation type (i.e. forest, grassland or
desert) is conditioned by humidity, either in the form of rain-
fall or, more rarely, as moisture of the air; the flora (i.e. selva
or taiga, savanna or steppe) by temperature. There are excep-
tions to this general statement; the poleward limit of forest'
growth, for example, is set by temperature, and furthermore
soil conditions (edaphic factors) may overrule climate, causing
grassland to grow where the climate would indicate forest (e.g.
limestone soils) or forest where the climate would indicate grass-
land (e.g. water-holding clays), but in the main it expresses the
true 'relationship between climate and vegetation.
FOREST CLIMATES. The essential requirement for tree-growth
is a constant supply of moisture available for the roots. A
season of prolonged drought is a condition adverse to but not
prohibitive of forests, for the soil may store up sufficient moisture
to supply the tree through the dry season. Monsoon forests,
for example, survive a drought of four or five months with the
aid of the heavy rainfall which precedes it. Still longer drought
can be safely endured when adaptations exist for the economical
utilization of water supply. Various checks on transpiration,
CLASSIFICATION 59
such as reduced leaves,· thickened cuticle, woolly or hairy leaf
coverings serve this purpose; other trees store up water in
trunks or leaves, others dispense with leaves during the dry
season. But under such adverse conditions tree-growth seldom
attains the status of forest which is typically hygrophilous;.
woodland having these xerophilous characteristics Becomes in-
creasingly evident as the duration of the dry season increases;
such woodland, adapted to seasonal drought (tropophilous) is
known as scrub. The space between the trees and bushes opens
and becomes filled in with herbaceous plants, the trees become
rarer and rarer and a gradual transition occurs to grassland as
the drought grows longer ..
GRASSLAND CLIMATES. The life cycle of grass, on the other
hand, makes it indifferent tq prolonged drought, whose ill-effects
it escapes by seeding and dying down, to start life anew when
the rains return. Its rainfall requirements are small, provided
that the rain occurs over a period in which. temperatures are
sufficient for growth. The growing season is adapted to the
incidence of rainfall; on the steppes it is spring and early summer,
in the savanna summer, while in desert regions grasses and other
plants seize the opportunity presented by any adequate fall of
rain at any season to begin rapid growth. The summer rains
of continental interiors are well suited to grassland, but the
prolonged winter drought makes these areas unfavourable for
trees which, generally speaking, prefer maritime or sub-conti-
nental climates with their more regular moisture supply.
Clearly the length of the dry season rather than the actual
amount of rain is the chief factor in determining vegetation type;
forest, which is hygrophilous, passing into scrub, which is tropo-
philolls, and that into grassland and finally desert, which is
xerophilous, as the dry season becomes longer.
TYPES OF FOREST. The principal varieties of forest type
are:-
1., The broad-leaved evergreen.
2. The broad-leaved deciduous.
3. Th,e coniferous.
In the first of these growth is continuous throughout the
year and such forests can exist only where temperature and
rainfall conditions are favourable for continuous growth all the
year round; i.e. where the mean temperature never falls below
43° and where a well-distributed rainfall gives an unfailing water
supply.
In the second type there is a resting period, at the beginning
of which the tree sheds its leaves to grow a new leaf cover as
soon as conditions again become favourable. This resting peribd
60 CLIMATOLOGY
is imposed in hot climates by drought but in cold climates mainly
by temperature,! i.e. there is a cold season with temperatures
below 43° which checks active growth. The isotherm of 43°
(6° C.)· for the coldest month thus marks approximately the
boundary between the broad-leaved evergreen and the broad-
leaved deciduous forest. 2 .
The third type also undergoes a long period of enforced
inactivity imposed by the physiological drought (=cold) of
winter. But by the survival of the leaves through this resting
period the coniferous tree is ready to begin assimilating without
delay as soon as temperaturesJ)ecome favourable, whereas the
deciduous tree must spend some time in growing its assimilating
organs. Further, the coniferous type of fructification has this
a.dvantage, that it is pollinated one year and dispersed the next,
whereas the deciduous tree has to pass through the whole process
in one year. C0niferous forest is clearly able to exist with a
much shorter growing season than the deciduous type. Length

FIG. I9.-Southern Limit of the Taiga


Figures indicate the number of months at stations marked X with mean temperatures above 43·

of growing season, i.e. the number of months with temperatures


above the basal temperature for growth, is clearly the important
control of forest type. On the map (Fig. 19) are inserted figures
for numerous stations showing the number of months at each
with temperatures above 43° F., i.e. the duration of the growing
season. A strong coincidence is revealed between the southern
limit of the taiga and a six-month growing season. In other
words, the deciduous forest requires at least six months with
temperatures adequate for growth, and if this is not available
the deciduous forest gives way to the coniferous. The six-month
isopl~th runs from Bergen, through Oslo, Gefie, Helsingfors and

1 It is probable, also, that the darkness of winter in higher latitudes


has some influence on the adoption of the deciduous habit, for the photo-
synthetic function of the leaves must be fi\uch reduced in activity at this
season. .
2 See A. F. Schimper, Plant Geography, p. 4I7. Note that the figure 4Zo
has often been used as the zero of plant~growth, and is adopted by the Mete-
orological Office for the calculation of accumulated temperatures (Book of
Nprmals, Section II, 5.C). Nevertheless, Schimper's figure of 43° seems
to give better results in the case of tree-growth and will be adopted here.
CLASSIFICAnON 61
Leningrad to Kazan on the Volga. In North America, owing
to the extraneous factor of post-glacial dispersal, the deciduous
forest is not so clearly developed, but both the Pacific coniferous
forest and the mixed temperate forests of the Atlantic margin
contain deciduous species and correspond to the temperate
deciduous forests of western Europe. The sub-arctic forest,
which corresponds to the taiga, shows the same broad relation to
the six-month isopleth. It will be noticed, however, that there
are minor exceptions in the heart of North America, in the
extreme east of European Russia and in Manchuria (in fact, in
the more extreme continental climates); deciduous forest thrives
here with a growing season as short as five months. This is
doubtless to be accounted for by the combination during the
summer months of unusually high temperatures and adequate
rain. At Blagovestchensk, for example, although the tempera-
ture is above 43° from May to September only, the figures for
these five months are 50, 63, 7,0, 66, 54. This may be compared
with Riga where, though May to October have temperatures
above 43°, the figures for these six months are 51, 60, 64, 63,
55, 44· Adding together the excess over 43° for each month,
these might be stated as 88 and 79 month-degrees respectively
(d. p. 10).
Deciduous trees are essentially plants with two distinct and
strongly marked habits of life, hygrophilous and xerophilous,
and require a long transitional period in which to pass through
their rather long and elaborate stages of transformation. Such
prolonged spring and autumn are found only in marine climates,
the transition from winter to summer and back again in conti-
nental climates being sudden and complete, and consequently
better suited to the coniferous evergreen habit.
Further, the coniferous tree can exist on a much smaller
rainfall (about IS inches) than the deciduous. East of Kazan,
for example, the deciduous forest dies out altogether, not be-
cause of temperature, but because of deficient rainfall. East-
ward from this point, therefore, the southern boundary of the
taiga is set by aridity, approximately by the Is-inch isohyet
which divides it from the steppe. Kazan is therefore a meeting-
point of three vegetation types, the deciduous forest, the con- .
iferous forest and the steppe. There is, then, a double reason
for the replacement of deciduous by coniferous forest on passing
from western margin cool temperate climates towards the conti-
nental interiors, viz. the degeneration of the seasons of spring
and autumn and the increasing aridity.
Fig. 20 shows graphically the duration of the growing season.
at five stations in forest climates of different types. The upper
62 CLIMATOLOGY
curve is the curve of monthly temperature and the lower .,line
is the 43° line. The area enclosed between the two lines is
the growing season and is shaded except where, as at Athens,
absence of rain sets a limit to natural growth. Akassa illw;,-
trates the 'selva' type of evergreen forest with an abundant
margin of temperature in all months and an adequate rainfall,
well distributed. Athens illustrates the sclerophyllous wood-
land (evergreen) with !J.dequate temperatures in all months but

~III~IIIIIIIIIIIIIIIIIIIIIIIII ~.~
40
Akassa
40
Athens

70
60 ~ 60
50 ~IIIIIII"IIIIIIIIIII"~
40-:7-"'~UJ.IW.ww.:UJ.LJ.L.u.w.LJ.W.jJ~",- 50______~~~~wu~------
40
30 Paris 30
20
10
60 o
50
40--------~~~~~-----

30
20
10
o
fO
-20
-30
-40

FIG. zo.-Forest Type and Duration of Growing Season

a summer season of inactivity during drought. Paris illustrates


the typical temperate deciduous forest with a growing season
of eight months and tran:;;itional seasons with temperatures just
above the zero of growth. Harbin illustrates-the rather abnormal
deciduous forest of Manchuria with a growing season only about
five months long but with a large margin of temperature in the
midsummer months. Yakutsk illustrates the' taiga' type with
.a very short growing season and no transitional period; note
the steep rise of the temperature curve and its rapid passage
CLASSIFICATION
from temperatures below to temperatures above the basal tem-
perature for growth in May and back again in September.
This short discursion into plant geography brings to light
the vital part played by two very importan.t considera-
tions : -
1. The seasonal distribution of rainfall, and particularly the
length of the dry season, if any.
2. The seasonal distribution of temperature and particularly
the length of the cold season, if any.
The former is more important in low latitudes and will be
used to subdivide the hot climates, the latter is more important
in high latitudes and will be used to subdivide the temperate
and cold climates.
HOMOLOGOUS CLIMATES. A classification based on these two
considerations would have as its foundation the variation of
two 'climatic elements, but these result from the interplay be-
tween the two fundamental factors of the planetary circulation
and the distribution of land and sea, for the pressure and wind
belts, a,ctirtg on the continental blocks, will tend to produce
corresponding climatic types in corresponding latitudes and in
corresponding positions with relation to the coastline. The
trade winds, for example, blowing on-shore on eastern margins
in, say, lat. 20° S. give constant rains and high temperatures;
but blowing off-shore on western margins in the same latitude,
they give rise to deserts, whose temperatures are moderated
along the littoral by the upwelling of cold water which the winds
induce. Similar combinations of elements, in fact, will result
from similar causes, and it should be possible to devise a classifica-
tion based on a similarity of factors and having, therefore, a
similarity of elements, thus possessing the twofold advantage of
convenience and a foundation in first causes.

Subdivision of the Zones


Fig. 22, showing the duration of the temperature seasons,
and Fig. 2I, showing the seasonal distribution of rain, should
now be compared with maps showing the distribution of pressure
and the prevailing wind directions for January and July.
SUBDIVISION OF THE HOT ZONE. As defined by the mean
annual isotherm of 70° the hot climates, except where influenced
by altitude, never experience temperatures too low for active
plant growth; rainfall availability will therefore be the deciding
influence on vegetation type. Within the zone there are foqr
types of rainfall regime :-
CLIMATOLOGY
Rain at all seasons, with double maximum.
1.
Rain at all seasons, evenly distributed.
2.
3. Periodic rains with marked summer maximum.
4. Constant drought.
Turning to the pressure and wind map these are seen to
correspond to :- .,
I. Regions which are visited twice a year by the belt of
calms and rains as it follows the sun, but which are never far
enough from it to suffer drought. \
2. Regions where the trade winds are on-shore and therefore
bring relief rains, except when, locally and temporarily, the 10w-
pressure belt of calms brings convectional rains. .

I. Ram atall Seasons


a Evenly distributed
b. Double maximum
c. Summer.. ..
2.Penodicrains (mark.ed
dry season)
a. summerniaximum :...... .
b.Sprinq .. ..
c.Winter
~.
.. •
Constant drought
i
FIG. 2I.-Seasonal Distribution of Rain (Highly Generalized)

3. Interior regions visited by the doldrum belt of calms and


rains once a year, or, of twice, at intervals so close together as
to merge into one wet season. For the rest of the year they
are visited by the trade winds bringing a dry season. Beyond
the liinit of the swing of the rainfall belt these regions receive
summer rains from the trades, drawn further inland by a mon-
soonal effect.
4. Interior regions perennially under the influence of the
trade winds.
These correspond to four types 01 climate known as :-
1. Equatorial.
2. Tropical (Marine) or Trade Wind Coast.
3., Tropical (Continental).
4. Desert.
CLASSIFICATION
The first three types require further subdivision, for in
addition .to the normal type ~h_ey !pclude distinct varieties
charactenzed by a monsoonal regtme9
THE TEMPERATE ZONE. Subdivision of the hot zone has
been made on the basis of seasonal regime of rainfall, but in
the temperate zone the consideration of greater significance is
the seasonal regime of temperature. Within the tropical zone
the apparent movement of the sun results in only small changes
of temperature but in considerable changes of rainfall; the
seasons here are the' wet' and the' dry' seasons. In the extra-
tropical zones, on the other hand, seasonal changes of a planet-
ary nature produce changes of temperature which are often very
pronounced, and although there may also be important changes in

FIG. 22.-Temperature Zones

rainfall, temperature is the real determinant of season. Towards


the equa torward limit ofthe temperate zone the winter is only a ' cool
season', not severe enough to prevent plant growth completely,
but as we proceed polewards the winters become progressively
more severe .. Thus we may distinguish between :-(r) Temperate
climates without a cold season (Warm Temperate or Sub-tropical)
and (2) Temperate climates with a cold season (Cool Temperate),
separated by a temperature of 43° for the coldest month, a
value which, as we saw on p. 60, divides the evergreen from the
deciduous forest. The second type requires further subdivision
according to the duration of the cold season. We have seen
(p. 6r) that the restriction of the season of growth sways
the balance of favour from the deciduous to the coniferous
habit and that the transition occurs approximately when less
6G CLIMATOLOGY
"='
than six months become available for growth\ Further, the
increasing length of the cold season polewards restricts agri-
culture and gradually forces on man other means of living
such as fishing, hunting and stock-keeping. We may there-
fore subdivide the cool temperate zone into :-(1) Climate?
with a short cold season (1-5 months' below 43°), and (2)
Climates with a long cold season (6 or' more months below
43°)·
This last is by no means ' Temperate' although included in
Supan's temperate zone; it would be better desc:r:ibed as a
, cold climate', in which case Supan's Cold Caps may be promoted
to ' Arctic Climates '.
We have now recognized three subdivisions of Supan's tem-'
perate zone according to the length of the cold season, but each
of these embraces at least two different types according to dis-
tance from marine influence, which, in these latitudes, comes
chiefly from the West. This marine influence makes itself felt
in :-(~) A restriction of the annual and diurnal range of tem-
perature and (2) A modification of the seasonal regime of rainfall,
which in marine climates has a uniform distribution with a ten-
dencytoanautumnandwintermaximum(orographicandcyclonic),
but in continental climates has a spring and summer maximum'
(convectional) .
THE WARM TEMPERATE ZONE contains the following rainfall
regimes :-
1. Rain at all seasons, evenly distributed.
2. Periodic rains, spring and early summer maximum.
3. Periodic rains, winter maximum.
4. Constant drought.
Turning to the pressure and 'wind maps these are seen to
correspond to :-
1. A region with a restricted development on eastern margins
alternately under the influence of the trades which, coming off
the sea, bring rain, and the westerlies which, though coming
overland, bring a certain amount of rain from cyclonic
storms. The former are more effective as rain-bearers than
the latter and there is a tendency to a summer maximum,
which is exaggerated wherever there are monsoons (e.g.
China).
2. A region bordering the interior deserts on their poleward
margins and receiving cyclonic and convectional rains in spring
and early ·summer. Much of this is virtually desert, bordered
by steppe.
• 3. .A region with a resJricted development on western margins,
alternately under the influence of the calms of the high-pressure
CLASSIFICATION
belt (or the trades, both of which are desiccating agents) in
summer and the cyclonic westerlies in winter and therefore
enjoying winter rains and sununer drought.
4. A region in continental interiors, isola~ed from the sea
and receiving only scanty convectional rains. These correspond
to four types of climate known as :-
1. Eastern Margin Sub-tropical.
2. Steppe.
3. Western Margin Sub-tropical (Mediterranean).
4. Desert.
The eastern margin sub-tropical is sometime~ called' Chinese',
but since the essence of the Chinese climate is its monsoonal
character, the name' Chinese' is unfortunate as applied to the
normal facies. It is better to consider the Ihonsoonal variety
as a separate type of which South China may be used as the
example.
THE COOL TEMPERATE ZONE; and the COLD ZONE each contain
the following rainfall regimes :-
1. Rain at all seasons, evenly distributed.
2. Rain at all seasons, summer maximum.
As has been already pointed out these correspond to (I) Marine
and (2) Continental varieties.
THE ARCTIC CLIMATES (Supan's 'Cold Caps ') require no
subdivision, for the varieties differ little in the biological responses
they invoke, and in any case very little is known of the climatology
of the zone.
Two types of climate, although occurring in all of the fore-
going zones, are sufficiently distinct to receive separate recognition
and to be considered separately. These are :--
1. MOUNTA~N CLIMATES, i.e., those whose climatic and weather
conditions are determined chiefly by their relief and altit~de
and in which the influences of latitude and pn)ximity to sea are
strongly modified by these.
2. DESERT CLIMATES, i.e., those in whi(~h the dominant
characteristic is aridity. This character, with its concomitant
effects on temperature range, etc., gives a unity to deserts all
the world over in whatever zone they are found. The limit
may be roughly taken as the IO-inch isohyet, but this must not
be applied too rigidly, as IO inches in a cold climate is adequate,
while IS inches in a hot climate is inadequate for cultivation,
while IO inches nourish actual forest growth in the far north of
Canada and Siberia.
The desert climates may be subdivided into :-(a) Hot deserts,
with no cold season, i.e. no month below 43°; (b) Cold deserts,
with cold season, i.e. one or more months below 43°. Since the
68 CLIMATOLOGY
presence or absence of a cold season is the real distinction between
the two types, these may be conveniently divided by the isotherm
of 43° for the coldest month, the line already adopted as the
limit of the warm temperate climates. It is true that this line
has not the same significance here, since vegetation is scanty
or absent, but it is probable that deserts with such winter
temperatures will experience considerable night frosts in
winter.
We are now in a position to set down in tabular form the
classification of climates which will be followed. The figures
given are intended as approximate guides only and are not to
be accepted literally. It has been pointed out already that
the value of a single climatic element is seldom competent to
define the limits of a zone, while the combination of more
than one element, while giving a truer boundary, becomes
cumbersome.
A. Hot climates. Mean annual temperature more than
70°.
I. Equatorial. Double maximum of rain.
Im. Equatorial. Monsoon variety.
2. Tropical, marine. No real dry season.
2m. Tropical, marine. Monsoon variety.
3. Tropical, continental. Summer rain.
3m. Tropical, continental. Monsoon variety.
B. Warm temperate or sub-tropical. No cold season, i.e. po
month below 43°.
I. Western margin (Mediterranean). Winter rain.
2 .....Eastern margin. Uniform rain.
2m. Eastern margin. Monsoon variety. Summer
maximum of rain.
C. Cool temperate. With cold season, i.e. I-5 months below
43°·
I. Marine. Uniform. rain or winter maximum.
2. Continental. Summer maximum of rain.
2m. Continental. Monsoon variety. Strong summer
maximum. ,
D. Cold climates. Long cold season, i.e. 6 or more months
below 43°.
I. Marine. Uniform rain or winter maximum.
2. Continental. Summer maximum of rain.
2m. Continental. Monsoon variety. Strong summer
maximum.
E. Arctic climates. No warm season, i.e. I2 months below·
50°.
CLASSIFICATION 69

FIG. 23. -Distribution


,

Planetary
Controls

0;0

40

30

20

10

Wind Direction
.
Warm Currents . Types (cf. p. 68)
. t ClimatIc . r)
Letters indica e . all Regime (cf. FIg. 2
Cold Currents . . dica tes Ramf
Shading m . t
FIG. 24·-Schematic Distribution of Climatic Types on an Idea1 Contmen

5
70 CLIMATOLOGY
F.Desert climates. Less than IO inches of rain.
I. Hot deserts. No cold season, i.e. no months below
43°·
2. Cold deserts. With cold season, i.e. one or more
months below 43°.
G. Mountain climates.
The generalized distribution of these types is given in Fig. 23
and the schematic distribution on an ideal continent in Fig. 24·
CHAPTER V
EQUATORIAL CLIMATES
OT CLIMATES IN GENERAL. As defined by the mean

H annual isotherm of 70° the Hot Climates occupy a broad


. belt extending well beyond the tropics in each hemisphere
and occupying more than half the surface of the globe. Notwith-
standing the fact that three-quarters of this huge area is water,
such an enormous extension entitles these climates to full and
careful attention. Furthermore, this zone is one as yet very
under-developed and one wit:q. a great agricultural future whose
realization will be closely bound up with the study of its climate.
Unfortunately our knowledge of these climates is based on obser-
vations at comparatively few stations, huge areas still having less
than one station to 4,000 square miles, and much yet remains to
be done in the way of intensive study; but in this respect it is
fortunately one of the characteristics of the zone that climatic
phenomena possess a high degree of reliability and regularity,
and perhaps require less minute analysis than the more intricate
phenomena of temperate latitudes. The great extension of the
sea helps to preserve the uniformity of the zone, much of which
is necessarily marine or littoral, but two large continental
blocks, Africa and South America, project across the zone and
offer good opportunity for study of the continental type with,
the minimum modification by marine influences. In the
Indian and West Pacific oceans there is a monsoonal regime
which introduces peculiarities calling for special mention, and
these will be dealt with later. The symmetry of Africa about the
Equator, coupled with the fact that it is better known than
South America, makes it peculiarly suitable for illustration,
and for this reason examples will be taken chiefly from that
continent.
The hot climates lie within the sphere of the N .E. and S.E. trade
winds which meet in the equatorial belt of calms. The flanking
high pressures of the horse latitudes, the sources of the trade
winds, effectively exclude external influences from colder latitudes,
making for a stability of meteorological conditions which is a
marked characteristic of the zone. Both the seasonal and the
diurnal regime are controlled almost exclusively by the annual
71
72 CLIMATOLOGY
and diurnal march of the sun, so that it is sometimes said that
there is no weather, only climate.
:rhe annual migration of the sun between the tropics brings,
generally speaking, two maxima of insolation to the equatorial
zone and a single maximum to the tropical zone (see p. 34),
this single maximum being reached at about 12° N. and S. The
march of temperature, subject to certain modifications, follows
the march of the sun, and the rainfall belt, subject to a lag which
is often considerable, follows the sun also. Thus, broadly
speaking, double maxima of temperature and rainfall are charac-
teristic of the equatorial, single maxima of the tropical zone.
The Equatorial Climates
TEMPERATURE. Throughout this climatic type temperatur~s
are uniformly high and extremely constant, the excessive summer
temperatures which occur at the tropics are never recorded, nor
are their low winter temperatures, but the thermometer hovers
about the 80° mark throughout the year. Maximum tempera-
tures do not necessarily correspond to maxima of insolation, since.
cloudiness and rainfall considerably moderate temperature; and
in this zone, where the variation of insolation is never great,
their effects are particularly noticeable. Even at the tropics the
length of the day varies only between I~t and 13! hours and the
minimum midday elevation of the sun is 43°, while within the
limits of equatorial climates the variations of insolation intensity
and duration are considerably less even than this. To this the,
constant humidity and cloudiness add their stabilizing influence
and the annual range of temperature is consequently very small
indeed.
ANNUAL RANGE. The mean annual range near the Equator
is, on an average, about 5°, at Para (1° 27' S.) it is less than 3°,
at Akassa (4° 15' N.) it is 4°. Over the oceans it is less still and
many island stations have a mean annual range of less than a
degree (Jaluit, Marshall Is., 0·8°). Extreme temperatures show
the same constancy and the same small range as mean temperatures.
The thermometer rarely rises above roo o in the equatorial zone,
or falls as low as 60°, i.e. the extreme range seldom exceeds 40°.
DIURNAL RANGE. By contrast with the extremely small
annual range the diurnal range seems considerable, though it
may not be more than 15°. There is sometimes a nocturnal fall
of as much as 25° and the night temperatures (although, perhaps,
well above 60°) may appear uncomfortably cold, especially to a
native population inadequately provided with clothing and
prot~ction; wherefore it is often said that' night is the winter
of the trop!cs'. The native peoples are very sensitive to the
EQUATORIAL 73
slightest fall of temperature and will often light fires if the
temperature falls below 70°; Europeans, too, lose their power of
resistance to cold in these climates of uniform heat.
RELATION BETWEEN TEMPERATURE AND RAINFALL. At
Bolobo (see Fig. 25, a) the mean temperature varies only 2'2°
throughout the year and there are two maxima, reached at the
equinoxes. Rainfall and humidity, being remarkably uniformly
distributed over the year, do not affect temperature which follows
closely the vertical sun. But at Lagos (25, b) the curve of tem-
perature shows a marked drop in May and June when the rains
begin and actually reaches a minimum in August, although insola-
tion is then at a maximum. The drop in temperature is clearly
related to the cooling effect of the rains and of the clouds which
accompany them.

::·t::=::J~0.::c.::J20'
70 . . . . .- - _. . . . . .
~
70~--.
~

J F M A M J J A SON 0 J F M A M J J A SON 0
Bolobo Lagos
)
FIG. 2s.-Annual March of Temperature and Rainfall in Equatorial Climates

PRESSURE AND WINDS. Characteristically the equatorial belt


is a region of calms or very light winds; the evil reputation for
ill-healt4 and discomfort which it has earned is to be attributed
to the hot, moist, stagnant air. The reason for this light air
movement is not far to seek; on the monthly isothermal maps
the 70° isotherm never invades the tropical climates and the
belt with mean temperatures exceeding 70° is very wide. Within
this broad expanse of uniformly hot land small areas only exceed
goO and these are almost entirely the local products of the deserts.
Here, then, is the most uniform temperature distribution to be
found anywhere in the world; such feeble temperature gradients
beget a feeble pressure gradient and air movement is consequently
slight. The pressure gradient decreases towards the centre of the
zone and finally ceases in a region of calms, where escape is found
in upward ascent of an enormous body of air with resultant
expansion and cooling and consequent heavy precipitation.
Towards the tropics the barometric gradient is steeper and
vigorous air currents are directed equatorwards. These are the
trade winds, blowing with great regularity and fulfilling an import-
ant function by importing vast quantities of moisture from the.
warm oceans over which they have passed to be ultimately con-
74 CLIMATOLOGY
densed by the ascension at the belt of calms. The swing of the
wind belts (see p. 26) brings the beneficent influence of the trades
to different parts at different seasons, but wherever and whenever
they blow they are a welcome feature for the stimulation which
they give. Although they are warm winds their effect is pleas-
antly cooling, for they fan and ventilate the body, encouraging
evaporation from the skin. Exposure to the trade winds is a
desirable asset in a house in these latitudes and sites are chosen
with this object in view.
LAND AND SEA BREEZES. As the seasonal temperature
changes cause a seasonal swing of the wind belts, so the diurnal
changes bring about a diurnal change of wind direction, the land
and sea breezes (see p. 43). The mechanism, however, is not the
same in the two cases, for while the seasonal winds depend on
seasonal variations of insolation (Le. the solar regime), the diurnal
winds depend on the differences in behaviour of land and sea
towards insolation (Le. on physical influences), It is especially
in the belt of calms, where the pressure gradient is small, that
the diurnal changes have the best opportunity of producing
effects, and where, consequently, land and sea breezes are most
reliable. In the trade-wind belt the diurnal changes only modify
the prevailing winds, in the doldrum belt they produce an actual
reversal of direction. Although restricted to a narrow coastal
zone, the sea breeze has an importance out of proportion to its
limited distribution, for the majority of European settlements in
this zone are coastal trading settlements, and the residential
districts in these are usually those best favoured by the sea breeze;
whose daily arrival is impatiently awaited for its wholesome and
cooling influence.
RAINFALL AND SEASONS. In the absence of any marked
.temperature changes in this zone, rainfall becomes the deter-
minant of season. The principal rains are those of the convec-
tional type and reach a maximum shortly after the passage of the
zenithal sun, about April and November on the Equator. But
this simple scheme is subject to considerable modification for loca}
reasons. It frequently happens that one of the maxima is more
pronounced than the other (the' greater rains' and the' smalls'
of the Gold Coast) and often one is suppressed (see Para). In most
cases the spring maximum following the northward passage of
the sun is considerably greater than the autumn maximum follow-
ing its southward return. Entebbe has 25 inches during the three
months of the first maximum (March, April, May) and only 13
inches in the second (Oct., Nov., Dec.) ; Lagos has. 39 inches in
May, June and July and only 16 inches in September, October and
November. The graphs for Djole and Mwanza (Fig. 26, p. 76)
EQUATORIAL 75
provide more examples. Although not by any means universal,
this marked asymm~try of the curve seems to indicate a real
difference in the influences at work in the two rainy seasons.
It is very clearly marked in Africa where the contrasts in relative
proportions of land and sea in the two hemisphere~ are most pro-
nounced, and it appears to be the greater area of water in the
soutnern hemisp.i1:ere wh.ich causes the effect, since during its
passage over the southern hemisphere the sun has evaporated a
greater quantity of \\rater which is available for condensation in
the zone of ascension.
The equatorial rainfall type does not extend far on either side
of the Equator; witnin-a few degrees there enter two dry seasons
following the solstic~s and becoming gradually more and more
pronounced. At about 5° N. and S. these dry seasons are more
or less equal in duration and intensity, but polewards one becomes
gradually longer and the other gradually shorter . .In the coast
zone of British Guiana (7° ~.), for example, four seasons are
recognized :_ I

1. The long wet ~eason. Mid-April to mid-August.


2. The long dry Season. Mid-August to mid-November.
3. The short wet season. Mid-November to end of January.
4. ,-The short dry season. End of January to mid-April.
And of these the long dry season alone can be relied upon to
remain true to tYPe. Further polewards still the short dry
season becomes only a break in the rains and is eventually
eliminated; the lon~ dry season has assumed the proportions of
a drought and the equatorial climate has passed into the tropical.
Fig. 26 reproduc~s de Martanne's diagram showing the ideal
relationship of the rainy seasons to the zenithal sun. The region
with no dry season i; seen to extend only about 2° N. and S. of
the Equator; from_this point to nearly ISO N. and S. there should
be two dry seasons, aJter which the short dry season dies away and
there should be a single dry season and a single wet season in
which there may be lwo maxima of rain.
On either side of this ideal arrangement are placed some
actual rainfall graph~ for certain stations, African stations on the
left, South American. on the right, the base of each graph being
placed approximately on its own latitude. A warning should be
issued against a too literal acceptance of these graphs since they
are not strictly comIJarable, differing in altitude and in distance
from the sea; local factors, too, are often potent influences in
modifying the plan6,tary regime. They do, however, serve to
show cert~ importilnt departures from the theoreticaL
The most obviou~ of these is the entire displacement of ea~h
type some degrees te) the north of its proper latitude. The best
CLIMATOLOGY
examples of the equatorial regime in Africa are given by,Mobaye
and Bismarckburg, roughly 4 0 and8°N. oftheEquatorrespectively;
and by Georgetown (6 0 N.) in South America. Djole, almost on
the Equator, has a marked dry season in July and August, a
regime:such as one would expect in about 8 0 S. Bogota (50 N.) has
the double maximum of the equatorial type, but Quito on the
JFMAMJJASOND JFMAMJJASON~ JFMAMJJASOND
1,
Knartoum
Goree
I
5 long
Dry
, " A,'"
","
long 15
Dry
eason ' , SeasO!

-- -
---- ': Gambia

Bismarkburq
Io

:':s!i:<,:-::
10
Cartagena

--- -
_, ... J
~rqetown __

....
Mobaye
5 "

.N_'::'v'::i::::
" \ '5 Bogota

:::'J:,::': ','::: ", :,,~: :':'


Djolo
o 0
-
--
::+ ...

--
• oN ......
Para
.1!IIIIIIl! r" '0 " ,',' ,',: ~-: ,j
Mwanza c
.c
,I:
' 0

-.- --
--
o "' . . (quitos

--
Banana

Loanda
--- 5 ~ ~',
<f)

i:-', '
c'
.

... ..II1II
--
Sta, Anna de Sobradinho
1o ' 10

----
JFMAMJJASOND
Ft.Johnson

Leaui
.-
_1
'5: ',',

JFMAMJJASOND
long
Dry
Season
. ,,', 5
_Cochabamba
JFMAMJJASOND --
,

FIG. z6o-March of Rainfall in Equatorial and Tropical Climates

Equator, has a southern hemisphere regime. There is a double rain-


fall maximum at Mwanza (2°3r'S.), but by Tabora (sOr'S.) this is
scarcely recognizable. However it survives further south along the
coasts, e.g. at Banana (6 0 So) on the west coast and at Lindi (roO So)
on the east coast. By Loanda (9 0 So) the dry season is already
6 months long, but Gambia (nearly"rr ON.) has only 4 months
dry. The typical equatorial climate, in fact, is found, not along
the Equator, but between 2° and 8° N., while the full swing of
the rainfall belt is approximately between r8° N. and roo S.
This striking displacement of the equatorial climate springs
£rom 'the contrasts which the two hemispheres present in the
EQUATORIAL 77
distribution of land and sea. In these low latitudes, where
temperatures are constantly high, the land is nearly always
hotter than the sea, and since there is more land north of the
Equator in this zone, the northern hemisphere is hotter than the
southern. The S.E. trades are cooled in passing over seas, while
the N.E. trades, coming chiefly over land, are hotter and therefore
ascend sooner; the point of meeting of the two currents and the
upward escape is thus north of the Equator. A further factor,
though one of little importance, is the longer sojourn of the sun
in the northern hemisphere (in aphelion) than in the southern
(in perihelion) by about 8 days. This displacement of the heat
equator (see Fig. 17) has other important results which will be
dealt with later.
NATURE OF EQUATORIAL RAIN. Rainfall throughout the
equatorial zone is heavy and reliable. 50 inches to 80 inches is
usual, 100 inches is not rare and 200 inches is exceeded in places.
Grey skies are the rule during the rainy seasons; the morning
may dawn clear and sunny, except for morning mists which
quickly melt away, but as the day wears on dark clouds roll up
and violent storms occur, usually with an accompanime!1t of
lightning and thunder. The daily incidence of rain is strikingly
regular at any given place, often so regular that appointments
are ma.de for' after the rain' much as one might make arrange-
ments for ' after tea'. The time of this daily maximum varies
from place to place according to local conditions, but it is nearly
always between mi<;lday and midnight and usually about 3 or
4 p.m., i.e. following the greatest heat and convectional action.
In spite of the violence of these storms and the strong convec-
tiOl} which characterizes them, there is no violent vortical motion
such as occurs in convectional storms of higher latitudes, since,
by reason of proximity to the Equator, the deflection by the
earth's rotation is extremely feeble.
HUMIDITY AND CLOUDINESS are greatest during the wet season
and ierve to make the heat more trying; for although day tem-
peratures in the equatorial climates seldom exceed 100° and are
usually between 80° and goO, the conditions are the most oppres-
sive, unhealthy and enervating that can be experienced. Where
humidity is highest there is little relief even at night, the body
perspires freely, but there is little evaporation to cool it, and in the
sultry, hot-house atmosphere all strength of body and will seem
to ebb away. On the coast the arrival of the sea breeze brings
some relief, but when it dies away the sweltering heat seems to
settle all the heavier.
HIGH ALTITUDE EQUATORIAL CLIMATES. The most notice-.
able iDfiuence of altitude is a lowering of temperature, and for
CLIMATOLOGY
this reason, if for no other, the climates of high altitudes in equa-
torial zones would merit separate treatment. But there are also
changes in many other climatic elements which confirm the type
as quite distinct. At altitudes above 10,000 feet even on the
Equator the conditions are those of mountain climates and as
such will be treated in a later chapter, but at altitudes of 5,000
to 10,000 feet the climatic conditions are more truly described as
'modified equatorial'. The temperatures at these altitudes
are those of the temperate zone, it is the' tierra fria' of Latin
America (see p. 101).
Since each 300 or 400 feet makes a difference of about 1° in
the temperature, highlands offer refuges from the stifling heat of
the lowlands. Parts of South America are particularly fortunate
in the close proximity of highland to coast, for while commercial
activity is concentrated in the hot coastal zone, administration
can be centred quite close at hand in a climate with great
advantages of comfort and health. Caracas (4,000 feet) is only 6
miles distant (25 miles by rail) from its port, La Guaira, but
has a mean temperature nearly 10° lower, while the difference
of humidity makes the contrast seem even greater.
The mean annual range of temperature is typically small,
less even than at low levels, Quito varies only '7° between the
hottest and coldest months, Bogota 1·6°. The diurnal range,
however, is considerable (30° or 40°) and frost at night is a
common occurrence at altitudes of 8,000 and 10,000 feet, while
precipitation frequently takes the form of sleet and snow. But
the air temperatures give an inadequate idea of the physiological
sensations; for the clear and rarefied air promotes rapid radiation
at night and interferes little with the intense insolation by day.
The contrast between conditions in sunshine and shadow is a
revelation. These contrasts are exaggerated at still higher levels
and will be dealt with more fully under mountain climates.
The rainfall regime is true to type and the two equinoctial
maxima are usually clearly recognizable, but above the zone of
maximum precipitation the amount of rain is substantially de-
creas.ed; Quito has 42 inches compared with over 80 inches in
the lowlands which lie to east and west. There is the same
regular diurnal incidence of rainfall and a regular diurnal variation
of cloudiness, for the rainfall is of the same type (convectional)
and obeys the same controls. Mornings are usually fine and
clear but clouds gather in the late morning and early afternoon for
the regular afternoon storm, often a thunderstorm with hail.
VEGETATION AND CULTIVATION. The constant heat and
humidity of the equatorial climate supports a characteristic
vegetation type, the' equatorial rain forest' or ' selvas " with its
EQUATORIAL 79
astounding profusion and vigour of growth, especially of foliage.
Tall tree-tops form an irregular canopy beneath which grows a
layer of shorter trees and finally a tangle of creeper and under-
growth so dense as to be almost impenetrable, while parasitic and
epiphytic plants, often with gorgeous flowers, add to the display
of vegetational extravagance. A marked absence of seasonal
rhythm results from the monotony of the climate, flowering,
fruiting, seeding, growing and dying occurring side by side.
There is no resting period, no check to the rapid growth; no
clear annual rings appear in the wood of the trees.
There is little rhythm, either, in agricultural practice, but the
dry season (or the periods of minimum rainfall) become the
harvest seasons for most crops. Rubber is gathered in the dry
season, which in the northern half of the Amazon basin in August
to February and in the southern half is May to October.
Economically the wealth of vegetation is somewhat misleading,
for the profusion is chiefly of vegetative parts, but palm-oil and
rubber are important bases of industry and wealth. There is,
moreover, an unfortunate tendency to a sporadic occurrence of
species which much reduces their value; rubber trees, for example,
are scattered through the Amazon forests in such a manner that
the collection of wild rubber, for long Brazil's chief source of wealth,
was unable to compete with the concentrated cultivation of the
plantation product. Mahogany, greenheart and other timber
trees suffer from the same disadvantage.
The exuberance of growth of the forest makes clearance for
cultivation difficult to effect and expensive to maintain, especially
since there is no real dry season when the aid of fire can be
enlisted. Plantation cultivation, however, is rapidly spreading
over the zone and replacing the old wasteful exploitation. This is
chiefly a response to the demand from temperate countries for
tropical produce both for food and for raw material, a compara-
tively new departure which is accelerating development, and with
which is associated the growth of settlements, often of consider-
able size, for trading purposes. The difficulty of penetration
retards development; roads and railways are difficult and expen-
sive in construction and upkeep (the 225 miles of the Madeira-
Marmon~ railway cost over £5,000,000 in cash and incalculable"
loss of health and life). Penetration is chiefly by rivers, which
provide the only clear paths, and settlements are scattered along
their banks.
VEGETATION AND ALTITUDE. At higher levels the vegetation
type undergoes a steady change, but so much depends on aspect
in relation to sun and winds, on soil and other variables that
it is difficult to generalize. Up the Andean slopes the dense
80 CLIMATOLOGY
forests of the Amazon give way, at about 5,000 feet, to forest
of a sub-tropical nature with tree ferns and cinchona; this
degenerates into bush and scrub with bamb09 and fuchsia at,
about 8,000 feet and eventually into grassland at about ro,ooo
feet. In East Africa, where the rainfall is less, tropical grassland
begins at 3,000·feet, but is followed by sub-tropical forest above
and this by temperate grasslands; but in Kamerun, reached by
saturated air currents from the south-west, dense forest extends
up to 6,000 feet.

Regional Types
THE AMAZON BASIN, open to the east, and offering free acc~ss
to the trade winds, lias the widest expanse of uniformly heavy.
rainfall to be found anywhere in the wbrld, Para has 87 inches,
Manaos 66 inches, and more than 70 inches probably falls over
an area of more than 2 million square miles; but recording stations,'
are unfortunately few. 1 Westwards, as the Andes are approached,
the rainfall is heavier still since the easterly winds are forced to
ascend; Iquitos has over roo inches annually. This terrific pre-
cipitation (nearly 3,000 cubic miles a year) ensures an abundant
supply of water, and the discharge of the Amazon far surpasses
that of any other river. The double rainfall maximum which one.
expects to find is not recognizable and, as far as can be ascertained
from the scanty records available, a summer rainfall is general.
Most of the basin lies south of the Equator and the important
southern tributaries have a southern hemisphere regime with a'
marked dry season in August and September when the level of
the main river falls. In March and April these southern tribu,
taries flood and the level of the Amazon may rise as much as 40
feet, converting the whole basin into a vast swamp.
Separated from the Amazon basin by the Andean divide is·
another area of heavy rain, the west coast of Colombia, washed .
by the warm waters of the equatorial counter-current, which.
receives more than roo inches annually, and, in places, as much as
300 inches occurs; Buenaventura has 28r inches.
THE CONGO BASIN. Much of the Congo basin h. s an equatorial
climate, but the rainfall is considerably less than in the Amazon
basin, for the entry of the easterly winds into the Congo basin is
obstructed by the plateau of East Africa and they have lost"
much of their moisture before their ascent in the zone of calms.
However, 50 inches occur over a very large area (d. 70 inches or
1 While heavy rain is general, there appear to be " blind spots" where
the precipitation is remarkably small; Fonteboa (lat. 3° S.) records only
12·7 inches as the mean of observations kept from 1910 to 1919.
EQUATORIAL 8I
80 inches in the Amazon) and provides an adequate supply for the
river's discharge, which is the second largest in the world (Congo
average annual discharge 4I9 cubic miles, Amazon 528 cubic miles).
Thanks to the symmetrical arrangement of the Congo basin on
either side of the meteorologi<;al Equator, the river floods twice a
year, in May and December, these being the times when the
northern and southern tributaries respectively have their maxi-
mum precipitations. At no time does the level fall so far as to
cause serious inconvenience to navigation.
THE GUINEA LANDS, also, although between 5° and 10° N.,
have a climate which is virtually equatorial, but in which a certain
monsoonal influence is an essential constituent. The key to this
climatic province lies in the existence of the great land mass of
West Africa, always subjected to intense insolation. Even in
winter the southern Sahara and the Guinea lands are hotter than
the sea, consequently the doldrum belt never moves south of the
Equator but swings between the coast in January and 10° or
15° N. in July. In July the S.E. trades are drawn across the
Equator as a S.W. monsoon, and in January the belt of calms
lies over the coast, though even then the prevailing wind is
south-west. From June to November 75 per cent. of winds on
the Gold Coast are from the south-west, and from -December to
May 64 per cent. Thus the coastal lands never experience the
benefits of the N .E. trade wind to which they would be entitled
by virtue of their latitude, for these hardly ever penetrate south
of Freetown. From here northwards the coast is moderately
healthy, but" the south coast has earned the reputation of the
'White man's grave'.
Both winter and summer conditions bring rain to the coast
lands, Akassa (144 inches) and Lagos (72 inches) have the
double maximum, but in Liberia and Sierra Leone, w:here the coast,
backed by the Futa Jalon Highlands, is at right angles to the S.W.
wind, the regime is distinctly monsoonal.· Freetown (175 inches)
and Konakry (170 inches) have single maxima in July or August,
corresponding to the greatest vigour of the S.W. monsoon. The
most important rain here is, in fact, orographic rather than
convectional, and the same is true in Kamerun where again the
coast is at right angles to the S.W. monsoon and is backed by
high land. Kamerun Peak rises above 13,000 feet and on its
western (Le. windward) slope more than 400 inches of rain faU
in a year. -
There is a narrow strip of littoral east of Cape Three Points
where the rainfall suffers a startling diminution. In the midst of
a region with 80 inches or more this area has less than 40 inches,
and in places as little as 20 inches. The rainfall along the coast
82 CLIMATOLOGY
from west to east is as follows :-Axim 81 inches, Sekondi 40
inches, Cape Coast Castle 33 inches, Accra 27 inches, Christians-
borg 21 inches, K witt a 22 inches. But these peculiarly low
figures hold only for a narrow strip; Aburi, 30 miles behind
Accra, has 45 inches, and Kumasi, IOO miles further inland,
has 58 inches. The cause of this phenomenon is obscure, but
it is probably to be explained by the pull of the Guinea current
which causes the upwelling of cold water in lee of Cape Three
Points, thus decreasing the moisture capacity of the winds pass-
ing on-shore over it. At the same time the change in direction
of the coast makes the winds more oblique to the coastline.
The lower humidity reacts on temperature and Kwitta, becaus~
drier, is 5° hotter than Axim during the heavy rains.
The equatorial climates do not extend completely across
Africa, t)l:e eastern plateau having a continental type of climate
much modified by two influences, viz. its altitude and the prox-
imityof the Asiatic and Abyssinian monsoons. Much !_:educed
though the rainfall is, it still retains the typical double maximum
over much of this area. The resultant vegetation is grassland
(altitude savanna) and it forms a natural continuation of the
tropical savannas with which it is better described (see p. I02).
There is a return to equatorial conditions in the coastal
lowlands behind Pemba and Zanzibar, where, however, the effect
of the monsoons is clearly felt (d. p. 129). Zanzibar has a double
rainfall maximum, but with monsoon winds; Tanga shows a
tendency to a third maximum of rain in July between the normal
equatorial maxima which occur in April and November.

Monsoonal Variety
Situated between the two great monsoon foci of Asia and
Australia, the East Indian islands present a variety of the equa':
torial climate into which modifications are introduced by the
seasonal wind changes.
PRESSURE AND WINDS. In January (Fig. 27, a) there is a.
difference of pressure of 3 mm. between roo N. and IO o S. and the
so-called' West Monsoon' is at its height. By April (Fig. 27, b),
the isobars are more or less symmetrically arranged about a low
pressure trough just south of the Equator in South Celebes and
north New Guinea; light breezes and calms are now the rule.
By July (Fig. 27, c) the pressure difference is about 3 mm. in the
opposite direction and the so-called 'East Monsoon' is at its
height. By October or November (Fig. 27, d) there is a return to
the symmetrical arrangement about a low-pressure trough with
light and variable winds focusing on Celebes and New Guinea.
EQUATORIAL
Such, in broad outline, is the scheme of winds-two periods of
steady air flow (the monsoons) alternating with two periods of
indefinite circulation at the changes-over. But in actual practice
there is no such simplicity, for In the first place the change-over
of the monsoons is not simultaneous throughout the archipelago,
but follows the sun; for example, the south-east monsoon prevails
south of the Equator in April, but is not established north of it
until May. In the second place, so complex is the intermingling
of land and water and so varied is the relief that local influences
profoundly modify the circulation and even in places render it
completely unrecognizable.
The velocity of the monsoon is greatest and its direction most
constant in the west, over the open Indian Ocean, where friction
and interference are least; but even at its height, land and sea

Fm. 27.-Pressure Winds and Rainfall in the Malay Archipelago

breezes and mountain and valley winds are nearly always able
to make themselves felt in spite of the prevailing wind. The
direction and force of the wind depends largely on the direction
and strength of the relief; where the air flow is impeded by
mountainous islands set across its path (e.g. Java during the east
monsoon), the wind is 'stowed' and concentrated in gaps and
on the high plateaux where the velocity is greatly increased.
Warmed by descent on the leeward side, these winds have a
foehn-like nature, their dry heat doing considerable damage to
the more sensitive crops, especially tobacco. Such a wind is the
'Koembang' of Java, and the' Bohorok' which visits the plains
of Deli, Sumatra, during the east monsoon.
On the whole, winds in the East Indian Archipelago, though
extremely complicated, are regular and reliable at any given
locality, even to their diurnal variations. Once the local pecu:
CLIMATOLOGY
liarities, numerous though they are, are thoroughly mastered, the
general absence of strong winds favours shipping on the seas and
, agriculture on the islands. High wind velocities are very rare
and only the most northerly islands are visited by typhoons.
There are, however, local disturbances peculiar to limited areas,
such as the 'Sumatras' of the Malacca Straits, sudden squalls
with violent thunder and lightning and heavy rain which occur,
always at night, during the south-west monsoon (the East Monsoon
of the Dutch East Indies).
RAINFALL. Several factors combine to make the East Indian'
islands one of the largest areas of uniformly heavy rain in the.
world. They are most of them mountainous, they rise from a very
warm sea, they lie in the equatorial belt of ascending air currents
and they stand in the path of a double monsoon. Less than
40 inches is an extreme rarity, 80 inches is more the rule and ISO
inches is by no means unusual, while 268 inches is recorded from
Craggan in the mountains of Java. The air is so near the satura-
tion-point that practically any ascentional movement gives rise to·
precipitation at once. The ascent is brought about in one of two
ways:-
1. Local heating effects which give rise to local showers and
thunderstorms, especially during the calms which accompany the
changes of the monsoon. At this time of the year there is gener-
ally to be recognized a regular diurnal migration of the belt of
clouds. In the morning it lies in the low valleys where the tem-
perature is inverted, but the mists quickly vanish before the early
sun and the mornings are clear and fine. The moisture is carried
upwards by the ascending currents of daytime and condenses as
cumulus cloud with a flat base coinciding with the level of the
saturation temperature. Rising higher, the cumulus becomes.
cumulo-nimbus and thunder showers follow in the afternoon;
Buitenzorg is the most thundery place on earth. '
2. Forced ascent giving orographic rain, especially characteris-
tic of the monsoon months, distributed, naturally, on exposed
slopes and increasing with altitude up to about 3,000 feet. These
rains are more continuous than the other type, they give the
greatest daily amounts and account for most of the floods. They
do not, of course, possess the same regularity of diurnal incidence.
SEASONAL DISTRIBUTION. Thus we have a combination of
equatorial and monsoon regimes. The former would tend to give
a typical double maximum, but such is not often recognizable
owing to interference by'the latter. The double maximum may,
however, be identified fairly clearly at several stations in the
_immediate vicinity of the Equator, e.g. Pontianak and Padang.
At southerly and easterly stations in this zone there is a tendency
EQUATORIAL
for the maximum at the earlier change of the monsoon (April)
to exceed that at the later change (October), e.g. Manokwari.(see
Fig. 28), Tondano and Bandoeng, while at northerly and westerly
stations the opposite obtains, e.g. Penang (see Fig. 28), Kuta
RaJa, Pontianak. The autumn' maximum, in fact, tends to pre-
dominate over the spring maximum in each hemisphere. Any
departures from this formula, and they are the rule rather than
the exception, are accountable by exposure to one or other mon-
soons, and numerous interesting examples of compound rainfall
regimes occur. Still further from the Equator a single maximum
becomes the rule, a season of distinctly lower rainfall appears (e.g.
Batavia), becoming progressively more and more marked until there
is an actmil dry season, e.g. Koepang, Timor (see Fig. 28). It is
to be noted, however, that the dry season is July to September (Aus-
tralian type) even north of the Equator (e.g. Menado, Celebes), the
Asiatic type with the dry season from January to March being
practically limited to the north of Sumatra and the Chinese Sea.
In other words; the east monsoon is the dry season, the rains
coming more particularly from the west monsoon. This is easily
understood when it is realized that these islands are much nearer
to the Australian than the Asiatic monsoon centre; the east
monsoon is thus a descending wind, passing overland from the
Australian high pressure, and is therefore doubly dry, while the
west monsoon is an ascending wind passing almost entirely over a
warm sea towards the Australian low and is fully saturated. The
importance of a passage over water as a source of moisture is
well illustrated by a comparison of the figures of Koepang, Timor,
whose monsoon travels over Northern Territory, with Toeal in
the Kei Islands, whose monsoon comes over the Gulf of Carpen-
taria; the former has five months with less than I inch of rain,
the latter none.
Summarizing, we may say that there are four rainfall regimes
represented, the last two of which are more properly tropical :-
1. Double maximum with the earlier one predominating (in
the south and east).
2. Double maximum with the later one preponderating (in
the north and west).
3. Single maximum with dry season July to September
(Australian type).
4. Single maximum with dry season January to March
(Asiatic type).
TEMPERATURE. As would be expected in a thoroughly marine'
equatorial climate the sea-level temperatures are monotonously
high (about 80 0 throughout the year) and humidity is constantly
excessive. The highest monthly averages of wet-bulb tempera.:
6
86 CLIMATOLOGY
ture at Batavia are in the neighbourhood of 76° and the mean
maximum about 79°. These do not appear excessive when com-
pared with,. say, Hanoi, whose June mean is 80° and which reaches
83° in the heat of the day, but there is no relief throughout the
year. The conditions at Batavia, then, though always enervating
and unpleasant, are never unbearable, and moreover they can
be avoided by retreat to near-by hill stations. The temperature
decreases steadily with altitude and it is the hill station which
has made the energetic Dutch colonization possible. Night
frosts are experienced on the high plateaux, especially in the dry
season when nocturnal radiation is excessive and inversions occur,
but not on the mountain tops where air movement is freer. The
annual variations of temperature are so small as to be of more
82
80
78

J JFMAMJJASOND
IloilO (Philippines) 2. Penang ( Straits Settlements)

"~
82

80l _____------------~
80
781-
7 8 :

3 Manokwari(NewGuinea) 4 Koepanq (TimOr)

FIG. 28.-Temperature and Rainfall Regimes in the East Indies. For explana-
tion, see text

meteorological interest than geographical importance, but three


main types may be recognized corresponding to : -
1. A northern hemisphere regime with a maximum in the
northern summer, e.g. Iloilo, Philippines (see Fig. 28). This is
actually tropical, not equatorial.
2. A central equatorial regime with two maxima approxi-
mate~y corresponding to the zenithal sun, when the heat com-
bined with the high humidity and calm air is most oppressive,
e.g. Batavia. .
3. A southern hemisphere regime with maxima in the southern
summer, e.g. Koepang (see Fig. 28) and Port Moresby. This also
is actually a tropical climate.
As was the case with the rainfall in the region of the double
maximum, one maximum usually preponderates over the other
and we may subdivide (2) into : -
(a) An area in the north and west in which the first (April
EQUATORIAL
and May) maximum is greater, e.g: Penang (see Fig. 28), Fort de
Kock and Padang.
(b) An area in the south and east in which the second (Sep-
tember and October) maximuJ? is greater, e.g. Manokwari (see
Fig. 28), Soerabaja and Bandoeng.
A comparison of the temperature and rainfall curves in Fig.
3 1 emphasizes the cooling effect of rain, the greater maximum of
temperature coinciding with the lesser maximum of rain, and vice
versa. Close examination shows, too, that in all cases the maxima
do not occur at the time of greatest insolation but at the end of
the dry season when the cloudless sky permits free inward radia-
tion, but when the humidity, daily increasing as the rains approach,
appreciably impedes loss by outward radiation. The greatest
heat is recorded when the rains are late and their cooling influence
is delayed.
For notes on reading, see end 0,£ next chapter.
8R CLIMATOLOGY
TEMPER,ATURE
: -
Station. Lat. Long. Alt. (ft.) J F M A M J J A S 0 N
-- c-- - -
- - - - - - - -
OCEAN ISLAND . 1° S. 170° E. 177 81 81 81 81 81 81 81 81 81 81 81
MALDEN ISLAND 4° S. 155° W. 21 82 82 82 82 82 82 82 82 83' 82 82
SEYCHELLES 5° S. 55° E. 15 80 80 81 82 81 80 78 78 79 79 80
JALUIT 6° N. 170° E. Coast 81 81 81 80 89 80 80 80 80 81 81
AKASSA. .' 4° N. 6° E. 20 78 79 80 80 79 77 76 76 7 6 77 "7 8
CAPE COAST
CASTLE } 5° N. lOW. Coast 80 79 81 8r 80 78 77 76 77 79 80

DUALA 4° N. roo E. 26 80 80 So 79 79 77 75 75 76 76 78
YAUNDE 4° N. 12° E. 2,461 74 74 74 72 72 71 70 7 1 7 1 7 1 72
LULU'ABURG 6° S. 22° E. 2,034 76 76 76 77 77 76 77 76 76 76 77
NEW ANTWERP 2° N. 19° E. 1,230 79 80 79 78 79 78 77 76 77 77 78
MOMBASA 4° S. 40° E. 50 80 80 S2 81 78 77 75 76 77 78 79
ZANZIBAR 6° S. 39° E. 56 83 83 83 81 79 78 77 77 78 79 81
PARA. 1° S. 49° W. 42 78 77 78 78 79 79 78 79 79 79 80
MANA.OS. 3° S. 60° W. 144 80 80 80 80 80 80 81 82 83 83 82
IQUITOS . 4° S. 73° W. 328 78 7 8 76 77 76 74 74 76 76 77 78
GEORGETOWN 7° N. 58 °W. 6 79 79 80 81 81 80 81 82 83 83 82
NAIROBI 1° S. 37° E. 5.495 64 65 66 65 63 61 59 60 63 66 64
ENTEBBE 0° 32oE . 3,84 2 71 71 71 70 70 69 69 69 69 70 70
QUITO 0° 79°W. 9.35 0 55 55 55 55 55 55 55 55 55 55 54
BOGOTA. 5° N. 74°W. 8.730 58 58 59 59 59 58 57 57 57 58 58
COLOMBO 7° N. 80° E. 24 80 80 82 83' 83 82 81 81 81 81 80
SINGAPORE. 1° N. 104° E. 10 80 80 81 82 82 81 81 81 81 81 81
BATAViA 6°
, S. 107° E. 23 78 78 79 80 80 79 79 79 80 80 79
FOR'f DE KOCK. 0° 101° E. 3,5 00 69 69 70 70 71 70 69 69 69 70 69
PADANG 1° S. 100 0 E. 3 79 80 79 80 80 79 79 79
79 79 7f.!
BANPOENG . 7° S. 108° E. 2,3 66 72 72 73 72
72 72 72 72 72 72 72
SOERABAJA. 7° S. 1I3° E. 16 79 79 79 80 80 79 79 79 80 80 89
KUTA RAJA 6° N. 95 E. Coast -
0
-
- - - - - - - - -
P,ONTIANAK 0° 109° E. 10 78 79 79 79 80 80 80 79 79 79 78
MENADO 1° N 125° E. 28 77 77 78 78 79 79 79 80 So 79 79
EQUATORIAL
RAINFALL

D Yr, Ra, J F M A M J J A S 0 N D Total,


- - - -- ---- -- -- -- -- -- -- -- -- -- - - -
8! 81 0'3 II '5 8'9 8,6 8,! 5'6 5'1 6,8 3'9 5'2 5,6 5'7 8'9 83'9
82 82 1'1 4'1 2'0 4'7 4'2 1'9 2'0 4,6 0'9 1'0 1,6 0,8 0,8 28,6

80 80 3,8 16'9 12'5 9'1 7'0 6'0 4'5 2,6 2'3 5'7 5'0 9'4 13'5 94'5
81 81 0,8 I I '5 II'9 17'9 14'0 19,8 15'8 15'5 13'8 13'6 II'4 14'0 17'3 17 6 '5
2,6 6'5 10'0 8,6 17'0 18,6 10'1 9'3 19'3 24'7 10,6 6'5
79 78 4'0 143'8

80 79 5'7 0'5 1'3 2'4 3'5... 7'2 10'2 2'2 0'9 0'9 2'4 2,8 0,8 35'1
,
79 7 8 4'7 1'9 3'7 8'0 8'9 12'0 21'5 29'3 27'2 20'7 16'9 6'3 2,6 159'0

73 7 2 3'7 1'6 2'7 5'9 9'1 8'1 4'5 2,6, 3'3 7,6 8'9 5'9 2'0 62'2
7'2 5'4 7'9 6'1 3'1 9'2 0'1 2'5 6'5 6,6 9'1 6,6 60,8
77 77 1'5
I
5,6 6,6 2,6
78 78 3,8 4'1 3'5 4'1 6'2 6'1 6'3 6'3 6'3 9'3 66'9
80 79 6'5 0,8 0'9 2'3 7,8 13'7 3'6 3'5 2'2 1'9 3'4 5'0 2'2 47'3
83 80 6'1 2,8 2'2 6'1 14'1 10'7 2'0 2'4 1'7 2'1 3,6 7'5 5'4 60'2

79 78 2'7 12'5 14'1 14'1 12,6 10',2 6'7 5'9 4'5 3'5 3'4, 2,6 6'1 96:2
81 81 3'0 8'3 8'0 8'1 8'4 6,6 3'9 1,8 1'3 1'4 4,6 4'5 .8'2 65'1
9,8 12'2 6'5 10'0 6,6 4,6 10 3'1
7 8 77 4'0 10'2 7'4 8'7 7'2 8'4 II'5
81 81 4'<1 7'9 4,6 7'2 6'0 II'I II'7 9'9 6'5 3'1 2'9 6'7 11'1 88'7
63 63 6'4 1'9 4'2 3'7 8'3 5'2 2'0 0,8 0'9 0'9 2'0 5,8 3'5 39'2

70 70 2'7 2,6 3,6 5'8 9,8 8'5 5'1 3'0 3'0 3'1 3'5 4'9 5'1 '5 8 '0

5~ 55 0'7 3'2 3'9 4,8 7'0 4,6 1'5 1'1 2'2 2,6 3'9 4'0 3,6 42'3
58 58 1,6 ' 3'7 3'5 4'5 9,6 6'5 3'2 2,6 3'3 2'9 8'4 9,6 5,6 63'4
80 81 3'2 3'3 1'9 4'3 9'7 10'9 7'3 4'4 3'2 4,8 13'4 II,8 5'1 80'1
80 81 2'3 9'9 6,6 7'4 7'6 6'7 6,8 6,8 7'9 6,8 8'1 9'9 10,6 95'1
79 79 1,8 13'0 12,8 7,8 5'1 4'0 3'7 2,6 1'7 2'9 4'5 5'5 8'5 7 2 '1
69 69 1,8 9'2 7'0 9'2 10'2 7'6 5'6 3,8
6'4 6'4 9'0 9'0 IO'4 94'0
79 79 1'0 13'5 9'9 14'0 12,6 13'0 II,8 13'7 16'1 20'0 20'7 19'4
I I '9 177,6

72 72 1'0 7,6 7'1 9,6 9'0 5'2 3'6 2,6 2'3 3,6 6'7 8'9 8'5 74'7
80 79 2'0
- -5'9 - 3'5 3'5 4'2 6'1 3,6 4'3 4'4 6,8 7'2 7'2 8'3 65'0
8
7 79 2'2 IO,8 7'9 9,8 10,8 10'7 8'7 6'3 8'9 8'4 14,8 15'7 13'2 12 5',?
78 79 2'3 18,6 14'4 10'3 8,0 6,6 6'5 4'9 3,8 3'4 4,8 8,6 14'7 104'6
CHAPTER VI
TROPICAL CLIMATES
EYOND the limits of the swing of the equatorial rainfall

B belt, in the latitudes where the trade winds b.low through-


out the year, are the greatest deserts of the world. Be.,
tween these and the equatorial climates lies a belt which,' for
part of the year, is under the influence of the trades, but for the
rest of the year experiences an invasion by the belt of convec-
tional rains; here are found the tropical climates with alternate
trade-wind and doldrum influences. In continental interiors
and up to the seaboard on western margins the season of trade-
wind influence corresponds to the dry season; but the trades,
though generally drying winds, are rain-bearing where they blow
off the sea; consequently there are certain areas on eastern
margins which do not suffer drought from their sojourn in the
trade-wind belt.
Here, then, are two fundamental types of tropical climate,
continental and marine, the one with, the other without, a pro-
nounced dry season. The latter are restricted to a narrow strip
on eastern margin, but extend some distance beyond the swing
of the equatorial rains, as far as the. trade winds are found (see
Fig. 24). In lee of these areas which lie beyond the poleward
extensions of the equatorial rains, the rainfall decreases steadily
westward (away from the marine, influence). and in this direction
the savanna passes gradually into desert. But this transition
area of gradually decreasing rainfall experiences a regime almost
identical with the tropical continental climates and can be
included with them, although its causes and the mechanism
of its rainfall are different. In South Africa, for example, the
equatorial rainfall belt reaches only <j.bout as far south as the
Zambesi and most of the rain south of this comes from the
trades. But the seasonal regime of this rainfall is controlled
by a form of monsoonal action,. for in winter the pressures over
the land are high and the trades cannot penetrate, while in
summer the continent develops a low pressure which reinforces
the trades and draws the rain-bearing currents far inland. Thus
there is a marked summer maximum and a winter drought
90
TROPICAL 9I
which is practically as complete as that of the normal tropical
type (e.g. Kimberley).

Marine Type
RAINFALL. Near the poleward margin of this type rain is
derived almost entirely from the trades, but stations nearer
the Equator receive. orographic rain from the trades for part of
the year and convectional rain from the low-pressure belt of
calms for the remainder. In having no dry season the rainfall
regime agrees with the equatorial, although from different causes,
and it produces similar reactions. Much of the east coasts of
Brazil and Africa, most of Central America and the windward
slopes of islands in the trade-wind belt enjoy a rainfall of over
50 inches and are heavily forested. Havana, on the latitude of
the driest part of the Sahara, has over 50 inches and the driest
month claims nearly 2 inches.
JFMAMJJASOND
10

5
0 _ _" .

5
o . . . .~~~~.I ..
10
5
o ......-1111111!1
FIG. 29.-Rainfall Regimes on Trade Wind Coasts

Naturally the aspect of a station with regard to these rain-


bearing winds plays a great part in determining its rainfall. The
extraordinary contrast between the windward and leeward coasts
of Hawaii has already been quoted (p. 39), but nearly every
island shows the same character in a mQre or less marked degree
according to its relief. The Blue Mountains of Jamaica rise
above 7,000 feet, Port Antonio on the windward shore has I40
inches, Kingston on the leeward shore has less than 40 inches;
Colon; at the Atlantic end of the Panama Canal, records I30
inches,. Balboa, at the Pacific end, only 70 inches. The lower
islands have a lower and less reliable rainfall since the trades
pass over without forced ascent and therefore without much
precipitation (the Bahamas have only about 50 inches and in
some years less than 30 inches), which is rather unfortunate a.s
the lower and flatter islands are usually better suited by their
92 CLIMATOLOGY
relief to agriculture, but often suffer from drought. Being
chiefly orographic in nature, these rains do not show the same
regularity of diurnal incidence as the convectional rains of the
doldrum belt, sometimes they are heavy at night when the
land surface is colder, sometimes during the day when the sea-
breeze strengthens the prevailing trade winds on the eastern
side.
The seasonal incidence sometimes shows a winter maximum
when the trades are strongest (e.g. Honolulu, Fig. 29), but more
usually an autumn maximum while sea temperatures are high
and land temperatures are falling (e.g. Nassau and Tamatave,
Fig. 29). On the equatorward edge of the belt these are usually
overshadowed by a heavy summer rainfall due to invasion by
the doldrum belt (e.g. Trinidad and Samoa, Fig. 29). Further-
more, the rainfall is not entirely orographic,; the land surface
sets up convection currents which bring rain and this action is
naturally strongest during the period of strongest insolation, a
factor which further conduces to a summer maximum. In
winter this is inoperative and lee coasts, especially, are fairly
dry.
TEMPERATURE. In temperature, as in rainfall, the tropical
marine type resembles the equatorial. Temperatures are uni-
formly high (Kingston and Mozambique each have a mean
annual of 79°) and the yearly range is small (Kingston 6°, Mozam-
bique 9°). But the uniformly high temperatures are less trying
than in equatorial climates because of the steady fanning of the
trade winds for most of the year, and only bring discomfort
during the season of calms with high humidity which comes
about the change of the sun.
WINDS. The prevailing wind is the constant trade wind,
modified by land and sea breezes. The exposure of east coasts
to the prevailing wind has the result of silting up the harbour
mouths and the surf makes them difficult of approach; the
best ports. are therefore to be found on the west coasts, e.g.
Kingston (Jamaica), Castries (St. Lucia), etc.
HURRICANES. It is during the season of calms that there
are apt to occur the violent storms known as hurricanes, typhoons
or tropical cyclones, a feature of the trade-wind belt which,
though irregular and infrequent, is very important, resulting in
considerable loss of property and life. '
Although liable to occur at any time of the year, there is a
danger period which, as the accompanying table shows, follows
after. the furthest poleward migration of the belt of calms, i.e. in
autumn. In the Indian Ocean, the necessary conditions for these
storms are fulfilled during the period of calms which come at
TROPICAL 93

J F M A M J J A S 0 N D
- - - - - - - - - - -
Northern Hemisphere I I I I 2 4 16 25 25 15 6 3
Southern Hemisphere 26 22 23 9 4 0 0 0 0 I 4 II

Bay of Bengal and Arabian


Se1 .,. 3 0 I 12 19 19 2 3 5 17 13 6

AVERAGE ANNUAL FREQUENCY OF TROPICAL CYCLONES. PERCENTAGE


• OF TOTAL.

the changes of the monsoon, i.e. there are two annual maxima.
Fig. 30 shows the distribution of these storms and their most .
frequent track~: It will ·be seen that there are six principal
centres in which about 90 per cent. of tropical hurricanes originate
and these lie in the belt of calms at its furthest limit. Here is
found the necessary combination of conditions, viz. : -
I. a}lm air which allows intense heating of the lower layers,
thus giving rise to a condition of instability followed by rapid
inversion.

FIG. 3o.-Distribution of Tropical Cyclones

z. Distance from the Equator sufficient to produce the neces-


sary vortical movement.
In each ocean it will be noticed that the storm area is on
the western side, since here the currents have piled uJl the warm
waters of the equatorial belt and thus provide a generous supply
of water vapour which nourishes the storms.
It will be noticed, also, that there is no storm area in the
South Atlantic off Brazil, where it might be expected, as the
belt of calms never passes into the South Atlantic. The tract<
of the storm in each case is a simple parabolic curve, first west-
94 CLIMATOLOGY
ward, then poleward, then eastward. By reference to the isobaric
map this curve will be seen to round the western ends of the
permanent oceanic 'highs'; the storms make use of weak
points in the high-pressure belts, occasioned by the seaward exten-
sion of the continental summer ' lows', to break through into
the westerly circulation where they can often be traced as normal
cyclonic storms. The West Indies and Florida, Bourbon and
Mauritius, the'Philippines and Formosa suffer severely from the
visitations of these storms. Li~s are lost (5°,000 in the Swatow
typhoon of August, I922), plantations are ruined and extensive
damage to property occasioned by them. The growing of certain
delicate crops, e.g. cacao, is out of the question in areas frequently
visited, on account of the high wind velocities attained. The
torrential rain which these hurricanes bring contributes to a
second m·aximum of rainfall which can often be recognized. ·in
the rainfall graph.

Continental Type
In continental interiors and on leeward coasts a very different
type of climate is experienced. Here the trades are dry winds
and the season of their dominance is the dry season. Normally
there is a single rainy season and a single dry season of greater or
less duration; this marked seasonal incidence, judged by' the
vegetational responses it invokes, is the criterion of the climatic
type. .
THE DRY SEASON. After the rains the soil is saturated, the
air is moist and the nocturnal fall in temperature produces a
copious dew, but under the drying influence of the trades the
effect of the past rains is quickly obliterated, grass and shrubs
begin to wither and die down, the soil becomes dust, woodwork;
cracks, the rivers dwindle and the level of water falls in ponds
and lakes. Relative humidity is low (60 to 70 per cent.) and
may fall as low as IO per cent. when strong winds blow from
the deserts, such as the Harmattan of West Africa. This wind
has a'humidity of less than 30 per cent. and it often penetrates
into the humid coastal zone, where it is much appreciated in
spite of its heat; for its dryness is invigorating and it cools
by promoting evaporation. Here it is known as 'the doctor'
because of the improved health and· comfort which it brings,
but the dryness which suits man is inimical to plants, and trees
and crops may be dried up and ruined. Inland, where it blows
almost constantly in the dry season, the heat and dust associated
:with it are trying in the extreme, the skin dries and cracks, and
chills and colds are rife because, in the drier air, the greatly· in-
TROPICAL 95
creased diurnal range of temperature places a severe strain on
the constitution.
The natives take advantage of this dry season to burn the
bush and to prepare for cultivation; other bush fires are less
intentional, but the smoke of these fires, together with the dust
which the hot, dry wind often brings, fills the air and turns a
bright cloudless sky into a dark haze through which the sun
shows red and dull.
TEMPERATURE. Under cloudless skies and in the dry air
much higher te_mperatures are recorded than in the equatorial
zone. Mean temperatures of over 90° for the warmest month
are frequent, while daily maxima at the height of the hot season
JFMAMJJASOND J F M A M J J A SON 0
90'
85 85
80
75
70
5
L_____~. . . .~__~o

5 '

Cuyaba Zomba
FIG. 31.-Annual Match of Rainfall and Temperature in Tropical Continental
Climates

frequently exceed IIOo. At night the fall of temperature is


rapid, and before dawn '50° may be recorded, while night frosts
are by no means unknown in the ' cold ' season; such extreme
ranges, although stimulating, are somewhat trying. Tempera-
tures soar higher and higher as the season advances and as the
sun approaches the zenith, but the rains are approaching and
their arrival brings some alleviation of the extreme heat. At
Mongalla the mean for March, before the rains, is 83° and for
July only 76°. ;
The maximum temperature at Timbuktu (Fig. 3r) is reached
in May, after which the arrival of the rains drives the temperature
down nearly roO by August, in spite of increasing insolation.
When the rains cease there is a recovery to a second maximum
96 CLIMATOLOGY
in September, after which the falling off of insolation causes the
temperature to fall to a January minimum. At Cuyaba the
maximum is reached in October, after which temperatures are
remarkably uniform while the rains last; at Zomba the rise of
temperature is checked by the rains in October and falls more
than 6° to a second minimum in December, although insolation
is strongest then.
This fall in temperature with the arrival of the rains is prob-
ably not appreciated, for the increased humidity makes the
heat harder to bear, i.e. ' sensible temperature' does not decrease
although actual air temperature does. The season just pre-
ceding the rains is the most unpleasant of the year, for humidity
is steadily increasing and as yet the cooling and purifying
influence of the early showers is not felt.
THE TORNADO SEASON. The arrival of the rains is announced
by violent storms known in West Africa as tornadoes. In Siejra
Leone the first of these occur in April, three months before the
rains, and they increase in frequency until June when the real
rains begin. They are rare during the rains but recur when
these end in September, becoming less and less frequent until
December. They bring practically the only rain which falls
outside the rainy season. They are extraordinarily sudden and
violent and they usually occur at night, the darkness helping to
make them even more impressive, but they last only a short
time and affect a relatively small area. Although usually pre-
ceded by flickering lightning, they are sometimes entirely
unannounced; the calm air suddenly becomes a rushing wind.
the silence is rent with peals of thunder, and rain falls in sheets,
In half an hour it has passed and all is quiet again. Elsewhere
the season of violent storms is shorter, but everywhere thunder-
storms precede the rains. .
THE WET SEASON. The effect of the rains, when they come,
is magical; the dust becomes mud, the dry water-courses become
torrents, the withered vegetation bursts into life and in three
weeks a dusty grey landscape has clothed itself in the fresh green
of grass and leaves. The conditions during the wet season
may be described in almost the same words as the equatorial
climates; there is the same high humidity, the same sultry heat
with the same small diurnal range of temperature, but there
are usually spells of fine weather when there is a temporary
return to something like dry season conditions~
The duration of the rains decreases steadily polewards, so
also does the amount of rain and its reliability; 50 inches on
.the equatorial margin becomes IO inches on the fringe of the
desert; 20 inches to 40 inches is a usual amount, but a station
TROPICAL 97
recording a mean annual rainfall of, say, 30 inches may have
more than double this in one year and half this amount the
next. This unreliability is unfortunate since the tropical climates,
although at present largely pastoral, are otherwise well suited
to agriculture.
The fact that the rain all occurs in summer when temperatures
are high makes growth at this season very vigorous, but it also
results in a high rate of evaporation and a considerable loss by run-
off. Over the greater part of the Union of South Africa the evap-
oration is more than 90 inches, while the rainfall averages about
30 inches, i.e. evaporation is three times the rainfall. This dis-
advantage is increased by the intensity of the falls; four inches
often fall in 24 hours and 2 inches in an hour. At many stations
in Rhodesia the average fall per rainy day is half an inch, while
more than half the rain in the Transvaal falls at this rate. These
points must always be borne in mind in the selection of crops
for cultivation and it must b,e realized that only a small pro-
portion of the recorded rainfall is effective, 30 inches is frequently
inadequate for agriculture.
The departure 'of the rains, like their arrival, is marked by
violent storms, gradually decreasing in intensity.. Temperatures,
which had fallen with the advent of the rains, now probably
rise again, though not so high as before, since the angle of the
sun is declining; diurnal maxima rise higher and nocturnal
minima fall \ower. With the withdrawal of the sun to the
other tropic a minimum is reached soon after the solstice. The,
mean temperatures of the coldest months are usually in the
neighbourhood of 700, except on high plateaux, and it is only
relatively a 'cold season'. Conditions at this time are the
most pleasant of the year, for the freshening effects of the wet
season are still to be seen and felt, while the heat and· dust of
the height of the dry season have not yet brought discomfort.
Thus it is usually possible to make a threefold division of
the year comparable with that recognized in India, where,
however, it is much more clearly defined, into :-a hot season,
a wet season and a cool season. Although astronomically winter,
the sun being in the other hemisphere, the dry season is generally
spoken of as summer since it is bright and sunny and, in the
absence of cloud and rain, frequently hotter than the wet season.
Throughout Latin America it is ' verano' and the rainy season
is 'iverno', in which we may recognize the familiarity of an
exotic people with the summer drought and winter rain of their
native. Mediterranean regime.
HIGH ALTITUDE TROPICAL CLIMATES. As in ...equatorial
climates the effect of altitude is to change the value but not the
98 CLIMATOLOGY
distribution of climatic elements. The single maximum of
temperature and of rainfall at or near the solstice still holds
good, but temperatures are lower and rainfall, above a certain
level, is less. Mexico City (7,5°0 feet) has a mean annual tem-
perature of 60° and 23 inches of rain, Cochabamba (8,000 feet)
of 63° and 18 inches of rain. The lower rai~all gives a tendency
towards aridity which is increased by the clearness of the air,
the low atmospheric pressure and the tendency to strong winds,
all factors which promote evaporation. Further, the rainfall
has a high periodicity and falls in the form of heavy showers,·
thus increasing run-off and diminishing its efficacy; irrigation
is therefore generally necessary for successful cultivation.
The mitigation of temperatures by altitude has important
results on cultivation and settlements, Europeans can grow
sub-tropical and even temperate crops in tropical latitudes, hill
stations provide relief and renewed health for those compelled
to live for the rest of the year at lower levels. The Spaniards
found it possible to colonize Mexico and Peru, displacing the·
indigenous civilizations of the Aztecs and Incas; the Dutch
settled on the South African plateau, and Europeans are now
firmly established on the East African plateau even on the
Equator.
Temperatures here are practically those of sea-level at the
Cape, in fact over the whole of the plateau south of the Equator
there is a remarkable uniformity of temperature, decreasing
altitude compensating for increasing latitude, as the following
maximum and minimum figures show:-
Nairobi. . Lat. 1° 25' S. 5,450 feet. Max. (Mar.) 65. Min. (July) 58
Salisbury ,,17° 54' 4,880 " " (Jan.) 69 56
Bulawayo 20° 2' 4>47 0 " 72 57
Johannesburg " 26° I I ' 5,9 2 5 " 67 51
Graaf Reinet " 32° 16' 2,5 00 " 72 56
Cape Town. " 33° 56' 40 " 70 55

The great diurnal range makes frost a real menace towards


the margin of the zone and at higher altitudes, imposing a
limit on more delicate crops. Sugar-cane, whose upper limit
can conveniently be taken as the dividing line between the
, tierra templada ' and' tierra fria' (see p. 100) is grown up to
6,000 or 7,000 feet in equatorial latitudes in Peru, but at Jujuy
(lat. 23° 53' S.) it is not seen above 3,000 feet.
VEGETATION OF TROPICAL CLIMATES. The heavy and regularly
distributed rainfall and the constant high temperatures of the
marine type support a tropical rain forest practically indistinguish-
able from the equatorial, but as we go inland from the trade
TROPICAL 99
wind coasts or polewards from the Equator in continental
interiors forest growth becomes impossible as the drought
lengthens, and the selvas gradually pass into woodland and
scrub with scattered trees in a dominant setting of tall grass
(the savanna). By connast with the uniformity of the
equatorial regime the key-note of tropical continental climates
is the marked seasonal rhythm; and by contrast with the
perenn"ial activity of the selvas the savanna vegetation grows
with astonishing rapidity in the wet season and remains prac-
tically dormant through the dry. The trees of the savanna are
not forest trees, they are markedly xerophilous, hard, gnarled
and thorny, with reduced leaves (acacias) or with devices for

FIG. 32.-Duration of Rainy Season FIG. 33.-Vegetation of Africa


in Africa

the storage of water (bottle trees and baobabs); most of them


are deciduous, shedding their leaves at the commencement of
the dry season and passing a xerophytic existence until the rains
return. The heat and drought of the Harmattan often splits
the trunks of trees, and from these splits a gum is exuded; the
, gum arabic' exuded by certain acacias is an important article
of commerce in the Sudan. But it is the grass which is char-
acteristic, grass which grows strongly to a height of six feet or
more during the rains, but dies down to the roots in the dry
season.
The indigenous economy of tropical grasslands is pastoral
and, since pasture is poor in the dry season, partly nomadic.
The native tribes of the African savanna are dependent on their
cattle, extensive cattle ranching is practised on the campos of.
roo CLIMATOLOGY
Brazil and the llanos of Venezuela, while the monsoon grass~
lands of Australia, somewhat similar in nature, are the. home
of cattle industry on organized lines. There are attempts by
native tribes at agriculture on a small scale, maize or millets
being sown at the beginning of the rains, but as yet little has
been done to bring these lands under the plough. Yet with an
adequate water supply they will yield good crops of millet, maize,
tobacco, cotton, etc., and their future development is assured
along these lines; rapid progress is being made everywhere.
Th.e s~~m.er incidence of ra~n is a great advantage, but its "':ln~
rehablhty IS a drawback, whIch can, however, often be overcome-
by irrigation. It should be noticed that the incidence of rain
in tropical climates is similar to that in monsoon climates, but
that the amount is generally much less; thus they grow cotton,
but not jute; millets, but not rice; coffee, but not tea. The
torrential nature of the rain tends to an early exhaustion and
impoverishment of the soil, so that careful conservation and
replenishment are necessary for continuous cultivation.
The rhythmic climate imposes a rigid rhythm on agricultural
practice, water for growing plants is available, without irrigation,
only in the rainy season, while the dry season is the time of
harvest. Coffee, in Brazil, is harvested from April to September,
the dry season, and in Venezuela from September to May. Sugar-
cane in Rio de janiero, where there are summer rains, is cut
from March to October; in Pernambuco, where there are winter
rains, from October to March.
With its strong seasonal changes the climate is much more
stimulating than the dull monotony of the equatorial zone, and
the savanna tribes are vastly superior, mentally and physically,
to the forest dwellers. Poor pastoralists though they are, they
cherish their independence and do not readily accept subordina-
tion.
VEGETATION AND ALTITUDE IN TROPICAL CLIMATES. In the
tropics of Latin America, where the highlands rise from the
dense forests of a marine tropical climate, it is usual to recognize
three zones of climate and vegetation. The limits of the zones
differ from place to place, but the average altitudes are as
follows :-
1. The Tierra Caliente extends from sea-level to 3,000 feet
and embraces the hot steamy coasts of Mexico and Brazil with
their dense forests producing rubber and cacao.
2. The Tierra Templada, between 3,000 and 6,000 feet, has
mean annual temperatures between 65° and 75° and is charac-
terized by an unusually small annual range of temperature
'(often only 4° or 5°). Its altitude corresponds to a zone of
TROPICAL lor
cloud which forms at the saturation level when the air rises
up the slopes by day and a zone of mist at l'l:ight w~en the clouds
condense. This atmosphere of damp warmth keeps tempera-
tures uniform and supports a forest vegetation of extraordinary
density with tree ferns, lianas, etc. Under cultivation the zone
produces maize, coffee, tea, tobacco and cotton.
§. The Tierra Fria, between 6,000 and 10,000 feet, has mean
annual temperatures of 54° to 65°. The lower temperatures
and decreased rainfall bring about an impoverishment of the
forest which passes into scrub and grassland and the crops are
those of the temperate zone, maize, wheat, lucerne, etc. At the
same time conditions have become more favourable for settle-
ment and especially European settlement; the density of popula-
tion in the' tierra fria' of Mexico is four times that of the' tierra
caliente '.
A somewhat similar succession of zones is found in tropical
Peru, Bolivia, Ecuador and Colombia, where the trades, after
crossing the interior savannas, are again forced upwards by the
Andes and give rise to a strip of 'Montana' forest similar in
nature and origin to the forests of the marine tropical type.
But in East Africa the highlands, e.g. Kilimanjaro, rise from
the xerophilous 'altitude savanna' of a continental climate.
This savanna is followed by a wetter zone with temperate or
sub-tropical rain forest, clothing the slopes between 6,000 and
9,000 feet, above which is steppe with tussocky grass.
In a zone in which the trade winds play such an important
part it is easily understood that aspect with respect to these
will profoundly modify climate and vegetation. The rivers of
Colombia, the Magdalena and the Cauca, lie open to the north-
east winds and send arms of dense forest up into the grassland
and woodland of the plateau. The upland ridges of the East
African Plateau receive an adequate rain, but the north-south
depressions, sheltered from the rain-bearing winds, are often so
dry that they are basins of inland drainage, while some of them
are actually drying up and soda is obtained from the disappearing
lakes, e.g. Lake Magadi and Natron Lake.

Regional Types
AFRICA. North and South Africa present a striking contrast
in the latitudinal extension of their tropical climates; north of
the Equator the strip of savanna between forest and desert is
only about 600 miles wide, but south of the Equator it stretches
through more than double this width and, on the eastern side,
of the continent, almost down to its southern extremity. In
7
I02 CLIMATOLOGY
North Africa the extent of the tropical climates is the extent
of the swing of the equatorial rainfall belt, for the trade winds
have come overland all the way and bring no rain; the margin
of the desert is therefore sharply defined and runs almost due
east and west. In the southern half of the continent, bn the
other hand, the trades come over the sea, the continent is narroW'
and offers a high barrier to the passage of the winds, the eastern
side is therefore well watered and a belt of forested coastline
backed by savanna extends nearly the length 6f the continent.
NIGERIA AND THE GUINEA LANDS. Behind the coastal forest
belt the S.W. monsoon of the wet season (see p. 8I) alternates
with the N.E. trades of the dry season, the latter often stre1).g-
thened and blowing out from the desert as the hot dry harmattan,
sometimes temporarily bringing Saharan conditions, with humidi-
ties as low as 20 per cent., right down to the coast. A some-
what similar climate extends right across Africa as far as the
headwaters of the Nile and it is from the summer rains that
the White Nile gets some of its water. But the regularity of
the discharge of the White Nile, which is so important for Egyptian
agriculture, follows from its drainage of the lakes of the great
plateau where the regime is equatorial.
EAST AFRICA. There is no real dry season here but two
maxima occur (see Nairobi and Entebbe) , one'in February to
May, known as' the maize rains', the other in October to Decem-
ber, known as (the millet rains " but except locally round Lake
Victoria the rain is below normal (Nairobi 40 inches). Further
east the rainfall is still less (Makindu 20 inches) and near the
coast there is virtually desert (Kismayu I5 inches) which becomd
more and more complete to the north-east in Somaliland.
The whole region is affected by the proximity of the two
monsoon centres of Abyssinia and India, which receive much
of the rain which would otherwise fall here. In the northern
hemisphere winter the winds are north-east, reinforced trades
blowing parallel to the coast. They would not normally bring
much rain, but in actual fact they bring practically none, for
the African continent develops a high pressure of its own which
deflects these winds and gives them a direction which is actually
off-shore; not until the coast changes direction at Mombasa is
20 inches recorded in winter. In summer there is a strong air
flow to the focus of the Indian monsoon and winds are south-
west, again parallel to the shore, and the only rain which falls
is from the branch of this current which is drawn off to the
Abyssinian focus. It is during the change-over in May and
June, that this coast gets its rain (see Mogadiscio): Kismayu has
more than half its rain in these two months.
TROPICAL I03

The East African plateau is a highly favoured region with


an adequate but not excessive rain well distributed through
the year, a vegetation easily cleared for cultivation, and tem-
peratures modified by altitude to an extent that makes it a suitable
field for European settlement. The mean temperature of the
hottest month at Nairobi is 65°, not much hotter than London,
and of the coldest 58°, the equivalent of an English June, and
although the diurnal range is rather large, it is not excessive,
.ancf there is no danger from frost. The altitude and the bright
light are stimulating, perhaps too stimulating after a time, and
Europeans find the climate physiologically congenial as well as
agriculturally productive.
CONTINUITY OF THE AFRICAN SAVANNAS. By causing a break
in the fore?t belt the East African plateau completes the con-
nection between the northern and southern savannas, thus
facilitating free movement throughout the length of the continent;
a fact which has been of great I'ignificance in the spread of species
of animals and plants and which has also affected the migrations
of human groups.
DISCONTINUITY OF THE SOUTH AMERICAN SAVANNAS. The
Venezuelan and Brazilian savannas are isolated, since the oro-
graphic rains along the Andean slopes support a dense forest,
the Montana, which terminates the savannas on the west. This
fact may account for the poverty of the mammalian faunas of the
South American tropical grasslands as compared with the African.
For whatever may have caused their depopulation, and this is
clearly of quite recent date, their repopulation would be a slow
process in the absence of immigration from other regions.
THE TROPICAL CLIMATES OF SOUTH AMERICA are fuiI of
anomalies, imperfectly understood because of insufficient data.
North of the Equator the tropical climates are represented in
wide stretches of the Guiana Highlands and in the Llanos of
the Orinoco. South of the Equator there is a much fuller develop-
ment in the campos of Brazil (e.g. Cuyaba) as far south as Asun-
cion in Paraguay. A feature of the climate here, as in the middle
Amazon, is a winter anticyclone bringing cold waves known as
, friagems ' or 'surazos', when the temperature may fall below
50° and cause considerable discomfort.
The savanna type of grassland passes eastward into a more
xerophytic type, the caatinga or thorn scrub of the projecting
nose of Ceara and Pernambuco. Here is a region, centred round
the middle course of the Sao Francisco, with 30 inches or less,
but rendered still more inhospitable by the unreliability of the
rains. The dry season is normally six months long, but it often
happens that the rains fail to arrive to break the drought and'
I04 CLIMATOLOGY
famine results, with loss of most of the cattle and depopulation
of the area. This aridity is as yet not satisfactorily explained,
although it is generally attributed to the presence of high land
to the east; this high land, however, does not receive excessive
rain to compensate, as this explanation would lead one to ~xpect.
Along the east coast, from Bahia southwards, summer is the
rainy season, but between here and Cape San Rocque there is
an anomalous winter rainfall maximum. At this season, the
~S.E. trades are directly on-shore, while in summer, which is
relatively dry, the prevailing wind is north-east, i.e. parallel to
the coast.
CENTRAL AMERICA AND MEXICO. In the' American Mediter-
ranean' the prevailing wind throughout the year is the N.E.
trade, and the islands and coasts are wettest where most exposed
to this. But the simple circulation is modified by monsoonal
in draught which brings summer rains to the whole of the area.
In winter the warm waters of the Caribbean and the Gulf of
Mexico produce a low-pressure centre round which there is a
counter-clockwise circulation, more or less parallel to the coast-
line; rainfall is heaviest where the lie of the coast is across this
wind swirl, e.g. in Honduras. Winds along the Mexican coast
are northerly and sometimes strong (Nortes), a continuation of
the American ' norther', and they are occasionally felt high up
on the plateau (papagayos). These northerly winds bring cold
which is intense for the latitudes and the fall of temperature is
often sudden and unpleasant.

SUGGESTIONS FOR FURTHER READING


Besides A. Knox's Climate of Africa, I9II, and -the appropriate chap-
ters in the Oxford Survey of the. British Empire, the following articles, etc"
on the Climatology of Equatorial and Tropical Africa will be found
useful. H. G. Lyons, 'Meteorology and Climatology of German East
Africa', Q. J. Roy, Met, Soc" 1917; J. J. Craig, Rainfall of the Nile Basin,
Cairo, 19I3; C, E. P. Brooks, 'The Distribution of Rainfall over
Uganda with a Note on Kenya ',"Q. J. Roy. Met. Soc., I924; C. E. P.
Brooks, ' Rainfall of Nyassaland', Q, J. Roy. Met. Soc., 1919; C. E. P.
BrOOKS, 'Rainfall of Nigeria and Gold Coast', Q. J. Roy, Met. Soc"
19I6; N. P. Chamney, Climatology of the Gold Coast, Bull. 15, Dept. of
Agri., Accra, 1928; G. G. Auchinleck, Rainfall of the Gold Coast, Bull. 2,
Dept. of Agri,; H. Schmidt,. Der Jahrliche Gang der Niederschliige in
Afrika, Deutsche Seewarthe, I928.
For South America the following may be consulted :-C. de Carvalho,
Meteorologie du Bresil, I9I7; G. G. Chisholm, ' Meteorology and Clima-
tology of Brazil' (a review of the above), Scot. Geog. Mag., I9I7; 'Climate
in North-East Brazil', Geog. Teacher, I926 (a summary and review of
, Causas provareis das seccas do nordesti Brasileiro', J. de S. Ferraz);
: Climatological Data for Northern and Western Tropical South America "
Sup~lement to Monthly Weather Review, March, I928; B. Franze, Der
TROPICAL 105

Niederschlagsverhiiltnisse in Sud Amerika, Petermanns Mitterlungen, Suppl.


No. 193, 192 7.
For Central America and the West Indies :-' Climatological Data for
Central America', Monthly Weather Review, 1923; 'Rainfall Maps of
Cuba', Monthly Weather Review, 1928; 'Climatological Data for West
Indian Islands', Monthly Weather Review, 1926; 'West Indian Hurri-
canes and other Tropical Storms'of the North Atlantic Ocean', Supple-
ment to Monthly Weather Review, 1924; C. F. Brooks, ' Notes on the Cli-
mate of Panama " G. R., 1920; The Oxford Survey of the British Empire,
and Government Handbooks.
For the East Indian Archipelago :-Het Klimaat van Nederlandsh
Indie, by Dr. C. Braak, in three volumes, each with summaries in English.
C. E. P. Brooks, 'The Climate of the Fiji Islands', Q. J. Roy. Met. Soc.,
1920.
I06 CLIMATOLOGY

TEMPERATURE
Station. Lat. Long. Alt. (ft.) ] F M A M ] ] A S 0 N
-- - - - - - - - - - - -
HONOLULU. 21° N. 158° W. 38 7 1 71 7 1 73 75 77 78 78 78 77 75
PORT AU PRINCE 19° N. 72°W. 120 76 77 78 79 80 81 82 81 s,r 80 78
KINGSTON 18° N. 76° W. 24 77 77 77 78 80 81 82 82 82 .81 79
KEY WEST 25° N. 82 0 W. 22 69 71 73 76 79 82 84 84 82 79 74
COLON 9° N . 79° W. 36 80 80 80 81 81 80 80 80 80 80 80
RIO DE JANEIRO 23° S. 43° W . 201 78 78 77 74 71 69 68 69 70 71 73

ST. HELENA 16° S. 6°W. 1,900 64 65 66 65 63 60 58 57 57 58 59
DURBAN 30° S. 31° E. 25 0 76 77 75 71 68 64 64 65 67 69 72
KAYES 14° N . I2°W. 197 77 81 89 94 96 91 84 82 82. 85 83
BATHURST
MONGALLA
13° N. 17oW. 6 74 75 76 76 77 80 80 79 80 81 79
5° N. 32° E. 1.140 80 82 83 81 79 77 76 76 77 78 79
.
GONDOKORO 5° N . 31° E, 1,500 83 85 86 84 81 79 78 77 78 78 79
MOGADISCIO 2° N. 45° E. 59 77 79 81 82 8ci 77 75 76 76 77 76
LIND! 10° S. 40° E. 268 82 81 81 81 80 77 78 77 78 79 81
FORT JOHNSTON 15° S. 35° E. 1,55 8 79 77 78 77 73 69 68 71 75 80 82
SALISBURY 18° S. 31° E. 4,85 6 70 69 68 66 61 57 56 60 66 71 7 1
CARACAS 10° N. 67°W. 3.1 19 65 65 66 68 70 69 68 68 69 68 67
CUYABA 16° S. 56° W . 54 1 81 81 81 80 78 75 76 78 82 82 ii2
ASUNCIQN 25° S. 58° W . 383 81 80 78 72 67 63 64 66 70 73 76
MEXICO CITY 19° N. 99° W . 7AIl 54 57 61 64 65 64 62 62 61 59 56
PUEBLA. 19° N. 98°W. 6,9 87 54 57 61 64 65 64 63 63 62 61 58
OAXACA . . 16° N. 97°W. 5,080 63 66 70 73 74 72 70 70 69 67 66
SUCRE . 19° S. 64 °W. 9,190 55 55 56 54 51 49 49 52 56 56 57
TROPICAL r07

RAINFALL

D Yr. Ra. F 111 A M J J A S 0 N D Total.


- - - - - - - - - - - -- - - - - - - - - - - - - - -
3,8 2'3 1'9 1'1 1'3 1'5 1'9 4'2 4'1 3 1.6
7 2..- 75 7'7 3'7 4'3 1'5

77 79 5'9 1'2 2'5 3'7 6'5 9'4 4'1 2'7 5'4 7'3 6·6 3'4 1'3 54'1

78 79 4'9 1'0 0·6 1'0 1'2 4'3 4'1 1'7 3'7 4'1 7'5 3'1 1'0 33'9

70 77 14'4 1'9 1'5 1'3 1'3 3'5 4'3 3'4 4'5 6·8 5.6 2'3 1'7 3 8 '1

79 80 1·6 3'7 1·6 1·6 4'3 12'4 13'3 16'0 14.8 12'5 15'1 20'7 11'4 12 7'4

7 6 73 10'1 5'0 4'5 5'3 4'2 3'2 2'3 1'7 1'7 2·6 3'3 4'1 5'5 43'4

61 61 8·8 3'0 3.8 4'9 3'9 4'2 4'1 4'0 3'7 3'0 1'9 1'7 2'0 4 0 '2

74 7 0 12'9 4.6 4'9 5'4 3'4 1'9 1'2 1'2 1'7 3'2 5'1 5'0 5'1 4 2 '7

77 85 19'2 - - - - 0·6 3'9 8'3 8'3 5.6 1'9 0'3 0'2 29'1

75 7 8 7'2 .- - - - 0'2 2'9 10'9 19·6 10'0 3'7 0'2 0'1 47.6

79 79 6'9 0'1 0·8 1'5 4'2 5'4 4.6 5'2 5.8 4'9 4'3 1·8 0'3 38 '9

80 81 8'7 0'1 0·8 2'0 3'5 6'5 3'9 5'0 4'9 4'4 4'7 1'9 0'4 3 8 '1

77 7 8 6'9 - - - 7'0 2'2 3'5 2'0 0·6 0'5 0'7 0'4 - 16'9

82 80 5'7 6'1 4'1 7'5 4'7 0'1 - 0'3 0'5 0·6 0'7 1·8 4.8 32'0

79 7 6 13'0 8'5 7'0 4'0 2'9 6'3 O~I - 0'1 0'2 2'1 1'9 6'4 33'5

70 65 14.6 7'5 7'4 4'5 1'0 0'5 - - 0'1 0'3 1'1 3'7 5.8 3 1 '9

65 67 4'0 0'9 0'3 0·6 1·6 2·8 4'0 4'3 4'2 4'1 3.8 3'3 1·8 3 1 '7
81 80 6·6 9.8 8'3 8'3 4'0 2'1 0'3 0'2 1'1 2'0 4'5 5'9 ·8'1 54.6

80 7 2 18'0 5'5 5'1 4'3 5'2 4'6 2'7 2'2 1·6 3'1 5'5 5'9 6'2 5 0 '9

54 60 10'9 0'2 0'2 0'5 0·8 1'9 3'9 4'5 4.6 3'9 1,6 0'5 0'2 22·8

54 61 11'0 0'4 0'3 0'4 1'1 3'5 6·8 5.8 6'0 5'5 2'5 0'9 0'4 33'6

4 '3 0'1

I::
69 II'O 0'1 0'2 0·6 1'4 3'4 6'3 3'7 4'5 9 2'1 0'3 27'0

54 8'0 6'5 4.8 3.6 2'0 0'2 0'2 0'2 0'2 0'9 1'3 2'7 4'6 27'2
CHAPTER VII
TROPICAL MONSOON CLIMATES
LTHOUGH in matters of detail the monsoon climates

A present many complexities, the fundamental principle,


that of a land and sea breeze on a gigantic scale and
with a yearly periodicity, is simple and straightforward.
CAUSES OF THE MONSOONS. The excessive heating of the
great land masses in tropical latitudes while the sun is overhead
(Central Asia is ISO hotter than the normal for its latitude) sets
up low-pressure systems which exceed that of the equatorial
belt, and thus tend to become the focus of the trade winds of
the opposite hemisphere., The equatorial zone of calms is entirely
obliterated and a continuous current of air flo.ws from the southern
hemisphere Horse Latitudes to this continental low pressure.
Conversely, the excessive cooling during winter months (Central
Asia is 25° cooler than the normal for its latitude) sets up a high-
pressure system which exaggerates the nbrmal high pressure of
these latitudes and so gives added impetus to the outflowing
trade winds by increasing the equatorward pressure gradient.
The normal planetary regime thus disappears and its place is
taken by a seasonal and regional regime in which the contr,,!-st
between north and south, which normally controls the pressure
belts, yields place to a contrast between land and sea. Resulting
from th~s seasonal reversal of wind direction there occurs a seasonal
contrast of marine and continental influences, of excessive rain
and excessive drought, of abnormal heat and, usually, abnormal
cold, the change from one to the other being always distinct
and frequently abrupt.
With certain exceptions, namely, those stations such as the
west coast of Japan and Annam, whose situations cause them
to receive the winter monsoon over the sea, the maximum rainfall
is received during summer; in fact the winter is characterized
by partial or complete drought. In this respect the monsoon
climates resemble normal tropical climates, but they differ
markedly in the amount of rain received and in the incidence
of such rain within the rainy season. They are also peculiar
in that the heaviest rainfall generally occurs on west coasts,
especially when mountainous, e.g. the Western Ghats, Arakan and
108
TROPICAL MONSOON r09
Tenasserim, thus providing an exception to the general rule
that west coasts in the trade-wind belt are dry. A few regions
with peculiarly favourable situations receive rainfall during the
prevalence of both monsoons; thus Madras has nine months
of adequate rainfall. .
RAINFALL AND VEGETATION. The length of the dry season
thus varies according to local factors, especially the relation of
relief to wind direction, and this variation is reflected in a variety
of vegetation types, ranging from desert to dense forest. The
monsoon rain forest rivals the selvas of the equatorial zone in
density and wealth of foliage, and it yields many trees of com-
mercial value; where the dry season is longer, a deciduous forest
is found, adapted to more or less prolonged drought, a period
of enforced inactivity.
The coincidence of heat with adequate, or even abun-
dant rain' in summer makes these climates agriculturally
very productive, rice, oil-seeds, tea and jute being character-
istic monsoon products. A luxuriant vegetation of tropical
aspect flourishes far to the north, while rice is grown suc-
cessfully ~t Niigata (380 N.) and tea is an important crop in
Honshiu.
The success of agriculture is dependent on adequate rainfall
from the monsoon, adequate both in amount and distribution.
In this connection it is important to realize the phenomenal rate
of fall of much of the monsoon rain, downpours being usual in
which run-off is so rapid that only a small proportion is effective
for vegetation or irrigation. The average fall per rainy day at
Cherrapunji is 2·6 inches, in Bengal "7 inch, and in the Deccan
'3 inch or '4 inch. These should be compared with an
average intensity in the British Isles of '1 inch per rainy day.
The torrential falls implied by these figures are powerful agents
of denudation, causing great soil waste on slopes and sometimes
washing away roads, railways and bridges. Much of the rain
which falls is thus unnecess<;lry and incapable of utilization, but,
given storage, it provides a margin of safety in the event of
partial failure, while it is those areas which have just enough
which are the most subject to drought and famine when the rains
are below normal.
RAINFALL AND IRRIGATION. Since, for example, three-
quarters of the population of India is agricultural, and since.
rainfall means crops and crops mean food, the rainfall distribu-
tion determines the supporting power of the land. Where
rainfall is definitely,deficient there is a reliance on artificial water
supplies, and all cultivated land with less than about 12 inches.
must be irrigated. But where more than 20 inches occur there
no CLIMATOLOGY
is a tendency to rely on rainfall, a proceeding which is only
really justified by a rainfall of about 50 inches.
RELIABILITY OF RAINFALL. It is precisely in this zone of
just adequate rainfall that havoc may be wrought'by deficiency,
and the danger is greater or less in these areas according to the
reliability of the rain. The percentage variability is naturally
greatest where the rainfall is least, and stations in Sind some-
times receive three years' rainfall in 24 hours; but between the
critical figures of 12 inches and 50 inches there is a considerable
range of reliability. Thus:-
Lahore (15 inches June-September) has a percentage variation
of 38.
Delhi (24 inches June-September) has a percentage variation
of 29.
Calcutta (51 inches June-September) has a percentage varia-
tion of 16.
A deficiency of 25 per cent. will cause injury to crops, a
deficiency of 40 per cent. will spell ruin, and this is true whether
the normal rainfall is 15 inches or 50 inches, for the crop grown
will have been selected with a view to a normal rainfall.

The Asiatic Monsoon.


The interference with the planetary circulation exercised by
land masses is everywhere evident, but the extent of the inter-.
ference is proportional to the size and compactness of the con-
tinental block. It is natural, therefore, that Asia should be the
continent of monsoons par excellence. I ts enormous bulk, com-
prising about half the land mass of the world, its east-and-west
grain (see p. 41), its barren interior plateaux, its temperate and
sub-tropical latitudes, the huge expanse of warm seas which
flank it on the south and east, all combine to produce that contrast
between continental and oceanic influences which lies at the root
, of monsoonal phenomena.
Throughout the south-eastern angle of Asia, from Karachi
to' Vladivostok, and extending far into the interior, this seasonal
reversal of controls is the dominant feature of the climate, but
nowhere is the contrast so marked as in India. Lying between
the greatest land mass and the warmest sea, shut in by the
loftiest mountains, backed by the highest plateaux of the world,
it provides the best conditions for a monsoon. Although not a
complete barrier between India and Central Asia, the Himalayas,
the Sulaimans and the Burmese chains exclude external influences
to such a degree that the great summer low pressure of Sind
must be filled by winds from the south, and only from the south.
TROPICAL MONSOON III

India, and especially North-west India, thus becomes virtually


the terminus for ocean winds in summer and the centre of dispersal
of land winds in winter.
Within an area of varied relief embracing Ii million square
miles and extending through 30° of latitude there must naturally
occur a great variety of climatic types, ranging from continental
to marine and from arid to humid; but its almost complete
detachment, by virtue of its natural frontiers, from the rest of
Monsoon Asia makes India, as a climatic unit, capable of
independent treatmerH
India
THE SEASONS OF INDIA. Throughout most of India the year
can be divided into three seasons : -
q(~) The Cold SeasQn, from mid-December to the end of
February. .
(2) The Hot Season, from March to the end of May.
(3) The Wet Season, from June to mid-December.
The first of these is the season of the land monsoon and the
last that of the sea monsoon, while the second is in many respects
transitional. Rain falls, of course, in all these seasons, and
some districts actually have their maxima outside the 'Wet
Season " but the term' Wet Season' is well deserved, for 85 per
cent. of the rainfall of India is recorded during the summer
monsoon. 1
The Wet Se,ason is usually further divided into two: (I) the
season of the advancing monsoon, from June to mid-September;
and (2) the season of the retreating monsoon from mid-September
to mid-December. Since the rains leave the north-west plains
in mid-September, but linger in the southern peninsula until
mid-December, it is clear that no fixed date can be named for
the change from the Wet Season to the Cold Season, for this
event occurs later and later towards the south.
Thus we recognize a fourfold division of the Indian climatic
year, and it will be convenient to consider each of these in turn
and trace in each the changes of temperature, pressure and rainfall
and their interrelations.

Th(~
Cold Season
JANUARY PRESSURE AND WINDS. The main centre of the
1 The terms' summer' and' winter' monsoon are used here in prefer-
ence to ' South-west' and' North-east' because these directions are only
true over limited areas, the so-called South-west Monsoon, for example.,
blowing up the Ganges valley is a south-east wind.
IIZ CLIMATOLOGY
Asiatic high pressure in January, exceeding 30'5 inches, is
centred over Mongolia, and from it cold, heavy air is dispersed;
but very little of this reaches India, and such air as dQes descend
{rom the plateau is well warmed adiabatically in its descent of
the Himalayas. India, therefore, never experiences those waves
of intense cold which visit China, even as far south as Canton.
A secondary high-pressure system is over Kashmir and the
Punjab (see Fig. 34), from which centre pressure decreases fairly
uniformly southward to the low-pressure belt ,now south of the
Equator. It is this Punjab high pressure which controls the
'prevailing wind direction and causes north-west winds down
the Ganges valley, recurving to north-east in the Carnatic and
becoming almost easterly in Cochin. In Sind and Gujarat
winds are north-easterly; still controlled by the Punjab high
pressure, and in Central India they are light and variable. But
nowhere during the Cold Season are the winds strong or constant
in direction, light breezes being characteristic. Under this anti-
cyclonic regime the air is dry, fresh and invigorating, the sky is
clear and of a perfect blue and this is the most pleasant time of
,the year.
RAINFALL. The stability of the air during the Cold Seasop,
and the off-shore direction of the wind are factors unfavourable
to rainfall, and this season is, with few exceptions, a dry one.
These exceptions are (r) the tip of the peninsula and Ceylon,
and (z) along the foothills of the Himalayas and the margin of
the plains in the Punjab and extending into the United Provinces.
Rainfall in these two areas is fundamentally different in origin,
that of Ceylon being c~y-ec;.lional, that of the Punjab cyclonic.
Owing to its low latitude, Ceylon is not far removed from
the equatorial rainfall belt, and slight northward aberrations of
this bring rains to the island and to the tip of the peninsula.
This rainfall is particularly heavy on the east side of the island,
on which the N.E. monsoon impinges, moisture-laden from a
long journey over the Bay of Bengal. ,
The rainfall of the Punjab and the United Provinces is
,associated with shallow depressions whose nature and origin
still provide something of a mystery. Some have been traced
from the Mediterranean winter storm-track via Palestine and
Afghanistan, others may originate locally in the highlands to
the west of India. The Asiatic high pressure persuades them to
take a course well to the south, and the Himalayas later restrict
their movements and steer them into the plains. They travel
eastwards bringing one or two inches, of rain to the northern
plains and heavier falls to the higher slopes of the mountains;
Peshawar receives more rain at this time than from the i;)l,lmmer
TROPICAL MONSOON II3
monsoon, Murree receives 6 inches in January and February
and Simla 7 inches, while heavy falls of snow occur higher up,
this being the chief time of renewal in the zone of permanent
snow in the Western Himalayas. Southwards this rain extends
into Rajputana, Central India and the Central Provinces, but
the amount dec_reases markedly in this direction and the storms
become more violent; frequently they are accompanied by
thunder and hail which does much harm to crops. They die
out eastwards and bring very little rain to the Ganges valley.
Small in amount though this rain is, it is of the greatest
importance to the wheat and barley crops of the Punjab, and
g;ea~ distress is caused by its occasional failure. The comparative

FIG. 34.-Mean Pressure and Winds FIG. 35.-Cold Weather Rainfall (Jan.
(January) _ and Feb.) and Cold Weather Storm-
The length of the arrow is proportional to the tracks
steadiness of the wind. .

lightness of the fall increases its value inch for inch, as run-off
is slight-. and a high proportion is available for crops.
The barometric gradients of these storms 'are very slight, but
remarkable temperature changes are experienced during their
passage. The approach of the storm is heralded by a shift of
the wind to the south-east, bringing a current of warm, moist
air, whose temperature is further raised by the liberated heat
of condensation and by the blanket of cloud. With the passage
of the storm centre the wind veers to north-west and the tem-
perature may drop as much as 20° in 48 hours. The trajectory
of this north-west wind brings it froin the snow-clad slopes on
to the plains, and the temperature is further lowered by evapora-
tion. It is during these cold waves in rear of depressions that,
temperatures below freezing are recorded in the Punjab, and it
CLIMATOLOGY
is under these conditions that severe damage is frequently done
to fruit and crops.
COLD SEASON TEMPERATURE. Latitude is the principal con-
trol of temperature at this time, the January isotherms being
directed almost due east and west; the mean temperature of
day in Ceylon is about 80°, in the Punjab about 50°. But more
significant is the. diurnal range of temperature, strikingly related
to marine influence through the medium of relative humidity.
The thermometer often reaches 80° in the Punjab during the
day, while frosts at night are common occurrences, and fires1are
lit every evening well into February. This rapid nocturnal
radiation of heat is the direct result of the anticylonic conditions

FIG. 36.-Diumal Range of Tempera- FIG. 37.-l'IIean Daily Relative Hum-


ture (January) idity (January)

with concomitant outflowing winds, dry air and clear skies.


Round the coasts the diurnal range is less and frosts are unknown.

The Hot Season


TEMPERATURE. From January onwards, as the sun continues
its northward march, the temperature begins to rise, and a
tendency begins to be manifest for land and sea to assume control
and for isotherms to curve round over the land. Thus the
Februa,ry chart shows a circular isotherm of 80° in the centre
of the peninsula, and by March the peninsula is almost defined
by the closed isotherm of 82'5°, in the centre of which is a circular
area exceeding 87'5°. In April the 90° isotherm is found parallel
.to the coast and about 100 miles inland and in May a large
area in Central India and the Central Provinces exceeds 95°·
TROPICAL MONSOON lIS

:By now the heat is terrific, shade temperatures of 120° are fre-
quently recorded in Sind, and Jacobabad, which has the evil
reputation of being the hottest station in India, records a mean
daily maximum. above lI~o.
The heat and glare of the sun, the wafts of suffocating heat
reflected from the hard-baked ground or from walls compel
all outdoor life to cease while the sun is above the horizon,
and interior~ are only kept bearable by shutting all windows
and doors to keep out the stifling heat, and by the use of tatties,
matting sprinkled with water through which air is drawn and
cooled in passage by evaporation. Outdoor exercise is taken in
what is by courtesy called' the cool of the evening', but sun-

FIG. 38.-Mean Temperature (May) FIG. 39.-Mean Maximum Tempera-


ture (May)

stroke is the immediate penalty 'for venturing out during the


day without sunshades or ample covering for the head and
spine. All vegetation is withered and 'scorched by the desiccating
heat, all water is dried up and the rivers have shrunk to thin
threads lost in a wilderness of dazzling, shimmering sand and
stones.
On the coast such excessive max~mum temperatures are not
recorded, but the heat is no more bearable, because it is moister.
Relative humidities as low as I per cent. may occur in Sind, but in
the south the inflow of sea breezes keeps the air fairly humid.
For the same reason the loss of heat at night by radiation is
less, and nights are intolerably stuffy and airless. The diurnal
range in Sind in May exceeds 30°, while in Madras it is less than
20° and in Bombay only IS°.
PRESSURE AND WINDS. The gradual and steady rise of
II6 CLIMATOLOGY
temperature in the interior of India sets up convection currents,
which, as would be expected, attain a maximum during ,the after-
noon. The Punjab high presssure of January gradually breaks
down, and by April has entirely disappeared, its place being
taken by a low pressure extending throughout the plains. • By
May there is a deep low (29.6 inches) centring round Multan,
and during the daytime this intensifies to 29'5 inches, A further
intense low is established round ]ubbulpore, and these induce a
system of on-shore winds round the whole of the Indian coast.
It is during the climax of the Hot Season, more than at any
other time of the year, that India is an independent climatic
unit. This is a transition period between the winter and summer
monsoons, a time when local factors are able to make them-
selves felt above the planetary controls. Air circulation is
cyclonic and feeble, and conditions are ideal for the generation
of local thunderstorms. Temperature and winds are locally
controlled, and such storms as occur are of local origin, being
due to the great changes of humidity and temperature rather
than to outside influences. In this they contrast with the cold-
weather storms which enter from the western plateaux, and wiih
the storms of the rainy season, which originate in the Bay of
Bengal or in the Arabian Sea.
STORMS AND RAINFALL. Storms are fairly frequent and
widely distributed in the plains during the Hot Season, but it
is not everywhere that these storms bring rain. The dust storms
of the North-west, the 'Nor'westers' of Bengal and the rain
storms of Assam all appear to be similar in origin, since all
originate at the junction of two currents, a dry, cool upper
current from the north-west riding over the shallow surface
flow of air drawn in by the low pressure of the plain.
The arrangement is extremely unstable}l.and sudden inversions
are apt to occur, there is an uprush of warm air and thunder-
storms, often with hail, are the result. In the Punjab, Sind,
Rajputana and the western plains these storms bring no rain;
they are merely dust storms which, though bringing a welcome
cooling wind, darken the sky almost like night, filling the air
with blinding dust. In Bengal the sea breezes bring more mois-
ture because nearer to the sea; the ascensional movement here
frequently results in rapid condensation 'in the form of heavy
rain or hail with lightning and thunder. The hail stones are
sometimes of prodigious size, they often ruin the standing wheat
and have been known to kill men and animals. But it is' in
Eastern Bengal, and especially in Assam, that these storms are
valuable as rain-bearers; Dacca gets 18 inches in this season,
and parts of Assam as much as 20 inches, chiefly in the form of
TROPICAL MONSOON
afternoon thunderstorms. The economic value of this rain is
incalculable, it nourishes the spring rice crop in Bengal and it
brings the first flushes of green leaves to the tea plantations of
Assam. The storms become more frequent as the heat increases,'
as the Indian low pressure becomes more intense, and as the
inflow of sea breezes becomes stronger and their humidity greater.
Thus Calcutta receives I'4 inches in March, z inches in April,
and 6 inches in May; Cherrapunji has 9 inches in March, 30
inches in April, and 50 inches in May.
The only other region which experiences considerable rain at
this period is Ceylon and the tip of the peninsula, especially on
the western side. Colombo has 5 inches in March,' 8 inches in

FIG. 'to.-Hot Weather Rainfall (March-May)

April and I3 inches in May; in fact May is the wettest month


of the year, considerably exceeding the' monsoon' months of
June and July. A second maximum is reached in October and
November, during the retreat of the monsoon. These rainy
seasons may be considered as an invasion by the equatorial rainfall
belt. In contrast, Cochin, 8° further north, shows a more typical
monsoonal regime with a pronounced maximum in June, but
there is also a considerable fall (19'2 inches) in the three months
of March, April and May. These rains .are known as 'mango
showers' in Mysore and as ' blossom showers' in coffee districts,
and, as their names imply, their agricultural importance i~
great.
,8
II8 CLIMATOLOGY

The Rainy Season


PRESSURE AND WINDS. The prolonged heat and drought
from which the greater part of India has been suffering for some
months has generated increasingly low pressures over the plains
until June and July. Throughout all this time the form of the
isobars has changed scarcely at all, yet the great climatic event
of the year has arrived, ' the bursting' of the monsoon in early
June. In July, as in May, the isobars are arranged concentric-
ally around the north-western plains, and the wind direction,
controlled by this local low pressure system, is to all appearance~
the same in each month. Yet Bombay, which has had five
months without any measurable rain, and which records· only
t inch in May, is suddenly drenched with 20 inches in June and

FIG. 4I.-Weather on 14th May, 1897 FIG. 42.-Weather on 3rd July, 1889

24 inches in July. Examination of the isobaric map reveals the


information that although the direction of the pressure gradient
is the same in July as in May, ther.e is a marked difference in
intensity. The lowest mean pressure of May is 29'55 inches, but
in July it has intensified to 29'4 inches; the gradient from Sind
to Ceylon in May is '2 inches as tompared with '4 inches in July.
Thus winds are stronger and more constant and bring in more
moisture. But more significant than the pressure distribution
over India itself is that over the Indian Ocean. Figs. 41 and 42
show typical charts for May and July, and the essential differences
are at once revealed. In the May chart it is clear that India's
weather is the product of local conditions. There is still a
tendency to an equatorial (doldrum) low-pressure zone, which
a;ttracts degenerate trade winds from both hemispheres. Between
this and the Indian low pressure is an ill-defined area of higher
TROPICAL MONSOON
pressure which is the source of India's air supply. This air
current is bringing no appreciable rain, such falls as occur being
the products of local storms. Turning to the July chart it is
noticeable that the equatorial low pressure- has vanished, that
the gradient over the ocean is steady and steep and that there
is therefore a continuous stream of air from the southern Indian
Ocean to the Indian focus. The explanation, then, of the sudden
burst of the monsoon lies in the eventual overcoming of that
light, unsteady circulation in the Indian Ocean and the union
of the S.E. trades, deflected right-handedly on crossing the
Equator, with the local winds of India. This reinforcing current
from the southern hemisphere brings with it enormous reserves
of moisture, collected in its 4,000 mile journey over the warmest
ocean. The relative humidity at Bombay in May is 74 per cent.,
and in July 86 per cent. Further, the increased pressure gradient
gives increased wind velocity and therefore a continuous and
constantly renewed supply of. rain; the mean wind velocity at
Bombayis 7'4 in May and I4'2 in July. It is the sudden arrival
in India of this moisture-laden current which constitutes the
, burst' of the mOI!soon. Up to this point the intense heat and
dfyness of the air under the cloudless summer skies in India
have been capable of. absoroing the comparatively small supply
of moisture which the sea breezes could import.
BREAKS IN THE MONSOON. If, as sometimes happens, there
is a temporary reversion, after the arrival of the monsoon, to
'the pressure conditions of May, there will also be a reversion
.to the May rainfall, that is, a break in the rains. The rain-
bearing winds will be lost in the vague pressures over the ocean
and drought and famine will result in India if the conditions are
prolonged. It was a pressure distribution of this type which
led to the disastrous famine of I8gg. -
THE' BURST' OF THE MONSOON. Although there may be a
few days of light rain preceding the burst of the monsoon, this is
usually in the form of an intense cyclonic storm with thunder and
lightning and torrential rain, arriving with dramatic suddenness
to relieve the parched and thirsty earth. At sea the winds
attain great violence, native shipping does not venture out to
sea, while larger ships experience some very rough passages. All
the west-coast harbours, poor and few as they are, are closed,
with the exception of Bombay; sea-level rises all along the coast,
the water is driven up the flat shores of Cutch and Cam bay and
the Ranns of Cutch are flooded. The opening storm passes away
and may be followed by longer or shorter periods of fine weather,
to be followed in turn by further cyclonic storms. This pUlsatory
nature of the rain is extremely characteristic. It is to be attri-
r20 CLIMATOLOGY
buted partly to the essentially cyclonic nature of much of the
rain; and partly to a kind of self-feeding action, whereby the heat
liberated by the heavy precipitation lends further energy to the
system, increasing the ascensional movement and leading to a
crescendo of violence. This violence soon exhausts the moisture
reserve, and further rainfall has to await the arrival of fresh
moisture-laden currents from the sea. Slowly fresh energy accu-

J F M A M J J A SON 0 J F M A M J J A SON D

85
80
75
70

Bombay
75

55

Jacobabad (Quetta 1

FIG. 43.-Annual March of Temperature and Rainfall in India


(Note that the scale of rainfall used in the graphs for Jacohahad and Quetta is five times that used fOI
. Bombay and Nagpur) .

mulates as the humid currents come pressing inland and the pro·
cess is repeated.
EFFECT OF THE RAINS ON TEMPERATURE. The immediatE
effect of the rains is to lower the temperature, both by conduction
and evaporation, and to make the climate of the plains endurablE
again. The mean temperature at Bombay in May is 85·8°, and
in July 8r'4°, a fall of more than 4°; at Nagpur there is a fall
from 95° in May to 8r'7° in July, more than r3°, and falls of 6° 01
'}O are usual. Fig. 43 gives some idea of the relief which this droI=
TROPICAL MONSOON I21
of temperature affords .. The line of the temperature curve in
April and May for Bombay and Nagpur may be continued in
imagination to a hypothetical maximum in July, such as would
be reached were it not for the cooling effect of the rain. An
approximation to this 'typothetical curve is seen in the graph
for Jacobabad which receives only just over 3 inches of monsoon
rain. The curve continues to rise into June, when the mean
temp.erat!-lre reaches nearly to TOOo, and eV'en here there is a
'slight depression of the crest of the curve when the rains come.
At Quetta (whose climate is, of course, not monsoonal) there is
practically no June-September rain and therefore no cooling effect.
The curv~ continues to rise until a normal maximum is reached in
July, ne.atly 80° in spite of the altitude (5,502 feet). This flat-
tening of the top of the temperature curve in the rainy season has
been observed also in tropical climates, where rainfall, as here,
coincides with the overhead sun. Physiological comfort is not
increased to the extent that might at first sight be anticipated,
since the decreased temperature is coincident with increased
humidity; moist heat replaces dry heat and nights, in particular,
are oppressively sultry. Just as the arrival of the rains depressed
the temperature, so their withdrawal, three or four months
later, is often accompanied by a reassertion of the upward
tendency of the temperature graph, and a second maximum is
frequently attained in September or October. At Bombay the
temperature rises from 80·9° in September to 82~4° in October,
but this check in the fall of temperature is not so universal as the
earlier check, to the rising thermometer. The .hot sun beating
down on the saturated ground, .after the rains have left, draws up
mists and vapours from the swamps, with their attendant fevers
and rheumatism which, to an appalling degree, sap the energies of
the Indian peasant. The prevalence of irrigation works, tanks,
wells, canals and paddy fields increase the humidity and provide
greater space for the breeding of mosquitoes.
DATE OF THE' BURST '. The average date of the burst of the
monsoon in Malabar is June 3rd, at Bombay the 5th, in the Central
Provinces the TOth, in Bengal the I5th, in the Eastern United
Provinces the 20th, and at Delhi about the 30th; but its arrival
may be delayed as much as three weeks. By the end of the month
the rains are well established everywhere, maximum falls are
generally recorded in July and August and by the first week in
September the rains begin to diminish. The monsoon current
appears to be a shallow one, since the course it takes is profoundly
influenced by the shape of peninsular India, whose relief is not,
in fact, extremely mountainous. Coming from the sQuth-.west"
the air-flow divides into two branches, an Arabian Sea branch and
122 CLIMATOLOGY
a Bay of Bengal branch, and it is the relation between the courses
taken by these currents on the o'ne hand and the relief of the l~,nd
on the other that determines the distribution of rainfall at ~his
season. The rainfall is due almost entirely to ascent of m<;>istm:e-
laden air, and this ascent is due to one of three causes :-
(I) Relief-whereby there appear striking contrasts between
windward stations and stations in the rain shadow in
lee of mQuntains.
(2) Convection-whereby there is a tendency to afteriroon
storms. '.
(3) Cyclonic storms-which give the most evenly distributed
rainfall of the three. ,
ADVANCE OF THE MONSOON. The two currents into which
the monsoon divides (see Fig. 44) may be treated separately' and
the rainfall due to each described in turn. The goal in each ca.:_se
is the low pressure of the Indo-Gangetic Plain, culminating in the
intensely low pressure over Sind. But the heart of the system
varies in position from day to day and so influences the trajectory
of the winds and the courses of cyclonic storms. .
THE ARABIAN SEA BRANCH expends its full force on, the
Western Ghats which stand directly athwart its path. Here'it
must rise through a height of from 3,000 to 7,000 feet and the
resultant precipitation is enormous. More than 30 inches fall in
June along the coastal plain and more than 40 inches in July;
higher up the slopes the rainfall is still heavier, Mahabaleshwar
(4,540 feet) records an average of over IOO inches in July and over
250 inches during the five months of the monsoon. Here rain
falls, on an average, on rr6 out of the 122 days in June-September
and nearly 2! inches fall on each rainy day,-more than London
gets in the whole of July. But these heavy falls are experienced
only along a narrow strip, in fact along the western slopes; for
after crossing the summit the wind descends the eastern slopes as
a foehn wind. Thus there exists a remarkable rain shadow
immediately in lee; Gokak, only 65 miies from Mahabaleshwar,
has only 22 inches a year; Dhulia has 22 inches; Bellary has only
IS inches, of which only 9 inches fall in June-September and only
I'3 inches in July. The rain shadow effect is still more pronounced
in the south where the mountains are higher, Tinnevelly, in lee of the
CardaII!Q.I!!,Hills, which rise to Io,ooofeet, has less than I inch inJ uly
an«only 2'7 inches from June to September. At their northern
end the Western Ghats decrease rapidly in height and the rainfall
decreases coincidently (Surat 40 inches, June-September). At thE
same time the contrast between windward and leeward ceases tc
exist, as there is no obstacle to the rain-bearing winds; 40 inche~
v of Irain from the monsoon is recorded far inland, while 30 inche~
TROPICAL MONSOON 12 3

is recorded in a continuous strip from the Gulf of Cam bay to


the Gangetic Plain at Cawnpore. .
NORTHERN LIMIT OF THE RAINS. Northwards from the Gulf
of Cambay the monsoon rainfall decreases still further, 29 inches
at Ahmedabad, IS inches at Bhuj in the island of Cutch, 7 inches
at Karachi, and practically none in Baluchistan. The explana-
tion of this is to be found in two factor:s: (I) the trajectory of the
air currents and (2) the dryness of the air of Sind. The eastward
deflection of the air current in the Arabian ~ea, seeking the focus
of Sind, has become greater and greater further north, until it has
practically lost its northward component and the air is moving east
or even south-east. The rain-bearing winds thus reach only to the
south-west tip of Arabia and fail to reach Persia and Baluchistan
whose aridity at this season is complete. An east to west line
. drawn near Karachi divides those winds to the south which have
travelled over the Arabian Sea from those to the north which
have travelled over the arid lands of Arabia and Baluchistan, and
this line practically marks the limit of monsoon rainfall.
, THE BAY OF BENGAL BRANCH flows up between Ceylon and
Sumatra, swinging northward and westward under the attraction
of the Sind low pressure, and travelling up the Gangetic Plain
towards this destination. A branch of this current impinges on
the mountainous shores of Arakan and Tenasserim in a manner
analogous to the impact of the Arabian Sea branch against the
Western Ghats, and in a like manner there results torrential rain
along a narrow coastal zone. Akyab has a mean annual precipi-
tation of over 300 inch~s, of which about 170 inches fall in the four
monsoon months from June to September. As would be expected,
there is a region of pronounced rain shadow in the middle Irra-
waddy valley near Mandalay, which receives only 19 inches in the
four months from June to September; but a branch of the
Bengal current penetrates some distance up the Irrawaddy from
the delta, and brings 40 inches of monsoon rainfall as far north as
Thayetmyo. Under circumstances in general similar to those of
Arakan some of the heaviest rainfall in the world is recorded in
the re-entrant between the Khasi and Lushai Hills in Assam,
where the sudden forced ascent of air currents, confined between
the convergent mountain walls, is responsible for the prodigious
fall at Cherrapunji (4.455 feet) of 450 inches .a year, of which
nearly 350 inches occurin the four months June-September. Over
900 inches (25 yards) of rain have been recorded here in one year
and 40·8 inches in one day (June 14, 1876). Beyond the sum-
mit of the hills this figure falls off rapidly, and Shillong, 25 miles
away on the plateau-like top of the Khasi Hills, :p.as only 55 inches
in. June-September.
I24 CLIMATOLOGY
RAINFALL OF GANGETIC PLAIN. But the most valuable rain
in India results from that branch of the Bengal current which
advances up the Gangetic Plain towards the Punjab. The very
gradual rise in altitude along this track allows an even distribution
of rain, infinitely more favourable to man than the torrential falls
on the Western Ghats with their inevitable zone of deficiency in
lee. This rainfall is partly due to relief, but more e;>f\cially to
cyclonic storms. which follow the track' of low relief and low
pressure, particularly along the southern margin of the plains.
It should be noticed that this current is actually a S.E. wind,
although a direct continuation of the S.W. monsoon. The rain-
fall from this current decreases westwards and southwards;
westwards because further away from the source of moistvr~,
southwards because further away from the precipitating agent, the
Himalayas. There is a tendency for this current to bear against
the Himalayan foothills and so to bring heavier rains there at this
season; compare Benares (35 inches June-September) with
Gorakhpur (42 inches), Agra (23 inches) with Bareilly (36 inches)
and with Naini Tal (8I inches). The hill stations not only get
heavy rain (Darjeeling I02 inches June-September), but are also
enveloped in cloud throughout much of the rainy seas.on. The
zone of maximum precipitation lies at about 5,000 feet, beyond
which rainfall decreases, while beyond the first ranges the pre-
cipitation is very slight. Shrinagar has only 8 inches and Leh
only 2 inches. The decrease from east to west is still more pro-
nounced, Calcutta has 46 inches from June to September, Patna
4I inches, Allahabad 33 inches, Delhi 22 inches, Lahore I6 inches
and Multan 5 inches. The rainfall in the extreme north-west is
extremely variable, and its arrival or failure depends on the
strength of the Ganges valley branch of monsoon. This in turn
depends on the pressure gradient in the Gangetic Plain, the focus,
as in the case of the Arabian Sea branch, being Sind. When the low
pressure of Sind is well developed, rains extend far to the north-
west, but it occasionally happens that relatively high pressure
develops in the Punjab, and reverses temporarily the wind direc-
tion in the United Provinces and the surrounding country, which
now receives north-west winds of an anticyclonic type. There is,
in effect, a reversion to the Cold Season conditions, and drought
results. These dry north-west winds, which might be considered
as a prolongation of the Arabian Sea branch, which has travelled
ov~r Baluchistan and Sind, sometimes possess the plains as far
as Bengal and extend southwards into the Deccan as far as
Hyderabad. It was a pressure distribution such as this which
was responsible for the disastrous droughts of I876, I877, I880
and I883.
TROPICAL MONSOON 125

The tracks of the two principal currents of the monsoon have


now been traced, but there remains an interesting area at the
junction of the two.
STORM TRACKS. Throughout the peninsula the south-west
direction of the Arabian Sea branch holds sway, while in the
Gangetic Plain the air movement is from south-east to north-west;
the debatable ground between the two comprises the southern
margin of the Gangetic Plain from Orissa, through Allahabad
and Agra to Patiala. It is a favourite track of cyclonic storms
which originate in the head of the Bay of Bengal and pass west-
ward along the foot of the peninsular hills, accompanied,
especially on their southern side, by heavy rainfall, which often
causes sudden floods on the Narbada and Tapti. This rain is

FIG. 44.-Streamlines during Monsoon FIG. 45.-Monsoon Rainfall (June"'::Oct.)


(After Simpson)

of value since the southern side of the plains would be deficient


without it, and it is especially valuable to the rice-growing dis-
tricts of the Central Provinces.
THE SIND Focus. Bearing in mind that the goal of both
IJlonsooh .currents is the low pressure of Sind, and that this is the
temporary equivalent of the equatorial low-pressure belt, it is
not a little surprising that the ascensional movement here does
not give rise to heavy rainfall. But actually the site is one of
the world's most complete deserts, and to this condition a variety
of factors contribute. In the first place, of the air currents which
arrive here, those from the south-east have already shed nearly
all their moisture in their passage up the Gangetic Plain; those
from the south-west have passed from a warm sea to an intensely
126 CLIMATOLOGY
hot land, and so decreased their relative humidity from go per
cent. to about 55 per cent. But even this air would become
saturated and clouds would form at about 3,000 feet were it not
for another important factor, namely, the existence of an upper
current of hot dry air from the arid plateau of Baluchistan, which
greedily absorbs all available moisture and keeps the skies free
from the slightest cloud. Finally, the clear, cloudless air and the
scorching rays of the vertical sun on the sandy desert coflspire
to maintain that dryness of the air which makes rainfall almost
an impossibility.
RETREAT OF THE MONSOON. Within three weeks of their
appearance off the west coast in early June the rains have spread
over the whole of India, but the retreat is a much slower process
and occupies about three months. The rains slacken and cease
in the northern plains towards the end of September, but they
linger in the tip of the peninsula until the middle of December.
Slowly the northerly winds extend southwards, driving back the
southerly rain-bearing monsoon, and gradually from north to
south the Cold Season conditions are re-established. There is
none of that suddenness about the change which marked the
passage from the Hot to the Wet Season; for the burst of the
monsoon was due to the irruption of the rainy winds from the
southern hemisphere impatiently overcoming the resistance offered
by the feeble high pressure north of the Equator; but the retreat
is a long struggle between two well-matched but fteble currents,
the one becoming steadily stronger and the other steadily weaker
as the sun retreats towards the southern tropic. Thus dry winds
succeed to damp winds, anti-cyclonic conditions succeed to
cyclonic and diurnal range of temperature increases. The
advance of the monsoon is northwards and its retreat, from the
middle of September onwards, is to the south; the length of
the rainy season therefore decreases, as in normal tropical
climates, away from the Equator.
RAINFALL OF THE RETREAT. The same causes produce rain
in the same way as during the advance, but the area affected
grows less and less from north to south. The monsoon leaves the
Punjab about September 15, the United Provinces about Octo-
ber I, and Bengal about October 15. A premature departure of
the monsoon is a catastrophe greater than its delayed appearance;
for while the latter only prolongs the discomfort and expectant
inactivity of the Hot Season, the former dries up crops which,
have relied on a continuance of the rains. In 1883, for example,
the monsoon left Bengal a month too early; the rice crop was
ruined and serious famine resulted.
A change of wind direction from south to north-east accom-
TROPICAL MONSOON 127

panies the southward retreat of the low-pressure trough, and it is


these north-east winds which bring the October and November
rains to the Carnatic. These rain-bearing winds are not to be
considered as the N.E. monsoon, but as the retreating edge of the
S.W. monsoon; not as winds originating from the new continental
high pressure but as the northern periphery of cyclonic circulation
as')ociated with the retiring low-pressure trough. These winds,
.coming off a warm sea, bring Madras 25 inches of rain in October
and November, rainfall which is essential for the rice crop and
which fills the irrigation tanks against the dry months of
January-April.
HURRICANES. The isobaric chart for October shows a strik-
ingly uniform pressure distribution, varying little on either side
of 29·9 inches. There is a tendency to a low pressure in the Bay
of Bengal (29.85 inches), and tendencies to high pressures in
Peshawar, Rajputana and Khandesh (29·9 inches), but small
gradients resulting in feeble winds or calms are characteristic.
This is the time when violent storms brew in the Bay and travel
against the coast, doing great damage. The place of origin of
these storms'is located further and further south as the year
advances and as the calms move southwards. They follow the
recurved track, characteristic of tropical revolving storms,
travelling first westwards and then swinging round to the north
and. north-east. Storms at this time are of common occurrence,
but fortun~.ely severe ones are not frequent. The greatest
catastrophes occur when an intense cyclone arrives at the coast
while the tide is high. Then the sea water is piled against the
coast and floods alone may account for 100,000 lives, while the
epidemics which follow are even more destructive.
THE RAINFALL OF INDIA. Reviewing broadly the rainfall of
India, it is noticeable that while relief rains give the heaviest falls,
. it is the cyclonic type which, on account of its more even distribu-
tion, is of greater economic value, as, for example, in the Gangetic
Plain and the Deccan during the rainy season, and in the Punjab
and the United Provinces during the Cold Season. At no time
of the year is India entirely without rain. Taking India as a
whole: January and February are the driest months, but rain is
enjoyed in the north-west, and to a less extent in the Ganges
valley and Assam, from cyclonic storms, and in the southern:
peninsula and Ceylon from the equatorial belt .. In March, April
and May there is a dry triangle with its base from Bombay to
Karachi and its apex at Benares, but the peninsula has 10 inches
on its western side and Assam gets about 30 per cent. of its yearly
rain from thunderstorms. After the harvesting bf the 'rabi'
(spring) crop, the heat and drought make this a period of com:
128 CLIMATOLOGY
pulsory fallow in most parts, comparable to ' Charaqui ' in Egypt.
But Assam and Bengal grow rice and jute, relying on the rains.
From June to October there is rain everywhere, though less in the
north-west than elsewhere. This is the chief growing season of
India, when the land, cleansed by the Hot Season' fallow, and
watered by ,the monsoon, yields fruitfully rice, millets, maize or
cotton according to rainfall or soil conditions. These are the
, kharif' (autumn) crops, watered by the monsoon rains and
ripened by the hot, sunny weather which follows their departure.
In November and December there is a little rain all round the
shores of the Bay of Bengal, .while the £§:!natic and eastern
Ceylon gets more than 15 inches. This is a highly favoured
region, where abundant and well-distributed rain gives great
fertility, with resultant prosperity and leisure.

Abyssinia, Somaliland, Yemen


In the preceding pages India has been treated as a climatic
province virtually self-contained, but the western fringe or the
summer monsoon current is recognizable in the climate of East
Africa (see p. 82), Abyssinia, Somaliland, and Yemen. In
examining the earlier course of this current we obtain some useful
and instructive insight into certain aspects of the Indian climate
which may often be used for long-range forecasting. The July
pressure map shows that the Indian low-pressure system extends
westward about as far as the Nile, where winds are chiefly
northerly or north-easterly, while on the south side of the low-
pressure trough there blows the S. W. monsoon, the continuation
of the S.E. trades. In Somaliland (see p. I02), mainly owing to
the dryness of the air, it brings little or no rain, but the hot winds
give rise to violent sandstorms which halass man and beast and
even kill off the mosquitoes. While it blows off-shore it makes the
Arabian Sea so stormy that all native shipping is laid up; this,
is the season of 'Bat Hiddari.' or sea closed, the N.E. mon-
soon (November to March) being the season of ' Bat Furan' or
sea open. In the coffee country of Yemen, the mountains,
opposed to the wind, cause the precipitation of some 20 inches
of valuable rain, although the rest of Arabia, Persia and Balu-
chistan are beyond the influence of this current (see p. 123). A
further branch of the current supplies Abyssinia with 30 inches
or 40 inches on the highlands.
FORECASTING THE MONSOON. The S.E. trades, the Abyssinian
monsoon and the Indian monsoon are clearly all part of the same
great air flow and would be expected to wax and wane together.
It should be possible, therefore, by tracing this current backwards
TROPICAL MONSOON I29

along its course to obtain valuable foreknowledge of the all-


important monsoon rains of India and Abyssinia (the latter so
vital to agriculture in Egypt), A weakening of the system, from
causes largely unknown, will bring about drought and famine in
South-east Africa and in India, and a low Nile flood, with possibly
serious consequences, The following correlation table 1 shows
how close is the interdependence : -

Rainfall Variation,

Year, Nile Flood,


India ~,w,
(excluding Burma, India,
Burma), ,

189 1 Below Normal - 3'54 + 2'4 8 - 2'3 2


1892 Above Normal
, + 5' 0 9 - 7'28 + 6,88
1893 High, + 9'07 + 7' 0 4 + 7'53
1 894 High, + 6'47 + 11'47 + 8,84
1895 Normal - 2'19 - 11'63 - 5'20
1896 Below Normal - 4,83 ,+ 3'79 - 2,87
1 8 97 Normal ;,' - 0'15 - 0'13 - 2'03
189 8 Normal + 0'43 + 0'4 0 - 2,88
1899 Far below Normal, -11'14 + 6'33 - 15'5 6
1900 Above Normal , - 0'57 -~ 0'91 - 1'08
1901 Above Normal, - 4'13 + 0'07 - 8'18
190 2 Below Normal - 2'05 - 7'21 - 5' 1 7

It will be observed that Burma tends to vary in the opposite


direction from the rest of India especially from N,W, India, a
fact which, though not fully understood, is of some value in
forecasting, Observations over many years show also that a
high pressure in May over the Indian Ocean, as represented by
Mauritius, is followed by deficient monsoon rains in India, This
appears, at first sight, unreasonable, as a high pressure here should
strengthen the gradient and reinforce the monsoon, but experience
has shown that high pressure here coincides with high pressures
in India, which therefore offers less attraction for the monsoon,
Again it has been found that heavy rains at Zanzibar, and in
British East Africa generally, in April and May is prelude to a
poor monsoon, These rains are the convectional rains of the
doldrum belt at the change-over of the monsoons and they bear
witness to strong ascensional movement here, accompanied by a
low pressure, Now the arrival of the monsoon in India follows

1 Imperial Gazetteer of India, Vol. I, p, 128,


130 CLIMATOLOGY
the elimination of the low-pressure belt (see p. II9) and will be
delayed if this will not give way. In other words, the doldrum
belt is waylaying the S.E. trades and preventing their arrival at
the Indian focus. .

The Indo-Chinese Peninsula and the, Philippines


The position of this peninsula is somewhat analogous to that
of peninsular India and the climate presents many similarities.
The west coast and· the interior receive their heaviest rain from
the S.W. monsoo!}, but the air currents have parted witll their
rain in passing over the Malay Peninsula, and the Gulf of Siam
is not wide enough to replenish them; the rainfall is therefore
less in amount than in peninsular India. On the other hand, the
relief of the coast is much more favourable and the rain is much
more evenly distributed. The west coast of Siam from Cape

J F M A M J J A SON 0
30
J F MA MJ J A soN 0
25
20
15
10

5 ...........
Hue Madras
FIG. 46.-Rainfall at Hue and Madra.s

Cambodia to Bangkok has 10 or 12 inches in July, compared


with more than 16 inches along the Malabar cOast, but 8 inches is
recorded far into the interior, whereas in India the 4-inch July
isbhyet is not far from the dh:ide of the We~tern Ghats.
Except on the Annam coast the dry season is the season of the
N.E. monsoon and the duration of the dry season is greatest in
the north, where the rains arrive later and depart earlier. In
Cambodia there are only two or three months With less than I inch
of rain and it is a region of dense forest; the interior of Siam,
where the drought is five or six months long, is a region of tropical
grassland.- .
The Annam coast lies at right angles to the N.E. monsoon and
has its chief rains while this blows, but the rainfall regime is not
wholly accounted for by the on-shore wind, sin(~e the maximum of
rain (October and November) does not coincide with the maximum
force of the monsoon (January). By comparison with the similarly
TROPICAL MONSOON 13 1

placed Carnatic coast (p. 127) this rain would appear to be due
to cyclones associated with the retreating trough of low pressure,
but two other influences are at work :-(1) The higher temperature
of the sea with consequently greater moisture content of the
winds, and (2) the occurrence of typhoons. These typhoons
reach great violence in the Philippines which lie across their
favourite track; they are most frequent in September, Octol!er
and November, i.e. at the change of the monsoon, and add appre-
ciably to the autumn rainfall here, e.g. Baguio (see p. 17).
Chiefly owing to its more mountainous hinterland Hue (102 inches)
has more than double the rainfall of Madras (50 inches).
The great bend of the coast in the Gulf of Tongking brings its
northern shore at right angles to the summer monsoon, and the
more normal regime returns with a rainy season from May to
October and a maximum in July. But from this point northwards
the climate shows a strong continental influence, especially in
winter, and the effects of continental cyclones are felt. The
mean temperature range at Hanoi is 22° and extreme temperatures
exceed 100° and fall below 40°. These extremes ally the region
with sub-tropical China rather than with tropical Annam, and it
is more properly considered under that head (see p. 168).

Australia North of the Tropic


Matching the Asiatic continent across the Equator is the
island continent of Australia, whose monsoon is the complement
of Asia's; but the lesser bulk of the southern continent has its
sequel in a less clearly defined monsoon. Yet the climate of
Australia north of the tropic is essentially governed by the mon-
soonal wind reversal and bears a marked resemblance to India
in the following respects:-
r. A small annual range of temperature.
2. Clearly marked wet and dry season, with local exceptions
where the direction of the coast ensures rain from both mon-
soons.
3. Rainfall decreasing steadily towards the focus of the
monsoon, which, in each case, is a desert area beneath the tropic,
arid in spite of being the region of greatest convection.
The differences arise from the contrasts of size, position and
relief, as a result of which :-
r. The air flow is weaker and the rainfall therefore less.
2. The. trajectory of the air current is such that for much of its
journey it is travelling over an archipelago (contrast the Indian
monsoon which comes entirely over a warm ocean).
3. Northern Australia is a table-land sloping gently to the
I3 2 CLIMATOLOGY
north-west coast. No barrier such as the Western Ghats is
offered to the monsoon 'current, and rainfall, in consequence, is
cyclonic rather than orographic, with a more even distribution
accordingly.
It will be convenient to approach the study of the Australian
monsoon climates in the height of the dry season, and to trace
the successive changes of temperature, pressure and winds
throughout the year.
THE DRY SEASON. In July, when the Indian monsoon is at
its height, there is a continuous pressure gradient from the
southern hemisphere high-pressure ridge to the Asiatic low., The
axis of highest pressure, the source of the monsoon, crosses the
continent of Australia from west to east some distance south of
the tropic, the highest average pressure (30·I5 inches) being sh9wn
in the upper Darling basin. The outflowing air from this conti-
nental high pressure is characteristically dry, and no rain falls
except on the east coast of Queensland where the S.E. trade is an
off-sea wind. Even here the rainfall is small in winter, from
Rockhampton to Townsville the coast has less than 2 inches in
July, but at Halifax Bay the coast changes direction slightly, the
wind meets the coast at a higher angle and the rainfall is heavier;
Harvey Creek has over 4 inches. Apart from this coastal strip
the rest of tropical Australia is at the height of the dry season
with cloudless skies and a steady trade wind blowing, often
without interruption for many weeks on end.
Temperature is virtCily governed by latitude, the isotherms
running east and west, 75° in Cape York and Arnhem Land, 60°
along the tropic (d. India in the cold season, p. II4). The ,daily
range of temperature is considerable on account of the dryn~ss of
the air, and night frosts are apt to occur along the southern border
of the zone.
THE WET SEASON. As the sun returns to the southern hemi-
sphere temperatures steadily rise; Cloncurry, for example, records
72° in September, 85° in November, and 88° in December, while at
Wyndham both November and December have mean temperatures
above goo. Temperature is now governed by continentality,
the isotherms being arranged in loops round the heart of
Northern Territory (d. India in the hot season, fig. 38).
The continental anticyclone weakens coincidently and by
November has been replaced by a well-marked low centred over
the northern half of Western Australia where temperatures are
highest. Rain-bearing winds blow into this trough and gradually
the rains spread southwards, as shown in Fig. 47.
There is nothing here resembling the' burst' of the monsoon
in India, much more closely does it resemble the steady advance
TROPICAL MONSOON 133
of the equatorial rains into the savanna of Africa or South
America.
The monsoon reaches its height in January when the rains
extend almost to the tropic (compare the limit of the Indian
rain-bearing current at Karachi, p. 123). The lengths of the wet .
.seaso n as well as the actual amount of rain is thus greatest in the
north and decreases steadily towards the margin of the Australian
desert. The rain is mainly cyclonic in origin, associated with
shallow slow-moving depressions, thus resembling the rainfall of
the Gangetic Plain. But along the Queensland coast is a province
characterized by rainfall which is mainly orographic, thus resem-
bling the rainfall of the Western Ghats. The trade winds, drawn
in by the Pilbarra-Cloncurry low, here become more easterly,
imJ1inging on tJ:e coast ranges and bestowing 10 inches of rain in
January throughout a considerable length of coast. Harvey
Creek, backed by the Atherton Plateau and Mount Bartle Frere,

October
~ ':"" ~.,., ~~5"
November
.:
.,:
•••. :

December
. : ' . : ••.•• • • • •
.

January
. .3
':'1

FIG. 47.-Advance of the Monsoon Rains into Australia

has more than 30 inches in January and well over 100 inches in
the four months January to April.
The temperature at the height of the wet season everywhere
exceeds 80°, and reaches goO round Pilbarra, where the air is
clear and the sky almost cloudless. But it is in the somewhat
cooler but more humid localities further north that the heat is
particularly oppressive; the wet-bulb temperature exceeds 80°
in January over a considerable area round Wyndham (see Fig. 3,
p. 14)·
TROPICAL CYCLONES. The weather of northern Australia
during the rainy season is essentially cyclonic, the monsoonal lows
moving slowly south-eastward from the low-pressure centres of
Pilbarra and Cloncurry in obedience to the drift of the upper
air currents. More rarely rain-bearing cyclones travel south-
westward along the north-west coast of Western Australia, again
following the drift of the upper air. As a rule, though not
invariably, the depressions are of feeble intensity; but occasionally
there occur tornadoes, similar to those of the United States, highly
9
134 CLIMATOLOGY
localized, but very violent. In addition to the regular procession
of monsoonal lows there are two provinces subject to invasion by
tropical hurricanes; these are : -
I. The Queensland coast.
2. The northern coast of Western Australia.
The former group originate in the neighbourhood of the
Solomon Islands and travel south-westwards towards the Queens-
land coast. Usually they swing round to the south and thep. to
the south-east before reaching the coast, though their outer fringes
may bring rain to coastal stations, but when they do reach the
coas~ they often do considerable damage and bring torrential rain;
36 inches have fallen in 24 ttpurs at Crohamhurst and 20 inches
has freq~ntly been recordGd from stations in the coastal strip.
The west-coast hurricanes, known as ' Willy Willies'; appear
off the north-west coast and follow a track similar to that of the
monsoonal lows, recurving round the Pilbarra low and coming
inland in the neighbourhood of Onslow and the mouth of the
Fortescue River. This track is determined by the drift of the
upper air, by the repUlsion of the high pressure which lies 'above
the West-Australian current and by the attraction of the Pilbarra
low. In Northern Territory and Kimberley they give east winds
gradually veering to the north, but where the centre of the storm
strikes the coast great damage is to be expected and copious
rain results. Once it has left the sea, its source of moisture, the
cyclone rapidly loses its violence, but brings valuable and welcome
rain along its track to the Great Australian Bight. In both
localities the storms are restricted to the hottest months of the
year and reach a maximum in late summer. Between 1877 and
1912 they were distributed as follows : -

Nov. Dec.. Jan. Feb. Mar. April.


--- _- - - - -------
t2ueensland - - 6 2 8 I
West Coast I 3 9 6 6 6

SUGGESTIONS FOR FURTHER READING


The Climatological Atlas of India, issued by the Indian Meteorological
Department under the direction of Sir John Eliot in 1906, is a mine of
information on the climates of India. There are good summaries, too, in
Vol. I of the Imperial Gazetteer of India, the Oxford Survey of the British
Empire, and the Geographie Universelle (Tome IX, Asie des Moussons,
J. Sion). See also G. C. Simpson, 'The South-west Monsoon', Q. J. Roy.
Met. Soc., 1921. A full and readable treatment with many illustrations
, is;H. F. Blandford's A Practical Guide to the Climates and Weather of India,
Ceylon and Burma, 1889.
TROPICAL MONSOON I35
For Somaliland and Abyssinia there are good summaries in the Oxford
Survey of the British Empire, and for Indo-China in the Geographie Uni-
verselle.
For Australia, in addition to the Oxford Survey of the British Empire,
see Griffith Taylor's Australian Meteorology, 1920, and Hunt, Quayle and
Taylor's Climate and Weather of Australia, Melbourne, 1913. The rainfall
is treated by Wallis in the Scot. Geog. Mag., 1914.
CLIMAT«l.OGY
,'"
TEMPERATURE

Station.
1 Lat. '1 Long. A1t.(lt.) J F M .A- M J J A
_s_i~ N

LAHORE. 1320 N. 74° E. 702 53 57 69 81 89 93 89 87 85 76 63


JAIPUR • 27° N. 76° E. 1,43 1 61 65 75 85 92 93 86 84 84 80 70
BOMBAY 19° N. 73° E. 37 76 76 80 83 86 84 81 81 81 82 81
DELHI 29° N. 77° E. 718 58 62 74 86 92 92 86 85 84 78 67
'.
BENARES 25° N. 83° E. 26 7 60 65 77 87 91 89 84 83 8,3 78 68
CALCUTTA 23° N . 88° E. 21 76 76 80 83 86 84 81 81 81 82 81
CHERRAPUNJI 25° N. 92° E. 4,3 09 53 55 61 64 66 68 69 69 69 66 61
MADRAS. 13° N. 80° E. 22 76 78 81 85 90 90 88 86 85 82 79
AKYAB 20oN. 93° E. 20 70 73 79 83 84 82 81 81' 82 82 78
RANGOON 17° N. 93° E. 18 77 79 84 87 8+ 81 80 80 81 82 80
MANDALAY. 22°N. 96° E. 25 0 70 75 83 90 89 87 87 86 '85 83 76
NHATRANG. 12° N. 109° E. 'II 75 77 79 82 83 84 84 84 32 80 78
MONCAY
I
zzoN. 108° E. z9 61 61 66 73 80 82 83 83 8z 77 70
HONG-KONG 22°N. 115 0 E. 108 60 59 63 70 77 81 8z 82 81 76 69
MANILA.
.... 15° N. 121° E. 47 77 78 80 83 83 8z 81 81 80 80 78
KUPANG. 10° S. 1240 E. 48 79 79 79 79 79 78 77 78 '7? 80 81
DARWIN. IZo S. 131° E. 97 84 83 84 84 82 79 77 79 83 85 86
CLONCURRY. 20° S, 141° E. 69 6 87 85 83 78 71 64 61 67 72 83 85
WYNDHAM 16° S. 128° E. 23 88 88 88 87 82 77 76 79 85 89 90
,
BROOME. 18° S. 122° E. 63 86 85 85 83 76 71 70 73 77 81 85
CAIRNS 17° S. 146° E. Coast 82 81 80 77 74 71 70 70 73 76 79
SIMLA 31° N. 77° E. 7,23 2 42 42 50 59 64 68 65 64 62 58 51
DARJEELING 27° N. 88° E. 7,37 6 40 4 2 50 56 58 60 , 62 61 59 55 48
KODAIKANAL 10° N. 77° E. 7,688 55 56 59 61 62 59 , 58 58 58 57 55
ADDIS ABBABA. 9° N. 39° E. 8,000 60 62 65 64 66 64 62 61 61 62 59
1
· TROPICAL MONSOON 137
RAINFALL

D
I-
Yr.
-
Ra.
- _- - - - - _-
J F M A M
--
J J
------ ------
A S 0 N D
_- -Total.
--
0 0'9 1'0 0·8 0'5 0'7 1'4 5'1 4'7 2'3 0'3 0'1 0'4 18'1
55 75 4
2 0'4 0'3 0'4 0'2 0·6 2·6 8'3 7'3 3'2 0'3 0'1 0'3 24'0
63 7 8 3

77 81
10 0'1 0'1 - - 0'7 19'9 24'0 14'5 10·6 1'9 0'4 - 72 '4

60 77 34 1'0 0·6 0'5 0'4 0'7 2'9 7. 6 7'0 4'7 0'5 0'1 0'4 26'2

60 77 3 1 0'7 0·6 0'4 0'2 0·6 4.8 12°1 11 06 7° 1 2°1 0°2 0°2 4 006

10 0°4 1°0 1 °4 2°2 5° 6 11 °9 12 °7 13°4 10°0 4°9 0°6 0°2 64°3


77 81
2 °3 10 06 3 1 °3 50 °9 103° 6 10 7°5 81 °5 49°4 16 08 2 °3
55 63 16 0°7 0°3 457°3

1°1 0°3 0°3 0°6 1 08 2°O 3° 8 4°9 11 02 13° 6 5°4


77 83 14 , 4°5 49°5

72 79 14
0°1 0°2 0°5 2°0 13°7 49°4 53°7 4 2 °5 24° 6 II06 5° 0 0°6 20 3° 8

0°2 1°4 12°1 18 °4 21 °5 19°7 15°4 2 08


77 81 10 0°2 0°3 7°3 0°3 99°6

1°1 5° 8 4° 6 1 06 35° 1
7 1 82° 20 0°1 0°1 0°2 5"5 3°3 5°7 4°7 0°4

76 80 9 2 °4 1°1 0°9 0°9 2 °4 2°2 2°0 1°5 6 °9 10 06 13°9 9° 6 54°4


0 3° 1
63 73 22 1 °4 2°O 3° 1 4°4 10 8 17°9 20 °7 23° 6 12°0 5°4 1 °4 105°8

63 7 2 23 1 °3 1 08 2 °7 5"3 12°0 15°8 14°0 14°6 9"7 5°1 1 °7 1°1 85° 1

77 80 6 0°8 0°4 0 08 1°3 4°5 9°2 17°3 16 00 14°3 6"7 5° 2 3° 1 79° 6

81 79 4 15°7 14° 8 8 °7 2°5 1°2 0°4 0°2 0°1 - 0 08 3°4 10°0 57° 8

85 83 9 15°9 12°9 10°1 4° 1 0°7 0°1 0°1 0°1 0°5 2°2 4° 8 10 °3 61 08

88 77 27 5°1 4°9 2 °7 0°9 0°4 0°3 0°5 0°1 0°5 0°5 1°1 3° 0 20°0

go 85 14 9°7 5°9 4°3 1°0 0°4 0°1 - - 0°1 0°5 2°2 4° 2 28 °4

86 80 16 6 02 6 01 3° 8 1 °4 0°6 1°0 0°2 0°2 0°1 - 0°9 3°7 24° 2

81 7 6 12 3°°9 22°2 3 202 22°2 13°2 8°0 4° 2 5°4 3°7 3° 8 8 01 11 °7 165° 6

46 56 26 3° 6 3°7 3°3 2°7 3°9 8 08 21°1 20 °7 7°5 1 °4 0°5 1 °3 79°3

42 53 22 0 06 1°1 1 08 3° 8 8 °7 24°9 3 2 °3 26 01 18 °4 4°5 0°8 0°2 122 °7

55 58 7 2°9 1 °4 2°0 4°3 6°0 4° 1 5° 0 7° 0 7°3 9°7 8 02 4°4 62 °3

59 62 7 0 06 1°9 2 08 3°4 3° 0 5°7 II °0. 12°1 7° 6 0°8 0°5 0°2 49°f,j


...".,.
-
CHAPTER VIII
WESTERN MARGIN WARM TEMPERATE CLIMATES

T
HE WARM TEMPERATE CLIMATES. Situated in the lati-
tude of the oscillating front of divergence which divides
the spheres of influence of the trade winds and the
westerlies, these climates are characteristically transitional in
nature, enjoying for part of the year a climate which is typically
, tropical' in its constancy and for part of the year weather which
in its changeability is more closely allied with the temperate
zone. In the simplest form the warm-temperate (or sub-tropical)
climates derive their summer influence from the east and their
winter influences from the west; summer is therefore continental
on western margins and marine on eastern; winter conversely
is marine on western margins and continental on eastern. But
since the westerly-. circulation is less constant than the easterly
the continentality of the eastern margin winter is less pronounced
than that of the western margin summer. Clearly eastern or
western marginal situation must be a fundamental criterion of
subdivision of these climates.
This ideal simplicity is, however, not always realized, owing
to the disturbing influence of the continental masses on the
planetary wind circulation. Especially is this the case on the
eastern margins of large land masses where the trade wind of
summer is distorted into a monsoon and the westerlies of winter
are reinforced to become outflowing winds of a markedly con-
tinental nature. This interfere,nceis least marked in the southern
hemisphere owing to the small size of the land masses, and it
is here that the simplest form of eastern margin sub-tropical
climate is found. Both the great land masses of the northern
hemisphere generate monsoons and in the Eurasian block this
brings about so profound a modification of these climates as to
justify the creation of a monsoonal sub-type.

The Western Margin Type (Mediterranean)


The wide extension of this type of climate round the Mediter-
ranean Sea and the early familiarity with it in thi:;; region have
led, to the general adoption of the name to denote the climatic
138
WESTERN MARGIN WARM TEMPERATE 139
type. The term is a convenient abbreviation of the somewhat
lengthy full title, but the Mediterranean area actually furnishes
a number of highly complex varieties of the simple type which
appears in the ~ew World ~n a form much .more c1os~ly approxi-
mating to the Ideal. It IS the complexIty of relIef and the
confused intermixture of land and sea, peninsula, island and
gulf in the Old W?rld whic~ ma.k~s for the comple:xity_ of climate,
while the regulanty and SImplICIty of the coastlIne III the New
World allows the planetary regime to be established with the
minimum of interference. Yet the numerous variants in the
Mediterranean basin all agree in certain essential respects,
especially-
1. A winter incidence of rainfall and a more or less complete
summer drought.
2. Hot summers (warmest month usually above 70°) and
mild winters (coldest month usually above 43°).
3. A high sunshine amoupt, especially in summer.
DISTRIBUTION AND TRANSITIONAL TYPES. Since it is an
essential characteristic of these climates that the winter is mari-
time, receiving its influence from the oceans to the west, the
type has only a limited extension on the western margins of
continents, except in Europe where the Mediterranean Sea extends
its range for 2,000 miles into the heart of the land. Traced
eastward there is a rapid degeneration into steppe and desert
by the progressive diminution of the winter rainfalL The spring
rainfall survives furthest (steppe type), but eventually both
spring and winter rain dies out completely and such scanty
rain as does occur is the convectional summer rain of conti-
nental interiors. In South Africa and in South Australia and
Victoria the land is too narrow to support an interior steppe
climate, the western margin warm-temperate passing directly
into the eastern margin variety by a gradual increase of summer
rain, as the accompanying table and graphs show:-

Percentage of Rain in

D. J. F. III. A. III. J. J. A. S.O.N.

Cape Town 8 27 45 20
Knysna 22 21 25 32
Port Elizabeth 17 25 24 34
Durban 34 24 9 33

Equatorwards the winter rains begin later and later and


CLIMATOLOGY
cease earlier and earlier until there can scarcely be said to be
a wet season at all; this is the trade-wind desert.
Polewards the rainy season lengthens at each end until the
dry season can no longer be said to exist; this is the western
margin cool temperate climate: Tunis has five months with
less than one inch of rain, Palermo three, Naples one, and Genoa
none.
TEMPERATURE. The mean temperature of the coldest month
is usually between 43° and 50° and of the hottest between ~700
and 80°, so that the mean annual range is about 30°-consider-
ably greater. than that of tropical climates but somewhat less

7 7 7 7

"Harsnam
"Bendigo.

n
D
·Balmoral
77
"Ararat
53
'sallarat

FIG. 48.-Transition of Climates in S. Australia and Victoria

than that of most cool temperate climates. The range, however,


increases with distance from the sea: at Mogador it is only 12°
but at Morocco it is 34°, at Naples 29° but at Rome 33°, at
San,Francisco 10° but at Sacramento 27°. The highest tempera-
tures are recorded at the eastern end of the Mediterranean
furthest away from the Atlantic influence, the July mean at
Athens exceeds 80°, and Beirut, though on the coast, reaches 83°.
Stations on ocean margins usually have abnormally low summer
temperatures on account of the cold currents which set equator-
wards along the western coasts in these latitudes; the hottest
month at Mogador is only 68·5°, reached in September, and San
Francisco only reaches 59·3°, also in September. At Cape Town
and in Swanland, situated at the extremities of continents, the
WESTERN MARGIN WARM TEMPERATE I4I
cold currents are not so pronounced and summer temperatures
rise higher (Cape Leeuwin69°, Cape Town 70°). Insular stations,
since they are unaffected by cold currents, do not show quite such
low summer temperatures, but the range is very small, as the
winters are remarkably warm; Las Palmas (Grand Canary) has
a minimum temperature of 63° and a range of IIo, Funchal
(Madeira) of S9° and I3°.
The extreme range, too, is small at marine stations; frost
is unknown on the oceanic islands and is rare on the small islands
of the Mediterranean, but is fairly frequent, though not often
severe on the shores of the Mediterranean; ISo of frost have
been recorded at Rome. The rarity of frost encourages the

90
85
80
/ - \
75 /
70
/
/ /V
~ F-"
-
........

'\ N
\
65
60
a .......-:: It= 1'" 1\ ~ ~......
..2- ;7' /!/ \ \
55
50 l?' ~
45
:§._. d r\-
40
.~
FIG. 49.-Yearly March of Temperature in Mediterranean Climates
a. Funchal (Insular)
b. Mogador (Oceanic)
c. Nice (Marine)
d. Mosul (Continental)

cultivation of delicate fruits, especially citrus fruits, but its


occasional occurrence is a constant menace, met locally by ex-
tensive and expensive heating devices. At the other extreme
the highest maximum temperatures occur in the continental
variety and on the margins of deserts which lie equatorwards
and to the east. Algiers has recorded II2°, while inland in
Tripoli temperatures exceeding I300 occur.
The daily range of temperature is considerable, especially in
the dry summer· months when the conditions approximate to
those of the hot deserts; IOo or ISO in winter and ISO or 20°
in summer are usual. In the brilliant sunshine the heat of the
summer day is excessive and the midday siesta has become
confirmed as a habit. Houses are p~ovided with shutters to·
CLIMATOLOGY
keep out the heat and glare, and the design in general aims at
coolness in summer rather than warmth in winter. Tbere is,
however, often a breeze to temper the heat, especially near the
sea or mountains where sea-breezes begin to blow about midday
or where valley winds occur. The heat, too, is dry and not so
enervating as the moist heat which sometimes afflicts the later
months of summer and early autumn. The marked fall in tem-
perature after sundown causes a feeling of chilliness even though
the temperatures are not actually low. It also brings about a
great increase in the relative humidity, dew is copious and heavy
mists occur, which, however, quickly melt away before the
rays of the morning sun. Dew and mist are most frequent in
late autumn when damp airs are entering from the Atlantic
and when temperatures are beginning to fall rapidly. Tem-
perature inversions make the Lombardy plain (which is' not,
however, properly Mediterranean) a particular victim to dense
fogs. Rheumatism here is a common complaint and the death-
rate from pneumonia is exceptionally high.
The high humidity of autumn while the air temperature
still remains high sometimes makes this season unpleasantly
muggy and oppressive; conversely, spring with its delightful
freshness and warmth of the air is the most pleasant season;
to the charm of the climate at this time is added the beauty
of green fields and crops and a wealth of flowers and blossom.
RAINFALL. The occurrence of rain in Mediterranean climates
is usually associated with the passage of cyclonic storms and its
seasonal incidence corresponds to the equatorward migration of
the westerly circulation in which these occur, i.e. to the winter
months. Where mountains intercept rain-bearing winds oro-
graphic niin and even snow may make an important contribution,
as on the Sierra Nevada (= snowy mountains) which supply
California with valuable perennial water supply for irrigation
and power. But the orographic rain, like the cyclonic, has mainly
a winter incidence, since west winds (ocean winds) are mainly
winter winds and because the land, being cooler, acts more
efficiently as a precipitating agent.
The simplest rainfall regime occurs where the planetary
circulation 'is least affected by the configuration of the land,
e.g. California. Here the rainfall graph is a simple curve with
a January maximum and (at San Francisco) a five-month summer
drought centring on July. In the Mediterranean basin this
simple type is best seen on the southern shores in Morocco and
Tripoli. But in the northern Mediterranean, where the penin-
sulas and seas introduce complications, there is a tendency to
a double maximum in spring and autumn. Except in the
WESTERN MARGIN WARM TEMPERATE 143
Meseta, the Lombardy plain and the Balkan peninsula, whose
climates are not really Mediterranean but continental, the autumn
maximum tends to be the larger, for the seas are still warm and
nourish local storms, as well as yielding a good supply of moisture
for the air; e.g. Rome, Seville, Athens.
The yearly amount of rainfall is not large, 15 inches to 35
inches, but westward-facing shores, backed by mountains, may
receive 50 inches or rAore (e.g. Ragusa, 59 inches) ; the Dalmatian
highlands, behind Catarro, with over 180 inches, comprise one
of the wettest spots in Europe. In general, the rain decreases
eastwards, away from the oceans, and equatorwards, away from
the cyclones. The drought becomes longer in these directions;
Alexandria has only 8 inches, nearly all of which falls in the
three months, November to January. A little barley is grown
just west of here but is insecure without irrigation; southwards
of this is sheer desert.
Although winter is the rainy season it is by no means cloudy
or damp. Storms are not so numerous as in western Europe,
but such as occur usually bring heavy rain. The Scilly Isles
have rather less rain than Rome, but have more than twice as
many days with rain; Cape Town has more rain than Kew, but
the winter cloud amount is 5'1 compared with 7'4 at Kew and
the summer cloud is only 3, compared with 6'7 at Kew. Mediter-
ranean climates are famed for their blue and cloudless skies;
they are. . in fact, some of the sunniest parts of the world, and
are naturally suited to be the health and pleasure resorts of
every continent in which they occur. But the concentration of
the rains into a comparatively few heavy showers separated by
long intervals of fine weather is not an unmixed blessing. The
heavy falls cause considerable ~oil waste on the steep hill-sides
with their thin plant cover and this has been aggravated by the
serious deforestation of the past; rivers are subject to consider-
able fluctuations of level, prone to flood and then to dwindle
away. Thanks to the rapidity with which water drains away,
the soil is generally dry and healthy, but wherever soil or· relief
conditions are such that water stands and stagnates, e.g. on the
impervious volcanic floor of the Campagna, mosquitoes breed
and malaria is rife.
The Mediterranean climates lie near the limit of the cyclonic
rains and the precipitation in consequence is somewhat unreli-
able. The rainfall at Santiago is below normal in seven years
out of ten, the deficiency being made good by a few excessively
wet years; the range of variability here is from 18 per cent. to
226 per cent. In consequence irrigation is necessary everywhere
.north of 37° S. even with a rainfall of 40 inches. In the Mediter-
144 CLIMATOLOGY
ranean basin where the course of the cyclones is guided by the
natural features the rainfall has a much higher reliability.
Snow, though by no means unknown, is sufficiently rare to
call for comment when it falls, though on the mountains, of
course, even in North Africa there is a considerable yearly fall
which provides a valuable source of water for irrigation and power.
VEGETATION. The dominant characteristic of the Mediter-
ranean climates is the marked rhythmic recurrence of rain and
drought, and this naturally finds its echo in a marked rhythm
of plant growth. But the contrasts of growth and rest are not
nearly so startling as in that other essentially periodic type,
the savanna, since in the latter heat and moisture coincide,
while in the Mediterranean the rainy season is the coolest. Winter
temperatures are scarcely low enough to forbid growth altogether,
but they are too low for growth to be very vigorous. Summer
drought, while in general limiting activity, is not always complete
and is sometimes circumvented by local supplies of ground
water, either natural or artificial. Autumn and spring, with
their moderate temperatures and yet adequate rain, are the
seasons of greatest vigour. Growth is therefore steady, though
sometimes slow, throughout autumn, winter and spring, but
except under locally advantageous circumstances is checked in
summer by the excessive drought. Even where the rainfall
figures show a little summer rain this avails nothing, for in the
fierce heat and dry air evaporation is far in excess of any rain
which may fall. The summer landscape, dancing in the shimmer-
ing heat haze, is baked bare and brown, crudely coloured with
the tints of naked soil and rock, with the dazzling white walls
of houses and the grey of the olive trees-a scene of universal
drought, dust and glare.
The survival of plants is dependent on their ability to with-
stand the intense desiccation and the whole vegetation is charac-
teristically xerophytic. More hygrophyllous plants can grow
only where local conditions, generally edaphic but rarely climatic,
give a supply of summer moisture; e.g. the riparian groves of
Oleander, the irrigated orange groves and the Californian redwoods
fed on fog during the summer drought.
Structures to escape the drought are numerous and varied.
Almost every device for reducing transpiration is present, thick-
ened cuticle, spiny leaves, wax and hair coverings, etc. Annuals
run through their cycle quickly in spring and seeding is over
when the drought appears; bulbous and tuberous rooted plants,
many of them with beautiful flowers, e.g. tulips, gladioli, lilies,
narcissi and irises bloom in the early spring and die down, only
the bulb surviving.
WESTERN MARGIN' WARM TEMPERATE I45
MAIN TYPES OF VEGETATION. Where conditions are most
favourable, where rainfall is heaviest and there has been least
interference, the vegetation consists of evergreen woodland with
pine, cedar and evergreen oak; more rarely, where the drought
is less severe, deciduous oak may occur. The occurrence of the
cork oak in the midst of a wine industry is a fortunate circum-
stance; while the acorns provide valuable f()od for pigs. In
Western Australia the eucalyptus forests provide the durable
jarrah and karri woods. But the distribution of such forests is .
very limited and has been further seriously reduced by deforesta-
tion both by man and goats. Where conditions are less favourable
the forest degenerates into a low scrub, the macquis or macchia
of the Mediterranean, the chaparral of Calif()rnia, the M aZlee
scrub of Australia> a dense tangle of thicket made up of low-
growing evergreens, arbutus, laurel, myrtle, rosemary, etc., with
occasional taller trees. The degeneration of forest into scrub
is frequently to be attributed tq deforestation, especially in the
Mediterranean lands, but it is more usually the result of lower
rainfall or poor soil. In extreme cases of drought or soil poverty
the vegetation consists of scattered low-growing scrubs with
bare soil showing between, the garigue; it is especially charac-
teristic of limestone s()ils. The plants which make up the garigue
are highly xerophyllous with bright but short-lived flowers,
e.g. broom and gorse, and many of them are highly aromatic,
e.g. lavender, sage and thyme.
Difficult to clear for cultivation, the macquis is of little use
to man, though the mallee scrub, when cleared, makes good
wheat land in South and West Australia. The garigue is prac-
tically worthless and serves only to provide sustenance for the
ubiquitous and omnivorous goat.
The increase of rainfall with altitude brings about a mountain
forest zone, but here the trees are broad-leavE!d and deciduous,
beech and sweet chestnut, for the orographic rains are not so
restricted to the winter months and the winters are colder.
Above the forest is a zone of alpIne pasture, available during the
summer only, and therefore, to be useful, requiring to be supple-
mented by winter pasture on the plains, a seasonal transhumance
occurring across the cultivated slopes between. But grass is
unusual in Mediterranean climates since heat does not coincide
with moisture, conditions, in fact, are generally more favourable
for arboreal growth. In the Rhone delta, for example however,
the strong winds (mistral) discourage trees and grass is able to
establish itself. In the absence, or rarity, of g()od pasture, cattle
are rare, their place being generally taken by the less fastidious
goat. Thus there is throughout the Mediterranean lands a natural
CLIMATOLOGY
deficiency of meat, butter and milk, their place in the dietary
being taken by beans, olive-oil and fruit-juice,,-not altogether
satisfactory substitutes, especially for children, as witnesses the
high infant mortality. '
CULTIVATION. The climatic rhythm does not impose a
rhythm on agriculture, since fruits, either naturally drought-
resistant such as the olive, or irrigated such as the citrus fruits,
give employment during the summer, while cereals and vegetables
employ the labourer during the rainy months. The reliable
summer drought offers ideal conditions for the cereal harvest
and is further valuable for the drying of fruits such as currants,
raisins and figs. The long summer is a valuable asset for fruit
growing and especially for the vine; maximum temperatures
at many stations are delayed until August and even in September
the temperature is still high in the sixties.
The native fruits are all well adapted to stand the drought,
especially the olive, the fig and the vine with their long root
systems; though the latter may require careful pruning and
wide spacing to economize moisture. The imported fruits, how-
ever, are less suited and generally require artificial watering, e.g.
peaches, oranges, lemons, limes. These are so thoroughly estab-
lished in Mediterranean climates all over the world to-day that
they have come to be considered as characteristic, yet most of
them belong more properly to the eastern margin warm temperate
or even tropical climates and are accustomed to summer rain.
The long sunshine hours of summer and the high temperatures,
however, ensure the success of the importation and the general
uniformity of the climatic type all over the world has made
their transference from one Mediterranean region and their
establishment in another a comparatively simple matter.

Regional Types: The Mediterranean Basin


The limits of the Mediterranean climates on the north and
west are practically coincident with the mountains which enclose
the basin, but the plateaux and basins of the Iberian and
Balkan peninsulas, together with the Lombardy plain, while
retaining certain characteristics of the Mediterranean climates,
have adequate summer rain which allies them more closely with
Central Europe. Furthermore, they fail to satisfy the require-
ments of the typical Mediterranean vegetation, of which the
olive is the most sensitive test. Southwards and eastwards the
boundary is vague, the Mediterranean climates grading in this
direction imperceptibly into desert and steppe.
WINTER CONDITIONS. During the winter months the warm
WESTERN MARGIN WARM TEMPERATE I47
waters of the Mediterranean intrude a ' lake' of relatively low
pressure between the North Atlantic high, which now extends
over the Sahara, and the Eurasian winter high which stretches
down the central axis of Europe to the Meseta. It becomes a
much used track for cyclones of the westerly circulation, but is
virtually separated from the more frequented path up the west
coast of Europe by the westward extension of the continental
high. Winds on the northern shore of the Mediterranean are
therefore northerly and in the southern Mediterranean westerly
in obedience to this pressure distribution. The actual direction
of the wind varies during winter from day to day in accordance
with the position 10f cyclonic centres in their passage, but it is
in general, from land to sea, i.e. from high pressure to low.
There is a tendency for the three peninsulas to intrude tongues
of high pressure southwards while the intervening seas tend to
generate local lows. As a result of these pressure distributions
there is a tendency for warm" wet south-west winds to occur
on the west coasts of the peninsulas and cold, dry north-east
winds on their eastern shores. Thus Lisbon has 30 inches while
Murcia has only I5 inches, Ragusa has 59 inches and a January
temperature of 480, while Athens has only I5 inches and a January
temperature of 46°.
CYCLONES. Although cyclones of local origin do affect the
weather of the Mediterranean they are usually feeble, the domi-
nant type belonging to the westerly circulation and entering the
area either by the Straits of Gibraltar or from the Bay of Biscay
via the Gate of Carcassonne. Their passage from west to east
is connected with many important local phenomena. On the
advancing front the winds are southerly, coming from the deserts
of North Africa and often excessively hot and dry, sometimes
laden with red penetrating dust .. This is the Sirocco of Algeria,
the Leveche of Spain, the Klzamsin of Egypt and it receives other
names in other parts. Its hot breath dries and cracks the skin,
bringing considerable bodily discomfort and sorely trying the
nerves .and will; it withers the vegetation, often doing per-
manent damage, especially if it comes when the vines and olives
are in blossom. Where it is a descending wind, e.g. on the
Algerian coast or on the north coast of Sicily, its heat and aridity
are further increased and maximum temperatures of over IrO o
often occur. In its passage across the warm seas it sucks up
water, and is consequently less desiccating on the northern
shores, but the high humidity, combined with the heat, make it
now enervating and depressing in the extreme.
The rear of the depression is associated with northerly winds
from the cold interior of Europe, reinforced to give rise to the'
CLIMATOLOGY
mistral when a high pressure over the continent combines with
a low pressure in the Ligurian Sea to bring a torrent of bitterly
cold dry air from the plateau of the Cevennes and concentrates
it in the narrow passage of the Rhone Valley. In spite of the
warming by its descent it is a biting wind with a temperature
often below freezing. The mean minimum at Marseilles is 22°
and the extreme minimum n°, figures which are not' matched
elsewhere in Mediterranean climates. Hedges of cypress are
planted to protect orchards and gardens from its icy breath and
the sites of houses are chosen for their protection from it. The
Bora (see p. 40) is a similar wind in the Adriatic.
SUMMER CONDITIONS. In summer the Azores high extends
along the now relatively cool waters of the Mediterranean which
is thus largely responsible for the maintenance of Saharan o;m:-
ditions. Winds ar:e northerly over the whole area and are
equivalent to the trade winds, being directed towards the great
trough of low pressure which extends. westward from India and
Arabia. In the eastern Medite~ranean where the gradient is
steepest these winds reach great force and constancy; these are
the Etesian winds, well known to the ancient Greeks. From
mid-May to mid-October they blow with great regularity with
a velocity of IO to 30 and sometimes 45 miles an hour; they
increase in force during the day as the daily convection'in the
hot lands to the south and east temporarily increase the pressure
gradient, but at night they weaken and often die away. On
land they bring clouds of dust which make the summer at Athens
an unpleasant season, at sea they raise foam-crested waves
strangely contrasting in their storminess, with the deep blue
dome of a cloudless sky.
For the Etesian winds, travelling towards warmer lands,
are dry (relative humidity 20 to 30 per cent.) and in spite of
their heat are physically cooling and refreshing. Sailing is
dangerous, especially to windward of exposed rocky coasts, but
the wind is reliable, the conditions are familiar and accidents
are less frequent than during the squally cyclonic storms of
winter. The force and dryness of the wind is sufficient to pro-
hibit tree-growth in exposed places, and orchards have to be
protected by a row of cypresses as wind-breaks on the north
side.
In their passage over the Mediterranean the north winds
pick up moisture and although they seldom cause rain they
bring mist and fog which may last for days in Algeria and Tunis.
They assist, too, in moderating the temperature along the African
coast; should the wind blow from the south, the heat increases
'at once. At Benghazi (Tripoli) September is as hot as July
WESTERN MARGIN WARM TEMPERATE 149
(78°) and October' as June (75°), for the summer wind blows
from the north into the Arabian low (71 per cent. from the north),
but in autumn the Arabian low is weakening and the Mediter-
ranean low begins to draw south winds off the still hot Sahara
(34 per cent. from the south).

Spain
The climate of the Iberian Peninsula is a curious variety in
which the Mediterranean type is struggling against the effects
of continentality. The size of the peninsula is sufficient to
produce app:eciable monsoons which supersede the planetary
winds. In wmter the temperatures on the plateau are abnormally
low, in places below 40°, and frost is common; the resultant'
land monsoon partially excludes rain-bearing winds and the
winter fall of rain on the plateau is well below normal.
In summer temperatures .are abnormally high, Madrid,
though more than 2,000 feet above $ea-level, has a July mean
of over 75°. The days are excessively hot and dusty, the fierce
rays of the sun scorching through the thin air. Strong con-
vection is thus set up, and a steady flow of air, strengthened
by day, moves in towards the c~ntre of the plateau. Yet this
inflow of air brings no rain, July and August being almost com-
pletely rainless; for the heat of the plateau is so great that
the moisture capacity of the air is still further increased and
the rising air is carried away by an upper current before saturation
is reached (d. Sind, p. 125). It is during spring and autumn,
when the air currents are humid, and while the land monsoon
is not strong enough to exclude them, that most of the rain
falls. This is augmented by local thunderstorms, especially in
the spring. These features, a large temperature range, strong
winds, spring rains and hot summers are steppe characters and
much of the vegetation is of the steppe type.

The New World


The uninterrupted line of the west coast of the Americas in
these latitudes enormously simplifies the relationship of land
and sea influences, the resultant climates being correspondingly
simple. The rainfall regime shows a single winter maximum
and a single summer minimum, a steady decrease in amount
equatorwards and a steady increase in that direction of the
duration of the drought. The presence of a coastal chain of
mountains causes the rainfall to increase inland up to a point,
after which it diminishes ..
10
ISO CLIMATOLOGY
CALIFORNIA, WINTER CONDITIONS. The wind system of Cali-
fornia is closely bound up with the circulation over the North
Pacific, being virtually independent of conditions east of the
Sierras, which act as an efficient climatic barrier, especially in
winter. In January the chief influences are the Aleutian low,
the North Pacific high and the continental.high. The westerly
winds on the north flank of the Pacific high impinge on the
coast in about latitude 40° N. and are turned north and south
by the double obstruction of the mountain barrier and the con-
tinental high. The northern air stream is drawn past British
Columbia into the Aleutian low, the southern turns to the south-
east and then to the south and finally to the south-west to become
the N .E. trade wind. The prevailing wind of California is thus
north-west or north, but the passage of storms causes consider-
able variations from day to day. The majority of the cyclones
pass some way to the north, over Oregon and Washington (see
p. 191) and only their southern edge affects California. They
are felt less and less further south and rainfall diminishes in this
direction, as the following table shows:-
Winter Rain Dry Months
Station. Lat. Jan. Rain. (Oct.-March). (less than I
Inches. Inches. inch).
Eureka . 41° N. 8 36 3
San Francisco 38° N . 5 19 5
Los Angeles. 34° N. 3 15 6
San Diego . 33° N . 298

The typical temperate cyclones, in fact, scarcely reach southern


California and most of the storms are thunderstorms (Sonoras).
As in the European Mediterranean, the passage of the cyclones
often draws hot winds off the deserts. The Santa A nnas of
southern California and the Northers of the Sacramento Valley
are hot, dry winds, charged with dust, and, like the sirocco at
its worst, owe their high temperatures largely t<;> adiabatic
heating during their descent of the mountain slopes. The
effects are the same as those of the sirocco, acute discomfort
tp man and disastrous drying up of the vegetation. Being
due to cyclones they are chiefly winter phenomena. but the
worst damage is done if they come in spring when the fruit trees
are in blossom or when the young fruit is formed.
Mean temperatures in winter increase steadily from north
to south (San Diego is 5° warmer than San Francisco). More
extreme temperatures occur inland, especially in the valley
bottoms where inversions occur.
SUMMER CONDITIONS. Although coastal temperatures are
kept low by the cold current, inland temperatures rise rapidly in
WESTERN ~ARGIN WARM TEMPERATE 151
spring, and by April there is a conspicuous northward bend of the
isotherms. In July the isotherms run north and south; con-
tinent and ocean are the controls. The temperature of the North
Pacific, thanks to the cold current, is only 57° and, the wind
being on-shore, these low temperatures are carried on to the
land. Temperatures along the coast are below 60° (e.g. Eureka
56°, San Francisco 57'3°), but these phenomenally low tempera-
tures are confined to a narrow coastal strip and rise rapidly
inland: Mt. Tamalpais, although 2,375 feet above sea-level, is
I3° warmer than San Francisco, while in the Great Valley mean
temperatures exceed 80°. San Francisco thus lies between a
cold sea and a hot land; while the sea breeze blows, the tem-
perature may fall to 50°, when the land breeze blows it may rise
above 90°. The very low mean temperature shows that the
sea breeze blows nearly all the while, in point of fact 98 per cent.
of the July winds are westerly (S.W., W. or N.W.). The striking
persistence of this wind is due to the suction effect of the heated
interior, by which air is drawn into the Great Valley and con-
centrated on San Francisco by the relief of the Coast Ranges
and the funnel of the Golden Gate. The westerly wind is only
a shallow current, easterly return winds are frequent on the
summit of Mount Tamalpais.
Temperature in the Great Valley itself is governed by prox-
imity to the funnel entrance; thus Sacramento, opposite the
gap, is 9° 'cooler than Red Bluff which actually lies 1° further
north. Not until the heating of the Great Valley diminishes
does this ventilation of San Francisco cease; but by September
the conditions are weakening, westerly winds now make up only
88 per cent. of the total and their velocity has decreased. San
Francisco, freed from the cooling draught, begins to warm up
and reaches its maximum temperature (59'9°) in September.
The south-west wind at San Francisco has a high humidity
(85 per cent.), yet duting the six summer months rain is very
rare, which, at first sight, is surprising; but a combination of
conditions prevents precipitation :-(I) The sea off which the
wind comes is cold, (2) the land on to which it blows is hot,
(3) the air is extremely stable since the temperature lapse is
inverted. But though no rain falls, fog is most persistent from
May to Cktober, rolling.in from the sea every afternoon as the
strength of the sea breeze increases; Port Reyes has 1,860 hours of
fog a year. It affects only a limited area where the air current
flows and the fog-free areas are the popular residential districts
(Berkeley and Oakland). The fog supplies moisture during the
summer months for a hygrophilous vegetation (the Californian
redwood and the Californian laurel), contrasting strongly with
I52 CLIMATOLOGY
the bracken and berry shrubs of the hill-sides and inland districts
which lie beyond its reach.

Chile
From Coquimbo (30° S.) to Concepcion (37° S.) the climate
of Chile qualifies as Mediterranean, the type thu'.) occurring at
rather lower latitudes than in the northern hemisphere. The
same controls are at work here as in California, viz. : -
1. The sub-tropical ridge of high pressure, swinging south in
summer and becoming an independent anticyclone over the sea
as the continental low develoDS to the east.
2. A cold current off-shore~
3. A straight coastline backed by a mountain range. The
same simplicity is therefore noticeable especially as regards the
rainfall regime, thus:-
No. of Months
Lat. Yearly Rainfall. with less than
Inches. I inch.

La Serena 4'3 9
Valparaiso 20 8
Puerto Carranza 28 5
Concepcion 53 I
Valdivia. 105 o

But unlike California, where the only gap in the coast range
is the Golden Gate, the coast ranges of Chile are low and broken,
thus permitting marine influence to penetrate much more freely
inland. Thus although on account of the cold current coastal
temperatures are low in summer (Valparaiso 69°) and maxima
are delayed, there is nothing comparable to the conditions at
San Francisco. The inland temperature gradient is also much
less steep and the longitudinal valley has nothing approaching
the furnace heat of the Sacramento-Joaquin valley. Santiago
(1,703 feet) has a January temperature below 70°. The greatest
heat, as in California, occurs under foehn conditions; with winds
blbwing down from the Sierras over 100° has been recorded at
Punta Tumbez, near Concepcion. Although in 37° S. latitude
the extreme temperatures here are considerably higher than
in tropical Chile, where conditions are not so favourable to foehn
winds.
Rainfall, low on the coast on account of the cold current
(Valparaiso 20 inches), increases up the slopes of the coast ranges
(Quilpue 27 inches), decreases again in the rain-shadow of the
. l<;mgitudinal valley (Santiago 14 inches), increases again up the
slopes of the Andes (Portillo 60 inches) and finally decreases and
WESTERN MARGIN WARM TEMPERATE I53
practically ceases at high altitudes and across the watershed
in Argentina. (All those stations are in the same latitude,
namely, 33° S.) The scanty rainfall in the valley makes irrigation
necessary for most crops and fruits, but the Andean slopes, with
their heavier rainfall and their snow, are a valuable source of
water.
With its dry bracing air and with the heat of summer tempered
by altitude and the cold current, the climate of this heart-land
of Chile is almost ideal for man; and to the advantages of health
and comfort it adds considerable agricultural and horticultural
productivity. Good crops of wheat, barley, lucerne, etc., are
grown and fruit flourishes, those va~ieties being especially gr?wn
which lend themselves to long dIstance transport. e.g. vmes
(as wine) and nuts (especially walnuts).

Cape Town
The area with a Mediterranean climate in South Africa is
very small-from the Olifants River in the north to the Breede
River in the east-but small as it is its importance is great owing
to its suitability for wheat and fruit growing. As in California,
Chile, Morocco and West Australia, the summer temperatures of
the coastal zone are kept down by a cold current (the Benguela
Current), off which blows the prevailing wind. Cape Town has
a January mean below 70°, but temperatures are higher inland
and a shift of the wind to this quarter means a marked rise in
temperature, especially when winds blow off the high plateau.
These winds, known as Berg Winds (see also p. 236), are especially
characteristic of the winter months when there is a strong anti-
cyclone on the plateau and low pressure out to sea from a cyclone
passing to the south. As a result of adiabatic heating their
temperature may exceed rooo, the winter temperature thus tem-
porarily exceeding those of midsummer. They are analogous to
the Santa Annas of California and are equally damaging. They
occur on all sides of the plateau edge, but the season of their
occurrence varies from place to place.
Less than 6 inches out of a total of 25 inches falls in the six
summer months at Cape Town and four months each have less
than I inch. The depressions of winter generally pass south of
the continent, and it is the south-west wind in rear which brings
most of the rain. The rainfall is extraordinarily variable within
small distances, different stations within Cape Town itself record-
ing means of 18 inches and 40 inches. There is a marked increase
on. hill slopes; parts of Table Mountain have 80 inches while
over 200 inches is recorded in small areas near by.
154 CLIMATOLOGY

Australia
Swanland and the south Australian littoral may be included
among the Mediterranean climates, but Adelaide is much less
characteristic than Perth. Swanland is subject to controls
almost identical with those of Ca-pe Town and the climates and
resultant cultivation are very similar in the two cases. The
west Australian plateau is, however, much lower and less steep-
sided than the South African and there is no 'Berg Wind'.
In Australia, too, the climatic type has a much wider extension,
the Is-inch isohyet cutting off the corner of the continent from
Geraldton in 28° S. on the west coast (cf. 33° S. in S. Africa)
to Esperance on the Bight. The rainfall, too, is extremely
reliable.
Dairies _
Wheat IZZZl
Oranges •••••
Vines YVv.

Apples AU.

FIG. so.-The Rainfall Control of Vegetation and Crops in Swanland


(Griffith Taylor)

The winter temperature of Perth is 55° (equivalent to May


in London) and frost is unknown. The heat of summer is tem-
pered during ·the day by a regular sea-breeze known as 'the
Doctor' and cloud amount in summer is less than 3. This is
one of the most favoured regions of Australia with a healthy
and enjoyable climate admirably suited to dairy farming, wheat
cultivation and fruit growing.
The shores of the Great Australian Bight, lying parallel to
the rain-bearing winds, have only about 10 inches of rain and the
desert reaches down to the coast, but the Eyre's Peninsula has
15 inches and the isohyets are carried far to the north by the
Elinders Range. Adelaide has 21 inches, of which 70 per cent.
falls in the six winter months, but there is a small fall in spring
WESTERN MARGIN WARM TEMPERATE I55
and autumn and even a little in summer; December, January
and February each have less than I inch, but are not quite dry:
The winter rain is due to the cyclones of the temperate zone,
that of the summer half of the year chiefly to tongues of low
pressure intruded from the tropics through the sub-tropical
high-pressure belt (see p. I62).
Eastwards the summer rains begin to increase in importance;
Robe 'has 25 inches, of which 78 per cent. falls in the winter half
of the year and may still be referred to the Mediterranean regime,
but by Melbourne the incidence of rain is uniformly distributed
over th'e year; the dividing line coincides roughly with the
north and south range of the Grampians (see Fig. 48).

SUGGESTIONS FOR FURTHER READING


For ,the Mediterranean basin see A. Philippson, Das Mittelmeergebiet,
1904; Atlas of Normal Monthly Values of the Meteorological Elements of
the Mediterranean Sea and the Adjacent Lands, Met. Off., 224, London, 1919;
H. A. Matthews, 'Mediterranean Climates of Eurasia and the Americas,'
Scot. Geog. Mag., 1924; W. W. Jervis, 'The Mediterranean Climate and
its Variants: Geog. Teacher, 1925; Notes on the Climates of the Eastern
Mediterranean and Adjacent Countries, LD. 1117, H.M. Stationery Office;
E. G. Mariolopoulos, Etude sur le Climat de la Grece, Paris, 1925;
R. de C. Ward, 'Climatic Notes on Palestine, Mesopotamia and Sinai,'
Nature, 1918; 'Bewolkung und Sonnenschein des Mittelmeergebietes,'
Hamburg, Arch. D. Seewarte 35, 1912, NO.2.
For California, see Matthews, op. sup. cit.; C. E. P. Brooks, Variations
I

of Temperature at San Francisco,' Geog. Teacher, 1927; R. de C. Ward,


Climates of the United States; R. J. Russell, 'Climates of California,'
Univ. of Cal. pubs. in Geog., Vol. 2, NO.4.
For Chile, see Matthews, op. sup. cit.; R. C. Mossman, 'Climate of
Chile, J. Scot. Met. Soc., 19I1; M. Jefferson, 'Rainfall of Chile,' Am.
Geog. Soc. Research Ser., NO.7.
For South Africa, see Oxford Survey of the British Empire; and for
Australia, in addition to the above, see G. Taylor, Australian Meteorology,
and Hunt, Quayle and Taylor, Climate and Weather of Australia, Melbourne,
19 13.
CLIMATOLOGY

TEMPERATURE

Station. La!. Long. Alt. (ft.)1 ]. F M A M ] I] A S 0 N


- - - - - - -.-
-

PUNTA DEL GADA 38° N . 26° W. 73 58 58 57 59 62 66 70 72 70 66 62


FUNCHAL 33° N. 17° W. 82 59 ,59 60 61 64 67 70 72 71 68 64
LA LAGUNA. 28°N. 16° W. 5 54 55 57 58 62 65 69 72 70 66 60
LISBON. 39° N . 9°W. 20 9 51 52 54 58 60 67 70 71 68 '62 57
GIBRALTAR. 36° N. 5° W . 53 55 56 57 61 65 70 73 75 72 67 60
MOROCCO 32° N. 7°W. 1,54 2 52 55 59 67 69 77 82 85 76 70 62
MOGADOR 32° N. 9°W. 33 57 59 60 63 65 68 68 68 69 67 63
ALGIERS 37° N. 3° E. 72 53 55 58 61 66 71 77 78 75 68 62
MARSEILLES 43° N . 5° E. 24 6 44 46 50 55 61 68 72 71 66 59 51
ROME 42oN . 12° E. 20 7 45 47 51 57 64 71 76 76 70 62 53
PALERMO 38 ° N . 13° E. 23 0 51 52 55 58 64 71 76 77 73 67 59
ATHENS. 38° N. 24° E. 35 1 48 49 52 59 66 74 80 80 73 66 57/

ALEXANDRIA 31° N. 30° E. 105 58 60 63 67 72 76 79 81 79 75 68
, 1

SMYRNA 38° N. 27° E. 65 46 48 53 59 68 75 80 79 72,. 66 56


JERUSALEM. 32° N. 35° E. 20454 44 48 51 59 66 70 73 73 7 1 67 56
SAN LUIS OBISPO 35° N . 121° W. 201 52 53 54 55 57 61 63 1i4 64 6r 56
SAN FRANCISCO 38° N. 122° W. 155 49 51 53 54 56 57 57 58 60 59 56
SACRAMENTO 39°N. 121° W. 71 4 6 50 54 58 63 69 73 72 69 62 53
RED BLUFF 40 0 N. 122° W. 33 2 45 49 53 59 66 74 81 79 72 63 53
SANTIAGO 33° S. 71°W. 1,70 3 67 66 .62 56 51 4 6 46 48 52 56 61 ~ 1
VALPARAISO 33° S. 72°W. 135 67 66 65 61 59 56 55 56 58 59 62:
I"
CAPE/TOWN 34° S. 18° E. 40 70 70 68 63 59 56 55 56 58 61 64,I
GERALDTON 29° S. II5° E. 13 74 75 73 69 64 60 59 59 61 64 68
I
PERTH 32° S. II6° E. 25 74 74 71 67 61 57 55 56 58 61 66'
EUCLA 32° S. 129° E. 30 71 71 69 66 61 56 54 56 59 63 66 1

ADELAIDE 35° S. 139° E. 140 74 74 70 64 58 54 52 54 57 62 ~71


ROBE 37° S. 140° E. coast 65 65 62 59 56 52 51 52 54 57 60 1 \
. I
WESTERN MARGIN WARM TEMPERATE 157

RAINFALL

Yr, Ra, J F M
D
- - -- -- -- -- -- -- -- -- --
-
A M
--
J J A S
-_
0 N D Total.

60 63 14 2,8 3'1 2'3 2'0 2°2 1'3 0,8 r05 2'4 3'3 3,6 3'1 28'4
61 65 13 3'4 3:6 3'4 2'0 1'1 0'4 0'1 0'1 1'2 4'0 4'7 3,'2 27'2
56 62 ;8,
4'1 3'5 2'9 1,8 0,6 0'2 0'2 oor 0'5 2'4 3'4 3,8 23°5
52 60 20 3,6 3'5 3'4 2,6 2'0 q'8 0'2 0'2 1'4 3'3 4'3 4'1 29'4
56 64 20 ' 5'!' 4'2 4°8 2 °7 1'7 0'5 - 0°1 1'4 3'3 6'4 5'5 35"7

54 67 33 '1'3 1'2 1'4 1°1 0'7 0'3 0'2 - 0'3 0'5 1 °5 0'9 9'4
59 64 12 2°2 1'5 2'2 0"7 0,6 o'r - - 0'2 1 °3 2 °4 2'0 13'2
57 65 25 4° 2 3'5 3°5 2 °3 1'3 0,6 0°1 0'3 1'1 3° 1 4° 6 5°4 30 °0
46 57 28 1'7 1'4 1'9 2'2 1'7 1°1 0'7 0°8 2'4 3,8 2,8 2'1 22,6
46 60 3 J 3'2 2"7 2'9 2,6 2°2 1,6 0'7 1'0 2'5 5°0 4°4 3'9 3 2 '7
53 63 26 3'2 2"7 2,8 1 °9 roI 0'7 0'2 0°4 1,8 3'2 3'3 3'6 25'0
52 63 3 2 2'0 1"7 1'2 0'9 0,8 0'7 0°3 0'5 0,6 1 06 2,6 2,6 15°5
61 70 23 2°2 0°9 0°5 0°2 - - - - - 0'3 1 °4 2,6 8'1
49 63 34 4'3 3'3 3° 2 1 °7 1 °3 0,6 0'1 - 0"7 1 °7 3,6 5'2 25°7
49 61 29 6'2 4'6 3'5 1'5 0'3 - - - - 0'4 2'5 5°7 24'7
53 58 12 5'0 3'9 3'5 r'4 0°6 0°1 - 0'3 0°3 0'9 1,6 3,8 21 °4
51 55 II 4° 8 3° 6 3'1 1°0 0"7 0'1 - - 0°3 1'0 2'4 4° 6 22°2
46 57 27 3° 8 2'9 3'0 1,6 0°8 oor - - 0°2 0'9 2'1 4° 0 19'4
46 62 36 4,8 3'9 3°4 1 06 1°1 0°5 - - 0°8 1'4 2 08 4'4 24'7
66 56 21 - 0°1 0'2 0°6 2 06 3'2 3'2 2'1
1°2 0°5 0'3 0°2 14'2
64 61 12 - - 0,6 0°2 3°5 5°8 4° 8 3° 2 0,8 0°4 0°1 0'3 19'7
68 62 15 0'7 0,6 0°9 1'9 3° 8 4°5 3'7 3°4 2 °3 1,6 1'1 0,8 25'3
72 67 16 0°2 0'2 0'4 1 01 2,6 4° 6 3° 6 2 °9 1°1 0'7 0'3 0°1 17'8
71 64 19
, 1,6 4'9 6'9 6'5 5'7
0°3 0°5 0°7 3°3 2'1 0,8 0,6 33'9
?9 64 17 0"7 0'5 0'9 1°2 1'2 1°1 0'9 1'0 0,8 0'7 0°7 0°4 ro'r
71 63 22 0'7 0'7 1'0 1 08 2,8 3° 1 2"7 2'5 2'0 1'7 1'2 1°0 21°2
62 58 14 0,8 0"7 1°2 1'9 3'0 4° 0 4° 0 3,6 2 02 1'7 1°1 1'0 24°7
.
CHAPTER IX
EASTERN MARGIN WARM TEMPERATE CLIMATES
HE transition zone between the trade wind and westerly

T circulations is represented on the eastern margins of


continents by a type of climate which, while generally
sharing the mild winters and hot summers of the Mediterranean
climates, differ fundamentally in the 'amount and distribution of
rainfall. The trade winds, or winds closely related to them, which
bring summer drought to western margins are here rain-bearing,
because on-shore; the westerlies which bring cyclonic rain to
western margins are here continental and cyclones are less vigorous
though by no means extinct. Winter rain is, for this reason,
relatively less than in Mediterranean climates, but is amply com-
pensated by generous summer rains, as the following comparable
figures show:-

Temperature. Rainfall.
Lat.
July. 6 Summer 6 Winter Total.
Jan. Months. Months.

Valparaiso 33° S. 69 53 'Z 18 20


Montevideo 35° s. 72 50 18 21 39
Cape Town 34° S. 70 55 6 19 25
Port Elizabeth 34° S. 69 58 10 13 23
Perth 32° S. 74 55 4 29 33
Sydney 34° S. 72 52 22 26 48
I

As might be expected in a region of transition, wind direction


,is somewhat variable, especially during winter when passing
cyclones cause frequent. and rapid veering and backing, accom-
panied by changeable weather.
WEATHER INFLUENCES. Both temperate and tropical storms
are elements of some importance in humid sub-tropical climates
for they lie within a zone common to both types. Visits by the
latter are comparatively rare in the southern hemisphere, but the
West Indian hurricane and the Chinese typhoon are fairly frequent
158
EASTERN MARGIN WARM TEMPERATE I59
visitors to the extra-tropical coasts of the United States and
China and Japan. The cyclones of the temperate zone bring some
remarkable weather tYp'es, with temperature changes of remark-
able amplitude and suddenness and with correspondingly import-
ant influences on life conditions. Sudden irruptions~ of polar
air may lower the temperature by 30° or 40° in 24 hours, damaging
fruit and crops and making fires and warm clothing a real necessity
for the shivering inhabitants, the more so as they are, as a general
rule, unprepared for a phenomenon so at variance with the average
climatic conditions and therefore inadequately provided against
it. Such are the Southerly Burster of New South Wales, the
Pampero of the Argentine and th~ Norther of the Gulf-Atlantic
States. Their arrival is usually sudden and squally and is often
accompanied by violent hailstorms and thunder, destructive on
land and a menace to shipping at sea. At the other extreme
invasions of equatorial air bring spells of unpleasantly hot weather,
especially during the sum~er months. The Brickfielders of
Victoria may"bring temperatures of I200 with clouds of choking
dust; hot spells with daily temperatures above roo o, often pro-
longed for days at a time, are an unpleasant feature of the climate
of Melbourne. The Zonda of the Argentine is a hot humid wind
which brings a feeling of complete prostration, leaving the body
an easy victim to disease and the mind in a state of dejection and
depression, sometiI.nes amounting to temporary insanity.
TEMPERATURE. Winters are mild, the mean temperatures
being about 50°, but there are considerable departures from the
mean figures, as described above, and frosts, though rare, occa-
sionally occur, esp.ecially inland. The temperature rises steadily
as the sun rises higher, the spring temperatures being about those
of a London summer. The maximum, usually between 70° and
75° in the southern hemisphere but from 75° to 85° in the more
continental northern hemisphere, is reached rather late, especially
near-the coast, owing to the strong marine influence in summer;
autumn, for the same reason, is always considerably warmer than
spring. The summer heat is oppressive since the humidity is
high and there is little wind to bring relief. The thermometer
rises above goO almost daily at midsummer and sometimes exceeds
rooo. This is the most unhealthy season of the year, the mortality
from dysentery and malaria rises rapidly as heat and humidity
increase and it continues to rise into late summer and autumn,
for the humidity remains high after the highest temperatures are
past. Even when health does not actually suffer there is a loss of
energy which makes such climates unsuited to manual labour by
white men, and it has been found more satisfactory, if not abso-
lutely necessary, to employ coloured labour in, for example, the'
r60 CLIMATOLOGY
cotton fields of the Gulf-Atlantic States and in the tea plantations
of Natal.
RAINFALL. The rainfall is characteristically adequate but
not excessive and its incidence is well distributed over the year.
The ideally uniform rainfall is realized, for example, at Dubbo in
New South Wales where a difference of half an inch only separates
the wettest from the driest month. But though it is characteristic
that the monthly totals are more or less equal, the nature of
the rain and its efficacy vary considerably in the different
seasons. The winter rain is chiefly of the cyclonic type,
occurring as light showers or prolonged drizzles; the summer
rain, on the other hand, is either orographic or instability rain,
occurring as heavy downpours, an hour of which may produce as
much in the gauge as a week of the winter type. Much of the

JFMAMJJASOND J F M A M J J A S 0 N 0

4
2
0

8 8
{> {,

4 4
2 (1

0 0
e Charleston (u.S.A.)
FIG. sr.-Rainfall Regimes in Eastern Margin Warm Temperate Climates
(For explanation see text)

summer rain is therefore incapable of utilization, being lost by


run-off and evaporation, while the winter rain possesses a high
degree of efficiency. The rainfall of the ~ummer half of the year
usually slightly exceeds that of the winter half, but the maximum
occurs at different times in different places.
Away from the marine influence which prevails along the
eastern coasts there is a tendency for the maximum to occur in
spring (see Fig. 5r, a and b), thereby showing a transition into the
steppe type of the interiors of continents in these latitudes.
Often a distinct minimum occurs in autumn (Fig. 5r, a), an arrange-
ment very favourable for the harvest. Along the eastern littoral,
however, the heaviest fall tends to occur in autumn, during which
season the off-sea winds are most humid (see Fig. 5r, c and d.)
Yet a third rainfall regime is found at such stations as New
brleans and Charleston (see Fig. 5r, e), where there is a pronounced
EASTERN "MARGIN WARM TEMPERATE r6r"
summer maximum, the result of a summer monsoon. This type
will be met with in a more pronounced development in the
monsoonal variety of South China, to be discussed presently.
VEGETATION AND CULTIVATION. As in the Mediterranean
climates, the temperature, even of the coldest month, is not so
low as to check entirely the growth of the plant, and perennial
activity is possible. Owing to the similarity of temperature
conditions many species of conifers (e.g. cypress), shrubs (e.g.
laurel) and other evergreens (e.g. evergreen oak) are common to
both climates, but the absence of a dry season makes the conditions
favourable for many handsome and important species, such
as the tree ferns, bamboos, lianas, magnolia, etc., which are
excluded from western margins by the summer drought. The
regular rainfall supports a forest vegetation which is usually
broad-leaved evergreen, but sometimes deciduous, while con-
iferous £c)rests also occur. Many of the trees are of considerable
economic value as timbers and .~specially cabinet woods, e.g. oak,
maple, walnut, hickory, tulip-wood, etc., while the mulberry and
tea grow abundant crops of valuable leaves during the long moist
summer.
There are, however, considerable areas, e.g. the Argentine
Pampas, which, though well enough watered for forest growth,
have a grassland vegetation. The reason for this is a matter for
speculation; it may be due to deforestation, or perhaps to strong
winds or perhaps to the geological history-an earlier episode of
greater aridity and· a lack of later colonization by forest trees
from areas outside. If given protection in early growth trees can
be readily established here and are frequently planted round
estancias, acting as wind-breaks.
But typically these climates are agricultural and horticultural
in their economy, growing valuable crops of tobacco, cotton,
maize," rice, tea, sugar-cane, oranges, etc. Their great virtue in
this <!Onnection lies in the length of the summer, combined with
its constant humidity; a growing season, free from frost, of :200
days or more is ensured, while the promise of a cool season to
come induces heavier fruiting in the autumn than in the peren-
nially hot climates of tropical latitudes. The typical cereal is
maize whith finds here the summer rain it requires, and, in the
wetter parts, rice. Other cereals do not as a rule thrive, because
of the absence of a hot dry season for ripening and harvest,
though there are many local exceptions to this general statement.
The warm winters make autumn sowing possible and two or more
crops may be raised should the density of the population demand
it; but the more usual practice is to grow a single crop which
may safely have a long growing season. Numerous varieties of'
CLIMATOLOGY
fruit can be raised, irrigation being generally unnecessary; but
there is a certain risk in damage to orchards from the cold snaps
which are to be expected in the winter; this applies especially to
the delicate citrus fruits.
Their high productive capacity makes these climates capable
of supporting a population of considerable density, but with the
exception of the monsoonal variety they are as yet not fully
developed. Where the stress of a dense population demands
that two or more crops shall be taken off the land the routine of
the agriculturist is arduous in the extreme; but where one crop
. only is aimed at the yearly round allows more leisure, and anxiety
is reduced to a minimum by the reliability of the climate.

Regional Types:
New South Wales and Victoria
RAINFALL. The south-eastern angle of Australia from Port
Macquarie to Cape Otway enjoys uniformly distributed rains,

FIG. 52.-Antarctic Depression Un- FIG. 53.-Antarctic Depression Fav-


favoura9Je to Inland Rains. No ourable to Good Inland Rains.
Tropical Connection Trough Connection with Tropical
Low

transitional between the tropical summer rains ofQueensland and


the cyclonic winter rains of Tasmania. Bass St;ait and the seas
to the south of the continent offer the easiest route for the cyclonic
storms of the westerlies and the storm centres seldom cross the
land, the rain being borne chiefly on the southerly and westerly
winds in rear of the centres. Such rain does not often penetrate
far inland unless the cyclonic centre is connected by a trough
through the high-pressure ridge with the tropical low pressures
lying to the north. The heavy autumn rains of Sydney (3I per
cent. in the three autumn months) come chiefly on south-easterly
, winds connected with anticyclones over Bass Strait or the Tasman
EASTERN MARGIN WARM TEMPERATE I63
sea. Similar anticyclones follow similar tracks in spring but
bring less rain, the decisive factor being the temperature difference
between sea and land. The high humidities which often obtain
in autumn make that season unpleasantly muggy at Sydney.
The rainfall decreases rapidly inland from 40 inches along the
coastal strip to 20 inches beyond the crest of the Australian Alps,
but there are numerous examples of local increase or decrease due
to relief. The Gourock Range has 30 inches which decreases to
less than 20 inches in the trough of the upper Murrumbidgee, but
the Australian Alps beyond have more than 50 inches.
STORMS. It is along the coast of New South Wales from Cape
Howe to Port Macquarie that the Southerly Burster is felt at its
worst. The weather chart of one of these storms shows a V-
shaped depression extending northwards from a cyclone over the
seas tb the south of Australia; winds in front are northerly and
warm, then, as the trough approaches, from the south or south-
west there appears a typical h;mg dark roll of heavy cloud, often
with a front thirty miles long. The wind drops to an ominous
calm and then, without warning, returns as a violent cold blast
from the south, whirling before it a cloud of blinding dust, and
accompanied by thunder and lightning and, more rarely, rain or
hail. The drop in temperature is sudden and considerable, a fall
of ISo in five minutes is not unknown and the total fall often
exceeds 30°, though 20° is a more usual figure. Two factors
encourage the best development of Southerly Bursters along the
New South Wales coast: (I) The depressions acquire renewed
vigour on reaching the warm eastern seas, and (2) the meridional
trend of the mountains exercises a selective influence over wind
direction, favouring particularly south or north winds.
The storms are most numerous in spring and summer, Novem-
ber and December being the favourite months.

Argentine, Uruguay and South Brazil


During the winter months the high-pressure ridge extends
almost continuously across South America just south of the tropic;
south of this the weather is affected by the procession of cyclones
which enter from the Pacific, cross the pampean plains and pass
out into the South Atlantic in the region of the Plate estuary,
accompanied on their south-western margin by south-easterly
rain-bearing winds. South-easterly winds also occur on the
advancing edge of moving anticyclones and these also are rain-
bearing (d. New South Wales, p. I62). The passage of storms
during the summer months and especially from October to
January, is often characterized by the sudden arrival of a cold,·
CLIMATOLOGY
dry, stormywind.of gale force (the Pampero),exactlycomparable
to the Southerly Burster, through rather less violent on q,ccount
of the less favourable relief.
During the summer the continental low splits the high-pressure
belt into separate Atlantic and Pacific anticyclones, and winds,
though by no means regular, are prevailingly easterly, winds from
the N.E., E. and S.E. constituting,about 60 per cent. of the total
from December to February. These are the equivalent of the
trade winds and are rain-bearing; thu~ there is well-distributed
rain at all seasons.
Much of this region is excellent wheat and maize country, the
wheat being autumn sown and utilizing the winter and spring
rain (12 inches at Santa Fe), the maize being spring or early sum-
mer so:wn and utilizing the summer and autumn rainfall (20 inches
at Santa Fe). The percentage of the rain falling during the sum-
mer months decreases from north, to south, and this, taken in
conjunction with the progressive southward decrease of tempera-
ture, gradually sways the balance of conditions from maize to
wheat. Rosario, with 24 inches in the six summer months and
a January temperature of 76°, has a maize climate; Bahia Blanca,
with only 12 inches in the six summer months and a January
temperature of 71°, is well suited to wheat. There is, moreover,
an increasing tendency towards the south for a second minimum
of rainfall to occur at midsummer, thus providing a dry season
for harvest. At Bahia Blanca November and March are the
wettest months, while December and January each have less than
2 inches. '

South Africa
The type has a very small distribution in South Africa since
the continent does not extend south of 35°, but the coastal zone
from Cape Agulhas to Port St. John's may be placed here. The
driest month at Knysna has nearly 2 inches and the wettest less
than 3'5 inches. Westward along the south coast ther,e is a
gradual passage to the Mediterranean type with summer drought
(see p. 153), northwards along the east coast to the tropical marine
type with a distinct summer maximum (Durban) and inland to a
steppe type on the plateau (Aliwal). •
Owing partly to the Agulhas current winter temperatures are
rather high; Port Elizabeth is 4° warmer in July than Cape
Town in the same latitude, while Berg winds, similar to those of
the Cape Peninsula" bring unpleasantly high temperatures at
times.
EASTERN MARGIN WARM TEMPERATE I65

The Gulf-Atlantic States of U.S.A .


. North America introduces a slight monsoonal complication
. into the normal Eastern Margin Warm Temperate type, but not
sufficient· to justify its inclusion in the monsoonal sub-type.
Charts of mean wind direction do actually show a complete
seasonal wind reversal, but this is rather misleading as the direc-
tion is by no means constant but manifests from day to day a
high degree of variability, and the weather types vary accordingly.
The pressure gradient is never comparable with that of Asia
(·2 inches between Nebraska and Cuba compared with '5 inches
between Ordos and Luzon), so that the influence of, for example,
passing cyclonic storms is sufficient to bring about a complete
obliteration of the' prevailing' wind. The prevailing wind direc-
tion brings to bear, however, a strong influence which makes itself
felt especially in the strong summer maximum of rain.
WINTER CONDITIONS. Winter winds are northerly along the
Gulf coast and north-westerly along the Atlantic littoral, but it is
not until December that these winds are firmly established.
Mean temperat)lres are rendered somewhat low by this prevailing
continental influence, as the following pairs of stations show. In
each case. the latitude, relation to surroundipg relief and proximity
to sea make the stations comparable:- .
Charleston: Coldest month, 49'3°. Pt. Macquarie, 54'5°.
Vicksburg: ,,47°. Parana, 54°·
New Orleans: 54°. Durban, 64.6.°
These mean temperatures are the resultants of considerable
variations of daily temperature, usually connected with passing
cyclonic storms which frequently travel along the 'southern
circuit' (see p. I9I) during the winter months. The southerly
winds of the advancing edge bring spells of warm muggy weather,
followed by cold dry spells as the wind swings to the north in rear.
The northerly winds are often strong since they reinforce the
prevailing wind and sometimes, under special conditions, they are
gales of bitterly cold air, the Northers of Texas in pjtrticular and
of the Gulf States in general. The special conditions consist of a
steep gradient between a deep cyclone and a following anticyclone
advancing across the plains; it is a ' line-squall' phenomenon
exactly comparable to the Pampero and the Southerly Burster
and is accompanied by the same roll of heavy cloud, the same
suddenness of onslaught and the same rapid drop of temperature,
often 50° or 60° in a few hours. The sufferings of man, beast and
plant are all the more acute on account of the muggy warmth of
the weather immediately preceding.
11
166 CLIMATOLOGY
The rainfall of the winter season is essentially cyclonic and
it is the southerly wind off the Gulf (the exact tJpposite of the,
prevailing wind) which brings most of the rain. Some stations
show a clearly defined tendency to a secondary maximum in late
winter when cyclonic activity is greatest. .
As winter gives way to spring and summer the cyclonic control,
weakens, and though cyclones passing along more northerly routes
may still draw occasional wet southerly winds off the Gulf, the
prevailing wind, still northerly, is less frequently interrupted;
hence the rainfall generally decreases until by April and Maya
minimum has been reached.
SUMMER CONDITIONS. The steady rise of temperature in the
continental interior .weakens the high-pressure syste~ and ulti-
mately substitutes a low pressure into which a southerly monsoonal
wind blows off the Gulf and up the Mississippi Valley, gradually
increasing in strength and reliability. It is a current of high
temperature and humidity, bringing abundant rain to the cotton
fields of the southern states and to the maize belt further north.
Temperature mounts steadily to July maxima of 80° which extend
as far north as the Ohio confluence, and this, combined with a
relative humidity of 80 per cent. or more, while encouraging
vigorous plant growth, makes the summer weather most unpleas-
antly enervating.
The prevalence of ?- sea wind, combined with thunderstorms,
is the main cause of heavy rainfall throughout the summer months
with a maximum coinciding with the greatest strength and 'fegu-
larity of the monsoonal wind. Along the Atlantic coast, and
especially in Florida, the maximum is delayed into September by
the occurrence of violent hurricanes with torrential rain (see
Fig. 54, Miami). They originate in the west Atlantic between
10° and 20° N. and follow the characteristically curved path·
passing up the east coast of Florida, Georgia and Carolina, but
seldom penetrating far inland. In the Miami hurricane of Sep-
tember, 1926, the barometer fell to 27.6 inches, a record for
U.S.A., wind velocities of over 100 m.p.h. were registered, causing
great havoc among fruit groves, crops and buildings, and 15
inches of rain fell, followed, naturally, by serious floods.
Autumn conditions are broadly similar to those of spring and,
like spring, the late autumn is generally rather dry; in fact,
irrigation is sometimes necessary at this season.
RAINFALL REGIMES. Reviewing broadly the rainfall of this
south-east quadrant of the United States two main types may be
recognized : -
I. Cyclonic rain with a winter maximum.
~. Monsoonal rain with a summer maximum.
EASTERN MARGIN WARM TEMPERATE r67
To these may be added two subsidiary types :-
3. Thunderstorm rain with a summer maximum.
4- Hurricane rain with a late summer maximum.
Types 2, 3 and 4 combine to give a predominant summer
maximum at nearly all stations, while Type I is generally recog-
nizable as a secondary winter maximum. Spring and autumn are
usually rather dry, occupying positions intermediate between the
cyclonic rains of winter and the monsoon rains of summer. This-
is the normal Gulf Coast type (see Fig. 54, Galveston).
- _ Stations round the southern epd of the Appalachians show a
maximum in late winter or early spring which corresponds to a
maximum of cyclonic activity (see Fig. 54, Montgomery). This
last shows a close relationship to the rainfall regime over the
greater part of the continental interior (Greely's Missouri Type 1).
CROPS. If the January isotherm of 43° is taken as the
6 10
4 8
2 6
o 4
Galveston 2
'6 '0
4
Miami
2
o
Montgomery
FIG. 54.-Rainfall Regimes in the Gulf-Atlantic States

northern limit, the boundary of the climatic province practically


coincides with that of the Cotton Belt as defined by O. E. Baker, 2
including the Carolinas, Georgia,' 'Florida, Alabama, Mississippi,
Louisiana, Arkansas and much of Oklahoma and Texas. This is,
however, mainly coincidence, since it is summer temperature
(the July isotherm of 77°) which delimits the cotton belt. Cotton
of the Sea Island type along the coast and of the American Upland
type inland, is by far the most important crop, the long moist,
hot summer giving suitable growing conditions, while the some-
what dry and cool autumn ensures good quality and reliable
conditions for picking. Further west, in Texas, the rainfall
decreases and Egyptian type is grown, often with irrigation.
Sugar, rice and other semi-tropical crops are cultivated, while the
especially mild area round the Gulf is noted for fruit and early
1 A. W. Greely, ' Rainfall Types of the United States,' Nat. Geog. Mag.,
18 93.
2 U.S. Dept. of Agriculture, Yearbook for 1921.
168 CLIMATOLOGY
vegetable farming on a huge scale. Florida, projecting 5° further
south and with winter temperatures 10° higher than the Gulf
.
coast, is the chief centre of the citrus fruit industry .

The Monsoonal Sub -Type


I t has been laid down as one of the characteristics of the normal
type that the wind is variable in force and direction owing to the
essentially transitional latitude, and it is in this fundamental
respect that the monsoon type varies from the normal; for its
winds, under the strong continental influence, exhibit a markedly
greater degree of reliability and regularity. In consequence of
this the seasons are much more clearly differentiated, the summer
being typically marine and the winter steadily continentaL This
finds expression particularly in the following ways : -
1. A considerably greater range of temperature, due especially
to the lower winter temperatun;s. Shanghai is 17° colder in
winter and 7° warmer in summer than Port Macquarie.
z. The rainfall is much more seasonal in nature: at Amoy
74 per cent. of the rain falls in the six summer months, at Port
Macquarie only 54 per cent.
3. The arrival and departure of summer are much more
sudden and in general the passage from one season to another is
much more clearly defined.
LIMITS. The element which perhaps most clearly dissDciates
the monsoonal from the normal climates is the winter temperature,
for this falls below freezing everywhere north of Shantung;
Peking, in almost the same latitude as Melbourne, has four months
with mean temperatures below freezing. Such a climate is clearly
not' Warm Temperate' even though its high summer: tempera-
ture allows the cultivation of rice to the north of 35°. But rice
is a monsoonal, not a warm-temperate' cereal, its growing season
being comparatively short and its prime requirements not winter
mildness (the criterion of warm-temperate) but adequate moisture
and ~eat during the short growing season.
THE NORTHERN LIMIT OF THE TYPE. 1 TheJanuaryisotherm
of 43° has been taken as the poleward limit of the warm temperate
climates (see p. 65) and this runs just south of the Yang-tse;
China north of this line would belong, by virtue of its cold winters,
to the cool temperate climates. But in practice this line is not
very satisfactory since it bisects the natural unit of the Yang-tse
Valley which is uniform in climate, vegetation and agriculture.
1 For a discussion on the climatic provinces of China, see The Climatic
Provinces of China, by Coching Chu, Mem. Inst. Met., No. I, N~tional
Research Institute, Nanking, April, 1929.
EASTERN MARGIN WARM TEMPERATE 169
The winters here, though in the northern part of the basin they
are as cold as those of England, are much shorter; too short, in
fact, to impose an effective check on the growth of trees, the
'majority of which are broad-leaved evergreens, though mixed with
deciduous species belonging to the more northerly climatic pro-
vince (chestnut and maple). It is the mountain rim enclosing
the Yang-tse Valley on the north, the Ta-pa-shan and the
Hwai-yang-shan, which function as the real, climatic 'and vegeta-
tional divide. The water-parting, in fact, being more palpable and
better defined, forms a more significant limit than the isothermal
line and m~y, with advantage, be adopted as the boundary
bet~een the 'warm- and cool-temperate climates .. The climate
usually called 'Chinese' is thus considered here under two
heads':-
1. A warm temperate type with a long growing season in
which two or three crops a year may be obtained; mulberry, tea,
sugar and other sub-tropical crops are grown.
2. A cool temperate type 'with cold winters and a relatively
short growing season, only one or two crops a year being possible
as a rule, and these are cool-temperate crops, e.g. wheat, barley,
maize, beans, etc.
The warm-temperate type, as delimited above, has a compara-
tively small extension in southern China only and does not extend
north of about 32°. Such a latitude elsewhere corresponds more
closely with the equatorward margin of warm temperate climates
since it roughly coincides with the lower margin of the transition
zone between the trades and the westerlies. Here, however, these'
elements of the planetary circulation have no significance, being
entirely obliterated by the alternating monsoons. The' pro-
nounced continentality of the winter, in fact, forces the winter
isotherms further south here than. anywhere else in the world and
repels all the climatic zones towards lower latitudes (see the
northern limit of the tropical monsoon, p. 131).
WINTER CONDITIONS. According to the latitude, South China
should lie in the high-press,+re belt and its winds should be directed
equatorwards; this tendency gives an added impulse to the
winter monsoon which consequently exceeds, both in strength and
duration, the summer monsoon which is established in the face
of the planetary circulation. Winds over southern China are
strong (10-15 m.p.h.) and northerly in direction, swinging to
north-easterly in the extreme south. The steepest pressure
gradient occurs between Japan and the Philippines ('5 inches in the
1,500 miles between Shanghai and Mindanao) and here are found
the strongest winds. Their arrival is sudden and violent; rough
seas confine native shipping to the harbours, and even the fastest·
170 CLIMATOLOGY
mail steamers may take five days to cover the 850 miles from
Hong-Kong to Shanghai in the teeth of the ga.les concentrated in
the Formosa Strait. To the northern shores of the mountainous
island of Formosa they bring 8 inches of relief rain in January,
but elsewhere the January rainfall is less than 3 inches alld
mainly cyclonic in origin.
THE CONTINENTAL CYCLONES introduce a second main factor
into the winter climate, temporarily interruPting the dominance
of the monsoon and upsetting the ideal regularity of the winds.
On the relative vigour of these on the one hand and of the Siberian
anticyclone on the other depends the weather experienced.
With an excessive development of the latter to the exclusion of
the former are associated seasons of excessiv~ cold and drought,
while an excessive development of the cyclones brings a winter of
rain and mist with mild periods interspersed with cold waves.
These cold-weather storms may be compared with those of
Northern India (p. IIZ) and, like these, theit place of origin is
uncertain. It is inconceivable that they should travel across the
whore wfdtli of Europe and Asia from tlie Atfqntic; it is, in fact,
a matter of observation that the Atlantic ~torms very rarely
reach as far as the Urals, while the rare survivors pass eastward by
the valleys of the Ob and Yenisei, and die out near Lake Baikal.
Very exceptionally, storms may survive into Mongolia, but these
are much too far north to affect southern China. Most would
appear to originate within the continent and to follow fairly well-
defined tracks (see Fig. 59, p. 198). The storms are frequent
throughout the cold season, but reach a maxirrmm in spring, when
the influence of the anticyclone is waning, January to April
being the favourite months. Along the mo:;t southerly track,
following the Si-kiang Valley, the frequency is as follows:-
October z, November 2, December 7, January 10, February 13,
March 18. '
The temperatures in front and in rear o~ these storms are,
as usual, strongly contrasted; warmth comparable to that of the
summer months and often exceeding 80° is often brought by the
southerly winds on the advancing edge and mcty endure for many
days since the storms usually travel but slOWly. The northerly
winds in rear are reinforced by the monsoonal pressure gradient
and blow with added force, bringing temperatures below freezing
almost as far south at Hong-Kong.
TEMPERATURE. Exposure to the north-w~st monsoon makes
China abnormally cold for its latitude in wiI1.ter, Hong-Kong is
nearly 7° colder in January than Calcutta in the same latitude, the
thermometer here has actually fallen to freezing-point and the
mean minimum is 43°. February on the coast is usually cooler
EASTERN MARGIN WARM TEMPERATE 171
than January, after which temperatures rise slowly as the year
advances. There is nothing in South China comparable to the
'.Hot Season' of India in the same latitudes, since winds from the
heart of Asia continue to cool the country right down to the coast;
Hong-Kong is 17° cooler than Calcutta in March and 15° cooler in
April. Temperature is largely dependent on exposure to or, pro-
tection from the cold monsoon winds; thus Szechwan, screened
by the high mountains on its northern rim, is, in spite of its alti-
tude, much warmer than the Yang-tse delta. The January tem-
perature at Chung-King is lIo higher than that at Shanghai. This
favoured basin can grow many sub-tropical plants (e.g. sugar and
tobacco) which the rigour of winter forbids elsewhere. In some
cases, especially on southern slopes which fortunately predomi-
nate in the basin, three crops may be obtained in the same year,
and this exceptional fecundity permits the basin to support
a population of about 50 millions at a density of some 500 per
square mile, with agriculture a? the chief means of subsistence.
RAINFALL. Winter rainfall in South China is almost entirely
cyclonic and its distribution is closely related to the most fre-
quented cyclone tracks. Central China has more than South
China since the Yang-tse Valley is a much used route; the coast
has more than the interior since the sea is the chief source of
moisture. Cyclones become more frequent and the rainfall
increases as the season advances and as the anti-cyclone weakens
(see data for Hong-Kong and Fuchau).
SUMMER CONDITIONS. WINDS AND RAINFALL.1 The con-
tinental anticyclone weakens in spring and by April has ceased to
be an important control; winds along the coast are now easterly
or south-easterly but cannot yet be called the S.E. monsoon since
they are extremely light and variable and bring but little rain,
By May the Asiatic low is developing strongly, the south-east
winds are gaining in strength and regularity and the heavy rains
have set in at Hong-Kong (April, 5 inches; May, 12 inches). The
effect is felt later and later northwards up the coast; at Shanghai
May (3,6 inches) is no wetter than April (3'7 inches), but the rain-
fall of June (7'4 inches) is double that ,of May. Sudden though
the increase of rainfall is, it is in no way comparable with the
, Burst' of the monsoon in India, so well displayed along the
Western Ghats, as a comparison of the rainfall of Bombay and
Hong-Kong clearly shows. This is explained by a variety of
causes : -
I. Owing to the continental cyclones there is no real dry season
in South China,
1 The rainfall regime of the Yang-tse valley is described in connexion
with North China on p. 199,
I7 2 CLIMATOLOGY
2. The monsoon current is of less vigour and lower humidity;
the pressure gradient in India is much steeper and the volume.of
air arriving is proportionately greater there.
3. The Indian monsoon has travelled over 4,000 miles of
continuous hot sea, while the Chinese monsoon has parted witn
some of its moisture to the East Indies and other islands which
lie in its path. .
The prevailing south-east wind raises the level of the sea along
the coast by an amount which varies from eighteen inches to five
feet. It imports high humidities and much cloud from the
Pacific Ocean, but rain does not generally fall until a cyclonic
disturbance, generally shallow but covering a wide area, upsets-
the equilibrium of the air. Thus the S.E. monsoon is not a
season of persistent rain, but a time of unsettled weather in
which heavy showers are interspersed with spells of fine weather.
Rain falls, as a rule, at least once a week and, in some places, on
one day in three during the wettest months, but occasionally the
fine spells are unduly prolonged.and crops suffer; for though there
may be no actual drought the rice crop is particularly.sensitive
at this stage and ten days or a fortnight without rain at the time
of planting or transplanting will have a disastrous sequel at the
harvest.
It should be noticed that the Gulf of Tongking renders to
China a service similar to that which the Gulf of Mexico renders
to North America, supplying moisture to the on-shore winds and
increasing the rainfall; but the Gulf of Tongking is not so warm,
for its latitude and its effect is not so great, a~ the restricted area
with 40 inches shows.
TEMPERATURE. The summer temperatures are monotonously
high (Hong-Kong has four months above 80°), and the humid
heat is disagreeably enervating, but since the wind is off the sea
unduly high figures are seldom reached along the coa!?t; the
mean maximum at Hong-Kong is only 4° higher than the mean
temperature of the hottest month. Inland it is hotter and the
85° isotherm makes a closed loop in the heart of central and
Southern China. Consequently the highest temperatures along
the coast are recorded when the wind is off the land. In late
autumn, for example, cyclones sometimes cause north-west winds
to descend the slopes from Yun-Nan into the delta of the Red
River as a foehn wind with temperatures of IOSo.
TYPHOONS. A characteristic element of the climate is the
tropical cyclone, here known as a typhoon, of which eight or nine
may be expected in a year. They are most frequent in late sum-
mer (85 per cent. occurring in July, August and September) and
"they affect chiefly Southern China. The course followed is
EASTERN MARGIN; WARM TEMPERATE 173

closely related to the pressure distribution of the season and


especially to the position of the high-pressure areas. In winter
they are repelled by the continental high and do not penetrate
inland, but in summer they frequently invade South China and
Tongking, gradually weakening as they pass inland but bringing
considerable rain to Central and Northern China. Usually, how-
ever, they follow the typical parabolic course, skirting the Pacific
a:nticyclone on its western margin and thus passing close to the
Chinese coast. During the disastrous Swatow typhoon of 1922
the barometer fell to 27'5 inches and the wind blew at 100 m.p.h.
for two hours. The shift of the wind to the south as the centre
passed brought a 'tidal' wave which inundated the town and
caused an estimated loss of 50,000 lives. Torrential rain accom-
panies these storms and materially affects the average figures for
coastal stations in this zone. Hong-Kong has recorded more than
27 inches in 24 hours, and nearly 20 inches in 8 hours. The
importance of this rain is, however, often over-stressed and it has
been shown that the maxima of rain do not coincide with the
maxima of typhoons; it is very highly localized and it concerns
only coastal stations; probably it does not add much to the
mean monthly figure at any given place inland.

SUGGESTIONS FOR FURTHER READING


For Australia, see works quoted in earlier chapters.
For South America, in addition to the general works already quoted,
there is The Climate of the Argentine Republic, by W. G. Davis, Buenos
Aires, 2nd ed., 1910. An article by N. A. Hessling on ' Relation between
Weather and the Yield of Wheat in the Argentine', Monthly Weather
Review, 1922, may be consulted.
For the United States, read Ward's Climate of the United States, and
frequent articles in the Monthly Weather Review.
For China, there is an excellent ·summary by Kendrew in Dudley
Buxton's China, 1928. See also ' Climate of China,' C. iE. Koeppe andN. H.
Bangs in the Monthly WeatherBeview, 1928; Etude sur la pluie en Chine,
E. Gherzi, Shanghai, 1928 (reviewed and summarized in the Monthly Weather
Review, 1929); 'Weather Types in East China', Coching Chu, Geography,
1928.
For Japan, see' The Climate of Japan and Formosa', E. M. Sanders,
Monthly Weather Review, 1920.
I74 CLIMATOLOGY

TEMPERATURE

Station. Lat. Long. Alt. (ft.) J F M


~I~ J I J A S O· N
I-t!
1
DUBBO 32° s. 149° E. 87 0 79 77 71 64 55 49 47 51 56 63 71
WILCANNIA. 31° S. 143° E. 26 7 81 80 74 65 58 52 50 54 60 68 75
PORT MACgU ARIE 31° S. 153° E. 49 73 73 71 66 61 56 55 57 60 64 68
SYDNEY 34° S. 151° E. 13 8 72 71 69 65 59 55 53 55 59 64 67
MELBOURNE 38° S. 145° E. 115 68 68 65 60 54 51 49 51 54 58 61
AUCKLAND. 37° S. 175° E. 260 67 67 66 61 57 54 52 52 55 57 60
WELLINGTON 41° S. 175° E. 10 63 63 61 57 53 49 48 49 52 54 57
PORT ELIZABETH 34° S. 26° E. 181 69 70 68 65 62 59 58 58 60 62 65
MONTEVIDEO 35° S. 56° W . 96 72 72 69 63 57 51 51 51 55 58 65
PARANA. 32° S. 61° W. 66 77 77 73 66 59 54 54 57 61 66 72
ROSARIO 33° S. 61° W. 95 77 76 70 62 56 49 51 52 57 62 69
BUENOS AIRES . 35° S. 58° W. 82 74 73 69 61 55 50 49 51 55 60 66
.
BAHIA BLANCA 39° s. 62 W.
0
82 74 72 67 60 53 47 47 49 54 59 66
NEW ORLEANS. 3ooN . 90 0 W. 53 54 57 63 69 75 80 82 81 78 69 61
GALVESTON. 29° N. 95° W . 69 54 56 63 70 76 82 84 83 80 73 63
MOBILE. 31° N. 88°W. 57 51 54 60 66 73 79 80 80 77 68 58
PINE BLUFF 34° N. 92o W. 21 5 43 45 55 64 72 79 83 81 76 63. 53 )
VICKSBURG. 32° N. 91° W. 247 47 51 58 65 73 79 80 80 75 65 56
HATTERAS 35° N. 76°W. II 4 6 46 51 58 67 74 78 77 74 65 56
,
ABILENE 32oN . 100° W. 1,73 8 45 47 56 64 72 79 82 82 75 65 53
FUCHAU 26° N. II9° E. 66 53 51 62 67 73 78 86 86 81 72 66
CHUNG-KING 30° N. 107° E. 750 48 50 58 68 74 80 83 86 77 68 59
>

KAGOSHIMA 35° N. 131° E. 394 45 45 51 60 65 71 78 80 75 66 57


EASTERN MARGIN WARM TEMPERATE 175

RAINFALL

D Yr J Ra,
F M A M J J A S 0 N D Total.

- ~ -- -- --
- --
-- -- -- -- ------
1,8 1'9 1:9 1'9 2'0 1,8 1'9 .1,6 1,8 2'0
76 63 3 2 2'1 1'9 22'6
0,8 r'r 0'7 r'o r'r 0,6 0,8 0'7 o'g 0'7 0,8
79 66 31 1'0 10'2

71 64 18 5'9 7'5 6'5 5'9 5,6 4,6 4'5 3,8 3'9 3'2 4'1 5'9 61'4
4,8 5,6 5'1 4,8 4,8 2,8 2'9
70 63 19 3'7 4'3 3'0 2'9 3'2 47'9
,.'

65 59 19 1'9 107 '2'2 2'3 2'2 2'1 1,8 1,8 2'4 2,6 2'2 2'3 25'5

64 59 15 2,6 3'0 3'1 3'3 4'4 4,8 5'0 4'2 3,6 3,6 3'3 2'9 43'8
60 55 15 3'3 3'1 3'3 3'9 4'7 4,8 5,6 4'5 4'0 4'r 3'5 3'2 4 8 '0
68 64 II 1'2 1'3 1,8 2'0 2'4 1'7' 1'9 2'1 2'2 2'1 2'1 1"7 22'5

69 61 21 2'7 2,8 3'2 4°5 3'5 3° 2 2'5 3,6 3'4 2,6 3'2 3'6 3808

76 66 23 4'4 3,8 3'7 3,8 1,8 o'g r'o 1 °3 1'9 4'4 3,6 4,8 35°4

75 63 28 3'7 3'2 5'3 3'1 I,d 1'5 1°0 1'5 1 06 3'5 3'4 5'3 34'9

71 61 25 3'1 2'7 4'4 3'5 2'9 2'5 2'2 2'5 3'0 3'5 3'1 3'9 37'3

71 60 27 2'0 2'2 2,6 2'2 lor 0'9 1'0 1°0 1,6 2'3 2'0 2'1 21'0

55 68 28 4'5 4'3 4,6 4'5 4'1 5'4 6°5 5'7 4'5 3'2 3,8 4°5 55"6

57 70 30 3'4 3'0 2'9 3'1 3'4 4'2 4'0 4'7 5"7 4'3 3'9 3"7 4 6 °3
52 67 29 4'7 5'2 6'4 4'9 4'4 5"4 7'0 7'1 5'3 3°5 3'7 4°9 62'5

'45 63 40 5'9 3'9 5,6 4,6 5'2 4'0 3'9 2'7 3"7 2'0 4,8 4'9 51'2

49 65 33 5'2 4,8 5'5 5'0 4'3 4'0 4,6 3'4 3°3 2,6 4°3 5"0 52 '0

48 62 3 2 4'3 4'1 4'5 3'9 3,8 4'5 5'S 5"8 5'2 5'2 3,6 4,6 55'0

46 64 37 0'9 o'g 1°2 2,6 3'9 2'7 2'0 2'4 2,6 2'5 1 °4 1'2 24°3

57 69 35 3'1 2'5 4,8 5"3 4,6 6'0 4'3 8"7 3'0 1 °3 0,8 1 °3 45°7

.
50 ·f)7 38 0'7 Oog r03 4'0 5'3 6'7 5'3 4'4 5,8 4'6 2'0 0'9 4 J '9
48 62 35 3'5 3'3 6'1 9'1 9,6 13'9 11 02 7"4 8'7 5'1 3'7 3'5 85'1
CHAPTER X
COOL-TEMPERATE CLIMA1:ES
HE quality which divides the cool temperate from the

T warm-temperate or sub-tropical climates is the possession


of a real colQ. season which retards or inhibits active
plant growth and places a check on agricultural activity. The
severity of the cold season increases from west to east, away from
the marine influence, and in this direction the seasonal control
of occupation becomes correspondingly more pronounced; the
autumn-sown wheat of western margins is replaced by the spring-'
sown wheat of continental interiors and there is, for example, a
considerable exodus from the prairie provinces when the dead
season of winter sets in. Winter warmth and rain come from the
oceans to the west and it is the relative position of land and sea
which, more than latitude, determines the seasonal distribution
of temperature and pressure and the position of forest, grassland
and desert.
THE CYCLONIC UNIT which exercised such a strong influence
on the winter conditions of the warm-temperate climates now
plays an important role throughout the year, though it still
assumes greater importance in winter than in summer. The
regular periodicity of prevailing winds, pressure, temperature and.
rainfall is masked and obscured by the non-periodic cyclonic
unit; the significance of climate decreases, and weather assumes
an importance such as it enjoys in no other climatic type.
While they exercise a considerable degree of individuality in
the· matter, these cyclones do tend to frequent certain fairly
clearly defined paths, especially following seas, gulfs, straits,
lak~s, river valleys, plains and humid areas of low relief generally.
They are repelled by the high pressures which develop ov'er
continents in winter and tend, therefore, to take a course along
continental margins at this season. The cyclonic control is
consequently more marked in the marine than in the continental
variety, whose winter conditions, under the influence of the anti-
cyclones, tend to be more reliable, more regularly periodic and
more constantly faithful to the climatic type.
CONTINENTAL AND MARINE TYPES. The distance to which
marine influence penetrates depends on the .configuration of the
176
COOL-TEMPERATE 177
western margin, as a com'parison of the new and old worlds well
shows. It makes itself felt in the following ways :-
r. A small annual range of temperature.
z. A higher humidity and rainfall.
3. Rainfall evenly distributed throughout the year, but with
a tendency to a winter' maximum (cyclonic) as compared with a
summer maximum (convectional) of the continental type.
4· An imperceptible gradation of the seasons with frequent
relapses from, for example, spring into winter or autumn into
summer.
PRESSURE AND WINDS. The prevailing westerly wind, though
fairly constant in the upper air, is disturbed near the ground by
tile frequent passage of cyclones and anticyclones and by the
yearly changes of the continental pressure systems to such an
extent as to be scarcely recognizable except by'records kept over
a period of time. The seasonal pressure changes over the two
great land masses of the northern hemisphere attempt to establish
systems of outflowing winds in winter and inflowing winds in
s\1mmer. In Eastern Asia this tendency is realized to a high degree,
the wind direction is almost completely reversed and a palp-
ably monsoonal regime is established. Elsewhere their success is
less complete and they are able only to modify the planetary regime.
During the winter of the northern hemisphere these continental
highs and the oceanic lows govern a system of winds which tend
to be south-west along the western margins of continents and
north-west along their eastern margins. The continental interiors
are under the dominance of the anticyclones, while south of the
axis of highest pressure the winds are northerly or north-easterly,
bitterly cold and e~cessively dry.
In summer the continental lows become the principal foci and
the oceanic lows virtually disappear; winds are, therefore,
mainly westerly on western margins and southerly, or even south-
easterly on eastern margins.
In the southern hemisphere the disturbance of the planetary
winds is much less; the' Roaring Forties' and the' Brave West
Winds' blow all the year round with considerable force and with
a high degree of dependability. In spite of the frequent passage
of depressions, westerly winds (S.W., W. and N.W.) make up
80 per cent. of the total winds at Punta Arenas.
TEMPERATURE. The designation' temperate' for these zones
is a decided misnomer, for within the limits of the cool-temperate
are found some of the most extreme temperatures in the world,
exceeded only in the cold climates further to the north. Mean
annual ranges exceeding 80° occur in Manchuria and Mongolia!
mean monthly temperatures IS° below zero at the one extreme
CLIMATOLOGY
and above 80° at the other are included in the zone, while extreme
temperatures of 105° and below- 20° find places in the climatic
type. However, these enormous ranges are limited to extreme
continental climates, the marine type well deserving the descrip-
tion 'temperate'. The annual range in the Scilly Isles~ for
example, is only IS°, frosts are rare and temperatures above
75° are exceptional.
The striking contrasts between oceanic and continental types
are a feature of the zone in the northern hemisphere, for here there
occur two of the largest land masses, separated by two oceans
which, for their latitude, are the warmest in the world. The North
Atlantic shows a positive anomaly in January of over 40° and the
North Pacific of over 20°, while both Siberia and North America
have negative anomalies exceeding 30° (see Fig. I, p. 12).
In the extremely oceanic southern hemisphere, by contrast, the
anomalies are less than 10° and the mean annual range seldom
exceeds 20°.
Temperatures below freezing are to be expected everywhere
away from the marine influence and river navigation is regularly
blocked by ice. The canals in Holland are frequently ice-bound'
and the flowing water of the Rhine is frozen on an average for
three weeks in the year at Cologne. Further east the closed
season is longer, the lower Danube is ice-bound for five or six
weeks and the upper Sungari for five months.
The prolonged cold of the continental winter makes some
efficient form of artificial heating necessary, and most houses,
especially in towns, are centrally heated; in the less extreme
winter of the marine climates there is not the same need of constant
heating and fires are generally lit every day, not in stoves but in
open grates whose bright glow helps to cheer the darkness of
cloudy winter days. ,
Maxima and minima of temperature are reached in January
and July, i.e. soon after the solstice, in the continental type, but'
there is frequently a retardation on western margins which may
amount to two months (e.g. the Scilly Isles where February is
the coldest month and August the hottest). Autumn is decidedly
warmer than spring in the marine type, but the difference becomes
less marked inland and finally almost disappears.
Mean figures, however, convey a very inadequate idea of the
temperature conditions, since owing to the irregularity of wind
direction resulting from cyclonic control rapid fluctuations of
considerable amplitude occur on eithet side of the mean. This is
a feature not found in hot climates since winds in low latitudes
bring warm air from whatever direction they blow; but in high
latitudes a change of wind direction imports air now from the warm
COOL-TEMPERATE 179
lands to the south (in the northern hemisphere), now from the
frozen north; in extreme cases a drop of 50° or 60° in 24 hours
may be recorded and of 30° within an hour. This demands
great adaptability on the part of plants, animals and man, and
survival«iepends on the degree of response to the demand. Up
to a poil).t the body is able to meet these extremes and adapt itself
thereto; in fact they are stimulating to both body and mind, but
powers of resistance, are weakened by the artificial conditions of
civilized life, heavy clothing, fires and central heating. Many of
the diseases characteristic of the zone, e.g, colds, influenza,
pneumonia, are more or less directly traceable to these sudden
temperature changes, while the lowering of vitality which results
from the strain on the system makes the individual more suscep-
tible to germ-borne diseases. Winter and early spring are
. generally the season of highest mortality since weather changes
are most frequent and severe at this time.
RAINFALL. Since winds anq. rain come mainly from the west
it)s natural that rain should be heaviest on western margins
and decrease eastwards. Where high relief presents a barrier to
the westerlies 80 inches or 100 inches may result and even 200
inches iocally. Hokitika (New Zealand), exposed to the full force
of the winds and backed by the New Zealand Alps, has 120 inches,
Bahia Felix, similarly exposed in South Chile, has over 200 inches,
while the Welsh mountains round Snowdon have a similar amount
under similar conditions. Such phenomenal figures are mainly
made up of orographic rain and are paid for by a rapid eastward
decrease and often a serious deficiency in lee of the mountain
barrier. But whether by coincidence or from more significant
geological causes, the west coasts in these latitudes are almost
everywhere highland coasts of the fjord type, of which British
Columbia, Norway, Western Scotland, South Chile' and New
Zealand are typical. The west coast of Europe is the most for-
tunate in this respect, for there are numerous breaks in the
fringing' Caledonian' Mountains which, moreover, reach no great
height south of the Skager Rak; thus it is only in Scandinavia
that there is any approximation to the suddenness of transition
from wet to dry, from marine to continental that occurs elsewhere.
In the Americas the 20-inch isohyet is within 200 miles of the west
coast, while in Europe 20 inches penetrate to Moscow, Kiev and
Bukarest-more than 1,000 miles from the ocean to the west.
But so great is the width of the Eurasian continent that rainfall
eventu.ally becomes too small to support even good grass, and
poor steppe and desert eventually supervene.
The rainfall thus decreases, suddenly or gradually according
to relief, to a minimum in the_interior of continents. On approach-
180 CLIMATOLOGY
ing the east coasts rainfall again increases, this rain being due to
the importation of moisture by easterly winds, generally in the front
of cyclonic centres. Thirty inches or 40 inches may occur here,
but again orographic influences may augment this considerably
(e.g. the Appalachians). In the southern hemisphere, where the
westerlies are stronger, more constant and less disturbed, relief
rain plays a more important part than cyclonic and eastern mar-
gins (e.g. Patagonia) are arid up to the coast.
A further source of rain on eastern margins is the monsoon
wherever this is well developed; since this is, by nature, summer
rain only, the seasonal regime is totally different and many
important consequences follow. This monsoonal sub-type, well
developed only in North China and Japan, deserves special
consideration and will be dealt with later.
TRANSITIONAL RAINFALL TYPES. In a meridional direction
the rainfall decreases polewards, in which direction the air is
colder and therefore carries less moisture, but equatorwards the
changes are less simple. On western margins there is a decrease
by loss of summer rain as these areas fall more and more under the
influence of the high pressures of mid-latitudes in summer, thus
providing a transition to the Mediterranean type. In continental
interiors there is a similar decrease, which, though small in actual
amount, is rendered more serious by the already deficient pre-
cipitation and by the higher temperatures which diminish the
efficacy of such rain as does fall. On eastern margins, on the
contrary, there is an increase equatorwards since the frequency of
rain-bearing easterly winds increases in this direction, especially
dliring the summer months. In this direction there is a gradual
passage to the humid warm-temperate type. Where there is a
monsoon the nature and distribution of the rain does not change,
but there is an increase in amount towards lower latitudes where
winds are warmer and carry more moisture.
SEASONAL AND DIURNAL DISTRIBUTION OF RAIN. It has been
pointed out on p. 18 that each type of rain tends to have its own
seasonal distribution; now all three types are represented in the
cool-temperate climates but with varying degrees of importance
in different places.
1. Relief rain which tends to have an autumn maximum, is of
greatest importance on western margins, is much reduced in
significance in continental interiors and attains renewed, though
secondary, significance on eastern margins. Where there is a
monsoon the maximum of relief rain coincides with the greatest
strength of the monsoon (i.e. midsummer). It tends to be
heaviest at night or in the morning when the difference of
temperature between land and sea is greatest.
COOL-TEMPERATE 181
z. Cyclonic rain attains its greatest importance in winter, at
which season the storms are most numerous and most vigorous;
nearly 70 per cent. of the storms over the North Atlantic between
45° and 60° N. occur in the four months of November-February.
The storms decline in force and precipitation value as they pass
inland and their 'chief influence is therefore felt on the west.
The almost featureless European plain illustrates this diminution,
for here there is virtually no complication by relief rain; convec-
tional rain is, however, increasing in this direction, and the net
decrease represents the decrease of cyclonic minus the increase. of
convectional. Furthermore, the cyclones, repelled by the con-
tinental highs, follow a marginal course in winter, while in spring
and summer they penetrate further into the heart of the continent.
3~ Convectional and thunderstorm rain, with its typical
summer maximum, is characteristically the rain of the interiors of
continents. The fall chiefly occurs in the afternoon: over 40 per
cent. of the rain at Berlin occurs between the hours of noon and
8 p.m. The rain, while it falls, 'is heavy, but does not last so
lo'ng as the persistent cyclonic and orographic rain of the west.
Bukarest and London have about the same amount of rain,
but rain falls on 167 days at Kew and on only 106 at Bukarest.
. RAINFALL REGIMES. As a result of the different distribution
in time and space of these three fundamental types we may recog-
nize four rainfall regimes as follows :-
I. A marine variety in which cyclonic and relief rain pre-
dominate. The rainfall is fairly evenly distributed over the year
but there is a tendency for a winter maximum (e.g. Glasgow).
At west-coast stations the predominance of relief rain is testified
by an autumn maximum (e.g. Valencia). This marine type,
while characteristic of western margins, reappears in a modified
form on eastern margins with the return of the relief and cyclonic
controls (e.g. New England), but is not found in Eastern Asia on
account of the monsoon.
Passing inland from west coasts convectional rain soon becomes
important and by Paris has brought about a summer maximum.
This becomes more and more pronounced, affording a gradual
passage into the second type. . J

z. A continental variety in which ,convectional rain predomi-


nates. Winter rainfall is small since pressure is high, the air is
cold and humidity is low. There is a similar regime in North
America (the Missouri province of Greely 1). Agriculturally this
and the following are ideal wheat climates since the rain comes at
the critical growing season and a small fall is therefore adequate

1 See reference on p. 167.


12
I82 CLIMATOLOGY
for cultivation. On its equatorward side this type grades into
desert by a transitional type which is :-
3. A steppe type in which the slight rainfall is largely convec-
tional but in which there is a spring or early summer maximum
to which cyclones contribute. These cyclones penetrate from the
Mediterranean region as soon as the weakening of the continental
high permits their ingress. The equivalent in North America is
Henry's 1 ( Rocky Mountain foothill type' (see p. I93).

4 4
2 2
0 0

6
4
2
0

6
4
b. Valencia
10
8
.. BIll
g. Denver -j
2 6
0 4
C. Boston (Mass.)
2
4 0
h. Peking
2
0 8
d. Vienna
6
4 4
2 2
0 0
e. Chicago j. TOKiO
FIG. 55.-Rainfall Regimes in Cool-Temperate Climates
a, b and c: Marine type; d and e.: Continental type; f and g: Steppe type; hand i: Monsoon type')

4. The monsoon type in which monsoon rains (relief and


cycl9nic) predominate, giving a summer maximum, generally some-
what suddenly reached, and a strongly marked seasonal rhythm.
SNOW. Even in the extreme south and west snow falls almost
every year, though the snow cover is only continuous and pro-
longed in the continental variety. In the marine type it does not
lie many days andin cities is reduced to slush almost immediately.
The number of days a year with snow decreases southwards and
westwards: Warsaw has 57, Berlin 34, Paris I4, Scilly 3.
1 A. J. Henry, Rainfall of'the United States. U.S. Weather Bureau
,BulL D., 1897.
COOL-TEMPERATE
SUNSHINE AND CLOUD. These are cloudy latitudes on account
of the frequency of cyclones, the seasonal variation of whose
paths is accompanied by a seasonal variation of cloudiness.
Cloudiness increases towards higher latitudes, from five-tenths in
Southern France to seven-tenths in Western Ireland and North-
west Scotland; Gascony has 2,000 hours of sunshine a year, parts
.of Scotland have less than half this. There is also an increase of
sunshine towards the east corresponding to the greater dryness of
the air,' but a decrease again as the east coast is approached
(d. the increase of rainfall). I

Fog is frequep.t especially near the coast, for the air is aJways
moist and a fall in temperature usually results in condensation.
Inl~nd, .under anticyclonic conditions, radiation fogs are charac-
teristic of early winter while the air is still moist.
STORMS. The pressure gradient in the normal temperate
cyclone is not, as a rule, sufficient to give rise to really dangerous
storms, but wind velocities of over 60 m.p.h. are not uncommon
in winter and do much damage to property. Some of the most
serious storms are the result of V-shaped depressions or line-
squalls similar to the Southerly Bursters of Australia, already
described. The' wind-shift line' in the trough of these depressions
is a region of great disturbance with strong squally winds and
often thunder and lightning. This is, in fact, the general form
of winter thunderstorm especially along western margins, but in
continental interiors they are generally heat thunderstorms due
to convectional overturning and are much more frequent during
the summer months and especially during the afternoon. The
number of days with thunder increases southward and inland:
Lerwick has only I day a year with thunder, Dublin 8, Kew 14,
Berlin 15, Vienna 18. Downpours of rain and hail accompany
the storms; the hail may do considerable damage to standing
crops, vineyards being especially vulnerable.
The violent tornadoes of the middle Mississippi Valley generally
occur along the trough of a V-shaped depression, but have some
connection with convection, as is shown by the fact that, though
they may occur in any month and at any time of day or night, they
are most numerous in the hottest months and just following the
hottest part of the day. Though affecting a very narrow area
only, they are one of the most violent and destructive meteorological
phenomena known.
THE SEASONS. Temperature is obviously the determinant of
season, rainfall being of very subsidiary importance in this respect.
In the extreme continental type where the annual range of
temperature is huge there are virtually only two seasons, summer
and winter, the change from one to the other being clear cut and
CLIMATOLOGY
well-defined. At Warsaw, for example, April is n° warmer than
May, and November is 10° colder than October. But in the
marine type, when the annual range is small, there is a gradual
passage extending over two or three months, the transition
seasons being spring and autumn. Owing to the considerable day-
to-day variation and the small annual range, temperatures at any
time during these seasons may revert to those of the season before,
or anticipate those of the season to follow. In Fig. 56 the five
curves represent :-
(a) Mean monthly temperature.
(b) Mean monthly maximum.
(c) Mean monthly minimum.
(d) Highest mean monthly maximum on record.
(e) Lowest...mean monthly minimum on record.
J F M A M J J A SON D
80·
:r 1'\
75
VL ~ ~ _\
70
65
I
/' V--
V l.- r-....
r'\.r\
~\
60
55
V j V r\ \\
50 <L ~ / / ~ I""-.. ~
b / / ~ 'f" ~
~ \ \~
45
F-'"
40 ./ // '\ :\. "-
~V )~
3S c
30 - -iLL ~
\
'"" '"i'-
~
25
;~ ~ .....
20
FIG. 56.-Maximum, Minimum and Mean Temperatures at Cantbridge

It will be seen that the coldest July was actually colder than
the warmest January, i.e. midwinter months may be actually
warmer than midsummer months. This unreliability of the
seasons and especially of the seasons of early growth and harvest
is a serious handicap to cultivation, for killing frosts may return
when blossom and shoot have been encouraged by weeks of
summer-like weather, or autumn fruit may be prematurely nipped
by an unseasonably early visitation of winter conditions.

Ve~etation and Cultivation


In the peculiarly favoured climates of western Europe with
their exceptionally mild winters evergreen trees of a sub-tropical
COOL-TEMPERATE 185
aspect. flourish; arbutus thrives in south-western Ireland,
myrtle, laurel, etc., in Cornwall, but over the rest of the zone
the winter season, lasting from one to six months, during which
temperatures are below the minimum for active growth (43°),
is g~rerally evaded by the trees by the shedding of their leaves.
The deciduous forest is characteristic of the western margins and
extends for a considerable distance inland, but the diminution
of rain in this direction places a limit at about 50° E. in Eurasia
(see p. 61). In North America the equivalent forest, here largely
coniferous (the Pacific coniferous forest), does not extend beyond
the Cordillera.·~ In an easterly direction this temperate forest,
usually deciduous but occasionally coniferous, gradually gives
way to grassland ancl ultimately to semi-desert. On approach-
ing eastern margins the forest reappears as the rainfall increases.
Polewards of these vegetation types lies the northern coniferous
forest (Taiga), stretching as' a broad continuous band across
both Eurasia and North America, belonging characteristically to
the cold climates with their long'winters.
THE BROAD-LEAVED DECIDUOUS FOREST. The extreme
western margins do not produce the best deciduous forest, for
the seasonal rhythm is scarcely well enough pronounced, and
moreover salt-laden winds are inimical to tree growth; it is in
the semi-oceanic climates of England, France and Germany
that the forest reaches its finest development. Oak, beech,
ash and maple are the chief forest trees, but chestnut, elm,
sycamore and lime are common. It is a feature of these forests,
as of the coniferous forests, that they often consist of almost
pure stands of a single species, a fact which adds considerably
to their economic value, for many of the trees make excellent
timber.
Deforestation has caused the disappearance of this forest
over huge areas and its replacement by grassland and meadow,
often cleared for cultivation. Thanks to the heavy rainfall, the
grass which is thus established is extremely rich and nourishing;
and thanks to the mild winters of western margins it continues
to grow practically throughout the year, thus providing excellent
fodder for stock, especially dairy cattle. Unfortunately those
same factors of high rainfall and humidity, together with the
cool summers 'in the extreme oceanic type, are unfavourable to
agriculture and especially to cereal cultivation. This climatic
type has a small extension in space and there is a scarcity of
economic plants adapted to life in iL Potatoes are practically
the only crop which really thrives, while oats is the most successful
cereal.
CONIFEROUS FOREST. Under less favourable conditions, and.
186 CLIMATOLOGY
especially on sandy soils, coniferous trees often displ"ace decidu-
ous, in some cases forming pure stands over very large areas.
The North American forests of this zone, for reasons which are
not climatic, are mainly coniferous. The centre of. the lumber-
ing industry in eastern Canada lies about the northern edge of
the 'Eastern Temperate Forest', along the banks of the St.
Lawrence, where almost pure stands of white pine (on the lighter
soils) and spruce and hemlock (on the heavier soils) provide
much of the timber and wood-pulp of commerce.
Deciduous trees are often mingled with the coniferous, especi-
ally sugar maple, elm, oak, yellow birch, etc., but they become
less and less important northwards. In British Columbia, under
milder and wetter conditions, the Douglas fir and Sitka spruce
grow to a prodigious size and provide fine large timber.
TEMPERATE GRASSLANDS. With less than 15 or 20 inches
yearly the forest passes into grassland as is clearly brought out
by a comparison of the mean annual rainfall and vegetation maps
of Eurasia or North America. The grassland climate is the
steppe variety with early summer rain and a considerable tem-
perature range. The heat of late summer scorches up the grass
which is a tougher, drier type than the meadow grasslands of,
say, Britain, and causes annual prairie fires whose ashes have
added fertility to the rich black soils. This grass is well suited
to stock-raising and its economy is mainly pastoral, but the
better watered parts have been largely brought under cultivation,
especially for cereals, for which the moist spring and the hot
dry period of late summer are ideal. Wheat is favoured on the
poleward borders, maize in lower latitudes where the summer
is longer and hotter and where the rains tend to be prolonged
later into summer. The richer parts of the steppe and prairie
are the great granaries of the world.

Regional Types: British Isles


The climate of the British Isles is characteristically mild,
the weather typically variable. The mildness of the climate is
due to the profound marine moderation exerted by the warmth
of the North Atlantic; the variability of the weather to the
frequent procession of cyclones and anticyclones accompanied by
veering and backing winds bringing weather now from the
ocean, now from the continent, at one time from the north, at
another from the south.
The British Isles lie in the debafable ground between three
great pressure and weather systems, the Azores high, the Icelandic
,low arid the continental pressures which are high in winter but
COOL-TEMPERATE 187
low in summer; as these systems wax and wane Britain may
be enveloped entirely in one or other of them. In the hot, dry
summers of I9II and 1921 the Azores high spread northwards
over us, in the cold winter of 1879 when the Thames was frozen
over and when - 23° was recbrded at Blackadder, Berwickshire,
it was the continental high that covered us. Normally the pre-
vailing wind is ·the result of the air-flow from the Azores high
towards the Icelandic low and is therefore south-west, but in
summer the continental low begins to exert an attraction and
winds become more westerly.
WINTER CONDITIONS. The Icelandic low is now at its deepest
and largest and the pressure gradient between here and the
Azores is at its steepest, the south-west winds blow, therefore,
with their greatest force. At the same time the continental
high is .strongly marked and attempts to repel the warm westerly
winds from the continental margin. Between the cold outflowing
continental winds and the warm, stormy westerly winds a constant
struggle for mastery is fought out over the British Isles. While
the ocean prevails, the weather is mild, but windy; while the
continent prevails, cold snaps occur; temperatures as low as
- 23° have been recorded, but it· is rare for the thermometer to
sink below 10°. That proximity to the ocean is m'ore important
than latitude at this season is shown by the north and south
run of the isotherms; tl~e east coast has 38° as far south as the
Thames- estuary 'in January, the west coast has 42° as far north
as the Outer Hebrides, while a gulf of warmth, a small-scale
replica of the North Atlantic, is thrust north into the Irish Sea
between the relatively cold lands of Britain and Ireland.
The cyclonic storms of this season tend to follow a marginal
course up the west coasts of Ireland, Scotland and Norway and
to brush the western seaboard with their wet southerly fringes.
Owing to the strength of the wind and the frequent cyclones,
the west coasts are receiving their heaviest rain: Ben Nevis,
with 18 inches in January, has II per cent. of its total in this
month; Seathwaite, with 12 inches, has 10 per cent. On the
other hand, the Midlands and East are under continental influence
and have their lowest rainfall at this season; East Anglia, with
less than-2 inches, has only about 7 per cent. of its total in
January.
. SPRING. As the continental anticyclone weakens, especially
in the south, storms begin to take a more southerly course, a
favourite track at this period being the warm Mediterranean
Sea. In this case the British Isles lie to the north of the cyclonic
centre and suffer from frequent easterly winds off the Continent.
These winds, especially in the eastern counties, are dry and
I88 CLIMATOLOGY
cold, chapping and roughening the skin and bringing a crop of
lung diseases and allied complaints. But anticyclonic weather
is common, with large diurnal range of temperature and night
frosts well into May. This is the driest season of the year,
most places having only about 6 per cent. of their total in April.

January

FIG.

SUMMER.
57.-Principal Tracks of Depressions in Western Europe
.
The Azores hIgh has now spread far to the north
,
and the Icelandic low has virtually disappeared as an individl,lal
system, owing to the growth of continental lows on either side.
Winds, therefore, blow mainly from the Azores high to the
continentallow and are westerly with very little of the southerly
component; the gradient is small and the resultant wind is
generally 'light. The track of cyclones lies directly acrosS the
COOL-TEMPERATE r89
British Isles or over the seas to north and south, but they are
~ess frequent and are not as a I1.1le deep when they do occur.
There are, however, occasional thunderstorms, often of con-
siderable intensity, especially in the eastern counties which
experience their heaviest rain at this season. The amount of
rain is not large: London has 2 inches in July, but this makes
up 9 per cent. of the yearly total. Places in the West have
five times as much (e.g. Seathwaite ro inches), but this makes
up only about 7 per cent. of their yearly total. The summer
rainfall of the eastern counties comes mostly in heavy showers
(coI?-vectional) which may give r inch or 2 inches in an afternoon
-a continental feature. Evaporation and run-off are rapid at
this season and it does not appear to be nearly so wet as winter
although the statistics show that it is actually wetter.
By contrast with the January conditions the isotherms now
run east and west, but have a tendency to form loops round
the larger land areas. The South and Midlands are enclosed by
the isotherm of 62°! while London is enclosed by the 64° isotherm;
marine influence keeps the coast somewhat cooler. The highest
temperature recorded is rooo at London in the hot summer of
r9II; this, however, is exceptional and temperatures above goO
are rare, occurring only under persistent anticyclonic conditions.
AUTUMN. These anticyclonic conditions of summer persist
into early autumn, but temperature falls and the nightS"" are
'positively cold. The strong nocturnal cooling of the still moist
air causes fog and mist to form in the evenings, but these are
soon dispersed by the morning sun.
By November anticyclonic conditions have ceased to be
common, a cyclonic control is established and the winter pressure
distribution is restored. Autumn is a wet season everywhere,
for the sea is still warm and the winds off it are very humid,
while the land is growing cold, especially at night. Rainfall is
chiefly orographic and cyclonic and is therefore heaviest in the
West and at night.
Europe
In the absence of any climatic barrier on tbe European plain
there is an almost perfect gradation from the humid, mild,
equable marine type to the dry, extrejne continental type, as
the following figures for stations, all in tHe same latitude, show :-
Temperature Percentage Rain
Station. Range. Rainfall. Wettest Month. in Winter
(six months).
Utrecht 28 29 August 47
Berlin 34 22 July 43
Warsaw. 40 15 July 43
Nikolaewskoe 61 14 June 40
I9 0 CLIMATOLOGY
But south of the plain the scattered blocks of the Her-
cynian Mountains and the more formidable loops of the alpine
chain cause the transition to be more spasmodic, the western
slopes being unduly marine, the basins to leeward being pre-
maturely continental. Not only do the western edges of these
blocks receive more abundant rain than the enclosed basins
which they protect, but also they are more ' marine' as regards
their temperature range and the transition of the seasons. Steppe
conditions occur in the Hungarian basin some 500 miles west of
the general boundary of the steppe. The heart of the Balkan
peninsula is 'cool temperate continental' far into latitudes
which would normally have a Mediterranean climate and summer
rains here nourish a dense oak forest, the home of herds of swine.
Even the Plain of Lombardy, shut off from the Mediterranean
Sea by the Apennin~s, has a climate which is typically continental
and has little affinity with the Mediterranean type. There are
summer rains here, of great value in supporting good summer
pasture for dairy cattle (rare elsewhere in the Mediterranean
basin) and for the peculiar crops of the Lombardy Plain, rice
and maize. Much of the rain is in the form of thunderstorms,
often of great violence with large hailstones which batter the
crops and damage the fruit. It is quite a general practice to
insure against it.
By latitude 30° E. the west winds are parched and dry and
steppe supervenes, but the Black Sea renews their moisture and
its eastern shores, backed by the lofty Caucasus, have 80 inches
of rain in places and dense forest as a result.
The Caspian has no mountains on its eastern shore, which·
is extremely arid, but the prevailing wind here is not westerly
but northerly, determined by the Mediterranean low of winter
and the Sind low of summer. Thus the south shore is the lee
shore, and, like the lee shore of the Black Sea, it is mount~inous.
Along the south Caspian littoral, backed by the Elburz Moun-
tains, is a region with 50 to 70 inches of rain; a swamp in
winter, steamy and hot in summer. Here are dense forests
(from which the Romans obtained the tigers for the gladia-
torial games) and sub-tropical cultures, rice, cotton, sugar
and fruit. . .
This is the last of the cool temperate climates in Eurasia,
north-east of this point there is sandy desert, east of it there
is high plateau, while to the north, where the rainfall is heavier
(Saratov, Orenburg), the long winters compel their inclusion in
the cold climates. When the type reappears in North China it
clearly belongs to the monsoonal sub-type.
COOL-TEMPERATE

North AmericJ
STORM TRACKS. Unlike Europ~ the continent of North
America presents· no obvious lines of least resistance for the
penetration of cyclonic storms, which are compelled to break
across high mountains by whatever route they enter. Yet there
are certain tracks followed with some degree of fidelity, the most
important of which are shown in Fig. 58.
The northerly route, taking advantage of the Great Lakes
and the St. Lawrence valley, is generally referred to as the
, Northern .Circuit ' and is the most frequented track during the
summer when, however, storms are fewest, feeblest and least
constant. puring this time the southern states lie almost entirely

FIG. 58.-Cyclone Tracks, N. America

to the south of the cyclonic Influence which does not disturb


the serenity of their summer.
As the winter approaches the cyclones begin to show an
occasional and increasing preference for the so-called ' Southern
Circuit' and by midwinter may reach to the shores of the Gulf
(see p. I6S). In addition to these two main routes there are
intermediate routes and ' SpUfc tracks', e.g. from Colorado and
Texas, joining up with the' Northern Circuit' in the neighbour-
hood of the Great Lakes. It is noticeable that most of the
storms enter near Puget Sound, while nearly all leave by New
England; these places, and especially the latter, are the storm
centres of North America and consequently have very changeable
weather. The storm tracks pay little attention to the continental
CLIMATOLOGY
winter high pressure which is, in fact, difficult to identify' on th~
daily weather maps, and only emerges clearly in the somewhat
artificial chart of mean winter pressure.
The weather associated with the cyclones depends on the
trajectory of the air currents (see p. 28) and therefore on the
position of the station with relation to the cyclonic centre.
Thus in the eastern states southerly and easterly winds are moist,
west~rly and northerly are dry. The northerly winds bring
cold waves, the southerly hot waves. Cold waves, three or
four of which may be expected each year, are usually sudden
in their arrival, hot waves are generally more gradual and cumu-
lative in their effect.
. DISTRIBUTION OF RAINFALL. The Cascade Mountains of
Oregon and Washington and the Coast Ranges of Canada limit
marine climates to 'a narrow coastal fringe, quickly followed to
the east by a rain-shadow area whose aridity must always con-
demn it to scanty population. With small local exceptions
due to relief this aridity persists as far as the IOoth meridian,
but east of this line America is highly favoured by an adequate
and well-distributed rainfall increasing steadily in amount to-
wards the east and south. This favourable treatment it owes
to a large degree to the presence of a great body of warm water
in the Gulf of Mexico and the flow of a warm curra.t up its
eastern coast. ,
As a result of the meridional trend of the relief umts the
well defined climatic boundaries in North America run from
north to south rather than from east to west, the changes in the
former direction being in the nature of gradual transitions and
changes~of~degree rather than of kind ..
THE PACIFIC PROVINCE. (British Columbia, WaShington and
Oregon.) During the winter months the three major controls,
exactly analogous to those of Western"Europe, are;-
1. The Aleutian low (d. the Icelandic).
z. The North Pacific high (d. the Azores high).
3. The Continental high (d. the Eurasian).
The resultant prevailing wind is southerly or south-westerly,
warm and wet; but owing to the weaker influence of the North
Pacific than the North Atlantic drift, it is neither so warm nor
so wet as the corresponding European wind. Coastal tempera-
tures a.re about 5° cooler, latitude for latitude, in the New than
in the Old World (d. Brest, Jan. 44° and Victoria, B.C., 39° ;
Blacksod Point 43° and Masset, B.C., 35°). Inland, away from
marine influence, temperatures fall somewhat rapic]J.y, even when
reduced to sea-level; the isotherms run north and south as in
western Europe, but the more continuous coastline and the
COOL-TEMPERATE 193
sudden rise of the mountains are reflected in the straighter and
more crowded isotherms. ~.,
The mountain screen behind the coast more than compensates
for any inferiority of humidity and the winter rainfall is very
heavy all along the coast.: 8 inches in J<.muary is recorded over
a long strip in Oregon and Washington. '
In summer, when the Pacific high has moved north and
tpe Aleutian low has faded and merged with the continental
low, the prevailing wind is north-westerly. Partly because it is
now less directly on-shore, partly because it comes from higher
latitudes, partly because the land is now warmer than the sea
and partly because cyclones are less numerous, the summer rain
is considerably less: Victoria has less than I inch in each of
the three midsummer months. The corre?ponding climates in
Western Europe are much wetter in summer. Further south
the summer rainfall decreases still more until in California the
drought of the' Mediterranean\' summer is complete. ,
THE BASIN AND THE PLAINS. The decrease of rainfall east-
ward bears a very close relationship to relief: Clayoquot, on
the west side of Vancouver Island, has II9 inches, Nanaimo,
on the lee side, only 37 inches; just across the strait the rainfall
at Vancouver has increased to 59 inches, but decreases inland to
. 10 inches at Kamloops and the maximum here occurs in early
summer. Thus within 200 miles of the coast there is a steppe
climate closely resembling that of south-eastern Europe. At
Spok.ane the temperature range already exceeds 40°, the rainfall
is only 17 inches, there is a tendency to an increased rain in
spring and the autumn is strikingly dry (July, August and Sep-
tember have less than .1 inch each); these are steppe charac-
teristics. A marine feature is, however, retained in a consider-
able winter precipitation from cyclonic storms. Thus there is
good rain in winter and spring which hinders spring sowing but
is ~useful for fall-sown wheat, the spring and early summer rains
being valuable for the growing plant and the autumn drought
suiting the harvest.
Across the Rockies the winter rain is no longer found, the
Spring maximum being dominant (e.g. Denver) and due to the
penetration of cyclonic storms (d. the steppes of the Black
Sea region). This is Henry's 'Rocky Mountain foothills' 1
rainfall regime-a true steppe type. Further east the maximum
tends to occur later into summer, the rainfall being mainly
convectional. This is the continental type/0f the plains, the
, Missouri type' of Greely.2
Winter temperatures in the plains are below freezing and
1 See reference on p. 182. 2 See reference on p. 167.
194 CLIMATOLOGY
there is a continuous snow cover in winter, but summers are
hot (over 70°). Thunderstorms are an almost daily occurrence
and the tornado, though a comparatively rare event, is more
frequent and more destructive here than anywhere else. They
are apt to occur along the trough of V-shaped depressions and
are a 'squall-line' phenomenon at the meeting line of two air
currents of different temperatures and humidities.
THE EASTERN STATES. Eastward again there enters a new
source of winter rain, viz. winds off the Gulf and the Atlantic.
At first these rains are less important than the convectional
type and the rainfall graph still shows a maximum in summer
(somewhat retarded) though winter is well supplied (e.g. Ports-
mouth, Ohio), but nearer the coast they assume greater import-
ance and rainfall is fairly evenly distributed over the ye<:tr (e.g.
Boston, see p. Z<?4). This evenly distributed rainfall confers
great advantages, risk of drought and flood are minimized and
the discharge of the rivers is not subject to great fluctuation,
a feature of considerable value in their utilization for power.
!Vinter precipi~ation is freq~ently in the form of snow..r which,
III the north, hes to a cO!.lslderable depth.
The climate of the eastern states is a curious mixture of
marine and continental, for the prevailing westerly wind tends
to carry continental conditions as far as the coast, yet the occa-
sional south-east wind in front of cyclones brings influences
from the sea. The Great Lakes are a further important influence
in moderating the climate: the Michigan and London peninsulas
being particularly mild and famous for their fruit, including
grapes and peaches. The high humidity, a marine trait due to
the proximity of ocean and lakes, makes much more unpleasant
the large (though diminished) temperature range, a continental
trait due to the prevailing land wind. The July mean for New
York is 75°, but temperatures above 95° are common in July
and August, over 100° is occasionally recorded and on 7th August,
1918, the mercury reached 104°. .
Heat waves such as this are slow to arrive but are cumu-
lative. The summer cyclones, with which they are associated,
move slowly and sometimes remain stationary for days while
temperatures mount higher and higher. Nor is there any relief
at night, for the high humidity of the air prevents nocturnal
cooling, night temperatures may not fall below 75° and sleep
becomes almost impossible. In agricultural regions the heat
and drought, if prolonged, may cause grave financial loss, but
it is particularly in the great cities, and more especially in the
overcrowded quarters, that the hot waves bring the greatest
suffering, and deaths from sunstroke and heat apoplexy are
COOL-TEMPERATE . I95
numerous. Light clothing is worn and quantities of ice and
cold drinks are consumed but bring little relief; all who can do
so leave for the seaside where hotels and boarding-houses are
full and the beaches are crowded all night long with fugitives
from the sweltering city. Not until the cyclonic centre has
passed and the wind swings round to the north is there any
relief.
The Southern Hemisphere
In the ocean hemisphere the planetary winds suffer no appreci-
able deflection by continental influences, and the Brave West
Winds impinge with full force throughout the year directly on to
the mountainous coasts which form the western seaboards of
southern Chile, Tasmania and New Zealand. The western coasts
are deluged with rain, the rain-shadow of the eastern slopes is
profound and is reached with striking suddenness. All three
regions have considerable area~ in the west with more than
IOO inches and all three have areas in the east with less than
20 inches.
SOUTHERN CHILE AND PATAGONIA. The western slopes of
the Andes in South Chile with their IDO inches or 200 inches of
rain, are a region of dense forests on saturated swampy ground
which militates against their economic exploitation. Rain falls
on six days out of seven at Evangelistas Island. As in North
America, where the disposal of relief is similar, there is a sudden
decrease in the rain-shadow in lee; most of Patagonia has less
than IO inches of rain fairly evenly distributed throughout the
year.
NEW ZEALAND. Chiefly owing to its oceanic situation New
Zealand, for its latitude, enjoys very mild winters; it i~ only
in the extreme south of South Island, in fact, that the mean
temperature of the coldest month falls below 43°. The genial
winter is one of its greatest assets as a pastoral country,
for winter grazing is available and the need for stall-feeding of
cattle is largely obviated. In spite of the absence, however, of
a real cold season, a characteristic of warm-temperate climates,
it would not be legitimate to consider the island group under
that heading, for in other respects it is typically cool temperate.
With the exception of the north of North Ireland, for example,
the prevailing wind throughout the year is westerly, and it is
only in the Auckland peninsula that the influence of the trade
winds is felt in summer. The weather throughout the year is
thus subject to cyclonic influences and consequently lacks that
reliability of summer conditions which the warm-temperate
climates expect to enjoy. Furthermore, the oceanic influence
196 CLIMATOLOGY
which kept the winters warm keeps the summers cool; Wellington
has about the same summer temperature as London, though it
is 10° nearer the Equator. The climate of New Zealand, then,
with the exception of the Auckland peninsula, is best described
as cool temperate marine, being, in fact, comparable with England,
but being warmer in winter, windier and sunnier.
Following the line of least resistance m:ost of the cyclones
follow a track to the south of New Zealand, though a few cross
South Island, especially in winter, moving generally from south-
west to north-east. The north winds on their advancing edge
are warm and muggy, but as the centre passes the wind veers
to north-west or west and reinforcing the normal westerly winds
blows strongly against the western slopes, rising and releasing
copious rain. On the eastern side it descends rapidly as a hot
foehn wind which is extremely trying on the Canterbury Plains.
Later the wind swings to the south and blows cool, damp and
refreshing.
In South Island the rainfall is fairly evenly distributed through
the year, both on the wet western and dry eastern sides, but
in North Island the influence of the anticyclone begins to be
noticeable in a summer decrease of rain. This becomes more
and more pronounced until at Auckland nearly 60 per cent.
falls in the six winter months, but even here the driest month
has nearly 3 inches of rain and there is no approach to a summer
drought. The rainfall, too, has a high degree of reliability and
New Zealand, unlike its nearest neighbour Australia (not, of
course, climatically comparable) does not suffer from drought.
One of the pleasantest characters of the New Zealand climate
is its sunniness. Considering that the islands lie in the cyclonic
westerly circulation and that they are surrounded by wide oceans
it is rather surprising that they should be able to boast a sun-
shine record comparable with that of the Mediterranean; yet
stations on the east coast show records of 2,500 hours a year.

The MonsooJ.}.al Sub-type


'Eastward from the Atlantic we have traced a progressive
degeneration of the climate by decreased rainfall into the steppe
and desert of Central Asia. On approaching the' Pacific the
rainfall increases again in a manner similar to the increase in
eastern North America (see p. 194), but whereas the eastern
margin of North America is characterized by an extremely high
degree of uniformity of seasonal incidence, the eastern division
of A.sia shows as its most striking characteristic a strongly marked
seasonal rhythm. At Che-foo 20 inches out of an annual total
COOL-TEMPERATE 197
of 24 inches, fall in the six summer months,and half the annual
total is recorded in the months of July and August alone: at
Tientsin 72 per cent. falls in _the three summer months (d. Wash-
ingtQn 30 per cent.) and only 2 per cent. in the three winter
month!? (d. Washington ?3 per cent.).
In the face of the constant off-shore winds it is clear that
the influence of the sea will not be able to make itself felt, and
the monsoonal sub-type is typically continental during the
winter months, the cold of the interior being carried right down
to the coast. China is, in fact, some 10° or IS° colder than the
United States in the same"'latitudes during winter; Mukden
(Jan. 8°) is IS° colder than Albany (23°), Shanghai (38°) is 12°
colder than Charleston (50°).
These features result froIp the steeper pressure gradient in
eastern Asia, for which reason not only are the winds more
regular, but also cyclonic disturbances are a less powerful in-
fluence. The swirl of air round these centres brings, from time
to time, warmth and rain from the Atlantic to eastern North
America, interrupting the reign of the continental anticykone.
The winter weather of the eastern states is therefore much more
variable and much more subject to sudden changes than that
of China; at Mukden the thermometer hardly ever rises above
59° in January, at Albany it frequently exceeds 60° and it is
the effect of these mild spells which explains the relatively high
mean temperatures in North America. Thus the difference
between the two continents lies rather in the greater persistence
of continentality in Asia than in its greater severity.

Regional Types: North China.


Although a degree of uniformity is lent by the monsoon to
the climates of the whole of eastern Asia, it is necessary to
recognize subdivisions on the grounds of differences of tempera-
ture. The tropical monsoon of Annam (p. 130) and the sub-
tropical monsoon of South China (p. 168) have already been
described. Northern China differs from these in the possession
of a pronounced cold season which, further north, increases in
length and severity until, north of about 45°, the length of the
winter places the climate in the category of cold climates (p. 217).
The cool temperate monsoonal climate therefore exists between
the latitudes of about 32° and 45° N. and includes the whole
of the great plain of China, much of the highlands which over-
look it on the west and the three peninsulas of Shantung, Liao-tung
and Korea.
WINTER CONDITIONS. Enclosed by mountains on three sides
13
198 CLIMATOLOGY
the air of the Asiatic high finds its chief escape to the east and"
impelled by the strong barometric gradient, cold dry winds sweep
'Over North China with great force and constancy; the mean
wind velocity in January is 10 m.p.h. Though the air is very
dry and the sky may be cloudless, the fine yellow dust which
these winds bring from the deserts hides the sun with a heavy
haze and carries bad visibility far out to sea. Closed windows

FIG. 59.-Tracks of Continental Depressions and of Typhoons


(Mainly after Gherzi)
Thickness of lines approximately proportional to frequency of storms

and doors in Peking are impotent to keep out the penetrating


grit which settles everywhere and irritates and inflames the eyes,
nose and mouth. The cold winds serve, however, to cleanse
and pulverize the soil of the plains and bring every year a fresh
surface dressing of -fine yellow soil. In many thousands of years
they have endowed Shansi and the neighbouring provinces with
a r~ch legacy of a mantle of loess hundreds of feet in thickness.
COOL-TEMPERATE 199
The temperatures which these winds import are phenomen-
ally low and proximity to the sea brings no moderating influence
since the land monsoon carries continental conditions right down
to the seaboard.
Everywhere in China winter is the season of least rain, and
this is particularly the case in North'China where the continental
influence is strongest; here the three winter months have gener-
ally less t1lan 5 per cent. of the yearly total. Such rain as does
fall in winter is generally to be attributed to the passage of
shallow depressions, the chief tracks of which and the relative
frequency along each are shown in Fig. 59. Since this rain
generally comes on south-east winds (i.e. off sea), the heaviest
incidence is in the coastal zone and there is a rapid decrease
towards the interior. The Yang-tse valley' provides a much
favoured path and Central China, in consequence, has heavier
winter rain than either north or south; Hankow has 9 per cent.
and Shanghai 13 per cent. in the three winter months and at
the ~atter place January is the only month with less than 2 inches.
The intensity of winter rain is much lighter than that of summer
and though the total amount is small the number of rainy days
is greater than might be imagined.
North of the Hoang Ho and Shantung the winter precipitation
is mainly in the form of snow and severe blizzards form a most
unwelcome feature of the climate of the plain.
SUMMER CONDITIONS. The passage from winter to summer
is remarkably sudden in North China, much more sudden than
in the corresponding latitudes in America; the land is warmed
with great rapidity and April at Peking is nearly IO o warmer
than March. At the same time the weakening of the anti-
cyclone allows marine influence to enter to an increasing extent,
the humidity rises and rainfall begins to increase both in fre-
quency and amount. These early rains are not the monsoon
rains as yet (see p. 17I), but they are of inestimable value for
the growth of the spring crops (e.g. wheat) which always suffer
from the disadvantage of inadequate moisture in the early stages
of germination and growth.
At the height of summer the temperature exceeds 80° nearly
everywhere south of Peking and North China is nearly as hot
as South. The monsoon rains have arrived and typically reach
a maximum in July. The Yang-tse Valley is an exception,
however, to the general rule in having a double maximum, the
first in June, the second in August and September with a some-
what drier spell between. The early maximum is apparently
attributable to shallow cyclones of continental origin moving
slowly down the valley and later across the Yellow Sea to Japan
200 CLIMATOLOGY
where they cause the Plum Rains'. As is usually the case
t

with the continental depressions the rain comes chiefly on south-


east winds and these are followed by northerly winds which
bring a marked drop in temperature, accentuated by a sudden
fall in humidity. The second maximum is connected with the
passage of tropical cyclones which, though they may not come
ashore, nevertheless cause rain over a wide area.
ECONOMIC EFFECTS OF RAINFALL IN CHINA. Summarizing
the facts of rainfall, it is clear that the length of the dry season
increases inland, for at one end the coast gets winter rain from
the continental depressions and at the other the summer rains
are prolonged by typhoons in the autumn when the monsoon
begins its southward retreat. Up to a point the summer maxi-
mum of rain is a valuable asset for crop production, but in many
ways the highly seasonal incidence is unfortunate, though the
absence of winter rain is less serious in North China where the
winter temperatures 'are in any c'ase too low for cultivation. It
has, on the other hand, supplied a valuable incentive towards
irrigation, in its turn a powerful stimulus to the growth of a
Chinese civilization.
Monsoon rainfall, largely cyclonic in origin, is characteristic-
ally variable in amount, and famine is the curse of monsoon
lands. The very fertility of the lands encourages the growth of
population up to the limit of supporting power and agricultural
practice is all too frequently based on the most optimistic ex-
pectation of rain. For this reason the disaster of a deficiency
is all the greater, particularly in North China where the annual
amount is lowest and the margin of sufficiency least. The
normal rainfall at Hangkow is 50 inches, but in any year the
total may fall short of this by more than 50 per cent. or exceed
it by an even greater amount; July, 18t)6, was practically rainless
at Shanghai, July, 1903, had 12 inches.
The heavy summer fall is often too torrential for full utiliza-
tion and the loss by run-off has been aggravated by deforestation.
A considerable rise of the rivers follows rapidly on every heavy
fall of rain and serious floods all too often result. The flatness
of the plains of Chi-Ii and Honan, coupled with the raising of
the river banks as levees, lends itself particularly to such disasters,
the distressing frequency of which is amply testified in Chinese
history and legend. Central China, though its rainfall is heavier,
is more fortunate in the possession of a better deyeloped drainage
and of large lakes, e.g. Po-yang and Tung-ting, which act as
reservoirs to equalize the flow. '
COOL-TEMPERATE 20r

Japan
Marine influence and the warm waters of the Kuro Siwo
combine to give Japan an adequate and well-distributed rainfall
and to moderate the extremes of temperature, but the effects
of insularity are diminished by proximity to land, especially
. during winter when the prevailing influence is off the continent.
Sakhalin, separated from the mainland only by the narrow Gulf
of Tartary, shares the extreme continentality of Amuria ; Yezo is
a little-more fortunate, but has four or five months below freezing-
point; Hondo, separated by the wider Sea of Japan, has much
milder -winters, though even here the January isotherms are
some roo south of their mean latitude. Shikoku and Kiushiu
are virtually sub-tropical, there is no real winter, the January

FIG. 6o_-Chief Storm Tracks, Japan


(Afler Sanders)

mean being everywhere above 40°, palms, camphor trees and


citrus fruits flourish and three crops a year may be obtained.
WINTER CONDITIONS. The Kuro Siwo, the Asiatic counter-
part of the Gulf Stream, sends a branch up each coast of Japan;
the smaller westerly branch bears against the Japanese coast,
but the more important easterly branch tends to swing away
into the Pacific, yielding place along the coast, as far south as
the angle of 'the Awa ,peninsula, to the cold Kurile current.
Thus the Kuro Siwo is less effective than might be expected in
raising the winter temperature of Japan; the January mean
is below freezing north of Niigata which is in the same latitude
as London, and frost occurs on more than 60 days in the year
at Tokio. The warm current does, however, strongly influence
202 CLIMATOLOGY
the course of depressions, the barometric minima closely follow-
ing the warm water. Fig. 60 shows clearly how the storm tracks
are related to the two branches of the warm current, the more
frequented track coinciding with the greater volume of the
current. The source of these storms is in the Asiatic continent
(see p. I70) and their destination is the Aleutian low. On leaving
the mainland they acquire renewed vigour from the warm waters
and cause the Japanese climate to be much more cyclonic, much
more variable and therefore much more stimulating than that
of North China.
The cyclonic winds bring adequate rain to both sides of
Japan, but the prevailing westerly (monsoon) wind, meeting
the mountains, brings heavy relief rain to the western coast;
Kanazawa has 32 inches in the three winter months and scarcely
a day passes without rain or snow in the winter. Much of this
precipitation is in the form of snow, especially on the mountains
where it lies to a depth -rarely equalled elsewhere. Though the
temperature is higher than on the mainland the cold is more
raw and unpleasant owing to the high humidity. 1'he east
coast, watered only by the cyclonic rains, is drier and more
pleasant; Tokio has only 7 inches in. the three winter months.
SUMMER CONDITIONS. In summer the conditions are reversed
and the east coast is the windward one, well watered by the
south-east monsoon; the west coast, sheltered by the mountains,
is drier though still well watered. Kanazawa, exposed to the
winter monsoon and protected from the summer monsoon, has
56 per cent. winter rain and 44 per cent. summer; Tokio, exposed
to the summer monsoon and protected from the winter monsoon,
has 39 per cent. winter rain and 61 per cent. summer. The
smallest rainfalls are recorded on the. shores of the Inland Sea;
protected from both monsoons, and there are places here with
less than 30 inches. .
The summer rains, like those of the Yang-tse (p. Igg), reach
a maximum in June or July (the Plum Rains), decrease notice-
ably in August and rise to a second maximum in September.
The earlier maximum is valuable for the transplanting of the
rice but is unfortunately unreliable; the June mean at Tokio
is 6 inches, but there have been years when less than half this
amount was recorded and in I9I7 there was less than ! inch.
The cause of these rains is somewhat obscure; they accompany
a temporary falling off of the monsoon and are thus clearly not
orographic. They are generally ascribed to the passage. of
shallow continental depressions (see p. I99) and this would
account for their variability. The autumn maximum is the
greater of the two and is due mainly to typhoons but partly to
COOL-TEMPERATE 203

:the relative difference in temperature of land and sea. Advan-


tage is taken of the drier period between the maxima to harvest
quick-growing varieties of rice, in fact quick-growing varieties are
in general use to ensure harvesting before the second rains set in.
Summer temperatures are everywhere high (the 70° isotherm
, for July passes only a little way south of the -Tsugaru Strait),
but owing to marine influence, not excessive. Marine influence
also prolongs summer warmth far into autumn, August is nearly
everywhere the hottest month and October is almost as hot as
May. The August maximum of temperature is, however, to be
ascribeci in some measure to the decrease of rainfall and humidity
during this month.
VARIABILITY OF THE CLIMATE. The climate of Japan shows
considerable variations from year to year, for the factors which
control' it are themselves subject to fluctuations: the strength
of the monsoon, the strength of the ocean currents and the
frequency of storms. In consequence agriculture is often in a
pre.carious condition. For example, the cultivation of rice, a
-cereal requiring heat and moisture, is grown in Japan, owing
to other favourable conditions, somewhat beyond its proper
range (the 75° isotherm for the four months of active growth).
A cold summer therefore means a decreased yield and wide-
spread distress. Such cold summers occur when the Kurile
current intensifies and extends its influence, which it does when,
arhong other causes, large quantities of: ice accumulate in the
Bering Sea. Excessive ice here is thus followed by poor harvests
in North Japan.
SUGGESTIONS FOR FURTHER READING
For the British Isles, in addition to the Oxford Survey of the British
Empire and the Geographie Universelle, see M. de C. Salter, Rainfall of the
British Isles, 1921, British Rainfall, published annually by the Meteorological
Office, and the Rainfall Atlas of the British Isles, published by the Royal
Meteorological Society, 1926. See also C. E. P. Brooks, ' Weather Influences
in the British Isles', S.G.M., 1924; R. H. Hooker, 'Weather and
Crops in England', Q.J.R.M.S., 1922; Cullum, 'Climate of Valencia',
Q.J.R.M.S., 1896; C. E. P. Brooks and J. Glasspole, ' The Drought of
1921', Q.J.R.M.S., 1922 .
For Europe, see A. Angot, Regimes de pluie de l'Europe Occidentale,
Paris, Ann. Bur. Cent. Meteorol., 1895; J. Glasspole, 'The Distribution
of Average Seasonal Rainfall over Europe', Q.J.R.M.S., 1929 ; A. Paulsen,
The Climate of Denmark, U.S. Weather Bureau Bull. II, 1893.
For N. America, besides the Oxford Survey (Canada) and Ward's
Climate of the United States, see F. N. Denison, 'The Climate of British
Columbia', M. W.R., 1925. The Climate of Long Island, Norman Taylor.
Brooklyn Botanic Gardens Contdb., No. 50. The Climate of Minnesota,
Bull. 12, Min. Geol. Surv., 1915. The Climate of Western U.S.A., Bull.
Amer. Geog. Soc., 1915.
CLIMATOLOGY
TEMPERATURE
Station. Lat. Long. Alt. (ft.) J F M A M. ] ] A S 0 N
- - - - - - - - ~

VANCOUVER. 49° N . 123° W. 13 6 36 38 42 47 54 59 63 62 56 49 43


PORTLAND 46° N . 123° W. 153 39 42 46 51 57 61 67 66 6J 54 46
DENVER. 40° N. 105° W. 5,29 1 30 32 39 47 57 67 72 71 62 51 39
OMAHA 41° N. 96° W .1, 105 22 25 37 51 63 72 77 75 66 55 39
ST. LOUIS 39° N. 900W . 5 68 32 34 44 56 66 75 79 77 70 58 45
CHICAGO. 42 0 N. 88°W. 82 3 26 27 37 47 58 68 74 73 66 55 42
36° N . 87°W. 54 6 39
1

NASHVILLE. 41 50 59 68 76 79 78 72 69 49'
CINCINNATI. 39° N . 85° W . 628 33 34 44 54 65 74 78 76 69 58 45
WASHINGTON 39° N . 77°W. II2 34 35 43 54 64 72 77 74 68 57 46
NEW HAVEN 41° N. 73° W . 106 28 28 36 47 58 67 72 70 64 53 42
BOSTON. 42oN . 71° W. 12 5 27 28 35 45 57 68 72 69 63 52 41
VALENCIA 52oN . 10° W. 30 44 44 45 48 52 57 59 59 57 52 48
ABERDEEN 57° N . 20W. 46 38 38 40
44 48 54 57 56 53 47 42
BREST 48° N . 5° W . 21 3 45 45 50
47 55 60 65 64 61 56 50
PARIS 48° N. 2° E. 405 37 39 49
43 56 62 65 64 58 50 43
BERLIN 53° N. 13° E. 196 30 33 48
38 57 63 66 65 58 49 39
MILAN 46° N . 9° .E. 49 0 32 3 8 46
55 63 70 75 73 66 56 44
BRESLAU 51° N. 17° E. 482 30 3 2 46
37 56 63 66 64 58 48 38
VIENNA. 48° N . 16° E. 664 29 33 40
50 59 65 68 67 60 50 39
WARSAW 52oN . 21° E. 43 6 26 29 46
35 57 63 66 64 56 46 36
BELGRADE 45° N . 20° E. 459 29 34 52
43 62 67 72 71 63 55 43
BUKAREST 44° N. 26° E. 269 26 3 1 41
52 62 69 73 72 64 53 40
KIEV 5ooN . 31° E. 590 21 23 31
45 57 64 67 65 57 46 34
ODESSA. 46° N . 31° E. 210 25 28 48
35 59 68 73 71 62 52 41
HOKITIKA 43° S. 171° E. 9 60 61 59
55 50 47 45 46 50 53 55
DUNEDIN 46° S. 171° E. 40 58 58 52
55 47 44 42 44 48 51 53
HOBART. 43° S. 147° E. 17,7 62 62 59
55 51 47 46 48 , 51 . 54 57
VALDivIA 39° S. 73° W . 141 60 59 54
57 51 49 46 46 49 51 53
PUNTA/ARENAS 53° S. 71° W. 92 52 51 48
44 39 36 35 37 40 44 471
ICHANG . 3 r o N . rIIo E. r67 40 43 62
51 70 78 83 83 75 65 54
ZIKAWEI. 31° N .121° E. 2~ 38 39 46
56 66 73 80 80 73 63 52
PEKING. 37° N . II6° E. 13 1 24 29 41
57 68 76 79 77 68 55 39
MUKDEN 42°N .123,° E. 144 8 14 30
47 60 71 77 75 61 48 29
CHEMULPO 37° N .127° E. 240' 25 29 36
49 59 67 73 78 69 57 4'2
NIIGATA. 36 ° N .139° E. 84 35 34 40
5f 59 67 74 78 70 59 49
NAGASAKI 33° N .130° E. 443 42 43 48
58 64 71 78 80 74 64 55
TOKIO 36° N .140° E. 70 37 39 44 55 62 69 76 78 71 60 51
MIYAKO. 40° N .142° E. 100 31 32 37 47 54 61 68 72 65 55 45
COOL-TEMPERATE 2 05

RAINFALL
~

D Yr. R,: J F M A M J J A S 0 N D Total.

- r- - -- -- -- -- ---- -- -- ---- -- ---


48 48 27 8·6 6'1 5'3 3'3 3'0 2'7 1'3 1'7 4'1 5'9 10'0 7·8 59.8
41 53 28 6'7 5'5 4.8 3'1 2'3 1·6 0·6 0·6 1'9 3'3 6'5 6'9 43.8
32 50 42 0'4 0'5 1'0 2'1 2'4 1'4 1·8 1'4 1'0 -1'0 0·6 0'7 14'3
27 51 55 0'7 0'9 1'3 2·8 4'1 4'7 4'0 3'2 3'0 2'3 1'1 0'9 29'0
36 56 47 2'3 2·6 3'5 3.8 4'5 4.6 3·6 3'5 3'2 2·8 2'9 2'5 39.8
30 50 48' 2'1 2'1 2·6 2'9 3·6 3'3 3'4 3'0 3'1 2·6 2'4 2'1 33'2
41 59 40 4.8 4'2 5'1 4'4 3.8 4'2 4'1 3'5 3'5 2'4 3'5 3'9 47'4
36 56 45 3'5 2'9 4'0 j'I 3·6 3'7 3'4 3'3 2·6 2'5 2'9 3'1 38.6
36 55 43 3'2 3'0 3:5 3'3 3·6 3'9 4'4 4'0 3'1 3'1 2'5 3'1 40 '7
32 50 44 3.8 4'0 4'0 3'5 3'9 3'2 4'3 4'5 3"7 3'7 3·6 3'7 45'9
II

32 49 3'7
45 3'5 4:1 3.8 3'7 3'1 3'5 4'2 3'4 3'7 4'1 3.8 44.6
46 51 5'5
15 5'2 4'5 3'7 3'2 3'2, 3.8 4,8 4'1 5'6 5'5 6·6 55'7
39 46 2'2
18 2'1 2'4 1'9 2'3 1'7 2·8 2'7 2'2 3'0 3'0 3'2 29'5
,4 6 54 2·6
20 2'4 2'2 2'1 2'4 1'5 1'3 J'9 2'5 3'4 3'1 3'7 29'1
38 50' 28
1'5 1'4 1,6 1'7 1'9 2'1 2'2 2'1' 1'9 2'3 1'9 2'0 22·6
33 48 36
1'7 1'4 1·6 1'5 1'9 2'3 3'0 2'3 1'7 1'7 1'7 1'9 22'7
36 55 43 2'4 2'3 2'7 3'4 4'1 3'3 2,8 3'2 3'5 4'7 4'3 3'0 39.8
32 47 36 1'3 1'0 1,6 1'5 2'4 2'4 3'4 2·8 2'0 1'5 1'5 1'5 22'9
32 49 39 1'5 1'3 1·8 2'0 2·8 2'7 3'1 2'7 2'0 1'9 1,8 1·8 25'4
30 46 40 1'2 J'I 1'3 1'5 1'9 2,6 3'0 2'9 1'9 J'6 1'5 1'5 22'1
34 52 43 1'2 1'3 1·6 2'3 2·8 3'2 2'7 1'9 1'7 2'2 1'7 1'7 24'4
31 51 47 1'3 1'1 ' 1,6 1'7 2'5 3'5 2'7 2'0 1·6 1'7 1'9 1·6 23'1
24 44 46 I'I 0,8 1'5 1'7 1'7 2'4 3'0 2'4 1'7 1'7 1'5 1'5 21'
31 49 4 8 0'9 0'7 1'1 1'1 1'3 2'3 2'1 1'2 1'4 1'1 1·6 .1'3 16,
58 53 J6 9.8 7'3 9'7 9'2 9·8 9"7 9'0 9'4 9'2 II·8 10·6 10·6 II6'I
, 56 51 16 3'4 2'7 3'0 2'7 3'2 3'2 3'0 3'1 2·8 3'0 3'3 3'5 36'9
6,0 54 16 1·8 '1'5 1'7 1'9 1,8 2'2 2'1 1'9 2'1 2'2 2'5 2'0 23'7
57 53 14 2'9 3'2 6'4 9'3 I5'3 17'5 15'4 13'5 7'3 5'0 4'4 4.8 105'0
50 44 17 1'4 1'2 1'7 1·6 1·6 1'2 1'2 1'2 1'1 0·8 1'1 1'4 15'5
44 62 43 0·8 1'1 2'0 4'2 5'0 6'2 7·8 6·8 4'0 3'6 1'5 0,6 43. 6
42 60 42 2'0 2'4 3'4 3'7 3·6 7'4 5'9 5'7 4'7 3'1 2'0 1'3 45'2
27 53 55 0'1 0'2 0'2 0·6 1'4 3'0 9'4 6'3 2·6 0·6 0'3 0'1 24'9
14 44 69 0'2 0'3 0·8 1'1 2'2 3'4 6'3 6'1 3'3 1·6 1'0 0'2 26'5
.)0 51 53 0·8 0'5 1'0 2·6 3'2 4'3 8·6 8'2 4'2 1·6 1'7 0'7 37"4
, 39 55 43 7'7 4'9 4'1 4'2 3'7 5'2 6'2 5'2 7'4 5'7 7'2 9'1 70 .6
46 60 38 3'1 3'5 5'2 8'1 7'4 13"2 9'3 7"3 8,6 4'6 3'3 3'3 76 '9
Iii 41 57 41 2'2 2·8 4'4 4'9 5'7 6'5 5'3 5'7 8'7 7'4 4'2 2'1 59'9
36 50 41 2'7 2,6 3'4 3'9 4'7 5'0 5'3 7'0 8'5 6'7 3'2 2'5 55'5
CHAPTER XI
COLD CLIMATES
YING within the sphere of influence of the westerly

L winds the cold climates are subject to controls differing


little from those of the cool-temperate climates, and
much that has been said of the latter is equally applicable here.
There is the same eastward gradation from marine to continental
type, making itself felt in the same way by an increase in annual
and diurnal range of temperature, a decrease of rainfall and an
"increasing tendency to a summer maximum of rain. From the
meteorologist's point of view there is little need for separation
of the two, but from the geographer's point of view there is every
need; the differences are of degree rather than of kind, but the
environments they present differ widely. The essential distinc-
tion is the greater severity and longer duration of the, winter
season which puts a stop to agricultural duties in winter and
restricts outdoor activities to trapping, fishing and lumbering.
The land masses of the southern hemisphere do not extend
far enough south to experience this type of climate and only
the northern continents of North America and Eurasia need be
considered. _
It is worthy of note that both North America and Europe
in these latitudes present to the westerly winds high and rugged
coasts deeply indented with fjords. The mountains confine
marine influence to a narrow coastal fringe and beyond them
we are plunged at once into extreme continental climates. J:his
arrangement of relief units is unfortunate, for the life-giving rain
which might spread so far across the wide and open plains which
lie to the east is extravagantly squandered on the unresponsive
western slopes. The marine type accordingly has a very limited
distribution and is fairly clearly marked off from the continental
along the line of the topographic divide and it will be convenient
to describe the two types separately.
On approaching the east coast in North America a modified
continental type is encountered in which the annual range of
temperature is reduced and the precipitation is evenly spread
over the year. But the eastern margin of Asia in these latitudes
is under the influence of the monsoons and has their typical
206
COLD CLIMATES 207

regime of temperature and rainfall. Since the two continental


masses,_lJ,ave little in common on their eastern margins they will
be described separately under' regional types '.
PRESSURE AND WINDS. The principal controls of the pre-
vailing winds, as in the previous type, are the Icelandic and
Aleutian lows and the continental winter highs and summer
lows (see p. r92), but as in temperate climates, these are subject
to complete temporary obliteration by the passage of cyclon~s
and anticyclones. The weather element is, however, less insistent
and the climatic element more persistent than further south;
continental interiors especially lie under the influence of the
winter anticyclones for long periods at a time, meanwhile enjoy-
ing cold, clear, calm and settled weather with light outward-
blowing winds. From time to time, however, depressions with
their anti-clockwise system of often violent winds and their
sudden and severe temperature changes succeed in penetrating
into the hearts of continents, but they are the exception rather
than ~he rule, pursuing, more usually, a course marginal to the
high pressures. In summer the continental interiors are regions
of low pressure with light inflowing winds; subject to sudden
disturbances and convectional overturnings.

Marine (Norwe~ian) Type


TEMPERATURE. Both in the North Atlantic and the North
Pacific drifts of warm water bathe the eastern shores (= Western
Coasts) and weaken the grip of winter far to the north. The
North Pacific drift carries the January isotherm of 32° as far
as Sitka (Alaska) in latitude 57° N., while the still warmer North
Atlantic drift carries it beyond the Lofoden Islands roo further
north still. The Norwegian coast is kept ice-free all the year
round, although extending far beyond the Arctic Circle and the
January mean at Bergen is above freezing, although there are
six months below 43°. This phenomenal mildness, it is true,
applies only to a narrow coastal zone and the fall in tempera-
ture away from the marine influence is very rapid. Even at
the heads of the fjords temperatures are 5° or roo colder and ice
forms here every year, though the mouths are ice-free.
Characteristically the minimum temperature is delayed, Feb-
ruary being little warmer and, in N .W. Europe, generally actually
,colder than January. The sun, low in the sky, is slow to melt
the snow; spring is late in arriving and summer is also delayed,
the maximum at many coastal stations not being reached until
August. In the north spring and autumn virtually do not exist;
the long winter gives way suddenly about June to the cool
208 CLIMATOLOGY
summer and in September winter returns as sudde:rff. In the
south winter departs earlier and returns later ancr there are
fairly recognizable seasons of spring and autumn. The July
mean is below 60° in the south and below 50° in the north,
temperatures sufficient to ripen only the least exacting of crops,
but the long days of summer daylight partially offset this dis-
advantage and permit the ripening of hardy cereals even within
the Arctic Circle. At North Cape the sun does not sink below
the horizon between 12th May and 29th July; at Bergen June
days are nearly 19 hours long and twilight lasts throughout the
night. The angle of the sun is, of course~ low and there is great
advantage to be derived from a south aspect.
August is usually as warm as July, and at typically marine
stations is actually warmer (e.g. Sitka, Alaska; Port Simpson,
B.C.; Christiansund, etc.). September remains warm, but in
October the long winter sets in again.
RAINFALL. The west winds, blowing over the warm waters
of the North Atlantic and North Pacific drifts, are heavily
charged with moisture which they precipitate on the mountain-
ous western seaboards, but there is a further important source
of rainfall in the cyclones whose chiM tracks are shown in Figs.
57 and 58. It should be remembered that apart from the cooling
by ascent there is a difference of some IS° or 20° in winter be-
tween the temperature of the sea and the temperatures only
a short distance inland, even when reduction is made to sea-level
equivalents.
Rain may be expected on two days out of three at Bergen
and an umbrella is part of the equipment of every man, woman
and child; in fact it is alleged that ponies in Bergen shy if they
see a man without one. The high humidity of winter detracts
from the benefit of the mild temperatures; the air is often damp
and chilly, the sky leaden and cheerless and fog is a frequent
occurrence; these conditions are typical only of the coastal
zone, a short distance inland rain, cloud and fog decrease rapidly.
Partly because cyclones are most frequent in autumn and
winter and partly on account of the steep temperature gradient
at that season the winter half of the year is considerably wetter
than the summer half and there is a clearly marked peak of
rainfall in the autumn when the differences of temperature and
humidity are greatest and when the marginal cyclone track is
particularly favoured (see Figs. 57 and data at end of chapter).
This rainfall regime is exactly the same as that of the most
typic?lly maritime type of cool-temperate climate (p. 18I) and
for exactly the same reasons.
The annual amount decreases to the north since the air is
COLD CLIMATES 20 9
colder here and its moisture capacity therefore less; it is mainly
the winter precipitation that suffers the decline, as the following
fi~res show:-
Total
Lat. Precipita- Percentage, Percentage,
tion. June-Aug. Dec.-Feb.
Bergen . 60° N. 81 21 29
Christiansund 63 47 20 28
Bod6 ' 67 36 22 26
Gjesvar 71 29 23 25
The .winter precipitation is, of course, mainly in the form of
snow which lies to a considerable depth and which is perennial
above about 4,500 feet in the south and 2,500 feet in the north.
As a result of the heavy winter snowfall and the low summer.
temperatures Norway has the largest ice-fields in Europe (Joste-
dalsbrae, 300 sq. miles) and glaciers which descend to within
ISO feet of sea-level and actually reach the sea in the extreme
north.
Continental (Siberian) Type
TEMPERATURE. The marine influence along western margins
does not penetrate very far, especia,lly in winter, when the con-
°tinental high pressures, though acting in opposition to the pre-
vailing westerlies, tend to keep out oceanic winds. Along eastern
'margins the marine type virtually does not exist, since here the
winter continental w_inds reinforce the planetary winds and are
competent to exclude completely the influence of the sea.
The January mean at Bergen is 34°, at Oslo it is 24°, and
approximat~ly along the same parallel of latitude are Helsingfors
(20°), Leningrad (IS°), Tobolsk (- 30) and Olekminsk (- 310).
To state the same thing in another way, the January isotherm
of 20 0 mns near Vard6 (lat. 700 N.), Helsingfors (60 0 N.), Charkov
D 0
'(S0 N.), and Kazalinsk (45 N.).
The cold of the continental interiors is phenomenal, yet even
where it is most extreme (Verkhoyansk - 59 0 ) life is not im-
possible for either animal or plant. The plant, and some animals,
seek refuge in hibernation, while man and other animals find
conditions far from unendurable, since the air, under the influence
of the anticyclone, is generally still and excessively dry, so that
the body loses little heat by conduction, while the bright sun,
shining out of a clear blue sky, warms the skin, though the
temperature of the air may be far below zero. Heavy clothing,
chiefly furs which the forests of the zone furnish in abundance,
must, of course, be worn, and.are often not removed throughout
. the winter; the cold discourages the use of water for washing
or shaving, fat and oil being superior substitutes. Personal.
210 CLIMATOLOGY
cleanliness is, in fact, at a discount, but health does not suffer,
for the dry cold air is healthy and germ-free.
Occasionally the peace of the anticyclone is broken by the
invasion of cyclonic storms, in rear of which are northerly winds
which may blow with velocities of So or 60 m.p.h. and at tem-
peratures of 20° or 30° below zero, laden with powdery snow
or sharp ice particles. Such are the buran of Siberia and the
blizzard of Canada, especially formidable on the open steppe
lands where there is no wind-break; from them man must find
shelter or perish. When the cyclone has passed and the gale
has' blown itself out' the anticyclone resumes its sway; the
temperature generally falls still lower, but is no longer unbearable,
as the air is calm again.
The lowest temperatures in the world are recorded in this
zone, not in the Arctic climates to the north; the mean at
Verkhoyansk, 'the pole of cold', is nearly - 60°, and - goO has
been recorded. Canada, owing to its smaller size, is less extreme,
but means of - 30° and extremes of - 70° occur in the Mackenzie
Valley.
The higher angle of the sun in February quickly makes itself
felt in a rise of temperature. We have seen that along the
western margin February is as cold as, or colder than, January,
and this is true also of most stations round the Baltic; but at
Helsingfors February is ·5° warmer than January, at Leningrad
1·5°, at Tobolsk 7° and at Olekminsk 13°. Temperature rises
rapidly in March and April, but does not reach temperatures of
growtll before May, when winter suddenly leaps into summer.
The ice and snow melt rapidly, the surface of the ground is
thawed and converted into an impassable morass, the ice breaks
on the rivers and, with the rains that occur at this sea?on, cause
widespread floods, especially since the larg~ rivers of the zone,
the Yukon, Mackenzie, Ob, Yenisei and Lena, flow from south
to north and their upper reaches are set free while the mouths
are still ice-bound.
Winter transport is by sledge on the universal highway
provided by the snow-covered ground. and frozen river, lake
and even sea; summer transport is by road. Between the two
is the period known as ' Rasputitsa ' in Siberia when for some
weeks flood and mud bring all transport to a standstill. But
the rapid rise of temperatur_e as the sun climbs higher soon dries
the ground and the melting of the river mouths ~llows the floods
to subside.
The summers are really hot, <the June mean at Tobolsk is
SgO and July 6S·7°-warmer than a London July-while maxima
above <)0°, very rare at London, are recorded almost every year.
COLD CLIMATES 2II

Added to these genial temperatures is the benefit of long hours


of sunlight, and the effect on vegetation is surprising.
In August the temperature begins to fall markedly, in Sep-
tember night frosts occur and by the middle of. October the
land is fast in the ir~:m grip of the winter frost.
RAINFALL.. So effective is the barrier of the western moun-
tains that the continental type hardly anywhere receives more
than 30 inches of rain and in most places has less than 20 inches.
By further decreasing rain the climate degenerates eastward
into steppe. and, in Central Asia, actual desert. Cyclonic and
relief rain decrease in importance inland and convectional rain
increases proportionately. As. the following table shows, there
is at Helsingfors still a trace of marine influence in the autumn
m~ximum of rain, but by Leningrad summer is the wettest
season with nearly two-fifths of the yearly total and by Tobolsk
the summer share has increased to half the total.

Percentage Rainfall occurring in


Total
Rain.
Spring. Summer. Autumn. Winter.
M.A.M. J.J.A. S.O.N. D.J. F.

Helsingfors . 24 19 29 31 21
Leningrad 19 19 39 27 15
Tobolsk. 19 15 51 23 II

. . The air of winter is so cold as to be incapable of holding


much moisture, and the winter anticyclone supplies conditions
unfavourable for precipitation; it will be noticed that the
drought of winter becomes more complete as the winter tem-
peratures fall and as the anticyclone becomes more intense.
The winter' precipitation is, of course, in the form of snow
which in spite of the dry winters faUs surprisingly frequently.
There are 65 days with snow in the year at Irkutsk, though the
six winter months during which mean temperatures are below
freezing are credited with a total precipitation of only 3 inches.
The snow cover in regions such as this is not always continuous,
since occasional strong winds strip off the protecting mantle,
piling the powdery snow into drifts and exposing large areas
of bare ground which are thus robbed of the moisture which the
melting of this snow would have provided for the spring-sown
crops. The prevention of this,drift by autumn ploughing is one
of the practices of dry farming; the snow collects in the furrows
and so is conserved.
212 CLIMATOLOGY

Vegetation and Cultivation


Two main types of vegetation characterize the cold climates;
where the rainfall is adequate the coniferous forest reigns supreme,
where it is deficient grassland takes its place. The sub-arctic
forest stretches continuously across Eurasia and North America
and abuts on its southern edge against the temperate forest,
which differs from it in the greater abundance-in the Old World
the predominance-of deciduous types and in the possession of
larger species, e.g. red pine, white pine, and Douglas fir. But
in the interior of continents the great grassland formations
(prairie and steppe) replace the temperate forest and penetrate
into the sub-arctic zone (see Fig. 19 on p. 60).
The boundary line between grctssland and desert, both in
temperate and sub-arctic zones, is difficult to demarcate. In
a region where the climatic control is so clearly transitional it
is left to local conditions to determine the plant associations.
Along the river banks tongues of forest, nourished partly by
underground water, penetrate far into the grasslands, e.g. along
the valleys of the Dneister and the Bug; where there is shelter
from desiccating winds outlying patches of forest are established
or have survived. Doubtless the hand of man, particularly
aided by fire, has brought about considerable readjustments,
his activities tending in general to drive the forest northwards.
and replace it by grass, sometimes by design and with a view
to cultivation, more often unintentionally.
Strong winds, and especially salt-laden winds, are a factor
inhibiting forest growth, for which reason the extreme western
margins of the zone (e.g. in Norway), though abundantly watered,
have trees only in sheltered places, the open country being bleak
moorland.
THE CONIFEROUS FOREST. These ·comprise the' Northern
Coniferous Forest' of the' sub-arctic zone' in Canada and Alaska
and the typical ' Taiga' of Siberia. The actual amount of rain
that this forest demands is very small, 10 inches being sufficient
if, as is usually the case, it is concentrated into the season of
active growth. Economically the sub-arctic forest is of less
importance than the temperate forest, for the trees are smaller
in size, becoming more and more stunted to the north towards
the limit of tree-growth; the commercially valuable species of
the temperate forests, too, such as the white pine and red pine
of eastern Canada and the Douglas fir of British Columbia,
do not occur. A few hardy deciJiuous trees persist, e.g. poplar,
lar,ch, birch and willow, some of them extending actually to the
liD;lit of tree-growth, but the cbmmercial hardwoods of the tem-
COLD CLIMATES 213

perate forest have been "eliminated. Spruce is an important tree


which extends as far as the limit of tree-growth, and since it
provides good pulp it will probably have an important bearing
on the future development of these forests.· At present the forest
is of little economic importance, trading in little else but furs
and skins.
GRASSLANDS. Grassland is often the dominant association
even where climatic conditions favour fores~, but except in
continental interiors it can always be ascribed to inhibitions,
e.g. strong winds or fires. It has been claimed that even parts
of the prairie and steppe could support forest, but that the
trees have been destroyed by fire and bison. Along the forest
margin the grass is fairly rich, but in the direction of greater.
aridity the cover becomes less continuous, open patches of bare
soil appear and widen and there is a steady deg~neration towards
desert. At first the stronghold of pastoral nomadism, these
st~ppes and prairies have beel} recently brought under cultiva-
tion, especially where good communications allo,," extensive
cereal growing, e.g. along the trans-Siberian and the ,Canadian
trans-continental railways. The trend of economic develop-
ment and the great increase of population during the last century
has focused attention on the problem of cereal food supply,
and it is particularly in. these sub-arctic grasslands that the
greatest advances have been made. Several qualities of the
climate combine to give them great advantages for cereal growing:
the early summer incidence of rainfall, the rapid rise of tem-
perature in May and June, the hot summers with July means
of 65°":'70° and mean maxima of 80 0 _g0 0 , the long hours of day-
light, the sunny skies of summer and the dry air of .autumn.
The aridity, the length and severity of the winter and the
shortness of the summer, despite its warmth, were elements
which did not offer great possibilities for wheat cultivation under
the conditions that obtained fifty years ago, but the almost
yearly improvement in quick-growing and drought-resistant
varieties promotes an ever-increasing area to the position of
potential wheatland.
Regional Types: Alaska, and Canada
The marine type of cold climate has only a very small develop-
ment in North America, viz.: along the Alaskan coastal strip
from Sitka to Cook Inlet. The rainfall of the Pacific slopes is
prodigious (Latouche records 172 inches), but the decrease across
the ranges is striking and tpe interior of Alaska has only 10
inches in places. The mountain barrier is more formidable
and complete than in Europe and this tends to make the tran-,
14
214 'CLIMATOLOGY
sition to the continental type more sudden, bat as far as tem-
perature is concerned the effect is offset by the adiabatic warming
of the westerly. winds in their descent of the eastern slopes.
Thus Calgary is ISo warmer than Winnipeg in the same latitude
although it stands nearly 3,000 feet higher. The high plains
at the foot of the Rockies, both in Canada and U.S.A., owe
their comparatively mild winters to the frequency of the Chinook
winds, giving spells with temperatures well above freezing. The
Chinook typically occurs when pressure is high over the plateaux
to the south and when a depression is crossing Canada,-a common
combination of conditions during winter and spring. The result-
ant wind is generally south-westerly, but like the foehn wind,
is much modified by local topography. Being the result of a
passing cyclone it shows no preference for any time of the day
or night. Under Chinook conditions heavy rain falls on the
western slopes, the liberation of heat by condensation checking
the fall in temperature as the air rises; a bank of black cloud
appears on the crest of the mountains and the Chinook descends
rapidly on to the plains. The rapidity with which the tempera-
ture rises is astounding, a change of 30° or 40° in a quarter of
an hour is not unusual and in extreme cases the thermometer
has been Known to rise 30° in three minutes. Although 40° is
seldom exceeded, the warmth seems like midsummer by com-
parison with the anticylonic cold that has gone' befo~e.
The economic results are of the first importance, the dryness
and warmth cause the snow (never very deep in these somewhat
arid regions) to disappear as if by magic. Thus winter pasture
is available almost throughout the year, especially further south in
Montana-an asset of great vallJ.e in an essentially pastoral region.
Further east and north the winter climate is more severe,
Winnipeg has two months below zero, while Dawson City has
four months, a January mean of - 24° and extremes as low as
-70°. Anticyclonic conditions ar,e more persistent here and in
the still, clear air radiation reduces the temperature to very low
figures. But the anticyclone is neither so intense nor so per-
sistent as in Siberia and while the temperatures are not so extreme
there is greater risk from blizzards and sudden changes of tem-
perature in which comparatively mild periods are followed by
intense cold and icy winds.
East of Winnipeg the winters are not quite so cold, the
moderating influence of the Great Lakes begins to make itself
felt and, further east, the effect of the sea. It has bee;n pointed
out (p. 41) that the Great Lakes ,and the St. Lawrence lowlands
constitute the path by which most of the cyclones leave North
America and these cyclones bring heavy winter snow and rain
COLD CLIMATES 21 5

to the .eastern parts of Canada, which has precipitation all the


year round, thus sharing the regime of the eastern states of
U.S.A. (p. 194).
The climate of Newfoundland and Labrador shows evidence
of the cold Labrador current. This has little effect in winter,
when Labrador is actually war~er than the interior of Canada
(i.e. the current is actually warming), but it keeps the summers
unseasonably cool; the warmest month at St. John's, in the
latitude of Paris, does not reach 60°, while at Nain, in the same
latitude as Edinburgh, it is only 47°. The cold current carries
the July isotherm of 50° as far south as Hamilton Inlet, in the
latitude of Dublin, north of· this the coast consists of barren
tundra land. Newfoundland thus shows a degree of similarity ~
with N orwa y, IS ° or 20° further north, in the following respects :-
I. The c001 and somewhat cheerless summer.
2. The relatively mild but damp and chilly winter.
3. The delay of minimum temperatures to February and of
maxima ·to August.
4. The heavy precipitation distributed evenly over the whole
year.
5. The frequency of fogs, especially in winter.
Scandinavia
The most noteworthy feature of the Scandinavian climate is
the extraordinary mildness of the coastal zone; Tromso (70° N.)
has a January mean higher than Bukarest, which is 25° nearer
the Equator, while Vardo, in the extreme north of Norway, has
. never experienced temperatures as low as the extremes at Paris
and Berlin. The North Atlantic drift which is responsible for
this phenomenal warmth, permits free navigation up and down
the coast and sea-fishing is pursued throughout the winter months.
Inland the temperature falls with great rapidity, for the con-
tinental anticyclone attempts with some degree of success to
establish out flowing winds which restrict the sphere of marine
influence. Along the Norwegian coast south-easterly or southerly
winds are normal and the mixing of these cold winds with the
warm air over the sea causes frequent 'smoke fogs'. These
outflowing winds are frequently replaced by south-westerly winds,
often of gale force, associated with passing depressions; three
or four such gales may be expected along the coast during each
winter month. February is colder than January, partly because
of marine influence, partly because the anticyclone intensifies
and so brings to bear a more. continuous continental influence.
Rain and snow are frequent; in the north snow falls on about
100 days in the year, but in the south, where precipitation may'
216 CLIMATOLOGY
take the form of rain even in midwinter, on only 30 or 40
days.
Sweden has a much more severe climate; the Gulf of Bothnia
is frozen over every year, and sledges can sometimes pass from
Finland to Sweden across the frozen Aland Sea and islands;
pack-ice is met with in most years north of Oland and Gothland.
Haparanda is closed to shipping from November to the end of
May, and Stockholm is only kept open by the use of ice-breakers.
Winter here is very dry, nowhere in Sweden having as much
as 10 inches in the six winter months. This applies still more
to late winter and spring when cyclones keep almost exclusively to
the west coast route and the Baltic lands are fine, dry ?-nd sunny.
In summer the prevailing wind, drawn into the continental
low, is westerly and generally light; the fjord inlets often modify
the direction of the wind which usually blows up them, especially
during the day. Depressions are few and shallow and rainfall
along the coast is less than in winter, but inland convectional rain
begins to contribute a preponderating share; in fact summer rain
exceeds winter rain immediately the mountain divide is crossed.
Siberia
In consequence of its great distance from the sea and of its
segregation by mountain barriers from the warmer influences to
the south, the heart of Siberia presents the extreme case of the
continental type. Verkhoyansk (Jan. - 59°) has the reputation
of being the coldest spot on earth and appears to be colder than
the North Pole itself. Yet at the other extreme the July mean
is 60° and forests flourish though the summer is only Jour months
long. Semipalatinsk has a July mean of 70°, although the
temperature is below 43° for seven months in winter. Winter
is extremely dry (3 inches to 4 inches in the six winter months),
a deficiency which brings no hardship as it is far too cold for
its utilization in any case, and the rainfall, small as it is, is
conveniently concentrated into the summer months. In the
south, however, even the small amount required is not available,
the/vegetation degenerates into poor steppe and desert. The
better watered parts of the steppe are, however, potentially
very productive, being generally blessed in addition with a rich
black soil, ideal for the cultivation of cereals (d. the prairies).
On approaching the east coast the rainfall increases and the
summer maximum becomes very well marked; this is the
monsoonal sub-type.
o

Monsoonal Sub-typ~
ln winter the monsoonal variety differs little from the conti-
COLD CLIMATES 217

nental in temperature and rainfall, for the prevailing influence


is continental, but in summer the differences are considerable,
for the prevailing influence is now marine. In winter Siberian
cold extends to the shores of the Pacific where some of the lowest
mean temperatures for their latitude in the world are recorded,
but in summer the south-east monsoon tempers the heat, while
in Siberia the thermometer rises unchecked; the July isotherm
of 70° passes near Vladivostok in latitude 43° N. and then swings
due north to Olekminsk in 60° N. Again, Manchuria is almost
as dry in winter as Siberia, but its summer rainfall is more than
double the Siberian. The essence of the Manchurian climate is,
then, the alternation of a continental winter and a maritime
summer and especially the great contrast in humidity between
the seasons.
WINTER. The Manchurian winter is very severe, the rivers
are frozen for five or six months, snow lies everywhere, sledges are
the regular means of transport) skins and quilted clothing are
worn.· Everywhere north of Vladivostok and the great bend
of the .Sungari the January mean is below zero and temperatures
of - 40° occur in northern Manchuria. To this bitter cold must
be added the discomfort of the strong wind, often of gale force,
of the winter monsoon. .
SUMMER. In April comes the thaw, the Siberian high is
yielding, and by May warm winds are coming in from the sea
bringing the first of the summer rains. Crops, wheat or barley,
are sown in early April as soon as the frost is out of the surface
of the ground, and the further thaw provides water for the
germinating seed. Temperature rises rapidly and aided by the
rains the plant makes rapid growth; by June the earliest of the
cereals are ripened and a second crop (buckwheat) may be
snatched before October frosts put an end to the growing season.
The July isotherm of 70° encloses Manchuria, though since
much of the land is high the actual mean temperatures do not
reach this figure. The monsoon rains last from May to Septem-
ber, reaching a maximum in July, the amount decreasing from
40 inches near the coast to about 20 inches or less at the scarp
of the Great Khingan, beyond which the influence of the monsoon
is scarcely felt.

SUGGESTIONS FOR FURTHER READING


Many of the references quoted at the end of the last chapter deal also
with these climates. See also Atlas de Climat de Norwige, nouvelle edition
par A. Graarud et K. Ingens. Geofis. Pub. II, 7. Kristiania, 1922; , Le
Climat de la Siberie Orientale', A. Woeikof, An. de Geog., 1897; 'The
Climate of Alaska', E. M. Filton, M. W.R., 1930.
218 CLIMATOLOGY
TEMPERATURE

Station. Lat. Long. Alt. (ft.) J M A M J J A S


-- 0
F
- - - - - - - - - - - -- - - - - - - -
DUTCH HARBOUR 54° N . 167° W. 4 32 32 33 35 41 46 51 51 47 42
KODIAK. 58° N . 152° W. 6 29 31 34 36 43 50 55 54 50 42
EAGLE. 65° N. 141° W. 4 -15 -4 7 27 45 55 59 53 42 25
DAWSON 64° N . 139° W. 1,052 -23 - I I 4 29 46 57 59 54 42 25
C,ALGARY 51° N. 114° W. 3,3 89 12 15 25 40 49 56 61 59 51 42
EDMONTON. 53° N. 114° W. 2,158 5 I I 24 41 51 57 6J 59 50 41
QU'ApPELLE 51° N. 104° W. 2,115 0 2 ,16 38 50 59 64 62 52 40
WINNIPEG. 500N . 97° W. 760 -4 0 15 38 52 62 66 64 54 41
MARQUETTE 47°N. 87°W. 734 16 16 25 38 49 59 65 63 57 46
TORONTO , 44° N . 79° W . 379 22 21 30 42 54 64 69 67 60 49
MONTREAL. 46° N . 76° W . 18 7 13 15 25 41 55 65 69 67 59 47
SABLE ISLAND. 44° N . 60 0 W. 25 30 29 32 38 44 51 59 63 60 52
ST. JOHN'S. 48° N . 53° W . 125 24 23 28 35 43 51 59 60 54 45
VESTMANNO 63° N . 20oW. 43 35 35 36 40 45 50 53 52 47 42
THORSHAVN 62 0 N. 7°W. 84 38 38 37 41 44 49 51 51 48 44
BERGEN 600 N. 5° E. 72 34 34 36 42 49 55' 58 57 52 45
TRONDH]EM 63° N . 10° E. 70 26 26 31 39 46 54 57 56 49 41
BOD6 67° N . 12° E. 7 30 28 30 36 43 50 55 54 48 40
COPENHAGEN 56° N . 13° E. 16 32 32 35 42 51 59 62 61 55 47;
UPSALA. 60 0 N. 18° E. 4 24 23 27 38 49 57 62 59 50 41
KONIGSBERG 55° . N 21° E. 7 27 27 32 42 51 60 63 62 56 46
HAPARANDA 66°N. 24° E. 8 12 I I 18 29 40 53 59 55 '46 35
HELSINGFORS 60° N. 25° E. 37 21 20 25 34 46 57 62 60 52 42
LENINGRAD. 60 0 N. 30° E. 16 18 18 24 36 48 59 63 60 51 41
Moscow N
56° . 37° E. 4 80 12 15 . 24 38 53 62 66 63 52 40
ARCHANGEL 65° N . 41° E. 22 8 9 18 30 41 53 60 56 46 34
KAZAN. 56° N . 49° E. 25 0 7 10 20 38 54 63 68 63 51 39
ORENBURG. 52°N. 55° E. 360 3 6 17 38 58 66 71 67 55 39
BARNAl-lL 53° N . 84° E. 53 1 0 3 14 34 52 63 68 62 '51 35
URUMTSI 43° N . 87° E. 2,875 5 9 23 47 61 67 72 70 59 41
HARBIN 46° N .127° E. 525 -2 5 24 42 56 66 72 69 58 40
YAKUTSK 62°N .130° E. 330 -4 6 -35 - I o 16 4 1 59 66 60 42 16
VERKHOYANSK. 68°N .133° E. 328 ~58 -4 8 -24 9 36 56 60 52 36 6
VLADIVOSTOK 43° N .132° E. 50 5 12 26 39 48 57 66 69 61 4"9
OKHOTSK 59° N .143° E. 30 -I 1-7 7 21 35 45 55 55 46 27
NEMURO 43° N .146° E. 80 "23 22 32 37 44 50 58 63 59 50
COLD CLIMATES ZI9

RAINFALL

~
D Yr. Ra. J F M A M J J A S 0 N D Total.
- - - - - - - - - - - - - - - - - - - - - -I - - -
32 40 19 5'4 7'1 5.6 3'4 5"0 2'7 2'3 3'1 5·8 8'4 6·8 7'2 62·8
35 30 41 26 4.8 4.6 3'9 3'9 5'3 4'7 3'4 5'4 5"6 7'3 5·8 6'4 61'1
, 2 - I I 24 0·8 1·8 2'0 1'3 0·8 0'5 0'5 10'9
74 0'5 0'4 0'4 0'4 1'5
I -13 23 82 0·8 0·8 0'5 0'7 0'9 1'3 1·6 1·6 1'7 1'3 1'3 1'1 13.6
28 19 38 49 0'5 0·6 0'7 0·8 2'3 2'g 2·6 2'5 1'3 0'7 0'7 0'5 16'1
25 14 37 56 0'9 0·6 0'7 0·8 0·8 3'2 3'5 2'4 1'4 0'7 0'7 0·8 16'5
122 9 35 64 0·8 0·8 1'0 1'1 2'3 3'5 2·8 2'0 1·6 1'1 0'9 0·8 18'7
. 21 6 35 70 0'9 0'7 1'2 1'4 2'0 3'1 3'1 2'2 2'2 1'4 1'1 0'9 20'2
33 23 41 49 2'2 1·8 2'1 2'3 3'1 3'5 3'1 2·8 3'2 3'0 3'0 2'5 32.6
37 27 45 48 2·8 2'4 2'4 2'3 2·8 2'7 2·8 2·8 2'7 2·6 2·6 2'5 3 1 '4
33 19 42 56 3'7 3'2 3'7 2'4 3'1 3'5 3: 8 3'4 3'5 3'3 3'4 3'7 4 2 '7
I 44 36 45 34 5'4 4. 6 4'4 3'5 3'4 3: 8 3. 6 3'7 3'4 4'9 5'7 5'5 51 '9
37 29 4 1 37 5'4 5'0 4.6 4'3 3.6 3'6 3.8 3'7 3.8 5'4 6'0 5'4 54.6
I 37 35 42 18 5.8 4.8 4'4 3.8 3'2 3'3 3'1 3'1 5'7 5.8 5'3 5'5 53.8
41 38 43 13 6'7 5'3 4'9 3'7 3'3 2·6 3'2 3.6 4'7 6'1 6'5 6·6 57'2
3'9 3 6 45 24 9'0 6·6 6'2 4'3 4'7 4'1 5'7 7.8 9'2 9'3 8'5
J
8'9 84'3
34 28 41 3 1 4'3 3'0 3'4 2'5 2'2 1'9 '2·8 3'4 4'4 5'0 3'9 3'4 40 '2
34 29 40 25 3'3 2'7 2'3 2'0 2'0 2'1 2·6 3'0 4'5 4'0 4'0 3'1 35'5
39 34 46 3 0 1'5 1'3 1'4 1'4 1'5 2'0 2'4 2·6 2'1 2'2 1'9 1'7 22'0
32 25 41 39 1'3 1'1 1'2 1'2 1'7 2'0 2'7 2·8 2'0 2'1 1'7 1·6 21'4
35 29 44 36 1'7 1'4 1'5 1'5 2'0 2'4 3'4 3'5 3:0 2'4 2'3 2'3 27'4
23 . IS 33 47 1'5 1'1 1'0 1'0 1'2 ,1'5 1·8 2'1 2'4 2'2 2'0 1'4 19'2
32 25 40 41 1·8 1'4 1'4 1'4 1·8 1·8 2'2 2'9 2'5 2·6 2'5 2'4 24'7
30 22 39 45 1'0 0'9 0'9 1'0 1·6 2'0 2'5 2·8 2'1 1·8 1'4 1'2 19'3
28 17 39 54 1'1 0'9 1'2 1'5 1'9 2'0 2·8 2'9 2'2 1'4 1·6 1'5 21'0
22 12 33 52 0'9 0'7 0·8 0'7 1'2 1·8 2'4 2'4 2'2 1·6 1'2 o·g 16·8
25 I I 37 61 0'5 0'4 0·6 0'9 1·6 2'2 2'4 2'4 1·6 1'1 1'0 0'7 15'4
24 I I 38 68 1'1 0·8 1'0 0'9 1'4 2'0 1'7 1'3 1'3 1'2 1'2 1'2 15'2
17 6 33 68 0·8 0·6 0·6 0·6 1'3 1'7 2'2 1·8 1'1 1'3 1'1 1'1 14'2
I22 10 40 67 0'5 0'3 0'5 1'5 1'1 1'5 0'7 1'0 0·6 1·6 1·6 0'4 II'3
121 3 38 74 0'1 0'2 0'4 0'9 1'7 3.8 4'5 4'1 1·8 1'3 0'3 0'2 19'3
:-21 -4 1 12 II2 0'9 0'2 0'4 0·6 1'1 2'1 1'7 2·6 1'2 1'4 0·6 0'9 13'7
-34 -51 3 II8 0'2 0'1 0'1 0'2 0'3 0'9 1'0 1'0 0'5 0'4 0'3 0'1 5'0
14 40 64 0'1 0'2 0'3 1'2 2'2 3'5 2'4 1·6 0'5 0'2 14'7
I "306 -8 22 66 0'1 0'1 0'1 0'2
1'3 1'5
1'1 1·8 2'1 0'7 0'2 0'2 7'5
iI 0'5 0'5
39 29 4 2 40 1'3 1'0 2'2 2'9 3', 3'7 3.8 4'3 5.6 3.8 3'3 2'3 37'9
L
I
CHAPTER XII
ARCTIC CLIMATES

T
HE isotherm of 50° for the warmest month which has
been adopted as the lower limit of these climates
follows a regular course in the southern hemisphere,
encircling the globe at about 55° south latitude, but in the northern
hemisphere its course is much less regular. Being a summer iso-
therm it would be expected to extend polewards over the land and
equatorwards over the sea; actually it reaches well into the
Arctic Circle in Alaska and Siberia but is carried by the cold
currents down the coast of Labrador almost as far south as
Belle Isle Strait and through the Bering Strait as far south as the
Aleutian Islands.
The climates included within this zone might be divided by
the 32° isotherm for the warmest month into two types, namely : -
I. Tundra climates with a summer above freezing-point,
however short, during which the ground is free from snow for a
sufficient period to allow the growth of the typical tundra
vegetation.
2. Perpetual frost climates in which the growth of vegetation is
impossible. .
The former are habitable and though scantily populated
supply some interesting examples of climatic control of occupation
and habits; the latter can never be inhabited and are of little
geographical interest though they present many interesting
meteorological problems and are, of course, of vital importance
in polar exploration. The climates of the former are much
b·etter known, but in neither case have we much satisfactory
material' on which to base climatic generalizations. There are
very few stations keeping regular records and such as there are
have been established for a comparatively short time only. The
records of explorers, while providing valuable samples of polar
weather, are not sufficiently systematic and are naturally mainly
restricted to the summer months, the most favourable season for
travel and exploration. When two or l!1ore expeditions have been
engaged at the same time, e.g. the Scott and Amundsen expedi-
tions of iI9II-12, the synchronous records at different points are
~specia:Qy valuable.
220
ARCTIC ,CLIMATES 22I

. J?RESSURE AND WINDS. The poleward decrease of pressure


from the Horse latitude highs is not ~ontinuous as far as the poles,
but appears to reach a minimum at about 60° S. in the southern
ocean and along a more irregular and sinuous line in the north
polar regions. This girdle of lowest pressure r,epresents the most
frequented track of the cyclonic storm centres which, it has been
suggested (p. 28), occur along the polar front, or line of convergence
of the polar and westerly winds. Beyond this low-pressure
trough pressure increases again to reach a maximum which in the
Antarctic is probably in the neighbourhood of the Pole, but
which in the less regular Arctic appears to lie between Alaska
and Greenland. Out-blowing winds from these high-pressure
centres take on, as a result of deflection by the earth's rotation,
an easterly direction which has been found to be prevalent where-
ever the polar anticyclone is well established; their direction is
more regular in the Antarctic than in the Arctic. The Greenland
ice-cap· produces its own anticyclpnic circulation which is only
rarely invaded by winds from without. Nearly 80 per cent. of
the winds on the western side blow from the east or south-east,
and nearly 70 per cent. of the winds on the eastern side blow from
the north or north-west.
The anticyclones are most likely the direct results of the
chilling of the lower air layers by the polar cbld and there is a
great deal of evidence to show that their extent is affected from
year to year by the extension of the pack-ice, the limits of the
anticyclone being more or less coincident with the limits of ice.
They are comparatively shallow phenomena and are replaced
above by a general cyclonic circulation with westerly winds, as is
shown by the drift of high clouds and of the smoke of Mt. Erebus.
From these anticyclonic centres the surface winds gravitate out-
wards, aided in some cases, and especially in Antarctica, by the
creep of cold heavy air down the slopes of the ice-cap.
The downward creep is sometimes convertedjnto a downward
rush of air whenever the pressure gradient is increased, as, for
example, by the passage of a deep depression along the edge of the
polar anticyclone. Then .the air may be strongly warmed by
compression and arrives at the coast as a wind of foehn character;
such winds are met in Greenland blowing down the fjords, espe-
Cially of the east coast and also along the coast of Antarctica. In
general the air movement in anticyclonic areas is characteristically
light and the polar anticyclones are no exceptions, calms being
frequent, but on the other hand severe blizzards, sometimes with
wind velocities of ISO m.p.h., ,are distressingly frequent, especi-
ally in certain localities. It is a curious fact that while at certain
stations blizzards are comparatively rare and innocuous, at
222 CLIMATOLOGY
others (e.g. Adelie Land and Cape Evans) they are numerous and
very severe. During Sir Douglas Mawson's expedition of 1912
the wind velocity in Adelie Land averaged 50 m.p.h. for the whole
year, 85 m.p.h. for 24 hours and 107 m.p.h. for 8 hours .. Adelie
Land has well earned the name ' the home of the blizzard', but
it is an exceptional station; at Amundsen's base at Framheim the
average wind velocity from April 19II to January 1912 was only
10 m.p.h. The cause of these contrasts is apparently to be found
in the local configuration, especially in the existence of large
areas of high ground and of steep slopes which add a large gravita-
tional force to the pressure gradient. The winds of gale force
which accompany the blizzard break doWn inversions of tempera-
.ture and by causing the mixing of air layers usually bring about a
considerable rise in temperature, further increased by foehn effects.
Physiologically, however, the still, anticyclonic weather that
precedes is much to be preferred in spite of the intense cold.
Cyclonic storms are unable to invade the polar high pressures
and are concentrated round the margins, especially in the southern
ocean where there is a constant procession of depressions which
make these latitudes the stormiest on earth. But in the north
polar regions the depressions travel far into the Arctic Circle
along the gulf of warmth and associated low pressure of the
Norwegian Sea; Spitzbergen experiences as many westerly as
easterly winds, for the depressions pass chiefly to the north of the
islands.
TEMPERATURE. The inequality of the length of day and
night reaches its maximum at the poles, where there is a six-
month day and a six-month night, while everywhere within the
polar circles enjoys at least one day on which the sun does not
set. This mathematical factor introduces a new conception in
climates, for diurnal range has now little meaning, since insolation
is absent in midwinter and continuous at midsummer, however low
the angle of the sun. The annual march of temperature, too, is
profoundly altered, for temperature represents the balance of
heat, received over heat radiated away; elsewhere this generally
reaches a minimum in January, after which the higher angle of the
sun and the greater length of day "cause the temperature to rise,
but at the North Pole the sun does not rise until the spring
equinox and loss by radiation goes on continuously up to this
point where the minimum is reached. In lower latitudes where
the sun rises earlier, the minimum is reached earlier, but every-
where in the zone the rise of temperature in early spring is slow.
The sun is low in the sky and its fee,ble rays, largely reflected from
the. white surface, have little power to melt the snow. Most of the
.heat available is used up in this process, for the specific heat and
ARCTIC CLIMATES 223
latent heat of snow are high, and little is availahle to raise the air
temperature.
The ground, deeply frozen in the long cold winter, thaws only
to a depth of a few inches in the short cool summer. The result
is a surface layer of mud, saturated with ice~cold water, over-
lying an impervious hard-frozen subsoiL In the tundra zone of
the northern hemisphere the temperature does not rise above
freezing-point until June, the mean of the hottest month does
not reach 50° and by September the thermomet~r is below freezing
again and winter has taken hold.
Although air temperature is low in summer the sun's rays are
not without power;· where they shine on solid objects they may
raise their temperature to 60° or more and they give a feeling of,
warmth and comfort to the body; the black-bulb temperatures,
in fact, may exceed roo o , though the air temperature is below
freezing. The average maximum temperature reaches about 60°
and occasionally exceeds 70°, and thanks t~ the anticyclonic
condftfons and the dryness of tIle air the sky is generally clear
and a high percentage, I)f the possible hours of SUnshine is attained.
In the warmth of the sun and aided by the long hours of day-
li.ght pl~nt &ro~th is fairly vigorous, the most fa:rourable condi-
tIons bemg realIzed where the slope of the ground IS such that the
rays of the sun meet the surface at a high angle, for these condi-
tions give both warmth and good drainage. Plants, especiitlly
flowering plants, thrive in such spots with a success that is
surprising.
Autumn is warmer than spring, for the :Snow cover which
delayed the advent of the latter is no longer operative, or at least
not so continuous. September, especially in rnarine· climates, is
often as warm as June (e.g. Jan Mayen,June,36·9°; Sept.,37·4°).
The winter temperature depends partly on 1:he distribution of
land and water; Jan Mayen feels the influ~nce of the North
Atlantic drift and has a minimum (in Marth) only 8° below
freezing, Spitzbergen (in February) of - 2·4°. More continental
stations have lower minima, e.g. Sagastyr, on the Siberian coast,
- 36°. At oceanic stations the annual range of temperature is
considerably diminished (d. Sagastyr 70° an<i Jan Mayen lIO).
That the winter temperatures are but little below freezing-point,
is, however, of small account, for they are too low for plant-growth
in any case; it is the low mean temperature of summer and the
absence of really warm days that is the seriol1s drawback. The
mean temperature of the warmest month at S1)itzbergen does not
reach 42° and the highest temperature recorded is 58°; actually,
then, these climates, in spite of their mild winters, are bleak and
inhospitable in the extreme.
224 CLIMATOLOGY
The permanently frozen seas behave more or less as continental·
areas, and have a larger though not extreme range, for sea ice
cannot become very cold since the temperature of the water
below is about 29° and ice is a good conductor of heat .. Absolute
minima are very low, most stations having recorded 40° or 50°
below zero, but nowhere are such excessively low figures returned
as under the extreme continentality of the Siberian forest zone
(Verkhoyansk, p. 216). Winter is generally calm and the cold
is by no means unbearable; but the occasional blizzard is fatal
unless shelter can be reached, and the darkness, added to the
numbing cold, plays havoc with health and spirits.
PRECIPITATION. Though rain sometimes occurs in the Arctic
,summer, the usual form of precipitation is snow. The actual
amount which falls is difficult to measure because of drifting,
especially since much of the snowfall is connected with blizzards.
In fact it is often impossible to tell whether snow is falling at all,
for the wind drives along a flurry of snow and small crystals of
ice which may be falling from the clouds or may be swept up from
the ground. What is quite clear, however, is that the amount
of precipitation is very slight, probably 10 inches or 12 inches in
most places. This is quite in accordance with expectations; in the
first place the anticyclonic conditions are unfavourable for pre-
cipitation, the air being dried by its descent and further dried by
the adiabatic warming due to gravitational descent from the high
plateaux (especially in Greenland and Antarctica); in the second
place the air, being cold, can hold but little moisture, even when
saturated; and in the third place there is a general absence of
thunderstorms and other convectional effects which are such a
fruitful source of precipitation in other zones.
Under the circumstances it is at first sight sur.prising to find a
permanent snow cover and such huge glaciers and ice-caps as
exist in Antarctica and Greenland; but 'if the accumulation is
slow the dispersal is even slower, and the dimensions of the
glaci~rs and ice-caps are to be as~ribed mainly to the slow rate of
melting. The air temperature is never much above freezing-
point....,in Antarctica hardly ever-and snow disappears by subli-
mation rather than melting. In any case the cold air has little
capacity to take up water vapour. 'Absorption thawing' occurs
round solid objects on the ice, e.g. morainic debris, but is not an
important agent in disposing of snow and ice on a large scale.
It is likely that high plateau ice-caps recruit their moisture
from the upper layers of air into which they rise; the cyclonic
westerly winds of this upper circula.tion (see p. 25) must import
considerable moisture which could be precipitated as hoar frost
py contact with the cold mountain tops and ice-caps. In this
ARCTIC CLIMATES 225

connexion it is interesting to note that neither the summit of the


Greenland ice-cap of to-day, nor the summit of the Scandinavian
ice-cap of the Pleistocene concides with the highest point of the
underlying land.
Twelve inches is" only exceeded where cyclonic storms invade
the polar zones, e.g. Jan Mayen (I4 inches), the east coast of
Greenland (Angmagsalik, 36 inches), ~erguelen (33 inches), South
Georgia (35 inches). Cyclonic winds sometimes break through
the Greenland anticyclone, blowing from coast to coast, and bring
heavy falls of snow.
In regions such as these the maximum fall is in winter when
cyclones are deepest and most frequent, but in regions influenced
by the anticyclones alone there is ~ tendency to a maximum in.
summer when the air carries most moisture.
Fog occurs frequently, both"·in the Arctic and Antarctic,
. especially where warm water meets cold water or cold land, e.g.
Labrador and the South Orkneys.
180' 180·

90E 90'W

o· o·
FIG. 6I.-Approximate Positions of the Summer Isotherms of 32 0 and 500 in
the Polar Regions

The Arctic and Antarctic Compared


Whether from coincidence or for some fundamental reason
connected with the structure of the earth, the Arctic is an almost
land-locked sea, while the Antarctic is a sea-girt land. From this
there result some important and interesting climatic differences.
The seas which surround Antarctica, being continuous, are
uniform in temperature, in which respect they differ strongly
from the independent oceans aJ;ld semi-independent seas which
surround the north polar regions. As a result of its greater regu-
larity, of its uniform enclosure by sea and of the uniformity of
that sea, the Antarctic shows a greater uniformity of climate
(see Fig. 6r).
226 CLIMATOLOGY
Through the only considerable breach in the land girdle of the
Arctic Ocean, marine influence, strengthened by the warmth of
the North Atlantic drift, penetrates deeply, allowing the zero
isotherm of January to spread to the tip of Spitzbergen in latitude
77° N. Antarctica, on the other hand, is strongly held by the
polar anticyclone and warm influences are unable to penetrate;
nowhere does the mean of the warmest month rise above freezing-
point and only in favourable sites on the tip of Grahamland have
flowering plants been found. In the north polar regions, on the
contrary, with the exception of the Greenland ice-cap and some of
the islands, the land has everywhere summer temperatures
adequate for lowly plant-growth and the polar limit of the tundra
climates is set, not by cold, but by the sea. Thus the mainland of
Antarctica belongs almost entirely to the 'perpetual frost
climates' and the 'tundra climates' are virtually confined to
the northern hemisphere.

Life in Arctic Climates


PLANTS. The tundra vegetation has to contend with
extremely adverse circumstances and the number of species that
can endure the conditions is small. The rhythm of the seasons
compels a strongly rhythmic habit in the plant, but the rhythm
is very unevenly balanced; a long phase of inactivity is followed
by a short season in which life must be lived at high pressure;
The growing season is short, only two or three months perhaps,
and the plant must pass through its life cycle with all possible
speed, in which it is aided by the more or less continuous daylight.
The seeding habit is generally discarded in favour of vegetative
reproduction, the fruits being often barren; the plant grows
actively, almost feverishly, until overtaken by the winter cold, and
when it is nipped by the winter frosts it oears the last of its flowers
and fruit still unready on its branches .
.The second disadvantage with which tundra vegetation has to
contend is the physiological drought. Though its roots may be
in a waterlogged soil the plant may suffer from drought, for the
soil-water may be ice-cold, the soil may be sour and acid with
decomposing vegetation, while drying wind and strong sunshine
may be demanding rapid transpiration from its leaves. Against
this risk of drought tundra plants employ all the usual adaptations,
especially an extended root system .(but not extended downwards
because the lower layers of soil are frozen) and a dwarfed sub-
aerial growth (the cushion habit 1;>eing characteristic).
The drainage is generally bad, for the soil is permanently
frozen below a shallow surface layer, there is surface water for
ARCTIC CLIMATES 227

only a few months and under these conditions erosion is slow and
ineffective; the normal geographical cycle is thus scarcely opera-
tive. The chief agents in fashioning relief are landslip and the
downhill creep of half-thawed ground on a frozen subsoil. Pools
and marshes of stagnant water lie throughout the summer,
waterlogged areas growing only sphagnum, lichen and sedges.
Where drainage is better the .tundra is a dreary stretch of lichen
and coarse grass with occasional hummocks of a brighter green
where the home of an arctic fox or snowy owl enriches the soil
with its refuse. Dwarf willows, birches and alders occur in shel-
tered hollows which are' better drained and protected from the
wind; .they seldom grow more than two feet high, but their
shoots and branches, rich in protein, provide nourishing food for,
herbivorous creatures. But it is on the sun-warmed southern
slopes, when the sun melts early and the ice-cold water drains
away from the roots, that the flower carpets of the tundra
brighten an otherwise monotonous landscape. These are veritable
garden plots bright with the blossom of campion, rock rose,
monkshood, purple saxifrage, Iceland .poppy, forget-me-not,
th~ift yarrow, geum, willow-herb and numerous other flowers
with most of which we are familiar in our own latitudes.
ANIMALS. The rhythm of the seasons impresses itself on the
animal life as on the plants; most of the birds leave at the end
of summer for a warmer land, the reindeer retreats to the margin
of the taiga, the arctic wolf follows, and other animals make
long journeys in search of winter food supply. Hibernation, a
common escape from winter scarcity, is not practised, the winter
is too long and the summer too short to build up the necessary
reserves of fat or food supply. A surprisingly large number of
aJlimals, including the musk-ox, arctic hare and lemming, brave
the rigour of winter, seeking their food in the vegetation (e.g.
reindeer moss), which is buried under only a thin mantle of snow,
since precipitation is everywhere light. The sea is a great source
of food and many animals, e.g. polar bear, derive their winter
food supply from this source; even the reindeer is said to make
sp.ift with seaweed when nothing else is to be had. But the
winter is hard and when the polar night is nearing its end, when
food is scarcest and the cold is greatest, they are reduced to a very
poor condition and numbers perish.
Insects, like plants, begin their life with the advent of summer,
mosquitoes, especially, appearing in swarms to the sore discomfort
of man and beast. The reindeer suffers most acutely, for his
antlers, at this season in velvet, are a prey to the insect swarms.
Beca:use of this the Samoyedes take their herds away from the
swamps to higher and drier ground.
CLIMATOLOGY
MAN. The life of man is a constant struggle against nature;
by necessity a food-collector he is often a fisher as well as a hunter,
for the summer is too short to store up food against the long
winter. Fishing, in fact, plays such an important part in his
life that settlement in the higher latitudes is almost restricted to
the coast. The seasonal change of habits necessitates a nomadic
existence, the summe,r home is the portable tent of skins, the
winter home a more permanent structure of earth or snow. It
should be remembered that constructional materials, especially
wood, are scarce and the native must utilize what is to hand,
hence the 'igloo' of the Eskimo. In summer the rivers and the
sea are the chief media of transport (in canoes), for much of the
land is swamp; in winter the sledge drawn by dogs or reindeer,
for the frost converts river, land and even sea into a uniform
medium.
The physical hardships and risks that man is called upon to
undergo result in a high death-rate from accident and exposure;
frostbite and snow-blindness are maladies directly attributable to
climate, but otherwise the polar regions are healthy. Cold is a
preservative and putrefactive organisms are inacthre at low
temperatures, germs and germ-borne diseases are practically
negligible risks-it is an old joke that polar explorers do not
suffer from colds until they return home. Scurvy, -once the
scourge of the lands of winter night,' is now known to be due
to a lack of the vitamins that fresh food contains, and can be
avoided by careful attention to diet.

SUGGESTIONS FOR FURTHER READING


Articles on Arctic climates will be found in most books on the polar
regions, such as Rudmose Brown, The Polar Regions, 1927; O. Nordenskjold,
Geography of the Polar Regions, 1928; Problems of Polar Research, 1928,
published by the American Geographical Society and containing contri-
butions on climate and meteorology by H. H. Clayton and Griffith Taylor;
W., H. Hobbs, Characteristics of Existing Glaciers, 1911, etc. See also
British Antarctic Expedition, 1910-13. Meteorology, Vol. I, Discussion
by ,G. C. Simpson, Calcutta, 1919; Simpson, 'Antarctic Meteorology',
G.J., 1929.
23 0 CLIMATOLOGY

TEMPERATURE

Station. Lat. Long. Alt. (ft.) J F M A M ;1 J A S 0


- - - - - - - - -- -- - - - -- - - - - - - --
BARROW
POINT } 71° N. 156°W . 20 -19 -13 -14 -2 21 35 40 39 3 1 16 :
HEBRON 58° N. 63° W . 49 -6 -5 6 18 32 40 47 48 41 31
RAMAH. 59° N . 63° W . 16 -4 -9 3 19 36 44 48 46 42 33
UPERNIVIK. 73° N . 56° W . 65 -8 -9 -6 6 25 35 41 41 34 25
,

ANGMAGSALIK 66°N. 38 oW. 104 17 13 17 24 33 41 44 42 3 8 3°.


GRIMSEY 67° N . 18°W. 4 29 27 27 30 36 43 46 46 43 37
JAN MAYEN. 71° N. 8°.W. 76 27 26 24 28 3 1 37 42 33\ 37 31
SPITZBERGEN 78° N. 14° E. 37 4 -2 -2 8 23 35 42 40 32 22
VARDO 70° N. 31° E. 33 22 21 24 30 35 42 48 48 43 35
MALYE KAR-}
MAKOULY 72° N. 53° E . 49 2 3 5 14 25 35 44 43 35 26
SAGASTYR . 73° N. 1240 E. II -34 -3 6 -30 -7 15 32 41 38 33 6
EVANGELIS- 1 48 40
TAS Is. J 52° S. 75°W. 180 48 47 45 42 39 40 41 42
ANO NUEVO 55° S. 64° W. 174 46 46 45 42 39 37 35 37 39 41

NEYS ORX_~
SOUTH . 61° S. 45°W. 23 32 33 3 1 27 19 15 13 15 20 25 ,

SOUTH I
GEORGIA 54° S. 36°W. 13 41 42 40 36 32 29 29 29 33 35
McMURDO
SOUND 78° S. 167° E. Coast 24 16 4 -9 -II -12 -15 -1'5 -12 -2
ARCTIC CLIMATES

RAINFALL

N D Yr, Ra, J F M A M J J A S 0 N D Total,

- -- - - - - -- - - - - - - - - - - - - - - - - - - _- c - - -

59 - 15 10 58 0'3 0'2 0'2 0'3 0'3 0'3 1'1 0,8 0'5 0,8 0'4 0'4 5,6
20 4 23 S4 1'0 0'7 0'9 1'1 1,6 2'2 2,8 2,8 3'4 1,6 1'1 0'6 19'8.
25 13 as S7 0,8 1'0 2'S 4'3 1'1 2'4 3'S 1'7 z,o 4'1 S'4 4'2 33'0
14 I 17 49 0'4 0'5 0'7 0,6 0,6 o'S 0'9 1'1 1'1 1'1 1'1, o'S 9'1

23 19 28 29 3'5 1'7 2'2 2'4 2,8 2'1 2'1 2'S 4'0 6'3 3'4 2'7 3S'7

32 30 36 17 0,6 0'7 0,6 0'5 0,6 0'7 1'1 1'3 1'3 1'4 1'1 0'7 10,6
28 2S 32 15 1,6 1'5 1'2 0'9 o'S 0,6 0'9 0'9 1'9 1'7 1'3 1'2 14'2
II 6 18 44 1'4 1'3 1'1 0'9 0'5 0'4 0,6 0'9 1'0 1'2 1'0 1'5 II,8

28 24 33 27 2'7 2,6 2'1 1,6 1'4 1'5 1,8 2'0 2'4 2'5 2'5 2,6 25'7

13 6 21 42 0'3 0'2 0'3 0'3 0'5 0,6 1'2 1'7 1,6 1'2 0'5 0'4 8,8
-16 -28 I 77 0'1 0'1 -
- 0'2 0'4 0'3 1'4 0'4 0'1 0'1 0'2 3'3

43 46 43 9 12'0 9'7 11'3 11'3 9'4 9'2 9,6 8,6 9'3 9'9 9'9 10'0 Il9'4
42 44 41 II 2,6 ,2'5 2'7 2'9 2,8 1,6 1,6 I'S 1'3 1'3 1,8 2'2 24'0

28 31 24 20 1'5 1'5 1,8 1'7 1'3 1'2 1'2 1'4 1'0 1'0 1'4 0'9 IS'9

37 39 35 13 3'3 4'1 5'1 5'3 5'5 5'0 5'5 5'1 3'4 2,6 3'4 2'9 S1'2
14 25 I 40 - - - - - - - - - - - - -
CHAPTER XIII
DESERT CLIMATES
HE sole criterion of the desert climate- is aridity, but

T aridity in its turn produces a number of secondary


characteristics such as sunshine, temperature range and
even relief and soil types which help to conjoin deserts into a
single climatic group wherever and for whatever reason they
oq::ur.
MEANING OF ARIDITY. Now aridity, while mainly a matter
of rainfall, may be qualified by a number of other circumstances
which may serve to mitigate or increase it. For example, the
tundra lands have seldom more than 10 or 12 inches of rain, yet
far from being dry, the soil in summer is waterlogged because
run-off is slow (there is none at all for eight months) and there is
no loss by downward percolation, the subsoil being frozen and
impervious. Again, parts of Western Australia with little more
than IO inches of rain grow good crops of wheat, for the scanty
rain falls conveniently at the season when the plant most needs it
and when evaporation is least, and furthermore the rainfall
possesses a high degree of reliability. Along the banks of water-
courses or where underground streams exist vegetation (oases)
may thrive with no rain at all. Similarly an impervious layer
below the soil will conserve soil moisture, giving the plant a degree
of independence of rain-a condition artificially reproduced in
dry-farming areas by deep ploughing and the formation of a hard
, plough-sole' at a depth of 9 inches or so.
On the other hand there are climates with 20 inches 6f rain
which falls so extravagantly in thundery showers that most of it
is lost, by run-off on the hard-baked soil and by evaporation in the
hot-dry air and bright sunshine which follow. There are thirsty
porous soils which can absorb almost any amount-of rain and have
none to give to the growing plant. Numerous other examples
might be quoted, but enough has been said to emphasize the com-
plexity of factors at work in determining effective aridity.. The
real significance of the desert is its unproductiveness and its
consequent inability to support settled self-dependent c0II!muni-
ties. The best test of the desert is; therefore, its vegetation ~hich
is clearly not always determined by precipitation alone; .but
232
DESERT CLIMATES 233
rainfall is the most significant single element in the climate and
deserts may be expected wherever the rainfall is less than 15
inches and generally presumed where it falls below 10 inches.
A discussion on the limits of desert climates is given on
P·57· '
Tim SEMI-ARID TRANSITION CLIMATES. The tropical climate
degenerates polewards into desert beyond the limits of the migra-
tion of the equatorial rainfall belt, but retains its summer maxi-
mum of ril-in to the last; the Mediterranean climate degenerates
into desert equatorwards beyond the limit of the migration of
the stormy westerlies, but retains its winter ma~imum as far as the
rains occur; the trade-wind coasts degenerate into· desert west-
wards beyond the reach of the rains which the trade winds bring;'
the cool-temperate climates degenerate into desert eastward~ as
the westerlies lose their moisture; the MOJ;}golian desert lies
beyond the reach of the monsoon rains; the Great Basin is
robbed of its rains by the mountains which encircle it. In almost
every type of climate we may trace a progressive degeneration
into desert, but in each case the other climatic characteristics are
retained to the last, the essential change being the progressive
decrease 'of rain and the secondary characteristics which this
begets.
These transitional semi-arid areas of degenerate climate com-
prise the scrubland marginal to the great hot d~serts of the tropics
and the steppes marginal to the great cold deseJ:ts of extra-tropical
latitudes. Bhe dwell the nomadic pastoral folk, an unsettled
class, living, under conditions of climatic in~t:curity, subject to'
fluctuati<?1lS of prosperity, and prone, becausejI'ee and unattached,
,to great migrations, often to the discomfiture' ()f their neighbours.
Under modern economic conditions they' Il:lay become great
agricultural lands, growing millets and cotton in the tropics and
~ereals, especially wheat, in extra-tropical latitudes. They have
been described in the appropriate chapters in connexion with the
climatic type whose decay they represent.
THE CAUSES OF ARIDITY. The main Cause of aridity is
distance from marine influence, hence contin.ental interiors are
especially dry. But in evaluating distance from marine'influence
we must take into consideration the 'prevailing winds, their direc-
tion and their constancy; thus the deserts extend down to the
western shores in the trade-wind belt, but in the belt of the less
constant westerlies they do not extend to the eastern shores.
The one exception to this is Patagonia and in this last case, as in
the Great Basin, in Peru and· in Australia, the degeneration is
accelerated by the interposition of a mountait1 range.
The second great cause of aridity is directly connected with
234 CLIMATOLOGY
the planetary circulation. High atmospheric pressure is a condi-
tion unfavourable for precipitation; regions of seasonal high
pressure suffer from seasonal drought (e.g. the winter drought
of continental interiors), the permanent areas of high pressure
will normally be areas of permanent drought. . But ,it must not
be supposed that the converse is true and that deserts are neces-
sarily always areas of high pressure; in fact it often happens
that the desert gives rise by excessive convection to intense low
pressures, as, for example, in Sind in July and in Northern'Aus- .
tralia in January (see pp. II6 and 132).
The 'greatest deserts in the world occur benea,th the horse
latitude highs: the Sahara, Arabia, Australia, Kalahari, Atacama .
.These are the great trade-wind deserts, covering an enormous '
space of th~ earth's surface and doomed to backwardness on
account of their aridity. Because of the difficulties of, crossing
them they present one of the most significant barriers of race,
creed and civilization to be found on the earth. Because of their
unproductiveness and the harshness of the environment they
must be inhabite4 by nomadic peoples of great independence and
strength of character, self-reliant, proud and lawless. The,
desert is the stronghold of Mohammedanism, the simple, rigid
and severe creed of a monotonous and stern environment.
TYPES OF DESERT. Deserts in all latitudes are united by
their aridity, but they must be divided by their temperature.
As far as summer temperature is concerned there is little difference
between them despite differences of latitude. Luktchun (July
goO) is as hot as most places in the Sahara though 20° further
north, and the highest figure recorded here (II 8°) compares not
unfavourably with Saharan stations. Some of the highest tem-
peratures in the world are recorded in Death Valley (Cal.) in 36° N.
and therefore well outside the tropics. ~But deserts in high
latitudes are distinct in the possession of a well-marked winter
season 'which is sometimes very severe; four months at Luktchun,
for example, have means below freezing. We may therefore
divide deserts into two types (see p. 67) :-
1. . Hot deserts, with no cold season.
2. Cold deserts, with a cold season, i.e. with one or more
months below 43°.
The annual temperature range in the former is less than 50°,
in the latter, except in Patagonia, it generally exceeds that
amount. Such a large range is a feature generally associated
with extreme continentality, and in fact the cold deserts, with the
exception mentioned, are all exctssively continental; it IS to'
this that they owe their aridity.
DESERT CLIMATES 235

Hot Deserts
To all intents and purposes the hot deserts are the trade-wind
deserts, by far the most extensive being the Sahara, which, with its
direct continuation ·the Arabian desert, covers in all an area of
more than three million' square miles. Australia and North
Africa are unfortunate in having their greatest width in these
add latitudes, but North America is favoured in narrowing
rapidly south of 30° N. where the Gulf of Mexico actually functions
as a source of supply of humidity to the surrounding lands.
South Africa and Australia are unfortunate in presenting their
steepest slope to the east and consequently suffeJ;"ing from a long,
arid western slope; South America is better arranged, the· profile
being reversed, and enjoys a long, well-watered eastern slope, its'
tropical desert being confined to a comparatively narrow strip
between mountain ,and coast: ' .
Wherever the trade-wind deserts come down to the western
seaboard the coast is washed by a cold curreht,-the Equator-
seeking return current on the east side of each of the oceanic
anticyclonic swirls, reinforced (see p. 48) by the upwelling of
cold water from below as the off-shore winds skim off the surface
water. The chilling influence of these currents has a profound
effect on the climate of a somewhat narrow coastal strip. Here
is found a peculiar marine variety of the desert climate, with cool
summers, a marked diminution in temperature range, both
.annual and diurnal, a greater humidity and more mist and cloud.
It is best seen in the Atacama and Kalahari deserts, but it may also
be clearly recognized in Southern California, Rio de Oro and
perhaps Western Australia. In contrast with this the normal
hot desert is typically continental, with dry air and a huge
diurnal range of temperature and a considerable annual range.

The Marine Type of Hot Desert (Peruvian)


PRESSURE AND WINDS. It is perhaps surprising, at first sight,
that the cold currents along the coast should exercise so powerful
an influence over regions which, being in the trade-wind belt,
should receive their prevailing wind off the land to the east.
On closer inspection, however, it transpires that the local wind
along. the coast is an off-sea wind; thus at Iquique south-west
winds make up nearly 80 per cent. of the total, at Walfish Bay
50 per cent., in South California westerly or north-westerly winds
prevail throughout the year ~ and in Rio de Oro northerly winds
are the most frequent. To s6me extent these winds are integral
parts of the circulation round the oceanic anticyclones, but the:f
CLIMATOLOGY
are more properly local diversions of the planetary ~inds, due, in
large measure, to the attraction of the heated continental
interior and therefore to be classified as sea-breezes or even mon-
soons. The attraction is naturally greatest in summer and par-
ticularly during the heat of the afternoon; consequentiy it is
during summer that marine influence is strongest, which accounts
for the remarkably low summer temperatures. During winter
when pressure is higher over the continental interiors off-shore
winds are more frequent; in South-west Africa these. winds,
descending from the plateau, have a foehn character so' that,
curiously enough, the highest temperatures are recorded during
winter (cf. p. 153, Berg winds). There are somewhat similar winds
in Chile and California (see pp. 152 and 150).
. TEMPERATURE. Thanks to the influence of the cold currents
these regions are, for tropical deserts, remarkably cool; the mean
of the hottest month does not far exceed 70° (at Swakopmund it
is only 63°), and that of the coldest month is about 60° (at
Swakopmund it is only 55°). Extreme temperatures rarely
exceed 80°, never exceed rooo and never fall below freezing. The
annual range is in the neighbourhood of 10° (cf. 40° in the Sahara)
and the daily range is generally less than 20°. The crest and,
trough of the temperature curve are delayed until February and
August. In fact in all respects except rainfall the climate is
typically marine. I
The importance of the sea as a control of temperature is
demonstrated by the extraordinary uniformity of the means,
irrespective of latitude, of all coastal stations, as the following
table shows :-

Hottest Coldest Mean


Station. Lat. Month. . Month. Annual.

Callao , 12° S. 71 62'5 66'5


Arica 18° S. 71 62'5 66'0
Iquique .
Antofagasta . 20°
23°
S.
S.
7°'5
71
61'0
62'0
64'8
65'5

Wal:fish Bay 23° S. 67 57"3 62'5


Luderlitz Bay 27° S. 66'7 56'0 62'5

The isothermal maps bring this point out quite clearly, the iso-
therms running parallel to the coast through a considerable range
of latitude.
Inland the temperature rises fairly rapidly, and except wh~re
~limate is complicated by relief, there is a steady and somewhat
DESERT CLIMATES z37
rapid passage to the Saharan type. Windhoek, 400 miles inland,
has a mean annual tem,perature 7° hotter than Swakopmund,
9-lthough it stands more than 5,000 feet higher, and the hottest
month at Windnoek is more than IO o hotter than at Swakopmund.
When the wind'gets into the east these higher temperatures extend
towards the coast .and it is under these conditions that the
maxima are reached at coastal stations; the temperature at
Ca.~ Juby, for exal1?-ple, only exceeds 80° when the harmattan
bfows 'off the Sahara. '
'Thus in general the temperature extremes of the marine type
are much less severe than.in the continental type shortly to be
described;but the higher humidity makes them more uncomfort-
able to endure; particularly one misses the cool, refreshing
nights of the Sahara which are such a pleasant relief after the
burning heat of the day.
HUMIDITY AND RAINFALL. The presence of the cold current
~erves to increase the aridity since on-shore winds arrive cool,
and being warmed by contact with the land, hav~ their moisture
capacity: increased. In addition, Peru, South-west Africa and
South California lie in lee of high land, so that easterly (land)
winds are descending winds and therefore dry. These two influ-
ences, added to the normal high pressure of the latitude, ensure
that virtually no rains falls. The whole of the coast from Arica
to Caldera has, on an average, less than r inch of rain a year,
the mean at Swakopmund is 0'7 inches. Western Australia'and
Rio de Oro are unsheltered by high land to the east and further-
more the cold currents are less powerful, the aridity is therefore
slightly less; the Australian coast records IO inches as far north
as z6° S. (d. 3Zo in Chile and South-west Africa).
Very rarely a storm invades the desert and brings a short and
sometimes heavy shower of rain; Iquique had zt inches in a few
hours on zznd June, I9II. Showers such as this, spread over a
number of otherwise rainless years, serve to give Iquique an
annual average of 0'05 inch and definitely establish winter
as the 'rainy season '.
The scantiness of the rain is the more· striking in view of the
high humidity which prevails as a result of off-sea winds. At
Walflsh Bay themean relative humidity in January is 85 per cent.
and in July 77 per cent.; along the coast of Peru and North
Chile it is generally about 70 per cent. and seldom falls below
50 per cent.; at Cape J uby it is 8z per cent. in January and 9I
per cent. in July. Fog, mist and heavy dew are everywhere
characteristic and often persiste~t features. Iquique in winter is
clol1dj,er~ than England and the seaward slopes are often shrouded
for days 'on end with dense fog through which the sun cannot
CLIMATOLOGY
penetrate. The condensation is often .considerable, 'clothing may
be saturated, the ground i~ wet as after rain and the dried-up
branches of the tamarisks drip on to the desert soil.
The mists form over the cold waters near the coast as the air
flows landwards and roll up to the coast and up the slopes of the
land. Sometimes they lie at sea-level but more often the base of
the mist rises on meeting the land and hangs at about 2,000 feet
to 5,000 feet above it. In Peru there is a belt of vegetatio~ at
5,000 feet nourished by almost perpetual mist. The level tends
to rise by day and sink to sea-level during the night, following the
diurnal motion of the air. They do not sutvive far inland in the
face of the desert conditions, in South-west Africa not more
. than 70 miles; in Peru and Chile they fail to cross the Coast
Ranges and from the desert of the longitudinal valley one can
watch the heavy vapours rolling over the crest of the ranges but
melting away at once in the dry desert air beyond.
In view of the close relationship between these mists and the
sea breezes it might be expected that they would be most numerous
during summer and during the heat of the day, at which tim~s the
sea breeze is strongest; but actually winter is, in most places, the
foggiest season and night the foggiest time. The explanation is
to be found in the close balance between factors making for con-
densation and factors causing re-evaporation, and especially the·
greater heat in summer of the desert interior which is therefore
better able to dissolve the mists. In some places the coast may
be actually cloudier in summer than in winter (e.g. Callao), but
this never applies to the hill slopes which are comparatively
clear at this season. The uncondensed moisture passes over the
longitudinal valley to the Cordillera beyond and condenses there;
on these slopes summer is the cloudiest season and actual rain
may fall at about 8,000 feet. The scanty rainfall, in so far ,as it is
due to the same causes as the mist, has, like the mist, a winter
maximum along the coast, but a summer maximum on the Cor-
dillera; the latter is a tropical regime and very valuable for
irri~ation.

The Continental Type of Hot Deserts (Saharan)


The coastal strips just described are only narrow fringes and
grade rapidly eastwards into the true hot deserts, the climate
becoming more and more extreme into the heart of the continent.
The annual range of temperature on the west coast of Australia
is less than 20°, in the desert interior it is 30° or more; the coast is
frost-free as far south as Geraldton (29° S.), but inland night frost
occurs as far north as Alice Springs on the tropic. The annual
range at Alexandria is 22(), at Cairo 28°, and at Asyut 32°,
DESERT CLIMATES. 239
TEMPERATURE. The annual range of temperature is not very
great (30° or so) and is less significant than the huge diurnal
range; so great, in fact, is the difference between night and day
that mean values convey a very inadequate picture of the true
climatic conditions.' The dry air and the cloudless sky offer no
obstacle to the passage of the sun's rays, which make themselves
feft as soon a~ the sun shows above the horizon, which it does
with almost startling suddenness. As it climbs higher into the
sky the heat becomes rapidly more intense until by noon the
thermometer: stands, in summer, above the hundred mark; in
the afternoon it often rises above I200 and sometimes above I300.
A'ffer'sunset the heat is quickly lost by radiation and the tempera-
ture falls as rapidly now as it rose in the morning. The nights,
even in summer, are distinctly cool and in winter frosts occur.
Thus the daily range of temperature is great (2S O or 30°) and offen
excessive. In Death Valley, California, the mean diurnal range
for August, I89I, an exceptional month, was 64° and the greatest
diurnal range in that month was 74°. Thus the inhabitants must
be prepared for great extremes; the Arabs wear heavy clothing as
a protection against the chilly nights and coverings for the head
and face to protect them from the heat, glare and dust of the
daytime.
The rays of the midday sun, beating on the barren ground,
make the sand and rock so hot that they seem to scorch the feet
through the shoes. The air above them is heated by conduction
and a shimmering heat haze is set up, the differential refraction
in the heated layers resulting in the mirage. The heat gives
rise to !;trong convection currents resulting in strong, though
variable winds which whip up the dust and bear it along in clouds.
These may be only local swirls (dust devils), but sometimes they
are connected with cyclonic storms passing to the north and are
on a much greater scale. The simoom, fairly frequent in the
northern Sahara during the hottest months, is the most fearsome
type-a swirling rush of scorching air (I20 0- I35 0), laden with dense
clouds of blistering sand through which it is impossible to see
more than a few yards.
HUMIDITY AND RAINFALL. The prevailing wind, having come
over wide stretches of land and, in some cases high land and
mountain ranges, is normally very dry; at Aswan the mean
relative humidity for January is 46 per cent. and for July 30 per
cent., while the afternoon mean in, May is only IS per cent. Under
these conditions evaporation is often twenty times the rainfall.
To some extent the humidity ,varies according to the wind direc-
tion; relative humidity at Cairo is often more than 80 per cent.
with n.ortherlY winds, but when the Khamsin blows off the de~ert
240 CLIMATOLOGY
it is often below 25 per cent. and has been known to fall as low as
2 per cent. The temperature at the time is well over rooo and
the heat and dryness are most distressing; skin and finger-nails
crack, the hair is brittle and electric, and the whole body seems
desiccated. It was the great dryness of the air which allowed the
ancient Egyptians to mummify their dead. Scarcely a cloud
appears in the sky which is sometimes intens~ly blue, but more
often hazy with dust. The night sky is of a transparent' clearness
through which the moon and stars shine with a matchless bril-
liance. I t is not surprising that the ancient Egyptian,s ,and the
Chaldeans were well versed in astronomy, that the sun was wor-
shipped as a god and that the star and crescent have been adopted
, as emblems on their flag. - .
At night the air. often cools below the dew-point and copious
dew may be deposited, but the dew is quickly re-evaporated in
the morning sun. ,
Such is the dryness of the air that rain has little chance of
reaching the ground; sometimes clouds may be seen from which
rain is falling but it does not penetrate far in the thirsty air. The
chief rain occurs in the form of heavy storms which break with
great violence and suddenness, giving, perhaps, an inch or more
of rain in an hour. There is no vegetation to check the run-off
and in a few minutes the wadis fill with rushing water 1md
further down unsuspecting travellers may be overwhelmed by
the rapidity of the flood.
Higher land is' better watered, forming sometimes altitude
oases, such as the Ahaggar and Tibesti highlands, and the Macdon-
nell Range. The rainfall regimes of these altitude oases often
affords an interesting glimpse of the conditions up aloft. It has
been noted (p. Ig) that in the deserts of the high-pressure belt
the poleward margins get their rainfall from the westerlies, the
equatorward margins from the migration of the equatorial rain
belt. Further into the heart of the desert 'all rain disappears, but
the 'highlands allow us to trace the two types towards their
meeting-point. For example, the southern ranges of the Sahan~"
the .Air and the Tibesti, are watered by rain-bearing winds fro:m
the south-west, but further north, in the Ahaggar, the two
regimes overlap and rain falls from both sources, winter rains
from the westerlies and summer rains from the S.W. monsoon.

Cold Deserts
Only Eurasia and North America are wide enough, in inter-
mediate latitudes to bring about deserts, but Patagonia, in spite
of ~he narrowness of South America, is virtually desert becaus~
DESERT CLIMATES 24 I
of the completeness of the screen which the Andes provide against
the Brave West Winds. The last has certain maritime traits
which call for separate treatment, but the other two have strong
points of resemblance and can be considered together.
, Three contributory factors account for their aridity:-1. The
distance from the sea. 2. The basin form with surrounding
highlands. 3. The intensity of the anticyclone that covers them
in winter. In consequence few rain-bearing winds penetrate and
these must descend into the basins and are thus warmed and
dried.
The desert stretches from the Caspian Sea to the Khingan
scarp, but it is not continuous since numerous mountain groups
form altitude oases which break the desert up into a number of
more or less isolated basins, usually basins of inland drainage,
often at considerable altit"ude above sea-level. In America fhe
same climate appears in the Great Basin where the relief type is
reproduced with great fidelity. •
Such rain as falls round the Caspian and Aral seas is winter
rain (Merv 92 percent., Teheran 86 per cent., Quetta 82 per cent.),
showing that this region is a degenerate Mediterranean climate;
such rain as falls in Gobi and Takla Makan is summer rain (Urga
88 per cent., Kashgar 77 per cent.), showing that this is a degenerate
,cool-temperate continental climate.
-WINTER. The dominant influence at- tl;ds season is the con-
tinental anticyclone; in the heart of Asia its reign is absolute
and winter is a dry season; Kashgar has less than I inch in the
six winter months. But to the south and west it is disturbed by
the passage of occasional cyclonic storms bringing a little rain
and snow to Russian Turkestan and the high plateaux of Persia
and Afghanistan where snow lies throughout most of the year.
Here the rain persists into spring, in fact the spring months are
the wettest; Merv has 63 per cent. of its rain in the three
months February to April, Teheran 46 per cent. and Quetta 50 per
cent. Thus the rainy season coincides with the melting of the
snows on the mountains and the two combined cause sudden and
considerable floods, e.g. the floods of the Euphrates and Tigris
on which the ancient irrigation eIflpires depended.
In the Great Basin the rainfall is more evenly distributed, but
spring is here also the rainy season, 37 per cent. of the yearly total
falling at Salt Lake City in the three months March, April and
May.
In Turan the mean temperature of January is in the neigh-
bourhood of freezing-point (Samarkand 32°) and the ther-
mometer not infrequently falls below zero; across the Pamirs in
the Tarim desert it is colder still (Kashgar 22°). Partly thi~ is
CLIMATOLOGY
due to greater altitude, but mainly to the greater continentality;
in point of fact greater altitude is not altogether a disadvantage
and it is at the lower stations where, owing to the drainage and
settling of cold stagnant air, the lowest temperatures, both mean
and extreme, are recorded. Luktchun, in the Turfan depression,
50 feet below sea-level, has a January mean of I3°, Le. 9° colder
than Kashgar, although it stands more than 4,000 feet lower.
Warmer weather can, in fact, be found, as a rule, by ascending the
slopes, a fact well.known to the natives who take advantage of it
in many ways.
The duration of winter increases to north and east; in
Mongolia there are six months below freezing, in East.ern Turkes-
tan two or three, in northern Turan one or two, and in southern
, Turan none. The arrival of spring is everywhere very sudden, .
at'Samarkand April is lIo warmer than March and May lIo
warmer than April, at Kashgar April is I4° warmer than March, at
Luktchun 2IO. The daily range is enormous, often as much as
90°, and daily maxima are high; by June the shaded thermometer
is exceeding roo o in the heat of the day. The suddenness of this
rise of temperature is the result of the prevailing dryness. In
other climates much of the spring warmth is expended in melting
snow, warming water and drying the soil, but here there is little
snow to melt and virtually no water to warm and evaporate.
The result is that spring is actually warmer than autumn; April
is 3° hotter than October at Samarkand, 6° at Kashgar, and n°
at Luktchun.
In the Great Basin the continentality is less extreme and
autumn is warmer than spring, but the difference is only very
slight; at Boise City, for example, the April and October tem-
peratures are almost identical. .
SUMMER is almost as hot as in the Sahara; Petro-Alexan-
drovsk has a July mean of 83°, Tashkent of 8Io and LuktchJln
exceeds 90°. In the hottest hours.of July days the thermometer
ris~s above lIOO and Luktchun has recorded lISo. At the last-
narp.ed station local peculiarities of the site undoubtedly add to
the heat; the absence of cloud facilitates insolation, the bare,
dry ground is strongly heated and warms the overlying air, and
further, owing to its position in a deep depression, ascending air
must be replaced by air already heated by contact with the slopes
and further heated by compression as it sinks into the denser air
layers. Higher altitudes are, however, little cooler; even Lhasa
(lI,600 feet) has a July mean of over 70° and the temperature in
the sun is sometimes more than 140°.
The continental lows are now centred over these arid interiors
and air movement is somewhat ~ariable, but as in the hot deserts
DESERT CLIMATES 243
strong winds from allnost any point of the compass spring up by
day, carrying clouds of dust and sand which darken the sky.
The Karaburan is of this type, blowing strongly from the north-
east in the Tarim Basin. .

Patagonia
Patagonia, transitional between steppe and desert, differs from,
other cold deserts in that it is not typically continental"for on
account of the narrowness of the continent and the force and
regularity of the west winds there is always a certain marine
influence which prevents the winter temperature from falling
very low or the summer temperature from rising very high;
there is, moreover, at the foot of the Andes a ',foehn' effect which
increases the warmth and aridity, so that the isotherm of 32°
nowhere enters the continent. Owing to the dryness of the air
there is, however, a considerable diurnal range of temperature-
about ISO in winter and 20° in summer-so that temperatures
below zero are not unknown in the south and temperatures above
IOOo in the north. T~e strong wind, a characteristic feature of
the climates of these latitudes everywhere in the southern
hemisphere, increases the aridity and aggravates its ill effects by
z:aising storms of dust. Rainfall incre?ses to the north and east,
Neuquen has S inches, Bahia Blanca 2I inches, and there is a
gradual passage to the sub-tropical climates of the better watered
parts of the Pampa.

Desert Land Forms


The individuality of the desert environment is made up of
a subtle mixture of climate and relief, but the relief of arid lands
presents such a consistent and peculiar association of forms that
it might almost be considered as an element of the climate.
Exfoliation, frost-action and mechanical erosion by wind-blown
sand, typical desert agents, produce a hard, clear profile and
jagged lines in exposed rocks, exaggerated, in the absence of a
softening mantle of a vegetation cover, by the bareness of the
rock edge seen in a transparent atmosphere against a brazen
sky. One misses, too, the graded line, the steady fall of the
ground towards base-level, the ineluctable waste of land and the
transport of the debris to the sea. The absence of continuous
slopes and the ever-changing landscape of the sand-dunes give an
air of restlessness to desert scetlery that contrasts strongly with
the well-ordered purposeful lines of the normal fluviatile relief.
Watercourses (wadis or arroyos) that originate in the dest;rt
244 CLIMATOLOGY
contain water only once in a while and do not often escape from
the desert, but are lost in sand and stones. Those starting beyond
the desert, rising, perhaps, in the melting snows of mountains,
dwindle as they make the passage of the desert. Their thalweg
is graded, but not their valley sides; they flow in canons with
steep, rugged, irregular sides sculptured crudely along the lines of
weakness in the rocks. The Nile alone escapes from the Sahara,
tha Orange and the Kunene traverse the Kalahari, the Snake
River and the Colorado escape to north and south of the Great
Basin, but more often than not the rivers end in the desert in
basins of inland drainage. In many cases it is doubtless true
that the basin is tectonic and the inland drainage is not the result
, of the desert, but in any other climate the rivers would have
drained the interior basins. The very structures that produce
inland drainage tend to perpetuate it by erecting mountain
screens to keep out raip-bearing winds. .
Aided by the rivers which sometimes rise on the mountain
slopes which project into the higher rain-bearing layers of air, the
waste of the land creeps slowly down and buries the hills and
mountains, little by little, in their own refuse. Screes and alluvial
fans are built outwards and unite to form belts of piedmont'
gravel which may spread across the valley, so that the slopes rise
suddenly out of the flat aggraded floor. The water escapes as
springs, perhaps, along the foot of the gravel belt and here is
found a fringe of cultivable land, e.g. the line of oases in the Tarim
Basin, joined by the trade routes which pass through the basin on
their way to Ferghana and Western Turkestan.
In the absence of wood, stone or adobe form the prevalent
building material for the houses which are flat-roofed q,nd often
arranged to collect and conserve what little rain falls; in this way
they are the exact opposite of the houses of humid climates whose
sloping roofs are-designed to disperse the excessive rain.
Evaporation everywhere exceeds precipitation and lakeg are
highly saline, e.g. the Great Salt Lake (18 per cent. salinity),
the Dead Sea (24 per cent.) and Lake Van (33 per cent.). Their
shores are thickly encrusted with salt, indicating a once greater
extension, and in some cases the lake has altogether disappeared,
leaving a salina or salt-pan. .

Ve~etation and Cultivation


Without the aid of irrigation only the most xerophilous
plants can keep alive in the desert, which they do by means of
the usual devices for reducing transpiration to a minimum and
by; extending the range of their root system to a maximum.
DESERT CLIMATES 245
Otner plants, less xerophilous, have the power of remaining for
many months and even years in the resting state, waiting for
the occasional rains to waken them to active life. The incidence
of rainfall in the desert. is highly irregular and there is therefore
nothing in the nature of a seasonal rhythm to be followed; the
desert plant, like many desert peoples, is an opportunist, always
prepared to take advantage of what fortune provides. The
rate of growth of these plants, once germinated, is phenomenal,
and an ephemeral carpet of bl~ssom follows the shower with
surprising rapidity. Where the rainfall regime is more regular,
especially on the borders of the desert, there may be regular
seasonal pasture, or even crops, e.g. the iflfa grass of North
Africa, and it often happens that the seasonal supply of pasture
induces a seasonal migration of pastoral tribes. The irregular
imd usually torrential showers in the desert proper do more
harm than good;' they are not sufficiently reliable to be
put to use and th~ sudden floods to which they give rise are
a source of much damage and danger. In Peru and Chile,
where the rivers support a narrow ribbon of irrigated cultiva-
tion, the trail of sand, gravel, and boulders which these floods
leave in their wake may smother a whole season's carefully
tended crops.
Where the desert approaches the sea, e.g. in Peru and South-
"west Africa, there may be available an additional water supply
ip. the mist and fog which sometimes pass inland. Certain desert
plants possess the power of absorbing moisture from these mists
and dews through their leaves, and some exude hygroscopic
salts on their leaf surfaces which are thus salt-encrusted during
the day; the salts withdraw moisture from the cool night air
and deliquesce. The mists which daily rise from the Red Sea
. and bathe the slopes of Yemen provide not only moisture for
the famous coffee-trees but also shade from -the heat of the
midday sun.
The small amount of rain, or even dew with which crops
can be grown on desert margins is surprising, but broadly con-
sidered the cultivation of the desert is mainly a matter of irriga-
tion, either from streams or from underground sources. Streams
that originate outside the confines and lose themselves in the
desert give rise to marginal and terminal oases, but even when
the water has passed below the surface it is not lost. It may
be brought to the surface as .springs by irregularities of the
ground or of the impervious floor, or it may be tapped by wells,
often hundreds of feet deep, 3,nd sometimes, as in Australia,
artesian. The resultant oases are among the most isolated of
settlements, but they tend to become the foci of caravan routes
16 '
CLIMATOLOGY
and markets for exchange. One of the most important and cer-
tainly the most characteristic product is the date palm which
finds in the hot desert oases its optimum conditions. It is greedy
. for water at its roots but cannot stand rain which interferes
with pollination at this early stage and. spoils t.he fruit. lat.er.
If it is to bear good fruit it. demands a long hot. season (six
months above 64° and a large accumulated temperature above
this zero point); t.his last requirement is seldom available in
cold deserts. .
Maize, beans and millets are grown beneath the shade
of the date palm upon whose stems are often trained the
creeping branches of the vine. The desert gardens of Persia
,and Ferghana grow figs, pomegranates, mulberries, vines, apricots,
tobacco, opium, cotton, wheat, maize, barley, lucerne, melons,
ct~ •

FIG. 62.-Northern Limit of Productive Date Palm

Desert soils, when irrigated, are proverbially fertile, for the


long ages of drought have allowed the accumulation of sol~ble
salts which the plant requires. On the other hand, the salts
which rise by capillarity and effloresce at the surface are often a
hindrance to cultivation (e.g. the' reh' soils of India), either
because they are of a kind inimical to plant life or because they are
present in much too great concentration. In many places the
concentration of surface salts is sufficient to form the basis of
mining industries, e.g. the borax and nitre of Chile, the phos-
phates of Tunis and the salt that results from the evaporation
of lakes-often a government monopoly and sometimes a medium
of exchange. Also preserved by the aridity, though not due to
it, is the guano of Chincha Islands. Most of these salts are
valuable mainly as fertilizers and there is something rather
ironical in the spectacle of the' world's chief fertilizer being
obtained from the world's most complete deserts.
DESERT CLIMATES 247
SUGGESTIONS FOR FURTHER READING
Besides Knox, Climate of Africa; Ward, Climate of the United States,
and Taylor, Australian Meteorology, see H. Schirmer. Le Sahara, Paris,
1893; Woeikof, Le Turkestan Russe, Paris, 1914; Lespagnol, Sur Ie
caractere aesertique de l'Australie interieure, A. de G., 18g8; Lyons,
Physiography of the Nile Basin.
CLIMATOLOG Y

TEMPERATURE

~I
Station. Lat. Long. Alt. (ft.) J F M A 1.1 J J A S

SAN DIEGO 33°N. II7°W. 87 54 55 57 58 61 64 67 '68 67 63


YUMA 33°N. lI5° W. 141 55 59 65 70 77 85 91 90 84 72
CAPE ]UBY. 28°N. 13° W. Coast 61 61 63 64 65 67 68 68 69 68 I

INSALAH 27° N . 2°E. 920 55 59 68 76 86 94 99 97 92 80 I


:
ASWAN. 24°N. 33° E. 326 59 63 7° 78 85 90 91 90 88 82
CAIRO 300 N. 31° E. 380 55 57 63 7° 76 80 82 82 78 74
ADEN 13° N. 45° E . 94 76 77 79 83 87 89 88 87 88 84 I
BAGHDAD 33°N. 44°E. 12 5 49 54 61 71 81 90 95 94 88 80
]ASK. 26°N. 58° E . 13 67 68 73 80 85 90 91 89 87 83
KARACHI 25° N. 67° E. 13 65 68 75 81 85 87 84 82 82 80
]ACOBABAD. 28°N. 68° E. 186 57 62 75 86 92 98 95 92 89 79
IQUIQUE 20° S. 7° oW. 30 71 71 69 ~5 63 62 60 61 63 64
ANTOFAGASTA . 24° S. 700W . 13 72 7° 7° 67 64 63 62 60 61 62
WALFISH BAY. 23° S. 14° E. 10 65 66 66 65 62 60 59 57 58 60
PORT NOLLOTH 29° S. 17° E. 16 60 60 59 58 57 55 55 54 55 58
CARNARVON 25° S. 114° E. 15 80 81. 79 75 68 63 61 63 66 69
EUCLA • 32° S. 129° E. 30 71 71 69 66 61 56 54 56 59 63
ALICE SPRINGS. 24° S. 134° E. 2,000 84 82 77 68 60 54 52 58 66 74
HELENA 47°N. II2oW. 4,IIO 20 23 3 2 44 52 60 68 57 56 45
KAlI;1LOOPS . 5 Io N. 120° W. 1,193 23 26 38 50 58 64 70 68 58 48
ASTRAKHAN 46 °N. 48° E . -4 6 19 21 32 4 8 64 73 77 74 63 50
TEHERAN 36° N. 51° E. 4,002 34 42 48 61 71 80 85 83 77 66
IRGIS 49°N. 61° E. 360 3 4 19 44 63 72 76 73 59 42
PETRO-ALEX-
,
ANDROVSK } 41° N. 61° E. 295 23 29 43 58 73 80 83 79 67 52
KASHGAR 39°N. 76° E . 4,255 22 34 47 61 70 77 80 76 69 56
LUKTCHUN. 43° N . goO E. -56 13 27 46 66 75 85 90 85 74 56
URGA 48 ° N . 107° E. 3,740 -16 -4 13 34 48 58 63 59 48 30
SARMIENTO. 46° S. 69°W. 81 3 65 63 59 51 44 38 38 42 46 53
SANTA CRUZ 50° S. 69° W . 85 61 57 54 48 4 1 34 33 38 43 48 )
.
DESERT CLIMATES 249

RAINFALL

- ._ -
~

N D Yr. Ra. J F M A M J J A S 0 N D Total.


- -- - - - - - - - -- - - - - - ~- - - - -- - - -

59 56 61 14 1·8 1'9 1'5 0·6 0'3 0'1 0'1 0'1 0'1 0'4 0'9 1·8 9.6
61 56 72 3 6 0'4 0'5 0'4 0'1 - - 0'1 .0'5 0'2 0'2 0'3 0'4 3'1
65 62 65 8 0'5 0'5 0'5 - - - - 0'5 0'5 0'5 0'5 1'0 4'5
68 58 78 44 Pra ctica lly Nil
72 62 n 3 2 Pra ctica lly Nil
65 58 70 27 0'4 0'2 0'2 0'2 - - - - -
- 0'1 0'2 1',3
80 77 83 13 0'3 0'2 0'5 0'2 0'1 0'1 - 0'1 0'1 0'1 0'1 0'1 1'9
63 53 73 4 6 1'2 1'3 1'3 0'9 0'2 - - - -
0'1 0·8 1'2 7'0
76 70 80 24 1'1 0'9 0·8 0'2 - .0:1 - - -
- 0'3 1'1 4'5
67 78 22 0'5 0'5 0'4 0'2 0'1 0'9 2'9 1'5 0'5 -
0'1 0'1 7.6
174
68 59 79 41 0'3 0'3 0'3 0:2 0'1 0'2 1'0 1'1 0'3 - 0'1 0'1 4'0
67 6g 66 I I Pra ctica lly Nil
64 68 67 12 Pra ctica lly Nil
61 64 62 9 Pra ctica lly Nil
59 60 58 6 - 0'1 0'2 0'2 0'4 0'30'2 0'4 0'3 - 0'2 0'1 2'3
73 77 7 1 20 0'3 o·g 0'5 0:6 1'5 2·81'7 0'7 0'3 0'1 - 0'1 9'5
66 6g 64 17 0'7 0'5 o·g 1'2 1'2 1'1o·g 1'0 0·8 0'7 0'7 0'4 10'1
80 82 70 3 2 1·8 1'7 1'3 o·g 0·6 0·60'4 0'4 0'4 0'7 O'g 1'3 11'1
33 25 44 4 8 0'9 0·6 0·8 1'1 2'1 2'3 1'1 0'7 1'2 o·g 0'7 0·8 13.6
35 28 47 47 1'0 0·8
0'3 0'4 0'9 1'2 1'1 1'1 0·8 0·6 1'0 o·g 10'1
1 37 26 49 58 0'5 0'3 0'4 0'5 .0'7 0'7 0'5 0'5 0'5 0'4 0'4 0'5 5'9
15 1 42 62 51 1·6 1'0 I·g 1'4 0'5 0'1 0'2 - 0'1 0'3 1'0 1'3 9'3
26 I I 41 73 0·6 0'3 0'5 0'7 0·8 O'g 0·6 0'4 0'5 0'5 0'4 0'7 6'9

39 30 55 60 0'2 0'4 0'5 0·6 0'2 - - 0'1 - 0'1 0'1 0'1 2'4

40 26 55 58 0'3 - 0'2 0'2 0·8 0'4 0'3 0'7 0'3 - - 0'2 3'5
33 18 56 77 - - - - - - - - ~
- - - -
1

I 8 -7 ,
28 79 - 0'1 - - 0'3 1'7 2·6 2'1 0'5 0'1 0'1 0'1 7. 6
:57 61 51 27 0'1 0'3 0'5 0'4 0·8 0·6 0·8 0'4 0'4 0'4 0'2 0'2 5'1

i
55 55 47 28 0'5 0'3 0'2 0·6 0'7 0'4 1'1 0'4 0'2 0'4 0'4 1'0 6'0
CHAPTER XIV
MOUNTAIN CLIMATES
NCREASING altitude brings about a series of quite 'well-

I defined changes in the values of climatic elements which


have been discussed in a general way on pp. 35 to 41. At
first the effect serves only to introduce minor modifications of
the normal sea-level climate, such as the altitude variants of the
equatorial and tropical climates described on pp. 77 and 97, but
ultimately the alteration becomes so profound that we find
ourselves in a climate possessing such a number of individual
peculiarities as to require separate treatment. But the moun-
tain climate, being the direct result of great contrasts of relief,
is compounded of a great variety of contradictory types; of
dry burning heat on exposed slopes, of stifling heat in enclosed
valleys, of bitter cold in the same valleys on winter nights, of
fog and sunshine, of torrential rain and great aridity.
As in polar and desert lands the study of climate in mountain
regions is handicapped by the paucity of regular observations
arid we are driven again to seek information, especially for
higher levels, from the all too irregular visits of climbers. The
number of stations is, however, increasing and for the European
and, to some extent, the North American mountains there is a
fair quantity of material now available.
PRESSURE (see also p. 36). The~ decreased atmospheric
pressure is not of very great importance per se; there are, for
example, human settlements on the Andes and in Tibet where
the pressure is only about two-fifths of that at sea-level, but
it is of the greatest importance in that most of the phenomena
of mountain climates are the direct results of the rarefaction of
the air. In consequence of the fall of temperature which results
from this rarefaction the same temperature zones are passed
through on the slopes of a mountain up to the snow-line as on
a journey into high latitudes at sea-level as far as the line of
permanent snow. But the conditions in the corresponding tem-
perature zones are widely different; the characteristic feature
of insolation in the polar zone is t)).e long summer day of oblique
sunshine through a great thickness of atmosphere, the feature
of the insolation in alpine climates is the intensity of sunshine
. 250
MOUNTAIN CLIMATES 251

through a thin and often transparently clear atmosphere. The


far-reaching results of this' difference will ~ seen ~ater when
the relation of climate to vegetation and the habitability of the
zones' are discussed.
MOUNTAIN AND VALLEY WIND'S. Since mountains occur in
each and all of the great planetary' wind belts it is clearly im-
possible to generalize on the major winds of mountain regions,
except to say that on exposed slopes and summits, where the
conditions approximate to those of the free atmosphere, friction
is reduced to a minimum and winds are much stronger in pro-
portion to the barometric gradient than at low levels. But
there are a nutnber of local winds, characteristic of mountains
and due to their relief, which are of considerable importance.
Where the general air movement is not too strong to overcome
local influences there is a tendency to a diurnal reversal of wind
direction analogous to land and sea. breezes, the wind blowing
up the slopes during the day and, down the slopes at night. These'
winds are concentrated by the form of the land into the .valley?

FI~ 6,3.-Mechanism of Up-valley Winds in Mountains

where they are felt at their strongest. The down-valley wind is


the direct result of cooling by nocturnal radiation, the cold air
sliding down the slopes and into the valleys, but the cause of
the up-valley wind of the daytime is less apparent. One would
expect the air to rise vertically when heated by contact with
the ground during the period of insolation, but in actual fact
its movement is mainly horizontal. The explanation is as follows:
If the pressure distribution is stable and the isobaric surfaces
are horizontal (AB in Fig. 63) there will be no air movement.
When insolation begins the air, being warmed, will expand,
the amount of expansion being proportional to the length of
the column of air. Thus AC will expand by an amount AA';
BD, being shorter, by a shorter amount BB'. The originally
horizontal isobaric surface has now been tilted into the position
A'B' and air will begin to flow towards the mountain side.
H~re the air, directly warmed by contact with the ground, will
have a tendency to rise and to carry the horizontally moving
CLIMATOLOGY
air up the valley. Often these winds blow with considerable
force, especially the night wind; besides, being colder, it is
more generally noticed by mountain dwellers and visitors. In
the" neighbourhood of large snowfields and glaciers the force of
the night wind is strengthened by the cooling of the air in contact
with the ice and snow; in fact this effect is in places so profound
as to overcome the day wind and to give rise to a fairly regular
and persistent system of cold downcast winds, e.g. the Nevados,
an unpleasant feature of the climate of the higher valleys of
Ecuador.
THE FOEHN AND ALLIED WINDS. In mountain districts the
horizontal movement of the air is interfered with by the relief
and air is compelled to ascend and descend. In doing so it
'undergoes changes of pressure which result in changes of tem-
perature which, in turn, result in changes of relative humidity,
the ascending winds tending to bring rain, the descending winds
. tending to cause evaporation. These winds are well-known in
all mountain districts, but the best known is the foehn whose
name has been adopted as a general term for winds of this type.
It is an excessively hot, dry wind occurring typically in the
valleys of the northern alpine slopes, and though the wind itself
is not due to the ~ountains it owes its properties to the "relief.
It occurs when a depression moving to the north of the crest
of the Alps draws air from higher pressures lying to the south
across the divide. In ascending the southern slopes the air is
cooled below its dew-point, clouds are formed and heavy rain
occurs, the liberation of heat by the condensation checking the
fall of temperature, so that by the time it reaches the divide
it has lost most of its moisture but retained most of its warmth.
In its descent of the northern slopes it is adiabatically warmed
(in descending 5,000 feet it would be wa.rmed nearly 15°) and
blows as a hot, dry wind down the valleys. The foehn is an-
nounced by a crest of cloud along the mountains to the sOl1.th
while" the air is still, heavy and oppressive, yet so clear that
the distant mountains seem very near and tinged with a bluish
colour. Then come a few cold puffs of wind, then ominous calm,
then the hot blast of the foehn. The thermometer may rise 18°
or 20°, the snow vanishes in the hot, thirsty air and everything
becomes tinder dry. The wooden chalets are ready to catch
fire at a spark, but all fires are put out for safety's sake as soon
as the foehn appears.
The average yearly frequency of days with foehn winds in
northern Switzerland is as follows: Spring, 17'3; Summer, 4'9 ;
Autumn, 9;6; Winter, 9'1; year, 40'9. This is the same. as
the frequency of cyclones, i.e. a spring maximum (d. the Chinook),
MOUNTAIN CLIMATES 253
a fortunate circumstance as its effect in freeing fields and pastures
from their winter covering cif snow accelerates the beginning of
the agricultural year. But its occurrence in autumn and winter
have important consequences, the former helping to ripen the
harvest, especially the grape harvest which in some places is
dependent on the foehn for its success.
In general it is a southerly wind-it used to be thought that
it came from the Sahara, but this is neither true nor necessary
-:-but like most mountain ~nds it follows the relief lines, blowing
most strongly down those valleys whose courses coincide with
the pressure gradient, i.e. the north and south valleys such as
the Rhine from Chur t@ Lake Constance, the Aar to Lake Brienz,
and the Rhone from Martigny to Lake Geneva, but not in the
Rhone furrow from the Furka to Martigny, nor in the Vorder'
Rhein which lie at right-angles to the line of flow.
Similar winds to the foehn occur in all mountain districts
where cyclonic storms occur. Thqugh rarer than on tq,e northern .
.slopes it is by no means unknown in the Italian Alps in the
valleys of the Ticino and Toce ; the Chinook (see p. 214) is exactly
similar, so are the Samun of Persia, descending from the mountains
of Kurdistan, the Nor'-Westers of New Zealand off the New
Zealand Alps, and many others. . .
INSOLATION AND TEMPERATURE. The thin air of mountains,
almost free from dust and moisture particles, is extremely in-
effective in absorbing the sun's radiation which is thus allowed
to pass through with little loss of intensity. The short wave-
length violet and ultra-violet rays are absorbed first, those of
the red end of the spectrum penetrating the atmosphere most
successfully. Elster and Geitel showed that 40 per cent. of the
ultra-violet rays penetrate as far as the summit of Sonnblick
(10,000 feet), only 31 per cent. as far as Kolm Saigurn (5,250
feet) and only 16 per cent. reach sea-level. Insolation on moun-
tains i? thus not only very powerful but there is also a markedly
high proportion of the more refrangible ultra-violet rays. Now
these rays are of great value to plant life and are stimulating to
animals and man, they are powerful to burn the skin and, com-
bined. with the reflection from a white snow surface, quickly
produce a deep tan. The bright light, especially when reflected
from the snow surface, is trying to' the eyes and may cause
snow-blindness unless dark glasses are worn.
It is the avidity with which the 10wPf layers of the atmosphere
absorb the sun's radiation and especially the'infra-red rays that
accounts for their rapid warming up, and conversely it is the
trap.sparency of the upper layers which diminishes their rise in
temperature in spite of the power of the sun's rays. Rocks,
254 CLIMATOLOGY
stones and 'soil, on the other hand, absorb heat readily and their
temperature rises rapidly in the sun. The result is a surprising
difference between the temp~rature of the ground and the tem-
perature' of the air near it, between temperature in the sun and
temperature in th~ shade. Rann quotes an example froml Leh
(lI,500 feet) where, on lIth August; 1867, the shade temperature
was 70°, but the black bulb in vacu,o was registering 214°, which
is 22° above the' boiling-point for this altitude. As the sun
sinks in the sky and as insolatioll decreases and ceases the ground
rapidly loses its heat by radiation through the thin air, especially
as the convex form of the mountain presents a large surface for
radiation, and night temperatures of the ground surface falls to
a very low figure. Thus the diurnal changes of surface tempera-
'ture are very severe, an important factor in the disintegration
of rocks and in the production of the typical jagged peaks and
" aiguilles " of mountain scenery. But the air is as slow to lose
its heat at night as it was to absorb it by day and the air tem-
perature lags far above the ground temperature. Furthermore,
the air is usually in motion and fresh supplies are imported
before the air in contact with, the mountain can be chilled. The
diurnal motion of the air operates in the same direction; the
ascending currents of dayti,me, cooled by expansion, moderate
the daily maximum; the descending currents of the night-time,
warmed by compression, moderate the nocturnal minimum.
Thus a variety of factors combine to reduce the daily range of
air temperature which, as in marine climates, is very small;
Mont Blanc has a mean diurnal range in July of only 6'3 0 as
compared with 19° at Geneva.
EFFECT OF LOCAL RELIEF ON TEMPERATURE. Up to this
point we have considered only the changes of temperature on
continuous mountain slopes and have seen that in general high
level stations enjoy a diminished range of temperature both
diurnal and annual, but the form of the ground and the exposure
of the site and a number of other purely local factors modify
or even reverse the generalizations we have made. Exposure
with respect to the midday sun makes a very considerable differ-
ence to surface temperature and has a profound effect on the
habitability of a slope outside the tropics. Owing to the trans-
parency of the air the shaded slope lies in heavy shadow while
the sunny slope is bathed in brilliant light and warmth. The
,significance of the contrast 'is witnessed by the distinguishing
names which these receive: in German sonnenseite and, schatten-
seite, in French l'adret and l'ubac, in Italian ad,'etto and opaco.
The exposure may cause curious features in the daily march of
temperature; an east-facing slope will have warm mornings and
MOUNTAIN CLIMATES 255
cool afternoons, while the opposite will be the case on a west~facing
slope.
Just· as the convex form of mountain peaks is a condition
disposing to moder.ate temperatures, so the concave form of the
valley disposes to extremes. Valley air is warmed by day from
three sides, by contact with the bottom and both slopes, by
night it is cooled by radiation from three sides. By night, also,
the air, chilled by radiation, flows down the slopes and settles
in the hollows, being displaced on the slopes by warmer air.
Thus in calm weather, when the air layers are not disturbed by
winds, the stagnant air in the hollows may be many' degrees,
colder than at stations thousands of feet higher. It is under
these conditions that the lowest temperatures occur, the lowest
figure on record in U.S.A. is - 65°, reached at Miles City, Montana·
(2,37I leet), which lies in a deep hollow in the Great Plains;
Pike's Peak though II,OOO feet higher, has never recorded a
temperature below -40°. Hot by; day and cold by night, vaHey'
bottoms have a very considerable diurnal range.
Intermediate in form between the convex mountain and the
concave valley is the flat plateau. Here also we find the contrast
between surface temperature and air temperature, but the range
of air temperature is greater than on mountain slopes though
.less than in valley bottoms. #

The yearly march of temperature in mountain climates Closely


resembles the diurnal march, being characteri:z;ed by:-;-
I. A progressive decrease of mean temperature with altitude.
2. A small annual range on peaks and slopes.
3. A large annual range in valley bottoms.
4. A delayed minimum and maximum.
5. Autumn much warmer than Spring.
The last two features are particularly noticeable where there
is a heavy winter snow-cover whose melting absorbs the heat
of the spring sun. This is paralleled in cold climates and in
marine climates, the delay in the latter case being due to the
expenditure of heat required for warming the sea.
HUMIDITY AND CLOUDINESS. Most of the water vapour in
the atmosphere is. concentrated in the lower layers (at 6,500 feet
the amount is reduced by a half), thus the air of the mountain
tops is excessively dry and the low pressure further increases
physiological dryness by allowing water vapour to be dissemin-
ated rapidly into the air. Evaporation is rapid, perspiration
disappears at once, the skin is parched, face, hands and lips
become cracked and thirst is increased. On the high plateaux
of ;Peru the Incas were able ·to preserve their dead by allowing
them to' mummify naturally.
CLIMATOLOGY·
The daily march of humidity is fairly regular and is intimately
connected with the daily movement of the air. The day wind
carries moisture upwards,' cumulus and strata-cumulus clouds
form, their flat bases marking the level at which the dew-point
is reached; the night wind carries the moisture down again,
the summits are dry and clear and from aloft one may look
down on the billowy surface of the fog that fills the valleys,
condensed by the cold stagnant air that has drained into them
during the night. Thus at high levels the nights are fine and
dry and the best views are obtaineq in the early morning, there
is an afternoon maximum of humidity and of rain and there is
a tendency to afternoon thunderstorms; at low levels the n,ights
are often cloudy and foggy, but the mists melt before the morn-
'ing sun and rise again up the mountain-sides. The limits of
the migration of this zone of maximum cloudiness vary with
the season, rising in summer and falling in winter.. In winter
the alpine peaks and the higher valleys, wher~ are all the famous
centres for winter sports, rise into brilliant sunshine out of a
sea of cloud that enshrouds the lower slopes. In summer these
high valleys are cloudier and the lower levels are comparatively
free; on the Swiss plateau at r,500 feet winter is by far the
cloudiest season; on the summits of the passes at 7,000 feet
spring and on the mountain-tops at ro,OOO to r5,000 feet, summer.
PRECIPITATION. As has been described on p. 38, mountains,
by forcing the ascent of horizontally moving air, bring about
increased precipitation up to an indefinite level of maximum
precipitation, above which, owing to increasing dryness of the
air, there is a decrease of precipitation. This greater raininess
of high levels is of the greatest importance in arid lands since
here alone is water available; the rainfall of the high sierras
above the deserts of Peru nourishes the irrigated strips that·
run as green ribbons through the brown desert sands; the Tibesti
and Ahaggar highlands are well-watered. oases in the midst of
the Saharan desolation; the mountains that surround the Tarim
Basin provide a zone of sorely needed pasture between ro,ooo
and 'r4,000 feet. But the distribution of rain in mountain
regions is extremely complex because of the complexity of relief,
as an examination of a rainfall map of, say, the Alps will show
in no uncertain manner. In general the heights stand out as
areas of heavy precipitation and the valleys as drier strips;
thus the St. Gotthard mass has more than 80 inches, but places
in the Valais have less than 25 inches.
The daily and annual march ,,of rainfall accompanies the
march of cloudiness, rising in summer and in the heat of the
day, falling at ~ght and in winter. These tendencies, super.:.
MOUNTAIN CLIMATES 257
imposed on the normal regime for the climate in which the
mou,ntains are situated, produce an incidence of rainfall which
defies generalization.
SNOWFALL AND THE SNow-LINE (see also p. 37). The position
of the limit of permanent snow depends on a number of factors
both regional and local and only broad generalizations can be
made. The level of snow at any particular time clearly repre-
sents the line of balance between accumulation and dispersal,
Le. between the amount of precipitation ·in the form of snow
on the one hand and the rate of melting and evaporation on the
other. But both of these depend in turn on the aspect with
respect to the snow-bearing winds and to the sun, on the steep-
11,000
/
f-.,
10,000

9,000 I/' 1\
II- I ,\
8,000
7,000 / \ .
bOOO 1/ 1/ 'I'
5,000 J J ,\
1/ 17
1\
4,000

3,000
v v \\
2,000
a . / V'..,..
17 I,
~I-
lOOO
SL
JFMAMJJASOND
FIG. 64.-Mean Height of Snow-level in the Inn Valley in Northern Tyrol
a. On south-facing slope (
b. On north-facing slope

ness of the slope and other local influences. The line, of course,
migrates up and down the slopes in summer and winter. Fig. 64
shows the migration of the line on the two slopes of the Inn
Valley in the Northern Tyrol and illustrates the magnitude of
the difference that aspect makes. The term ' snow-line' refers,
however, to the line above which the snow does not melt in
summer, i.e. to the crests of the curves in Fig. 64 and is deter-
mined mainly by two values, viz. :-the winter precipitation
which determines the rate of accumulation and the summer
temperature which determines the rate of ablation. It is a line
of some biological significance, for plants can grow where the
snQw disappears in summer (cf. the tundra), no independent lif~
being possible above this line.
CLIMATOLOGY

Ve~etation and Cultivation


The variations of temperature and rainfall which mountains
bring about are sudden and abrupt, so that the climatic' optimum
may be displaced within a very short distance from one vegetation
type to another. Consequently there is a great·variety of plant
associations within a small space and a corresponding variety of
cultivation and of human occ!lpations. It is quite common for
one family in the French Alps to cultivate vineyards on the
lower slopes in an almost sub-tropical climate and wheat or barley
in a temperate climate at higher levels, to cut their wood from
the forest zone, to pasture their cattle on the alpine meadows
and to act as guides among the glaciers and peaks. The short
'distance between the zones encourages seasonal migration from
one. to another and the rigour of the higl;ter altitudes in winter
is easily escaped (contrast the tundra dweller). In general there
is a movement of animals and man up the slopes in spring follow-
ing the melting snows and down again in autumn; but some-
times the movement is in the reverse direction, e.g. in Central
Asia where the Kirghiz drive their flocks and herds above the
level of the winter clouds into the bright sunlight that bathes
the high pastures watered by the summer rains.
Whatever the vegetation zone in which the base of the moun-
tain stands the zone of maximum precipitation is nearly always
forest, deciduous below, passing upwards into coniferous. Be-
tween this forest and the snow-line, in the region of steadily
decreasing precipitation, stretches an ' alpine' zone of grassland,
narrow in the Alps where the snows are not far above the upper
limit of the forest, but very wide in the tropics, becoming drier
and drier towards its upper limit (see p. lOr). On the plateaux
of Arizona and New Mexico, a region 'Yhich would 'be desert
but for its altitude, the forest extends from 6,500 to 8,000 feet
and an alpine flora follows above this level. . The low temperature
and the high evaporation aided by low pressure and strong
winds here demand a xerophytic vegetation. Alpine pastures
are r~howned for the bright colours of their flowers and for the
tender succulence of their grasses, the basis of the Swiss dairying
industry. These properties they owe to the bright sunlight,
the warm soil and cool air, the fertile glacier silt, the adequate
rainfall and the supplies of moisture from melting snows. The
climatic conditions here are in strong contrast with those experi-
enced by the corresponding vegetation in Arctic climates, where
the sunlight from the low-angled sun is weak, where the soil is
poor and acid, ill-drained and saturated with ice-cold water,
The effect of decreased temperature at higher altitudes is to
MOUNTAIN CLIMATES 259
delay the function of plants by an amount which in the Eastern
Alps amounts to about ten days per I,OOO feet of altitude, as
is shown by the accompanying diagram based on the work of
the brothers Sch1?-gintweit, quoted by Hann. Thus at higher
levels the season available for growth becomes shorter and
shorter and crop after crop is eliminated. The limit of grain
is reached in the Alps at 4,000 to 6,000 feet according to local
conditions of rainfall, exposure, etc. On north-facing slopes the
vegetation is on an average about a fortnight behind that on
south-facing slopes .. Boissier has shown that on the southern
slopes of the Sierra Nevada in Spain, where conditions are African,
the limit of the vine is 7,000 feet higher than on the northern

450.0

May June J~ly Aug Sept Oct Nov Oec


FIG. 6S.-Altitude and the Growing Season in the Eastern Alps

slopes where the conditions are more like those of the Meseta,
that of the olive 1,700 feet and of rye 600 feet higher.
The phenomenon of temperature inversion has some important
effects on vegetation and on settlement in valleys and basins.
Houses and farms avoid the valley bottoms which are cold and
foggy and subject to severe cold snaps, preferring the slopes
and the alluvial fans raised above the valley floor which occnr
where tributary streams debouch into the main valley. The
liability to late and early frosts in such ' frost holes' or ' frost
pockets' shortens the growing season and prohibits the cultiva-
tion of delicate plants. The coffee plantations of Sao Paulo,
for example, are always situated on hill-sides for the drainage
of cold air. So sharp is the line of division between air above
and air below freezing-point in some cases that the bottom of
a shrub may be nipped by the frost, but its crown may be green,
260 CLIMATOLOGY
healthy and unaffected. The so-called thermal belt of some of
the Appalachian valleys is a zone about 300 feet above the floor
of the valley untouched by frost, which retains its verdant
green through the winter, separated by a perfectly horizontal
line from the blackened frosted vegetation of the valley bottom.
THE MOUNTAIN ENVIRONMENT tends to isolation because of
difficulties of transport and intercommunication and so to self-
reliance but also to insularity and suspicion; the climate dis-
poses towards versatility since the inhabitant gains experience
of several climatic zones and usually spends a part of the year
in each. He is agriculturist, pastoralist, forester and indoor
worker in one. Physically he is adapted in some degree to the
diminished pressure in which he lives; a larger lung develop-
. ment and wider chests begin to be noticeable above about 5,000
feet and at greater heights there is a marked increase in the
number of red corpuscles (see p. 36). Usually the physique is
good, e.g. the people of the South American ' Puna' and some
of the Kirghiz, but elsewhere there is a marked degeneration,
e.g. the Mexican Indians. Cretinism, prevalent in certain valleys,
has sometimes been attributed in part to the darkness of deep
north-facing depressions whicn never see the sunlight.

Regional Types
Out of the great number of regions with mountain climates
it will be sufficient to describe three only :-
I. The Alps, an example of mouhtain climates rising from
the cool-temperate zone.
2. The High Andes, an example in tropical latitudes; and
3. Tibet, rising from the arid interior steppes.

The Alps
Chiefly because they are best known and because they are
well supplied with meteorological stations the Alps have already
been used to illustrate many of the general points of the mountain
type of climate in the foregoing pages; the more particular
characters which they owe to their position and surroundings
will be discussed below.
The most arresting feature of the winter pressure distribution
in Europe is the ridge of high pressure-the climatic backbone
of Europe-an offshoot of the Asiatic high which extends from
the Black Sea to the Meseta of Spain and continues to the Azores.
It is this high pressure which gives Switzerland its fine wint~rs,
,with clear, crisp, cold air and cloudless skies. It allows local
MOUNTAIN CLIMATES
factors full play, permitting mountain and valley winds to
develop freely and encouraging inversions of temperature and
other climatic phenomena. .
In summer, too, the Azores high sends a ridge-like extension
eastwards into Central Europe, its axis again coinciding with
the axis of the Alps; again local conditions are favourable for
the free development of the mountain climate in its ideal form.
Thus l?y virtue of their position in the planetary wind system,
as welfas 'by virtue of their relief, the Alps function as a climatic
divide, separating the cloudy cyclonic climates of Central Europe
from the sunny Mediterranean province. One may recognize
an altitude variety of the latter, with the characteristic tendency
to summer drought over most of the southern slopes in Provence
and in the Maritime Alps, but the greater part of the Alps shares .
the Central European regime with summer rain.
The climate becomes more extreme to the east in proportion
to the distance from marine influences, the change being more
than normally rapid since the relief' accelerates aridity on the
leeward slopes. There is a decrease of rainfall eastwards and
an increase in the annual range of temperature; some of the
enclosed valleys of the Eastern Alps in Austria are bitt~rly cold
in winter, especially during the night an+d during anticyclonic
weather. Klagenfurt has a mean January temperature of 20°,
the same as on the Obir, more than 5,000 feet higher, and fre-
quently the lower station is the colder by some 20°.

The High Andes


The succession of temperature and vegetation zones, the
tierra caliente, tierra templada and tierra fria, has been described
on pp. IOO and IOI ; the first two are sc\!.rcely 11l0re than altitude
variants of the tropical climate, but by the level of the' tierra fria'
we have reached the land of hot sun and cold shadow-the
mountain type of climate. The temperatures are reduced by
the altitude to those of an English spring (Quito 54°-55°, La
Paz 44°-53°), but the conditions are in no way similar. In
contrast with the Alps aspect has here very little significance,
nor have the seasonal variations of temperature; the annual
range at Quito is less than IO, at Bogota less than 2°, at La Paz
less than 9°. The diurnal range is much greater (about 25°)
and of much greater significance than the annual, and the daily
procession of weather is strikingly regular. During the rainy
season the mornings dawn unpleasantly chilly, but soon warm
up, the latter part of the mcming being very pleasant with a
warm sun and a transparently clear air. But by mid~ay clouds
17
262 CLIMATOLOGY
appear and cover the sky throughout the afternoon; by two or
three o'clock the daily storm breaks, often with thunder and hail.
There is no real dry season on the high Andes in Colombia and
Ecuador, but southwards into Peru and Bolivia it increases in
length and at La Paz there are five completely dry months.
It is during this dry season that the greatest extremes of tem-
perature are recorded; the thermometer generally falls below
freezing at night and by day the sun scorches and blisters through
the thin air.
The conditions of the' tierra fria ' are exaggerated in the Puna
between 10,000 and 13,000 feet; the nights are .bitterly cold,
even in summer, while in winter the wind is so cold that it sears
the skin. Woollen masks are used to protect the face and the
, heaviest of ponchos and shawls are worn. The sunlight is
intensely brilliant and scorching hot, the shadow is dark purple
ana icy cold, for the clear air can neither hold the heat nor
disperse the light. The natives avoid the shadow and even do
their cooking out of doors in the sunlight. The daily winds
blow with great force", raising clouds of dust; in winter they are
of a penetrating cold that blows through almost any clothing
and brings a crop of ailments, especially to the lungs-at Oruro
they call them the 'Harvest of Death'. The rains last only
about three months and during this season. the lakes and rivers
expand and rise, but dwindle again in the dry season to shallow
salt-pans and dry stony courses which is their normal nature
for the rest of the year.
The xerophytic vegetation consists only of coarse tufted grass
and cushion-like plants with small leathery leaves and buds
sealed in hard cases. Agriculture is limited to the hardiest of
crops, potatoes, beans and barley grown vzherever water is
available. The staple industry is pastoral, depending on the
mountain wool-bearing llama, alpaca, chinchilla and vicuna.
Above the puna and up to the snow-line, intruded by the
tongues .of glaciers, stretches the bleak, uncultivable moorland
known as the paramos, a region of still more stunted vegetation,
lichens and mosses with, in favoured spots, mats of plants bearing
large bright blue and purple flowers, but barren and inhospitable
in the extreme.

Tibet and the Plateaux of Central Asia


The Himalayas, like the Alps, function as a climatic divide,
barring the way against the summer monsoon purely by virtue
of their relief. To the north of ,them lies the lofty plateau of
Tibet whose northern rim is the Kuen Lun, beyond which lie
MOUNTAIN CLIMATES
the desert depressions· of Eastern Turkestan and the Tsaidam.
The last vestiges of the monsoon climate are recognizable in the
south-east of Tibet, ,in the deeply trenched valleys of the upper
courses of the Tsanpo, Salween, Mekong and YaIig-tse, by way
of which the monsoon currents find access to the plateaux and
bring as much as 40 inches of rain to Lhasa. This is the most
favoured part of Tibet, fortunate in possessing an Alpine type of
relief, with fertile aggraded valleys and hill-sides which are densely
forested' but can be cleared for the cultivation of maize, millets
and wheat. It is the most densely populated part, and the only'
towns of any size are to be found here. r

The rest of Tibet is an arid plateau, virtually desert, most


of it with less than 10 inches of rain or snow, its aridity increased
by the diminished pressure and high winds. About z inches of
snow falls in th~ south-west, winter precipitation related to the
cyclonic storms which bring rainfall to the Punjab and Kashmir
(see p. IIZ). The Pamirs and the Tian Shan form altitude
oases with z5 inches to 30 inches (mostly summer rain) and
separate the deserts of Eastern from those of Western Turkestan,
but the rest of the plateau of Central Asia is so dry that the
snow-line is some 4,000 feet higher than on the south face of the
Himalayas (here at 16,000 feet) and the streams flow only in
summer when the snows are melting. The lakes are mostly
saline and there are expanses of dazzling white salt deposits
that mark the site of others that have vanished.
Winter is rigorous, the excessive cold (readings of - 30°
and - 40° appear to be not uncommon) being ,aggravated by
high winds during the day and sometimes by blizzards of great
violence. Summer is short, the growing season lasting only
from April to September, and no month i& free from the risk of
frost owing, to the large diurnal range of temperature. The
nights are generally calm and clear and the early mornings are
generally the pleasantest time of the day, for strong winds
which spring up in the morning are still, as in winter, one of the
most unwelcome features of the climate, bringing clouds of pene-
trating dust and sometimes even snow. They are not, however,
without their uses, for they help to sweep away the snow from
the pastures, especially in spring.

SUGGESTIONS FOR FURTHER READING


The influence of altitude on the various elements of climate are fully
treated in Hann's Handbook oj Climatology (Ward's translation); the
climate of Switzerland in Das Klima der Schweitz, 1864-1900, 2 vols.,
Frauenfeld, 1905, Maurer J. et al. • A good account is also given by E.
De Martonne in Les Alpes, 1926. - '
CLIMATOLOGY

TEMPERATURE

Station. Lat. Long. Alt. (ft.) J F M 'A M J J A S


-° -
N
-- ,--- - - - - - - -
OBIR 47° N. 14° E. 6,643 20 21 23 29 37 43 48 48 43 35 27
SONNBLICK 47° N . 13° E. 10,097 9 8 II 15 24 3°' 34 34 30 23 16
SANTIS 47°N. 9°E. 8,202 16 16 17 24 3 1 37 41 41 37 30 23
ANDERMATT 47° N . 9° E. 4,74 1 20 24 28 36 43 50 53 52 47 39 30
DAVOS. 47° N . 10° E. 5,121 19 23 27 36 41 50 54 52 47 38 301
BEVERS 47° N . 10° E. 5,610 14 20 24 33 42 49 53 51 46 36 26~
~

PIKE'S PEA 39° N : 105° W. 14,11 I 2 4 8 13 23 33 40 39 32 22 II,


I
LEH 34° N. 78° E. 11,5 03 17 19 31 43 50 58 63 61 54 43 32
AREQUIPA. 16° S. 72°W. 8,041 58 58 58 58 58 57 57 57 58 58 58
USPALLATA 33° S. 700W. 9,335 53 52 51 46 38 34 36 3 8 39 42 48
MOUNTAIN CLIMATES 2 65

RAINFALL

i';
/-
D
-
Yr. Ra.
-
J F M A
------ -- -- --
M J
-_ -_ - - -_ - - - - - - -
J A S 0 N D Total.

I 22 33 28 3'1 3'4 4·6 5'3 5'4 6·6 6'7 6'4 5'5 5'9 4'1 3'4 60'4
20 25 4'9 4'9 6'3 6·6 6'2 5"5 5.6 5'1 4·6 5'1 4·6 5'3 64'7
I II

17 27 25 5'7 6'7 6'7 8'1 7.8 11'2 12'3 10·8 8'3 7'2 4.8 6'1 95'7
'I
I'
22 37 33 3.8 4'2 3'5 3'2 3'4 3'5 4'3 4'7 5'5 5'6 3'1 2'9 47'7
i;
21
, i 1,8 2'2 2'2 2'2 2'3 4'0 4'9 5'0 3'7 2'7 2'2 2'5 35'7
I: 37 35
I ! 16 34 39 1'4 1'0 1,6 2'2 2·6 3'4 4'3 4'3 4'2 3'5 2'4 1·8 32'7
6 19 3 8 1·6 1'5 2'0 3'5 3.8 1·6 4'2 3.8 1'7 1'4 1'9 2·6 29·6
22 41 46 0'4 0'3 0'3 0'2 0'2 0'2 0'5 0'5 0'3 0'2 - 0'2 3'2

58 58 I 1'2 1'7 0,6 0'2 - - - - - - 0'1 0'4 4':;:

53 44 19 0·8 1'2 0'2 - 0'1 0·6 - 1·6 0'2 1'3 0,8 0'2 7'0
J I
CHAPTER xy
CHANGES OF CLIMATE
o branch of climatology has received so much attention
N during recent years as that of climatic change; geology,
botany, zoology, anthropology, m~teorology, astronomy
and other kindred sciences are daily supplying an almost over-
. whelming mass of fresh evidence Qearing on the subject; theories
as to its causes are being acclaimed, rejected, revived and modified
from day to day; few are without serious objections and none
find universal acceptance. It will only be possible here to give
an outline of the better established facts relating to the climates
of the past and to discuss briefly the means by which they may
have been brought about.
So long as the Nebular Hypothesis of the origin of the solar
system held the field, it seemed natural to suppose a certain
amount of progressive cooling of climates on the earth's surface,
and there appeared to be much confirmatory evidence in the
geological record for warmer climates in the past; but unfortu-
nately for this view some of the oldest rocks known contain records
of an Ice Age in what are now temperate latitudes and it is
clear that from very early times there have been climates of a
nature comparable with those of to-day. Life, evidenced by fossils,
has existed on the earth from before the Cambrian and probably
from a very much earlier period, from which it is apparent at
least that the somewhat narrow temperature limits of life have
not been overstepped at any time s_ince then.
EVIDENCE FOR CHANGES OF CLIMATE IN GEOLOGICAL TIME.
In the more remote periods the nature of the fossils is not of
great assistance in identifying climatic types, for most of them
are aquatic forms showing very little adaptation to climatic
conditions, but with the greater specialization of forms it becomes
possible, by Mesozoic times, to recognize differences between
boreal and tropical types. Still later, with the evolution of more
specialized land fauna and flora, much information can be deduced
from fossils as to the climatic conditions of their habitats, e.g.
the mammoth, the sabre-toothed tiger, the arctic willow, etc.
The evidence provided, by the" nature of the rocks'is mpre
satisfactory, e.g. boulder clays indicating glacial climates, salt
266
CHANGES OF CLIMATE
and gypsum deposits indicating aridity, coral limestones indicating
warm seas, coal seams which, from the absence of annual rings
of growth in the trees, have been interpreted a~ tropical forest
vegetation. Qther plants (see p. 272) did develop rings, demon-
strating the contemporary existence in other parts of the world
of seasonal changes, whi.,ch are further proved by alternations
of coarse and fine sedim'ents, resulting from seasonal flood and
low water (see p. 282), and alternations of rock-salt and gypsum.
Rock salt is more soluble in warm water than in cold, gypsum
more soluble in cold water than in warm; seasonal changes of
temperature 'res-qlt, therefore, in the crystallization from a
saturated solution of salt in winter and gypsum in summer.
Judged on such criteria as these the British Isles appear to have.
passed through many climatic vicissitudes, including tropical
rain forest in the Upper Carboniferous, sub-tropical arid' in
the Triassic, cool temperate in the Cretaceous, warm temperate
in the Eocene, and arctic in the Quarternary. .
EVIDENCE OF CLIMATIC CHANGE IN HISTORICAL TIME. There
can be no reasonable doubt that though the climates of geological
time differed in no fundamental respects from those of the present
time, the distribution of the climatic zones and their limits have
not always coincided with those of to-day. But with regard to
historic t~me (using historic in its widest sense to include the
record Of the last 7,000 years for which we have written records,
however scanty) there is less, though increasing, agreement
among authors. The evidence is collected from a great variety
of sources, few of them absolutely convincing in themselves, but
all pointing with such a high degree of accord in the one direction
that it seems impossible to escape the conclusion. Furthermore,
as will be shown later (p. 283), modern research tends to show
that much of the later 'geological time', for which climatic
oscillations are definitely proved, ~as contemporary with the
earlier 'historical time' elsewhere.
Early efforts to demonstrate these changes were stultified
by attempting to prove a gradual and progressive desiccation
-a wrong conception and one easily defeated.. At the present
moment the prevalent hypothesis, well supported by the facts,
is that of oscillation of climate above and below a certain mean;
periods of warm climate alternating with cooler, periods of dry
climate alternating with wetter. The evidence includes:-
Rainfall and other climatic records such as the meteorological
register kept by Claudius Ptolemaeus at Alexandria in the
first century and that of TYCho Brahe at Uranienborg in the
six.teenth.
Records of floods and droughts.
268 CLIMATOLOGY
Records of dates of sowing and harvesting of crops (there
are records of the wine harvest in parts of Europe since 1400).
Records of dates of freezing of harbours and rivers (winters
with ice off the Danish coast have been recorded since 1350).
Descriptions and comments on weather in contemporary
literature; usually relating only to phenomenal weather.
Legends, e.g. (The Flood' and (The Twilight of the Gods'.
Variations in width of spacing between the annual rings of
growth in trees, especially in the sequoias of California, some
of which are more than 3,000 years old.
Past distributions of plants sensitive to climatic limitations,.
e.g. the date palm and the vine.
Dead forests in lands with rainfall now inadequate for forest
'growth; peat bogs in lands too dry at present.
Evidence of settlement (e.g. ruined cities) where settlement
is now impossible because of climate, e.g. Palmyra in the Syrian
. desert which, it is claimed, must have had a population of more
than 100,000 but now has not enough water supply for 1,000.
Evidence of agriculture (e.g. wine presses and threshing floors)
where agriculture is now impossible.
Roads round dry lake basins and bridges over watercourses
now dry.
Irrigation works where rainfall is now adequate or old condutts
from sources now dry.
Records of lake levels, e.g. the Caspian Sea and Victoria
Nyanza.
Old strand lines of lakes; dried-up lakes and salt-pans.
Advance and retreat of glaciers.
Burials in Greenland: Coffins have been excavated from soil
which is now permanently frozen. The coffins were penetrated
by plant roots showing that the summel; thaw must have pene-
trated much more deeply at the time of burial.
Migrations of peoples on a large scale which may be ascribed
to, or at least correlated with, 'increasing drought of their hpme
region.
All this material must be treated with great care and cir-
cumspection, for there is a grave risk of discovering the changes
of climate for which one is seeking in phenomena really due to
other causes. Thus misrule may lead to the decay of irrigation
works and so to the abandonment of settlements; great invasions
and migrations may have their cause in personal ambition;
improvement of varieties may allow the extension of cultivation
into areas previollslyfconsidered unprofitable; irrigation works
may- alter the level of lakes and rivers; new crops may 'put old
ones, out of use; wind-blown sand may destroy oases; or a
CHANGES OF CLIMATE 269
few barren years may result in the evacuation of irrigated lands
which the desert then reclaims.

Causes of Climatic Change


Any change in the distribution of the climatic elements
presupposes a change in climatic factors, nearly all of which
have, at one time or another, been held entirely responsible>
for Ice Ages; but so complicated are the facts that it is unlikely
that it will ever be possible to explain them all on the suppo:;i-
tion of a single variable. Most of the earlier hypotheses assumed
a change of solar.climate, a change in the amount of the heat
received from the sun' or in its seasonal distribution.
CROLL'S THEORY. Thus Croll based his hypothesis on change's
in the eccentricity of the earth's orbit and supposed that at
periods of high eccentricity the long\ winter of the hemisphere,
in aphelion would not be balanced by the short hot summer in
perihelion. Ice would accumulate in winter and riot be dispersed
in summer and would ultimately grow into ice-sheets. The
precession of the equinoxes, which depends on the eccentricity
of the orbit, has a period of 25,000 years, and one would expect
.alternate glaciations in each hemisphere recurring with that
frequency. There is, however, a further effect, produced by
the bunching of the planets, which further increases the eccen-
tricity and gives a period of IOO,OOO years. The hypothesis
does not fit the facts as they are now known; the glaciations
in the two hemispheres were synchronous, not alternating, their
period is neither 25,000 nor IOO,OOO years and is not even
regular; there have been oscillations of climate between glacial
periods for which the theory offers no explanation and finally
the close of the Ice Age appears to have been only about
15,000 years ago, compared with Croll's calculation of 80,000
years.
DRAYSON'S THEORY was based on another variable, the
obliquity of the plane of the ecliptic which, at .present, is 23!0.
The amount by which this has varied is not agreed upon; Stock-
well and Milankovitch give the figure as 22° as a minimum and
24tO as a maximum, Lagrange gives 2Io and 28°, but Drayson
required n° and 35°-an amount no astronomer'is prepared to
allow. It is claimed that at maximum obliquity the polar regions,
large areas of which would experience the long polar night,
would be glaciated. Like other mathematically controlled
theories it demands regular,. recurrent glaciations, the interval
being a:bout 20,000 years and the last retreat being placed 7,000
years ago. The facts, however, are otherwise.
27 0 CLIMATOLOGY
TYNDALL, CHAMBERLAIN, HUMPHREY, FRECH AND OTHERS.
There are a group of theories which seek to explain Ice Ages
and other clima~ic changes by variation in the amount of carbon
dioxide, volcanic dust and other impurities in the air, which,
by checking outward radiation, help to maintain the earth's
temperature. Careful inquiry and experiment show that they are
quite inadequate alone to produce the known effects; furthermore,
Ice Ages do not follow great volcanic outbursts such as might
increase the carbon dioxide or volcanic dust content of the air.
THE SUNSPOT CYCLE. Huntington and Visher have elabor-
ated a theory which is partly based on the variation in intensity
of solar radiation with the development of sunspots. It is a
Gurious paradox that sunspot maxima, at which time the sun's
energy is greatest, are correlated with low temperatures on the
earth. They are also correlated with an intensification of pres-
sure gradients (the permanent highs are higher and the permanent
lows are deeper), a greater storminess, especially in certain belts,
and a slight redistribution of rainfall. It is probably the greater
storminess which accounts for the lowering of temperature, heat
being lost by convection and ventilation. The sunspot cycle
has a period of about I I years, a figure which emerges in many
meteorological phenomena, e.g. in the oscillations in level of the
African lakes, in the spacing of the annual rings in the Californian
sequoias and in the annuallaminre in the varve clays of North
America during the ice retreat at the end of the last glacial
period.
THE BRUCKNER CYCLE. There are other frequencies which
emerge from time to time in an empirical study of the periodicity
of phenomena, but whose causation remains enigmatical. The
best known of these is the 35-year or Bruckner Cycle, found by
Bruckner to recur, albeit very irregularly, in a variety of pheno-
mena including the advance and retreat of the Alpine glaciers,
variations in the dates of opening and closing of Russian rivers,
the level of the Caspian Sea and the rivers emptying into it,
the date of the grape harvest and the price of grain. There
would. appear to be irregular alternations of cold, wet periods
with hot, dry periods especially in the Continental Cool-Temperate
climates, which might be regarded as fluctuations in the distance
of penetration of marine influence. In hot, dry years the marine
influence is restricted to a narrow western zone and the con-
tinental type of climate spreads westwards, rain decreases by
about 20 per cent. in Germany and 30 per cent. in Russia. Con-
sequently harvests suffer in the E~st (where normal rainfall is
on the margin of sufficiency) and benefit in the West (where'rainfall
is usually excessive and sunshine deficien~). There is generally
CHANGES OF CLIMATE 27 I
a small lag of dry seasons after warm and of warm seasons after
cold, as the following figures show:-
Warm, I746-I755, I79I-I805, I82r-r835, r~5r-r870.
Dry, r756-r770, 178r-r805, r826-r840, 1856-r870. '
Cold, I731-1745, 1756-r790, I806-r820, I836-r850, I871-
1885·
Wet;1736-r755,r77r-1780,1806-r825,184I-r855,187r-r885·
KREIGHGAUER AND WEGENER. The most difficult single fact
to explain in all the history of climate is the glaciation of tropical
latitudes in Permo-Carboniferous times. A courageous attempt
to solve this was made by Kreichgauer, who supposed that the
position of the poles had varied through geological times. The
same idea is incorporated by Wegener in his ingenious theory
of Continental Drift and applied by Koppen and Wegener to
the climates of the past. The theory is carefully worked' out
and succeeds in explaining many hitherto unreconcilable anomalies,
of past climates and the past distribution of plants and animals.
Briefly the conception is one of rigid blocks of continental material
floating in a more or less fluid substratum and free to move
westwards and equatorwards in obedience to forces which, it
must be confessed, are generally agreed' to be inadequate to
overcome the resistance. The land masses of the world are
conceived to have been more or less united in Carboniferous
times into one large continent, Pangaea,"which contained the
pole of the period in what is now South Africa, but with which
Brazil, India and Western Australia were in close juxtaposition.
Even this compression is inadequate to bring together within
reasonable compass the great expanses in these areas which were
glaciated at the time and it is further suggested that the pole
migrated slowly south-eastwards from Brazil to Australia be-
tween the lower Carboniferous and the Permian, and that the
glaciated area and presumably all the climatic belts migrated
slowly in company with it (see Fig. 66). Fig. 66 reproduces
Wegener's diagram summarizing the evidence for the distribution
of climatic zones at this time; a concentric arrangement round
a pole in South Africa is clearly brought out, the equatorialforest,
represented by the coal measures, lying along a great circle exactly
90° from the Pole and extending through North America, across
Europe and Asia to China.
Subsequently rifts appeared in this great continental mass
of Pangaea, the component parts spreading and opening fanwise
to the south to allow the birth of the Indian and Atlantic oceans.
The palreoclimatology of. other periods is worked out with
equal ingenuity, but it is impossible here to enter into a description
of them. At first sight the theory appears to supply the k~y to
27 2 CLIMATOLOGY
all past climates, but on closer inquiry difficulties appear, many of
which are so grave as to compel postponement of its acceptance.
The Quarternary glaciation, for example, is a very recent episode,
yet Wegener requires the post-glacial separation of North-west
Europe and North America, thus demanding a westward migra-

o=
=
Glaciation representing Polar Climates
Coals and Lignite representing Sub· Polar
Peat· Bogs
Rock·Salt
Deserts
and Gypsum representing .

Coal Measures representing Tropical Rain


1- , Glossopteris' Flora representing Tundra"
Vegetation
Forest

/\ Wood with Annual Rings representing ~ Migration of the Pole


Taiga and Temperate Forest
Figures illllicate Age of Deposits
I. Lower Carboniferous
2. Upper Carboniferous
3. Permian

FIG; 66.-Evidence for Climatic Zones in Permo-Carboniferous Times


(Afle, Wegener)

tion of America at a phenomenal rate. Again, the 'squan-


tum tillite' in the Carboniferous of Boston, Mass., a boulder-clay
of obvious glacial origin, associa ted with ' varve ' clays (see p. 282),
occurs in close proximity, both in space and time, to coal-beds
which are considered to represent tlte tropical forest. Again, the
British Isles during the Cretaceous "are placed in latitude' 20° ~.,
but the chaIkin our islands contains far-travelled erratic pebbles,
CHANGES OF CLIMATE 273
indicating· transport by shore ice or icebergs and therefore the
proximity of glaciers reaching sea-level. .
RELIEF AND GLACIATION. It must be remembered that
altitude, as we.ll as latitude, brings about a fall" in temperature
and that the sequence of climatic zones up a slope is similar in
many respects to that along a meridian. Altitude has been
invoked to explain the Permo-Carboniferous glaciation of the
tropics, but it is difficult to believe that such a huge area can have
risen above the snow-line which must have lain at least 6,000 feet
up if conditions were in any way like those of to-·day. The alti-
tude of the land may well have been one of the contributory
causes of glaciation [it can be shown, for example, by means of a
buried channel of the St. Lawrence that the Great Lakes region
of North America stood some 2,000 feet above its present altitude
(related to sea-level) during the Pleistocene], but it can hardly
be held accountable for glaciation on a widespread scale.
CONTINENTALITY AND CLIMATIC CHANGES. But the indirect·
results of high relief are more far~reaching and may have a pro-
found effect on areas far removed by modifying the movements of
air and of ocean currents-factors which are powerful enough to
overcome the effects of latitude as the' Gulf of Warmth' demon-
strates in the North Atlantic. It is a significant fact, to be dealt
with in more detail below, that glacial periods are invariably
associated with strong relief in geological history, and Ramsay
has carefully developed this conception of. intimate relationship
between mountain-building and climate.
The heat received from the sun in the tropics is so far in excess
of that in high latitudes that the tropics would be unbearably
hot and high latitudes so cold as to be uninhabitable were it not
for the redistribution brought about by winds and currents.
Clearly any interference with the freedom of this circulation will
induce extremes of climate, while any change which encourages
the circulation and increases its efficiency will result in a more
even distribution of temperature over the earth's surface. By
analogy with the conditions in the North Atlantic and the
Barent's Sea we may imagine what would be the result of opening
wide the Bering Strait and giving free access for the warm North
Pacific Drift into the Arctic Ocean. The dimate of Alaska
would resemble that of Norway and there would be open water
in winter perhaps as far as the mouth of the Mackenzie
River. Conversely, the closing of the North Atlantic by the
emergence of the Scoto-Icelandic Rise would be followed by the
refrigeration of the Norwegian Sea and the rapid growth of ice-caps
in Norway and Sweden whicn, after all, are in the same latitude
as' South Greenland. The indirect results on pressure and winds
274 CLIMATOLOGY
would be very complicated, but not impossible to reconstruct
and the change~ of climate would be widespread and 'profound.
THE 'CHANGE of LEVEL' HYPOTHESIS. By a careful and
detailed application of such principles Brooks has given precision
to this' change of level' hypothesis which had previously suffered
from vagueness and loose treatment. His method is to work out
a basal temperature for each latitude (i.e. the temperature found
near the centre of an ocean in that latitude) and to reconstruct
the summer and winter temperatures for any given period by
adding to or subtracting from that basal temperature figures
which represent the effect of continentality. In the case of the
Pleistocene and later changes two wind systems only are con-
cerned,-the westerlies and, north of 70° N., the polar easterly
winds. In the zone of influence of the former the effect of land
to the west is to lower the winter temperature (vide Manchuria),
while the effect of land to the east is negligible (vide Norway) ; in
'summer the effect of land either to east or west is to raise the
temperature (vide Siberia); while at any season the effect of ice
is to lower the temperature. The amount of lowering or raising
of temperature is worked out from present conditions and applied
to well-established palreogeographical reconstructions of the
distribution of land and sea; the values arrived at are checked
over with values estimated by other authors on independent
bases and show, on the whole, quite good agreement. A warning
may be sounded here that there is a risk that the degree of accu-
racy apparently obtained may be misleading, for our knowledge of
past geographies at any precise time is not very reliable.
The reconstruction of precipitation is a more complicated ?-nd
difficult matter, but much can be done by applying certain
principles, e.g. :-
I. That convectional rainfall is mainly equatorial and con-
tinental, the amount of the latter type being proportional to the
distance from the ,moisture supply.
2. Orographic rainfall occurs mainly on windward coasts, the
amount being proportional to the humidity of the air and the
relief of the coast. ~
3. Cyclonic rainfall depends pn the facilities offered for the
passage of depressions; i.e. the storms are generated over oceans,
pass eastward, following water areas and low land, die out over
. land and avoid anticyclonic areas.
Each of the above hypotheses seeks to show how the variation
of one or more climatic factors might bring about variations in
the climatic elements, culminating in Ice Ages. None of them
can be rejected as having no influence at all, though many can
be dismissed as primary causes. We shall see later that changes
CHANGES OF CLIMATE 275
of relief and level.J)ffer the only really satisfactory explanation
of the major events, but that it is likely that some of the others
must be invoked to account for minor oscillations.

The Evolution of Climate


MILD GEOLOGICAL CLIMATES. Glaciation is the exception
rather than the rule in the climates of the past, the more normal
conditions, typified by the Silurian, the Lower Carboniferous, the
Jurassic and the Eocene, have been characterized by a mildness
and a uniformity of conditions which is surprising to us who live
in what is still a glacial epoch, though apparently on the wane.
Reef-building corals, which at present inhabit seas with a mean
annual temperature above 68°, occur fossil into latitudes where·
now the mean temperature is below 50°, where long, severe
frosts occur and where there is floating ice in ·the seas. Sub-
tropical plants such as magnolias, sequoias, etc., occur fossil in
the Eocene of the Arctic, the range of palms and cycads was very
widely extended. In short, the floristic and faunistic (and pre-
sumably the climatic) distinctions between intermediate and high
latitudes were much "less clearly marked than to-day, or, in other
words, theJemperature belts appear to have extended polewards
far beyond their present limits.
CAUSES OF MILD CLIMATES. These long periods of uniformly
mild conditions correspond to periods of peneplanation and
uniformly low relief. Denudation and deposition, working slowly
but incessantly, had reduced the mountains to stumps and base-
levelled the continents to flat and monotonous plains. The seas,
partially filled with sediment from the wastage of the land, had
spread in shallow sheets over the low flat margins of the continents.
The widespread seas allowed their equalizing influence to spread
unimpeded over the whole earth, perhaps even to the poles.
They provided a large surface for evaporation, the air was humid
and the high water vapour content minimized the escape of heat
out into space, especially in high latitudes whose atmosphere at
the present day can hold very little because of the cold. The low
relief offered the minimum surface for radiation (mountains are
very wasteful of heat in this respect) and the absence of mountain
barriers permitted moisture-laden currents to enter far into the
interior of the continents, which were small and low, the wide
sluggish rivers and swampy flat plains constantly renewing their
humidity. The ice-sheets and glaciers of the last mountain epoch
had disappeared, gradually retreating as the relief ,was lowered,
and their chilling influence 'Was withdrawn. The sea-bottom
temperatures were higher and there was little cold upwelling
CLIMATOLOGY
along the leeward shores of land masses; warm water, on account
of its lightness, covered the surface of the ocean and there were no
cold surface currents, such as the Labrador ~urrent, which depend
for their persistence as surface currents to-day on the ~ontinual
supply of fresh (and therefore light) cold water from the melting of
ice. In the absence of polar ice-caps the polar anticyclones had
ceased to exist and the warm westerly winds extended to the
poles. With the vanishing of the polar anticyclones passed the
, polar front' and with it most of the storms of this now turbulent
zone.
GLACIAL CLIMATES. This placid and uneventful regime was
brought to an end by the next orogenic revolution. The land.
began to emerge from the sea, hollows were formed in the sea-bed
into which the waters withdrew, the sea shrank back laying bare
the. margins of the lands. The air became drier, the climate
more continental. New land masses began to interfere with the
. circulation of the oceans, new mountain chains excluded the
moderating influence of the sea from the hearts of the continents,
where deserts came into existence. The polar regions, robbed of
warming currlmts, grew steadily colder, ice began to form in
winter, from year to year it appeared earlier and lingered later,
finally it survived the summer and ice-caps formed and grew.
For their nourishment water was withdrawn from the oceans
whose level fell further and whose influence was further restricted
As the poles grew colder the temperature gradient was increased
between them and the Equator, the polar anticyclone came into
existence, the ' polar front' was born and storminess increased ..
Floating icebergs carried arctic cold into lower and lower latitudes,
the melting of the glaciers a-nd ice-caps in high latitudes yielded a
constant supply of cold water which slowly sank and crept along
the ocean floor to the Equator, chilling the rest of the ocean by
contact and mixture. '
Now an ice-cap, once formed, is difficult to dislodge, and the
larger it grows the more firmly is it established. Its white surface
reflects much of the solar radiation and in warm weather a mist
tends to form which protects it from the sun's rays. As it grows
larger it develops its own anticyclone which forces Gyclonic storms
to pass along its equatorward margin, supplying it with fresh
moisture and helping it to grow by marginal accretion. As the
anticyclone extends, the polar limit of the westerlies is driven
nearer and nearer to the limit of the trades, the pressure gradient
becomes steeper, and storminess increases further.
CRISES OF EVOLUTION .. Thus from a single cause there follow
a great variety of results, which, "in turn, set in motio.n other
causes, all of which operate in the same direction. The new relief
CHANGES OF CLIMATE 277
calls into existence a variety of climatic types, often sharply
demarcated by mountain boundaries; new climatic environments
are created and to the stimulus of the new demands nature re-
sponds with marvellous adaptability. New species arise, narrowly
confined within climatic zones, there is a wholesale extinction of
unadaptable forms. These are the crises of evolution out of
which Rave arisen new virile groups, stimulated by hardship and
triumphing over difficulties, to become the dominant stock of the
following age. Out of the Hercynian revolution and the climatic
stress that followed arose the reptiles who dominated the Mesozoic
land fauna, from the Alpine revolution and the climatic changes
that followed man received the impulse that has brought him to
his present state.
FREQUENCY OF GLACIATION. The geological record shows,
that there have been at least four periods in the earth's history
when true glacial conditions existed and these follow close after
four of the greatest mountain-building eras. These may be .
dated as follows, the earlier ones by means of the' radio-active
clock', the last by other more accurate means (see p, 282) :-
Quarternary, 700,000,-20,009 years ago, following the Alpine
orogeny.
Permo-Carboniferous, 260,000,000 years ago, following the
Hercyntan orogeny.
Lower Cambrian, 500,000,000 years ago, following the Late
Proterozoic orogeny.
Lower Proterozoic, 750,000,000 years ago, following the Early
Proterozoic orogeny.
Thus there is a periodicity of about 250,000,000 years govern-
ing these major revolutions, a cause for which has been suggested
by Joly and Holmes as being due to the periodic melting and
resolidification of a basaltic substratum depending upon the
accumulation of heat generated by the disintegration of radio-
active minerals. In addition to these there have been minor
revolutions each followed by minor refrigerations, but not, so far
as is known, culminating in Ice Ages. It may be significant that
the most important of these, the Caledonian (Silurian to Devonian)
and the Laramide (Cretaceous) gave birth to mountain ranges·
oriented more or less meridionally and therefore interfering less
with the circulation between Equator and poles.
It is at first difficult to see why there should be such a consider-
able lag of the cold climates after the orogenic maxima; the
Alpine maximum occurred in the Miocene, yet warm climates
persisted through the early Pliocene and glaciation did not reach
a maxiplUm until the Pleistocene. But the maximum of relief
dC1es not coincide with the maximum orogeny; isostatic read-
18
CLIMATOLOGY
justments result in steady uplift for some time afterwards, and
there are other factors, as we shall see later, which cause a lag
of the glaciation behind the highest relief.
TERTIARY CLIMATES. The early Tertiary illustrates the
conditions that have already been described as ' mild geological
climates'. Fig. 67 shows the distribution of land and sea in
Eocene times, from which it is seen that there are four wide and
independent lines of sea communication between the Arctic and

FIG. 67.-Land and Sea during the Eocene

the seas to the south through which there was a free exchange of
water between the Tropics and the Arctic. Then came the
Alpine crustal movements, culminating in the ~iocene, but
followed by a long period of readjustments and vertical move-
ments, the highest relief being probably reached early in the
Pleistocene, since when erosion has probably succeeded in lower-
ing it to some extent. The resultant refrigeration begins .to be
noticeable in the Later Pliocene rocks of East Anglia,towards the

FIG. 68.-Land and Sea at Maximum


Glaciation
More important ice-caps shown'shaded

end of which period a prevailing north-easterly wind banked up


shells against the coast. This north-east wind may be the first
indication of a clockwise swirl round the glacial anticyclone
forming over Scandinavia. Fig. 68 shows the adhnced stages of
the process of elevation which finally established the Ice Age. .
GLACIAL AND INTERGLACIAL PERIODS.. At this point it may
be mentioned that the Pleistocene glaciation was by no means a
simple and single event, nor, it maiY be argued by analo&,y, were
previous glaciations.. There is abundant evidence that there
CHANGES OF CLIMATE 279
were' interglacial' periods within the Ice Age when temporarily
the climate improved and at times was milder than at the present
day. The evidence for this consists of :-
1. Alternations of glacial with non-glacial deposits, e.g.
boulder-clay with fluvio-glacial gravels or even with sediments
laid down in a fairly ,warm sea.
2. Alternations of arctic and temperate floras and faunas.
The typical glacial fauna of North-west Europe includes the
mammoth, the reindeer, the musk-ox, the woolly rhinoceros, the
variable hare, the Arctic lemming, etc., the typical interglacial
fauna includes the cave-lion, cave-hyena, the straight-tusked
elephant, etc.
3. Oscillations in the position of the jce front with important
r~sults in river drainage; advance and retreat of glaciers. '.
4. Alternation of glacial and fluviatile relief types in· the
evolution of valley profiles and other scenic forms.
Penck and Bruckner established the succession of four glacial'
periods in the Alps separated by three interglacials, but these
cannot always be recognized on the lower ground. In Britain
and on the North German plain only two glacial advances are
recognizable, represented by the Older and the Newer JJ.rift. In
North America five are accepted, generally correlated with the
Alps as follows:-
Alps. North Amenca. Date B.C.
Gunz. Nebraskan or Jerseyan
Mindel Kansan 430-37 0 ,000
Riss . IllinOian} 130-100,000
{ Iowan
Wurm Wisconsin 40- 18,000

CAUSES OF INTERGLACIAL PERIODS. To understand these


oscillations of climate it is necessary to examine carefully the
sequence of events which may be expected in an area covered by
an ice-cap and in aCIjacent areas. We have seen (p. 276) that as
the ice-cap grows by the congealing of precipitation the level of the
sea sinks and that of the land rises relatively; on a conservative
estimate of the dimensions of the ice-sheets this depression of
the sea-level has been estimated at about 400 feet. After a time,
however, the growth of the ice-cap constitutes 'an excess load qn
the surface and the glaciated area begins to sink. If isostatic
compensation were perfect the amount of depression would 1?e
about one-third of the thickness of the ice ac;cumulated (for the
ice has about one-third of the density of the substratum), but
because of the rigidity of the crust the compensation is not com-
plete aFld there is a considerab1e lag behind the theoretical amount.
The result of this sinking of the ice-caps was to bring them below
280 CLIMATOLOGY
the level of maximum precipitation and in some cases to bring
them below the snow-line so that ablation began to gain on accu-
mulation. Fur!hermore, the depression of the whole .area per-
mitted a partial return to freer circulation of currents and winds,
an effect which was rapidly increased when the melting of the
ice-caps returned water to the oceans and raised the sea-level.
There was, in fact, a temporary return to non-glacial conditions.
But the depressed areas, relieved of their load of ice, began to rise
again, eventually rising above the snow-line, accumulation of
snow began to exceed ablation and glacial conditions returned.
Thus as a result of the lag of compensation there would be oscilla-
tions on either side of the stable condition before equilibrium was
~eached. But it is not, of course, a perpetual motion cycle, the
amplitude of the oscillation must be smaller each time and sta-
bility must eventually result unless and until factors external
. to this process undergo a change.

The Climatic Belts at the Time of Maximum Glaciation



The glacial maxima were by no means of equal severity, nor
were the climatic conditions within or without the glaciated area
the same during each, but with each advance there occurred cer-
tain migrations of climatic belts, of which the following pages give
a generalized account.
EUROPE. Along the margins of the ice the climate. was of the
tundra type, though possibly the summers were warmer than in
the tundra of to-day. Survivors of this tundra flora are found
to-day at the higher levels in Scotland, Scandinavia and the Alps,
whither they retreated as the climate improved. The tundra
zone did not occupy all the European plain, in the east there was
a steppe climate of an extreme continental type dominated by the
east winds which prevailed on the southern edge of the glacial
anticyclone. Steppe animals and plants have been found fossil
as far· west as France, from which it appears that the influence of
the western ocean was not felt far inland.
PLUVIAL PERIODS IN THE GREAT BASIN. The southward
spread of the ice-sheets was accompanied by an extension of the
polar anticyclones with their prevailing east winds and the sphere
of influence of the westerlies with their cyclonic storms retreated
before their advance. Hence on the equatorward margin of their
present limit pluvial periods correspond to glacial advances.
The Great Basin received a much increased rainfall and large
lakes formed against the boundary scarps; the Great S<ilt Lake
represents the dwindled survivor of a once extensive lake; kno~n
as Lake Bonneville, whose shore line may still be seen about 1,000
CHANGES OF CLIMATE 281

feet above' the present level of the lake. Numerous small salt
lakes in Western Nevada, Humboldt, Pyramid, Walker, Winne-
mucca, Honey, and the Carson lakes are the remnants of another
known as Lake Lahontan.
PLUVIAL PERIODS IN NORTH AFRICA. The Mediterranean
became a much frequented track for cyclones and its shores were
well-watered, in summer as well as in" winter. Even in the
Northern Sahara: pluvial periods were contemporary with the
glacial periods in higl).er latitudes. Where now there are dry wadis,
or short streams losing themselves in the sand, there were then
continuous rivers who.se graded courses (in contrast with the
ungradell wadis) and normal fluviatile form with interlocking
spurs may still be recognized.
THE BIRTH OF THE NILE. To-day northerly winds prevail over
Egypt for most of the year, but above them are westerly ~nds
which may be met on Mount Sinai. During the pluvial periods
these westerly winds' were the surface winds and brought heavy
rain to the highlands which lie to the east of the Nile. Down
these there flowed torrential rivers which carried volumes of silt
and debris across what is now the valley of the Nile. But the
Nile did not exist then. At the present day it succeeds in crossing
the desert mainly by the aid of the Blue Nile flood waters from the
Abyssinian monsoon. The monsoon depends on the heating of
the Asiatic continent, but during glacial times a colder Asia
attracted a less powerful monsoon and Abyssinia had dry trade
winds. The White Nile lost itself in the desert and probably did
not succeed in crossing it to the Mediterranean until post-glacial
times, about I4,000 years ago, when the monsoon began to reach
the Abyssinian ·mountains. .
CHINA. But if the summer monsoon was weak the winter
monsoon was doubly strong, since the increased cold added to
the intensity of the continental high. Thus in China the glacial
periods of North-west Europe are represented by periods of
increased aridity and periods of heavy :eolian aggradation, when
the continental influence was redoubled and the winter winds
transported still greater quantities of loess from the arid interior
on to the plains.
THE SUDAN. On the equatorward side' of the trade-wind
deserts the climatic belts appear also to have shrunk towards the
Equator, so that here the desert gained on the savanna. Lake
Chad, for example, although it was once much larger, was also
once much smaller. The shrinkage was probably' glacial', the
expansion' interglacial' ; at the present day the lake is increas-
ing in £iize and in freshness-~ post-glacial recovery towards more
pluvial conditions. Similarly Lake Titicaca, occupying an
282 CLIMATOLOGY
analogous position in 16° S., is at present increasing in size and
drowning the valleys of the rivers that empty into it. There are
, fossil ergs', or'vegetation-covered dunes from Lake Chad, along
the upper course of the Niger, to the sea in Senegal, which testify
to a southward encroachment of the desert, presumably during
glacial maxima, and a northward post-glacial retreat.
THE EQUATORIAL REGIONS. The concentration of the pres-
sure belts doubtless strengthened the circulation of the trade
winds and gave rise to a deep and stormy equatorial low with
increased rainfall. The equatorial la]_(es of Africa swelled to
twice their present size, Lake Victoria was continuous with Lake
Kioga whose level was 600 feet above the present. The glaciers
on Ruwenzori and Kilimanjaro descended by 8,000 to 9,000 feet
to within 5,000 feet of sea-level, but the snow-line only descended
abol;lt 3,000 feet, a clear indication that the cause was increased
rainfall and not decreased temperature.

The Climatic Belts durin~ Inter~lacial Periods


During interglacial periods the climate reverted towards con-
ditions more or less similar to those of to-day, but in Central
Europe and North America the effect of the glacial anticyclones
to the north was felt as strong cold dry winds and an unpleasant
variety of steppe climate prevailed. These strong winds carried
clouds of dust derived partly from the moraines uncovered by
the retreat, partly from silt beds left by floods in basins of inland
drainage during the preceding glacial period. This dust now
forms the thick fertile cover of loess so plentifully distributed
about the limits of glaciation. During the Riss-Wurm inter-
glacial loess formation occurred on a large scale in Central and
Eastern Europe, and during the corresponding Iowan-Wisconsin
interglacial in the plains of the Upper Mississippi.

Climates durin~ the Retreat of the Ice


CHRONOLOGY. The dating of events in glacial and post-
glacial time is not always very reliable. Dates are based mainly
on the rate of growth of peat bogs and deltas and on the degree
of weathering of deposits, but from the last retreat onwards
we have an accurate method of computation by the actual
counting of the seasons represented by alternate layers of coarser
and finer laminre in the' varve' clays of Lake Ragunda in the
south of Sweden. The lake was drained in 1796, so that the top
layer of silt represents the sediments laid down in that year.
The coarser of the alternating layers beneath are assumed to
CHANGES OF CLIMATE
correspond to the summer floods while the glaciers which dis-
charged into the lake were rapidly melting, th~ finer layers to
winter conditions when melting was slower or ceased. The same
method has been employed in other areas, notably by Antevs in
North America.
By these means De G~er has estimated that by IO,OOO B.C.
the ice had left the Baltic and reached the present coast of
Sweden. If this is so, then the retreat has been surprisingly"
rapid-too rapid, in fact, to find universal acceptance, bu~ De
Geer attributes it to the sudden invasion of the Norwegian Sea
by the warm waters of the Gulf Stream.Drift. Certainly during
the latter part of the retreat there was an open sea, the' Yoldz'a
Sea,' across the Baltic lands. II) the comparatively short period
since then there have been -numerous changes of level of land
and sea in North-west Europe, frequent openings and closings of
the connections between the Baltic and the North Sea and
between the Baltic and the Arctic. Inevitably these brought
about oscillations of climate which have not ceased to the present
day; during periods of high relief, when marine connections
were severed, the climate of North-west Europe was severely
continental, during periods of submergence the climate was
milder, wetter and, in brief, more marine.
THE CONTINENTAL PHASE. It has been shown that the
rapidity of the retreat was related to the removal of the barrier
between the open ocean and the Baltic, but now (about 6000 B.C.)
an elevation of the southern shores of the Baltic restored the
barrier and gave rise to an enclosed lake, the' Ancylus lake'.
Marine influence was excluded and Central Europe, swept by
dry easterly winds, had a strongly continental steppe climate"
such as is not found to-day west of the Urals. The rainfall was
less than 20 inches. The Scandinavian anticyclone fended off
the cyclonic storms which were concentrated along the western
margin or else passed along the Mediterranean. Thus Britain,
especially Scotland and Ireland, actually had a heavier rainfall
than to-day with a damp, chilly climate in which peat bogs grew
(the 'Early Peat-bog Phase '). Elsewhere was a pervading
aridity from which the people of Asia and Eastern Europe escaped
westwards towards the better-watered lands. This is the Neo-
lithic invasion which put an end to the artistic palceolithic culture
of Western Europe. The' Nordics' came from Central Asia,
sweeping westwards to the shores of the Baltic, the ' Alpines'
fQllowed a more southerly course into France and to the shores
of the Atlantic. '
TfI.E MARITIME PHASE. • Meanwhile the melting of the ice-
caps was gradually raising the mean sea-level and about 4000 B.C.
CLIMATOLOGY
an important change came about:, namely the invasion of the
Ancylus lake by warm salt Atlantic water through the Sound
and the Belts. An open sea was produced, the' Littorina Sea "
by which marine influence penetrated to the furthest ends of
the Baltic and the Gulf of Finland. Everywhere the level of
the sea appears to have stood higher than to-day, as is shown
by raised beaches in Greenland, Spitzbergen, Franz Josef Land,
Norway and Scotland (the 2s-feet raised beach) ; and the climate
was milder, the mean annual temperature of the British Isles
being probably 3° or 4° higher than at present. Cyclones began
to penetrate along the warm waters of the Littorina Sea, bringing
much increased rainfall and much milder conditions to the steppe
lands of Russia; the volume of the rivers swelled and the lakes
(C!lspian and Aral) filled. The steppe ?-wellers must have pros-

Qland .~
_J"
t!11llllr StonntraCks

FIG. 6g.-Climatic Conditions during. FIG. 70.-Climatic Conditions during


the Continental Phase the Maritime Phase

pered and multiplied on the rich pastures. which resulted. This


period is often referred to as the 'Post-Glacial Optimum of
Climate', and, in fact, such mildness has never been repe?-ted
since,. but Western Europe, and especially the British Isles,
suffered from excessive rain, and from the point of view of human
settlement the following period was more favourable here. .
THE FOREST PERIOD. About 3000 B.C. an elevation of Europe
along the central axis caused a restriction of the seas here while
the Baltic remained depressed. Much of the Irish Sea and the
North Sea were dry land and El].gland was joined to France
where now are the English Channel and the Straits of Dover.
Cyclones began to follow the more northerly track across the
open waters of the Baltic and instead of, as now, crossing Eurasia
by the Black Sea and the Siberian ?owlands, entered the warmer
Arcti,c by way of the White Sea. The sub-tropical high pressures
CHANGES OF CLIMATE
extended northwards and at least during the summer months
covered most of southern Britain, whose climat~ thus approxi-
mated to the Mediterranean type. Ireland, temporarily relieved
of the embarrassment of too much rain, blossomed out into the
, Heroic Age', only to decline in the succeeding moister period.
The northerly track of the cyclones robbed the steppe lands of
their rain, the rivers dwindled and the level of the Caspian fell
to a minimum about 2200 B.C. ,The desiccation of these lands
drove out barbarian hordes who overran the better-watered
lands of Mesopotamia and the Mediterranean. About 2000 B.C.
the Hyksos, or ' shepherd kings' who, as their name implies,.
were a steppe people, invaded Egypt, conquering a sedentary
people by the greater mobility given by the horse. The
Canaanites migrated out of i\.rabia into Palestine; rather lat~l'

FIG. 7I.-Climatic Conditions during FIG. 72.-Climatic Conditions during


the Forest Period the Peat-Bog Phase

(between 1600 and 1300 B.C.) the Indo-Europeans entered India


by the North-west passes, the Hittites entered Asia Minor and
the Achreans settled in Greece.
THE PEAT-BOG PHASE. By about 1000 B.C. fresh submergence
had set in and drowned the marginal forests of the previous period.
The encroachment of the sea brought on a moister climate and
carried marine influence further inland. The western. margin was
now too wet for forests and peat-bogs grt::w over their decaying
stumps. Well-used cyclone tracks carried rainfall to two im-
portant areas :-to the Steppe lands and via the Gap of Carcas-
sonne to the Mediterranean; the steppe dwellers were settled
and contented, the Mediterranean was prosperous and its states
ambitious. In fact the misfortunes of this age are attributable
to excess of rain rather than-deficiency; the Bronze Age culture
of Norway perished in the damp cold; the lake dwellings of
286 CLIMATOLOG Y
Central Europe were drQwned by the rising of the lake levels.
The only migr,!-tions of the period are from moister into drier
lands-a complete reversal of the usual direction-e.g. the Celtic
movements eastwards along the northern foot of the mountains
from the Italian Alps to the Caspian Sea. In the Mediterranean
which enjoyed adequate rainfall and a stimulating cycloni~
climate, there grew up the great civilizations of Greece, Rome
and Carthage, for which the cultivation of grain supplied a sound
agricultural basis. Carthage found water supply a greater diffi-
culty than the other two and was compelled to import water
by aqueducts. The springs which fed these no longer flow and
Carthage could never have prospered as she did under the climatic
conditions of to-day.
DRY PERIOD, 200 B.C. The climatic arrangement outlined
abo.ve endured, and with it the prosperity high culture of the
Mediterranean peoples, until the third century A.D., but there
. was a temporary decrease of rainfall about 200 B.C. which brought
about an agricultural crisis in Italy and severely tested the
stability of Rome. This dry period was felt also in the steppe
and initiated an outward movement. Against this threat of
invasion the settled agricultural civilization of China built the
Great Wall, but without avail, the' hunger urge' was stronger
than the wall and China was overrun again and again.
WET PERIOD, roo B.C. The dry period in the Mediterranean
was only a temporary episode. A hundred years later agricul-
ture had revived (though it is true the basis of rural economy
was not grain but the more xerophylous olive) and Rome had
entered on a period of prosperity with wealth and luxury greater
than before. The journal kept by Claudius Ptolemaeus in the
first century A.D. at Alexandria throws an interesting light on
contemporary climatology. The record, describes a high per-
centage of south and west winds in summer instead of the
monotonous north winds of the present; the weather was more
changeable and thunder was more frequent. Altogether the
records indicate that the cyclonic westerly circulation was a
much more important element in the climate of Alexandria
than it is at the present day .
. DRY PERIOD, 200-1200 A.D. The revival of Rome lasted
only about 200 years and then began a period of renewed stress.
There was a further decay of agriculture, a shortage of corn
and a migration of rural population into the towns; the rivers
began to dry up, stagnant water supplied a breeding-ground for
mosquitoes and malaria became endemic, sapping the. vitality
of the people. The steppe tribes, • suffering also from drought,
became restless; by the end of the first century they were on
CHANGES OF CLIMATE'
the DanJlbe and Trajan was forced to conquer and annex Dacia
to secure the Danube as a frontier. But there was no stemming
the flood, Greece was invaded, Athens fell and then Rome. Orit
of Arabia came the Mohammedan wave of conquest, actually a
religious crusade, but indirectly encouraged by years of drought
and want. ,
Further north, too, the climate was drying; in Britain the
peat bQgs ceased to grow and forests took their place. The
climate here became more favourable for human occupation,
but the outer waves of the disturbance propagated from the
steppe prevented the establishment of settled order. The Huns
pressed against the Goths, who pressed against the Germanic
peoples who escaped westwards and invaded these islands. The
drought of the steppes is pr~>ved by the level of the Caspian Sea;,
In the fifth century the Red Wall was built as a bulwark against
the Huns, to-day th~t wall extends 18 miles into the lake below
the water.
Further south there is also evidence for the poleward shrink-
age of the climatic belts. The Mayan civilization of Yucata~
reached its zenith early in the third century and then declined,
overwhelmed, it is suggested, by the advance of the tropical forest
and the insidious attack of the monGtonously hot damp climate,
• Northern Yucatan apparently lay beyond the invasion of the
forest and preserved some vestige of its culture, but the collapse
in the south was complete.
The dry warm phase persisted into tp,e tenth and eleventh
centuries; Europe was il).vaded by Magyar horsemen and India
by the Mongols. The climate of Northern Europe was so mild that
Norway was able to plant and maintain colonies in Greenland
where cattle thrived and com was grown with success.
THE MEDIAEVAL WET PERIOD. The climate swung back in
the thirteenth century towards moister and colder conditions.
The glories of the Mediterranean were in some degree revived
by the city states of Italy and by the high Moslem culture in
Spain; the level of th~ Caspian rose until at the beginning of
the fourteenth century it stood nearly 14 feet above the level of
to-day; the steppes were well watered and its people were
settled and contented. But the gain of, the steppe and the
Mediterranean was the loss of North-west Europe which now
entered on a wet and stormy cycle. In the thirteenth century
the coastal defences of Holland were frequently breached and
the polders flooded. Sea-traffic was discouraged, Norway lost
touch with her Greenland colonies which were gradually frozen
to death .
. THE LAST CENTURIES. In the sixteenth century, the Eliza-
288 CLIMATOLOGY
bethan era of expansion and the age of maritime activity, the
curve had swul!g back to a drier, warmer and less stormy period
in North-west Europe and continued in that direction until the
end of the eighteenth century, since when there appears to have
been but little change.
THE ATTAINMENT OF EQUILIBRIUM. In the process of recovery
from the glacial maximum the amplitude of the climatic oscilla-
tions has grown progressively less; the earlier changes are beyond
reasonable question, but the later changes are so slight as to
be within the limits of the present yearly variability. There
are no records of weather after the Christian era which might
not apply to a single, perhaps very abnormal, year of the present
time. Thus allegations of climatic changes in recent times can
·generally be, and often are, refuted on these grounds. Yet it
seems unreasonable to deny that small oscillations may still be
occurring and the balance of evidence and the course of history
.at least lend some support to this hypothesis.

Civilization and Climate


In the outline account of the evolution of climate given in
the foregoing pages attention has been focused particularly on
North-west Europe. Similar stages might be d,escribed for North
America, the record of whose glacial climates is fairly complete;
but of the history of peoples and of their migrations in the New
World we know but little. In Europe and Asia, on the other
hand, thanks to many years of patient research into history
and pre-history, a wealth of material has been collected, though
there still remain many serious gaps to be filled. The Mediter-
ranean region is especially valuable because it has been con-
tinuously i1;lhabited from the beginning of civilization by cultured
peoples who have bequeathed fairly complete records. Further-
more., its transitional climate makes it peculiarly sensitive to
migrations of the climatic belts; it lies perilously near the limit
of sufficiency of rain, and if that limit is overstepped disaster
results.
From. the study of historical changes of climate certain
interesting relationships result. The steppes are the sensitive
trigger that fires the charge whose echoes are heard over a whole
continent. Even more than the Mediterranean the steppe is a
region precariously near the limit of rainfall sufficiency; a
decrease from 12 inches a year to 10' inches a year may mean
a decrease in the supporting power from 100 head of sheep to
10 head. When rainfall decreaseg> pasture must fail and man
must go elsewhere or perish. Fortunately for, him his mode 'of
CHANGES OF CLIMATE
life on the steppe makes that departure easy, he has few posses-
sions and no home ties. Outward he spreads, p~opelling others
before him, until the settled peoples of the well-watered agricul-
tural lands beyond feel the shock and perhaps go down before
it. He carries with him his language and culture which mingle
with those' with which he comes into contact. It may be that
the Indo-~uropean languages that extend from Western Europe
to India and from India to the North Cape desseminated
outwards from the European grasslands, while the Semitic
languages, south of the Alpine barrier, spread outwards from
the grasslands of Arabia.
The record of these movements seems to show a period of
about 690 or 700 years-300 years of unrest followed by 300
years of slow recovery-during which new civilizations emerge,
from the wreckage of the, old but rejuvenated by the infu?ioh
of vigorous barbarian blood. Looking down the corridor of
time we notice that as each disturbance settles down the seat.
of new power lies IlOrthward of the old. We are watching a
persistent, though oscillating, recovery from the Great Ice Age
which thrust plants, animals and man southwards before the
advancing cold. Several writers have stressed the importance,
of the cyclonic climate with its stimulating variety, its seasonal
rhythm and its insistence on constant and unrelaxing effort.
It would appear that power has followed the cyclonic belt in
its northward retreat.

SUGGESTIONS FOR FURTHER "READING


The literature on the subject of climatic change is immense, but two
books published recently by C. E. P. Brook's provide valuable summaries : -
The Evolution of Climate, in 1925, and Climate through the Ages, in 1926.
British Floods and Droughts, 1928, by the same author in collaboration
with J. Glasspole, is a treatment of the somewhat elusive material indi-
cated by the title. Geological climates are dealt with by C. Schuchert
in Climates of Geological Time, Carneg. lnst. of Wash., Pub. 102, 1914;
A. P. Coleman's Ice Ages, 1926, gives an account of these phenomena for
geological times, while the last of them is carefully dealt with by W. B.
Wright in The Quarternary Ice Age, 1914, and by E. Antevs in The Last
Glaciation, 1928, NO.7 of the publications of the Amer. Geog. So.c. Research
Series; and in Quarternary Climates, Carneg. lnst. of Wash., Pub. 352,
containing papers by J. C. Jones, E. Antevs' and E. Huntington. A
readable account of the evidence for climatic change in Central Asia is
given by Ellsworth Huntington in The Pulse of Asia, 1907, while his 'solar
cyclonic hypothesis' is treated, together with other interesting material,
in Climatic Changes, 1922, in collaboration with S. S. Visher. The
, Continental Drift Theory' as affecting past climates is dealt with by
W. 'Koppen and A. Wegener in Die Klimate der Geologischen Vorzeit,
1924. 'Ramsay's theory is most accessible in the Geol. Mag., 1924.
The following articles, etc., may be consulted : -
CLIMATOLOGY
Antevs: • Late Glacial and Post-glacial History of the Baltic', G.R., 1922.
Bishop: • The Geographical Factor in the Development of Chinese Civil-
ization " G.R., 1922.
Bovill: • Desiccation of North Africa in Historic Times " Antiquity, 1929.
Brooks: 'World-wide Changes of Temperature', G.R., 1916.
- - ' Secular Variations of Climate', G.R., 1921.
- - 'The Evolution of Climate in North-west Europe', Q.].R.M.S.,
1921.
- - 'Meteorological Conditions during the (Permo-Carboniferous) Glacia-
tion of the Present Tropics', Q.J.R.M.S., 1926.
Butler: 'Desert Syria, the Land of a Lost Civilization', G.R., 1920.
Coching Chu: 'Climatic Pulsations during Historic Times in China',
G.R., 1926.
Curry: 'Climate and Migrations', Antiquity, 1928.
De Geer: 'Geochronology', Antiquity, 1928.
Douglass: Climatic Cycles and Tree Growth, Vol. I, 1919; Vol. II, 1928,
. Carnegie lnst. .
Gregory (J. W.): 'Is the Earth Drying .Up ? ' G.]., 1914.
Gregory (Sir R): 'British Climate in Historic Times', G.T., 1924.
- - 'Weather Recurrences and Weather Cycles ' .. Q.].R.M.S., 1930.
'Hobley: 'The Alleged Desiccation of East Africa', G:]., 1914. .
Hume and Craig: 'The Glacical Period and Climatic Change in ~orth-
east Africa,' Rep. Brit. Ass., 19II. "
Huntington: 'Climatic Changes in America in Historic Times', S.G.M.,
1914·
- - 'Climatic Variati.ons and Economic Cycles " G.R., 1916.
Penck: • The Shifting of the Cli~atic Belts', S.G.M., 1914.
Shaw: Manual of Meteorology, II, pp. 320-325.
Simpson: 'Past Climates', Q.].R.M.S., 1927.
Taylor: 'Climatic Changes and Cycles of Evolution', G.R., 1919.
Walker: 'World Weather', Q.j.R.M.S., 1928.
Lake Bonneville, U.S. Survey Monograph 1.
Lake Lahontan, U.S. Survey Monograph XI.
~ Report of a Conference on Cycles', G.R., 1923.
INDEX
Aar Valley, 253 American Mediterranean, 10 4
Aberdeen, 204 Amoy, 168
Abilene, 174 Amur Basin, 51
Absorption Thaw, 224 'Ancylus lake', 2 8 3
Aburi,82 Andermatt, 264-
Abyssinia, 128, 135 Andes, 37, 41, 79' 101, 25 0
Acacia, 99 Anemometer, 23
Accra, 82 • Angmagsalik, 223' 23 0
Achaeans, migrations of, 285 Angot, A., 203
Actinic rays, 8 . Annam, 108, 1301 1.3 1
Addis Abbaba, 136 Annual rings, sp;lcmg of, 268, 270
Adelaide, 154, 156 Ano Nuevo, 230
Adelie Land, 222 Antarctica, 221, )!24, 225
Aden, 248 Antevs, E., 283, 2 89, 29 0
Adret, 254 Anticyclones, 27
A dretto , 254 Antofagasta, 235' 24 8
Advance of monsoon, III, II8 Aphelion, 77
Afghanistan, II2, 241 Apia, 91
African Lakes, level of, 268, 270 , Appalachians, 3 8' r67, r80
282 Apples, 154
Agra, 124, 125 Apricots, 246
Agulhas Current, 164 Arabia, 123, 128, 233
Ahaggar Highlands. 2{0. 256 Arabian Sea, 93, II.6 •
Ahmedabad, 123 Arabian Sea bra-nCh ot monsoon, r21,
A iguilles, 254 122, 123, 124, 12 5
Airways, I, 24 Arakan, 108,'12:7
Akassa, 61, 72, 81 Ararat, Australiil-, 140
Akyab, 123, 136 Arbutus, 145, 185
Alabama, 167 Archangel,2I8
Alaska, 213, 220 Architecture cIiJ)late and, 5, 244
Albany, 197 Arctic Circle: daylight at, 22
Alder, 227 Arctic climates, 66, 67, 220 et seq.
Aleutian Low, 31, 150, 192, 193, 202, Arctic hare, 227
. 207 Arequipa, 246
Alexandria, 143, 156, 238, 286 Argentina, 3, 16:7
Alfa grass, 245 Arica, 236
Algeria, 147, 148 Aridity, boundafY of, 57
Algiers. 19. 141. 156 Arkansas, 167
Alice Springs, 14, 238, 248 Arnhem Land, 13 2
Aliwal, 164 Arroyos, 243·
Allahabad, 124, 125 Artesian Wells, H5
Alpine orogeny, 277, 278 Ash, 185
Alpine pasture, 40, 145 Aspect, ClimatiC effect of, 51, 223, 254
'Alpines " 283 Assam, II6, II7' 12 3, 12 7, 128
Alps, 250 et seq. Astrakan, 248
Altitude, climatic effects of, 35 et seq., Asuncion, r03, fo 6
77, 97, 100, 250 et seq. Aswan, 239,248
Altitude, effect on vegetation, 79, 100, Asyut, 238
145, 258, 262 Atacama Desert, 49, 233, 235
Altitude .savanna, 82, 101 Athens, 61, 140, 143, 147, 14 8, 15 6
Amazon Basin, 79, 80, 103 Atherton Platea-u , 133
291
CLIMATOLOC;Y
Atmosphere, absorption of insolation Bergen, 60, 207, 208, 209, 218
by, 35253 , Berg Winds', 153, 164,236
Atmosphere, Circul~tion of, 24 et seq. Bering Sea, 203
Auchinleck, G. G., 104 Bering Strait, 48, 273
Auckland, 174, 196 , Berkeley, 151
Australia, 131, 135 Berlin, 181. 182. 183, 189.204
Australian Alps, 163 Bevers, 264
Australian Desert, 233 Bhuj. 12 3
Average Values, 4 Birch, 186, 212, 227
Awa PenjnsuJa, 20J BisJ:wp, 290
Axim,8I Biskra.I4
Ayau,51 Bismarkburg, 76
Azores' High', 148, 186, 188, 260, 261 Bjerknes, V., 29, 32
Aztecs, 98 Blackadder. 187
Azul,160 Black Hills, Dakota, 38
Black Sea, 190
,Baghdad, 248 Blacksod Point. 192
Baguio, 17, 131 Blagovestchensk, 61
.,Bahamas, 91 Blandford. H. F .• 39. 134
Bahia, I04 Blizzard. 199, 210, 214, 221. 222. 224,
Bahia Blanca, 164, 174,·243 2 63
Baker, O. E., 167 Blood. Oxygenation of. 36
.Baikal, Lake, 170 'Blossom showers'. 1 I 7
Balboa, 91 Blue Mountains (Jamaica). 91
Balkan Peninsula, 143, 190 Bodo. 209, 218 .
Ballarat, 140 Bogota, 76, 78, 281
Balmoral, Australia, 140 Bogra,39
Baluchistan, 123, 128 Bohorok,83
Bamboo, 80, 161 Boise City. 242
Banana, 47, 49, 76 Bolivia. 101, 262
Bandoeng, 85, 87 Bolobo.73
Bangkok, 130 Bombay, 18, lIS. 118, 119, 120, 121,
Bangs, N. H., 173 127. 136
Baobab,99 Bonneville. Lake. 280, 290
Barbados, 16 Bora. 40, 148
'Barbarians " 3, 285, 286, 287 Borax. 246
Bareilly, 124 Boston (Mass.). 182. 194. 204
Barley, 143, 153, 169, 217, 246, 262 Bottle trees. 99
Barnaul, 218 Bourbon. 94
Barrow Point, 230 Boumemoutlz,5 1
Bartle Frere, Mount, 133 Bovill. 290
Bass Strait, 162 Braak, C., 105
Batavia, 85, 86 , Brave \yest Winds', 177. 195
Bat Furan, 128 Brazil, 3. 37. 100. 10 3. 104
Bat Hiddan, 128 Brazilian Current. 46. 48
Bathurst, 106 Breede River, 153
Beachport, 140 Breslau, 204
Beachy Head, 51 Brest, 47. 192. 204
Beans,. 246, 262 , Brickiielders " 159
Beaufort's Scale, 24 Brisbane, 14
Beech, 145, 185 British Columbia. 49. 186, 192
Beira,48 British Guiana, 75
Beirut, 140 British Isles. 49, 186 et seq.
Belgrade, 204 Bronze Age. 285
Bellary, 122 Brooks, C. E. P .• 52, 104, 105. 155, 203,
Benares, 124, 127, 136 274. 289. 29 0
Bendigo, 140 Brooks, C. F .• 105
Bengal, 109, II6, 117, 126, 128 Broom. 145 .
Bengal, Bay of, 93. II2, II 6, 125, 128 Broome, 136
Benghazi, 148 Brown, Rudmose, 228
Benguela Current, 46, 153 Bruckner Cycle. 270, 271
Ben Nevis, 18, 187 Buenaventura, 80
INDEX 293
Buenos Aires, 174 Carthage. Civilization of, 286
Bug, River, 212 Carvalho, C. de, 104
Buitenzorg, 84 Cascade Ranges. 192
Bukarest, 177. 181. 182. 204 Caspian Sea, 50. 190
Bulawayo. 98 . - - past levels. of. 268. 270, 284.
Bulbous plants. 144 . 28 5. 28 7
Buran. 43. 210 Castries (St. Lucia), 92
Burma. 123. 1'29 Catarro. 143
., Burst of M0!lsoon " rI8, II9, 132, 171 Cauca, River, 101
Butter, 290 Caucasus, 190
Buxton, L. H. D .• 173 Cave hyena, 279
Cave lion. 279
Caatinga. 103 Cawnpore, 123
·Cacao. 94. 100 Ceara, 103
Cairns, 136 Cedar, 145
.Cairo. 238, 239. 248 Celebes, 82, '85
Calcutta. IIO. 116, 124. 'X36, 170, 171 Celtic Migrations, '286
, Caledonian' Mountains. 179 Central America, 104. 105
Caledonian orogeny. 277 Central India. II3, II4
Calgary. 214 218 Central Provinces, 21, 113. 114. 121.
Calicut. 18 125 .
California, 23, 45, 142, 150, 151, 193, 'Centres of Action', 126
235, 23 6 Cevennes. 148
Californian Current. 45. 48, 151 Ceylon. II2. 1I4, 116, 118. 123. 127,
Callao, 236, 238 128
~ Cambay. 119, 123 Chad. Lake. Changes of level. 281
Cambridge, 184 Changes of Climate, Causes of. 269 et
Campagna. 143 seq. .
Camphor trees, 201 - - - evidencefo~ 267. 268
Campos. 99. 10 3 Chamberlain's theory. 270
Canaanites. migrations Of. 285 Chamney. N. P., 104
I Canaries Current. 48 , Change of level' hypothesis, 274
Canary Islands, 141 Chaparral. 145
Canons, 244 Character, ,Climate and. 2
Canton, 112 Charaqui. 128
Cape Agulhas, 164 Charkov. 209
Cape Cambodia, 130 Charleston. 160. 165. 197
Cape Coast Castle, 82 Che-foo. 196
Cape Evans, 222 Chemulpo, 204-
Cape Guardfui, 46 Cherrapunji. 39. 109. 116. 123. 136
Cape Howe, 163 Chestnut. 169. 185
Cape Juby, 237, 24 8 - (sweet), 145
Cape Leeuwin. 141 Chicago. 182. 204
Cape Lopez, 49 Chile, 48, 152, 195, 236, 237. 238, 245
Cape Northumberland, l40 Chili,200
Cape Otway. 140, 162 China, Pleistocene climate of. 281 •
Cape San Lorenzo, 49 China Sea, 85
Cape San Rocque, 103 Chincha Islands, 246
Cape Three Points, 81, 82 Chinook, 17, 214, 252, 253
Cape Town, 14, 9 8, 139. 14 1, 143. 153, Chisholm, G. G., 164
156 • 158 Christiansborg, 82
Cape York, 18. 132 Christiansund, 208, 209
Caracas. 78. 106 Chronology, 282
Carboniferous, Climates of the, 267, Chu, Coching. 168, 173,290
27 1,275 Chung-King, 171, 174
Carcassonne. Gate of. 147 Cinchona, 79
Cardamom Hills. 122 Cincinnati, 204
Caribbean Sea, 104 Citrus fruits, 141, 146, 162. 201
Carnarvon, 248 Classification of Climates. 53 et seq.
Carnatic, 112, 127. 128 Clayoquot, 193
Carolin'a. 165. 167 Clayton, H. H., 31. 228
Cartagena. 76 Climographs. 14
19
294 CLIMATOLOGY
Cloncurry, 132, 133, 136 Cyclones, tropical, 19, 84, 92, 133,
Clothing, Climate and, 209, 239 158, 172, 200
Cloudiness, 21, 255' Cypress, 161
Clouds, absorption of insolation by, 35
Coal Measures, Climatic significance of, Dacca, 39, 1I6
267, 272 Dalmatia, 143
Coast Ranges of Canada, 192 Danube, River, 178
Cochabamba, 7~ 98 Dar-Fur, 38
Cochin, II2, II6 Darjeeling, 124, 136
Coffee, 23, 100, IOl, 245, 259 , Darling BaSin, 132
Cold Climates, 66, 206 et seq. , Darwin, 136
• Cold Wall " 21 Data of Climatology, 4
Cold Waves, 30, 43, II3, 192 Date palm, limits of, 246
Cold Weather Storms of India, II2, - - past limits of, 268
II 6, 127, 263 Davis, W. G., 173
Coleman, A. P., 289 Davos, 264
Colombia, 8o, IOl, 262 Dawson City, 214, 218
Colombo, II6 Daylight, duration of, 22, 33, 72, 208,
8010n, 91, 106 213, 222, 250
Colonization, Climate and, 3, 86 Dead Sea, 244
Colorado River, 244 Death-rate, Climate and, 3
Columbia River, 101 Death Valley, 234, 239
.Columbus, I Deccan, I09, 124~ 126 .
Complexion, Climate and, 2 Deciduous Forest, 61, 185
Concepcion, 152 Deforestation, 143, 145, 185, 200
Congo Basin, 86 De Geer, S., 282, 290, 212, 213
Coniferous Forest, 61 Delhi, IIO, 121, 124, 136
Continental Climates, 41 et seq. Deli, 83
Continental Drift Theory, 271 De Martonne, E., 58, 75, 263 .
• Continental Phase', 288 Denison, F. N., 203
Cook Inlet, 2 I 3 Denver, 18t, 193, 204
Cool Temperate Climates, 65, 176 et Desert Climates, 4, 64, 67, go
seq. Devon, fruit, 23
Copenhagen, 218 Dew, 24, 9r, 142, 237, 240
Coquimbo, 152 Dhulia, 122
Corals, limits of, 275 Diet, 2, 146, 228
Cornwall, 185 Disease, Climate and, 4, 81, 159, 179,
Cotton, IOO, IOI, 128, 160, '161, 190, 228
233, 24 6 Djole, 74, 76
Cotton belt, limits of; 167 Dneister, River, 212
Craggan,84 Doldrums, 26, 27
Craig, J. l, 10 4,29 0 Douglas fir, 185, 212
Cretaceous, Climates of, 267, 272 Douglass, 290
Cretinism, 260 Drayson's theory, 269
Crohamhurst, 134 Drift, older. and newer, 279
Croll's theory, 269 Dry farming, 211, 232
Cuba, 165 Dubbo, 160, 174
Cullum, 203 Dublin, 183
Currants, 146 Dunedin, 204
Currents, Climatic effects of, 45 et seq., Durban, 14, 106, 139, 164, 165
140 , 235 • Dust devils " 239
Curry, 290 Dutch Harbour, 218
Curtis, R. H., 10 Dysentry, 159
Cutch, II9, 123
Cuyaba, 95, 103, 106 Eagle, 218
Cycles, Climatic, 4, 270 • Early Peat-bog phase', 283
Cyclones, temperate, 27 et seq., 142, East African plateau, 3, 80, 98, IOI,
143, 147, 150, 158, 159, 162, 163, 102, 103
165, 166, 173, 176, 186, 187, 188, Eastbourne, 51
191, 192, 194, 195, 196, 197, 199, Eae.t Indies, 82 et seq., 105, 17,2
200, 201, 208, 214, ,215, 221,,222, East Monsoon, 82
253 Ecuador, rOI, 252, 262
INDEX 295
Edaphic faCtors, 58, 212 Frost, 10, 103, 132, '141, 201, 259, 263
Edmontdn, 2 I 8 Fuchau, 171, 174
Efflorescence, 246 Funchal, 141, 156 .
Egypt, 3, 239, 240 Futa Jalon Highlands, 81
Elburz Mountains. 190
Elements of Climate, 6, 8 et seq. Galveston, 167, 174
Eliot, Sir J., 134 Gambia, 76
Ellis, W., 9 Gambia, Mt., 146
Elm, 185, 186 Gangetic plain, 1I3, II4, 1~6, 123, 124,
Entebbe, 74. 102 125, 12 7, 133
Eocene Climates, 267, 275 Garigue, 145
Equatorial Climates, 64, 71 et seq. Geelong, 140
Equipluves, 19 • . Gefie,60
.Erebus, Mt., 221 Geneva, 254
Esperance, 154 Genoa, 140
.Esqtiimaux, 2, 228 Geography, Climatology and, 5
Etesian Winds, 148 Georgetown, .76
Eucalyptus, 145 Georgia, 166, 167
Eucla, 156, 248 Geraldton, 154, 156, 238
Euphrates, River, 241 Gherzi, E., 173
Eureka, 150, 151 Ghokak, 122
European plain, '181 Gibraltar, 156
Evangelistas Islalld, 195, 23 0 Gibraltar Straits, 147
Evaporation, 16, 97, 98, 239 Gjesvar, 209
Evesham, fruit, 23 Glacial periods, 277
Evolution, Climate and, 277 - - correlation of, 279
Eyre's Peninsula, 154 Glaciation, causes of, 269 et seq.
Glaciers, advance and retreat of, 268,
Factors in Climate, 6, 33 et seq. 270 '
Famine, 109, 124, 126, 129, 200 Glasgow, 47, 181, 182
Ferghana, 246 Glasspole, J., 203, 289
Ferraz, J. de S., 104 Glossopteris fiora, 272 /
Figs, 146, 246 Gobi, 2, 241
Fiji, 104 Gold Coast, 74, 81, 104
Fitton, E. M., 217' Golden Ga.te, 151
Fjord Coasts, 178, 206, 215, 221 Gondokoro, 106
Flinders Range, 154 Gorakhpur, 124
Floods, 80, 81, 143, 173, zoo, 210, 241, Goree (Dakar), 47, 76
245 Goths, 287
Floods, records of, 268, 287 Gorse, 145
Florida, 48, 94, 166, 167, 168 Gourock Range, 163
Foehn Winds, 17, 152, 172, 196, 221, Graaf Reinet, 98
243,25 2 Graarud, A., 217
Fog, 21, 148, 151, 208, :Zl5, 225, 237, Grahamland, 226 .
245 Grampians (Australia), 155
Fonteboa, 80 Grassland Climates, 59
Forest climates, 58 Great Australian Bight, 134, 154
• Forest period " 284 Great Basin, 233, 241, 280
Forest, types of, 59 • Greater rains " 74
Formosa, 94, 170 Great Lakes, 49, 191, 194, 214
Fort de Kock, 87 Great Lakes, Pleistocene level of, 273
Fortescue River, 134 Great Valley of California, 151
Fort Johnson, 76, 106 Great Wall of China, 286
• Fossil ergs', 282 Greece, Civilization of, 286
Frambeim, 222 Greeks, climatic zones recognized by,
Franze, B., 31, 104 54
Frech's theory, 270 Greely, A. W., 167, 181, 193
Freetown, 81 Greenheart, 79
Freezing of rivers, 178, 270 Greenland, 221, 224, 226
Freezing of harbours, records of d.:tes - burials in, 268
. of:268 - ice cap, 31
Friagem, 103 - Norwegian Colonies in, 287
19"
296 . CLIMATOLOGY
Gregory, J., 290 Hong-Kong, 136, 170, 171, 172, 17 8
Gregory, Sir R, 290 Honolulu, 91, 92, 106
Grimsey, 230 Honshiu, 109
Growing season, duration of, 10, 61, Hooker, R. H., 31, 203
161, 176, 226 . Horse Latitudes, 6, 25, 71, 108
Guano, 246 Horsham (Australia), 140
Guiana Highlands, 103 Hue, 130, 131
Guinea, 81, 102 Hume, 290
Guinea' Current, 49 Humidity, 14
Gujarat, 112 Humphrey's theory, 270
Gulf of Bothnia, 216 Hungary, 190
Gulf of Finland, ice in, 50 Huns, 287
Gulf of Mexico, 104, 172, 192 Huntington, E., 7, 270, 289, 290
Gulf States of U.S.A., t65 Hurricanes, 84, 92, 127, 158
Gulf Stream, 21, 27, 42, 46 Hwai-yang-shan, 169
Gum Arabic, 99 Hyderabad, 124
Gunz Glaciation, 279 Hyksos, 285
Gypsum deposits, 267, 272
Ice Ages, 266, 269, 270, 274
Hail, II6, 190 Ice Caps, 224, 225, 273, 276
Hair; cold'ur of, 2 Ice, effect on temperature, 50, 207, 222
Halifax Bay, 132 Icelandic' low', 31, 46; 186, 188, 207
fIankow, 199,200 Ichang, 204
Hann, J., 31, 40 , 52, 254, 263 Illinoian Glaciation, 279
Hanoi, 86, 131 Iloilo, 86
Haparanda, 50, 216, 2t8 Incas, 98
Harbin, 62, 218 India, 3, 108 et seq.
Harmattan, 94, 99, 102 Indian-Qcean, 92
• Harvest of Death ',262 Indo-China, 130, 135
, Harvey Creek, 132, 133 Indo-European languages, 289
Hatteras, 174 Influenza, 179
Havana, 91 Ingens, K., 217
Hawaii,91 Inland Sea of Japan, 202
Health, climate and, 3, 81, 121, 159, Insalah, 248
179,210,228 Insolation, 34, 72, 250, 25 1, 2'13
Heat rays, 8 Inter-glacial periods, 278 ~\
Heat waves, 194 Inversions of temperature, 51, 142, 255,
HeDroLl, 2]D 25 6,259
Helena, 248 Iowan Glaciation, 279
Helsingfors, 60, 209, ~IO, 211, 218 Iquique, 235, 236, 237, 248
Henry, A. J., 182, 193 Iquitos, 78, 80
Herbertson, A. J., 31 Irgis, 248
• Hercynian' MountaiI1S , 190 Irkutsk, 2II
Hercynian Orogeny, 277 Irrawaddi, River, 123
• Heroic Age' in IrelaIld, 285 Irrigation, 109, 127, 142, 148,200, 241,
Hessling, N. A., 173 245,25 6
Hibernation, 227 Isanomalous lines, 12
Hickory, 161 Isle of Purbeck, 51
Hilea (Hawaii), 39 Isohels, 22
Himalaya Mountains, 37, IIO, 112, 124, Isohyets, 19
262,2 63' Isonephs, 22
Hittites, migration of, 28S Isotherms, II
Hoang Ho floods, 200
Hobart, 204 Jacobabad, lIS, II9, 248
Hobbs, W. H., 228 Jaipur, 136
Hobley, 290 Jaluit, 72
Hokitika, 178, 204 Jamaica, 91
Holmes, A., 277 Jan Mayen, 223, 225, 23 0
Homoclimes, 14 Japan, 49, 108, 169, 200
Honan,,200 Jar1)ah, 145, 154
Hondo,'201 Jask,248
Honduras. 104 Java, 2, 40, 83. 84
INDEX 297
J ebel Ma~a, 3'i'> ~orea, 197
Jefferson, M,. 31, 155 1{reichgauer's theory, 271
Jerseyan Glaciation, 279 :Kumasi, 82 .
Jerusalem, 15 6 :Kunene River, 244
Jervis, W. W., 155 :Kupang, 136 .
Johannesburg, 98 :Kurile Current, 45, 48, 201, 203
Joly, J., 277 :Kuro Siwo, 27, 46, 48, 201
Jones, J. C., 289 :Kuta Raja, 85
J ostedalsorae, 20 9 :Kwitta,82
Jubbulpore; II6
JujJY,98 . . j..ahour, Climate and, 3
Jurassic chmates, 275 j..abrador, 48, 215,'220, 225
Jute, 100, 1 0 9 ~ Current, 21, 42, 47, 215, 220
j..agos, 73, 74 .
Kagoshim a , 174 j..agrange, 269
Kalahari, 21, 49, 235 )-a Guaira, 7'i'>
Kamerun, 80, 81 j..ahontan, Lake, 281,.290
Kamloops, 193, 24 8 j..ahore, IIO, 124, 136
Kanazawtt, 202 j..ake Constance, 49
Kansan Glaciation, 279 j..ake dwellings, 285
Karaburart, 243 j..ake Michigan, 49
Karachi, po, 123, 133, 137, ~8 j..akes, effects of, 49
Karri, 145, 154 ,j..a Laguna, 156
Kashgar, :Z4 1, 242, 24 8 j..and breezes, 43, 74, 92
Kashmir, 112 j..and forms of deserts, 243, 281
Kauai (Hilwaii), 39 f_andes, 52
Kau, desert of (Hawaii), 39 j..a Paz, 261, 262
I Laramide' orogeny, 277
Kayes, 10 6
Kazalinsk, 209 :L-arch, 212
Kazan, 61, 218 j..a Serena, 152
Kei Islands, 85 Las Palmas, 141
Kendrew, W. G" 3 1, 173 :L-atitude, effects of, 33
Kent, fruit, 23 :L-aurel, 145, 185
. Kenya, 104 :L-eake, H. M., 32
Kerguelen, 225 :L-eaui, 76
Kew, 143, 181, 183, 20 4 :L-egends of past climates, 268
Key West, 106 :L-eh, 124, 254, 264
Khamsin, 147,239 :L-emming, 227
Khandesh, 12 7 :L-emons,146
Kharif Crops, 128 :L-ena, River, 210
Khasi Hills, 12 3 :L-eningrad, 61,209,210, 2II, 2IS
Khingan jI.1ountains, 217 :L-eppan, H. D., 17
Khirgiz,4° :L-erwick, 183
Kiev, 179 :L-espagnol, 247
Kilimaniaro, 102, 282 f_eveche, 147
Kimberley, 91 :L-hasa, 242, 263
Kimberley (Australia), 134 Lianes, 161
Kingston (Jamaica), 91, 92, 106 Liaotung, 197
Kismayu, 102 Liberia, 81
Kiushiu, :Z01 Libreville, 49
Klagenfurt, 261 Light rays, 8
Knox, A., 104,247 Lime,'I85
Knysna, 139, 164 Limes, 146
Kodaikanal, 136 Lindi, 76, 106
Kodiak, :Z18 Line Squall, 183, 194
Koembang,83 Lisbon, 47, 147, 156
Koepang, 85, 86 • Littorina Sea " 284
Koeppe, C. E., 173 Llanos, 100, 103
Kolm Saigurn, 253 Loanda,76
Konakry. 81 Loess, Ig8, 281, 282
K,onigsDurg, 218 Lofoden Islands, 207
Koppen, W., 55, 57, 271, 289 Lomba.rdy pla.ill, 142, 143, 190
CLIMATOLOGY
London (see also Kew), 189 Melbourn{l, ISS, 168, 174
London peninsula, 194 Mellish, H., 31
Los Angeles, 150 .. Melons, 246
Louisiana, 167' Menado,85
Lucerne, 153, 246 Mendoza, 24
Luderlitz Bay, 236 Merv, 241
Luktchun, 233, 242, 248 Meseta, 143, 147
Lushai Hills, 123 Meteorology, Climatology and, 6
Luzon, 165 . Mexico, 98, 100, 104
Lyons, H. G., 104, 247 - City, 98, 106
Miami,166
McAdie, A., 9, 52 Michigan peninsula, 194
Macchia, 145 Migrations of animals, 227
Macdonnel Range, 240 - of ruan, 176, 233, 258, 268, 28 5,
Mackenzie Valley, 210 286, 287
McMurdo Sound, 230 Milan, 204
Macquis, 145 Milankovitch, 269
Madeira, 141 Miles City, 255 /
:lv):adras, 15, 109, 115, 127, 130, 131, Millet, !0O, 128, 233, 246, 2Q3
. 136 'Millet rains', I02
Madrid, 149 Mindanao, 169
Magadi, Lake, 101 Mindel Glaciation, 279
Magdalena, River, 101 Mirage, 239
Magnolia, 161 Mississippi, 117
Magyars, 287 'Missouri' rainfall type, 167, 181, 193
Mahabaleshwar, 122 Mist, .21, 142, 148, 237, 238, 245
Mahogany, 79 Mistral, 145, 148
Maidenhead, 17 \ Miyako, 204
Maize, 100, 101, 127, 161, 169. 186, Mobaye,76
190, 246, 2 6 3 Mobile, 174
, Maize rains', 102 Mogadiscio, 102, 106
Makindu, 102 Mogador, 140, 141, 156
Malabar, 120, 131 Moharumedanism, 233
Malacca Straits, 84 Mohammedan wave of conquest, 287
Malaria, 143, 159, 286 Mombasa, 48, 102
Malay, 130 Moncay, 136
Malays, 2 Mongalla, 95, 106
Mallee Scrub, 145 Mongolia, 1I2, 170, 175, 242
Malye Karmakouly, 230 Mongols, 287
Mammoth,279 Monsoons, 41, 45, 81, 90, 102, 104, lOS
Manaos,80 et seq., 138, 149, 166, 168, et seq.,
Manchuria, 61,62,177,217 177, 180, 196 et seq., 206, 216 et
Mandalay, 123, 136 . seq., 236, 281
• Mango Showers', 117 Monsoon forest, 58
Manila, 136 Montana, roI, 103
Manokwari, 85, 86, 87 Montana (U.S.A.), 214
Maple, 161, 169, 185, 186 Mont Blanc, 254
Marine climates, 41 et seq., 66 Montevideo, 158, 160, 174
MarioloJ)oulos, F. G., 155 Montgomery, 167
Maritime Alps, 261 Montreal, 218
'Maritime phase', 283 Morocco, 140, 142, 156
Marquette, 21S Moscow, 19, 179, 218
Marseilles, 148, 156 Mosquitoes, 121, 143, 227, 286
Masset, 192 Mossman, R. C., 155
Matthews, H. A., 31, 155 Mosul,141
Maurer, J., 263 . Mountain Climates, 4, 37, 67, 250 et
Mauritius, 94, 12 9 seq.
Mayan civilization, 287 - sickness, 36
Meadow grassland, 185 Mountains as climatic ~ivides, 40,261,
Mediterranean climates, 138 et s~. 262
- sea, 1138, 139, 141, 142 ' Moiambique, 92
Mekong V~lley, ~63 Mukden, 197, 204.
INDEX 299
Mulberry, 161, 169, 246 North Pacific Drift, 45, 46, 192, 207,
Multan; 40, II6, 124 208
Murcia, 147 - ~ , High', 151:>, 192, 193
Murphy, R. C., 52 , Nor'westers' of Bengal, 1I6
Murree, II3 - of New Zealand, 253
Murrumbidgee, 163 Nullagine, IS .
Musk ox, 227, 279 Nyasaland, 104
Mwanza, 74, 76
Mymensing, 39 Oak, cork, 14.5
Myrtle, 145 - deciduous, ~45, '161, 185, 186
- evergreen, 145, 161
Nagasaki, 204 Oakland, 151 '
Nagpur,119 Oases, 232, 244, 245, 256, 263
Nain, 47. 215 Oats, 185
Naini Tal, 124 Oaxaca, 106
Nairobi, 98, 102, 103 Ob, River, 170,210
Nanainio, '193 ObiI', 261, 264
Naples, 140 Obliquity of ecliptic, 269
Narbada, River, 125 Occupation, climate and,'99. 161, I7~'
Nashville, 160, 204 206, 220, 258, 260
Nassau, 91. 92 Ocean Currents, 45
Natron Lake, 101 Odessa, 204
Nebraska, 165 Oil seeds, 109
Nebraskan Glaciation, 279 Okhotsk,218
Nebular hypothesis, 266 Oklahoma, 167
Nemuro, 218 Olekminsk, 209, 210, 217
Neolithic invasions, 283 Olifants River, 153
Neuquen, 243 Olive, 144, 146, 147, 259. 286
Nevados, 252 Omaha, 204
New England, 181 Onslow, 134
Newfoundland,215 Ontario, wheat. 2'1, 51 '
New Guinea, 82 Opaco, 254
Newhaven, 204 ,Opium, 246
New Orleans, 160, 165, 174 Oporto,47
New South Wales, 162 Orange River, 244
New York, 47, 194 Oranges, 146, 154, 161
New Zealand, 178, 195 Oregon, 150, 192
- - Alps, 178 Orenburg. 190, 218
Nhatrang, 136 Orinoco Basin, 103
Nice, 51, 141 Orissa, 125
Nicolaievsk, 51 Oruro, 262
Nigeria, 102, 104 Oslo; 60. 209
Niigata, 109, 20i, 204 Oxygenation of the blood, 36
Nikolaiewskoe, 189 Ozark Mountains, 38
Nile, River, 102, 128, 244
- flood, 129 Pacific coniferous forest. 61. 185
- history of, 28r Pack ice, 216, 221
Nitre, 246 Padang. 84, 87
Nomadism, 99, 21 3, 228, 233, 234, Palaeobotanical evidence of climatic
245 change, 271. 272, 275, 279
Nordenskjold, 0., 228 Palaeolithic, 283 ,
, Nordics', 283 "- Palaeontological evidence of climatic
Norfolk, fruit, 23 change. 266, 271, 279
Nortes, 104 Palermo, 140. 156
North Atlantic drift, 45, 47, 207, 208, Palestine. 112
215, 223, 226 Palm oil, 79
North Cape, 208 Palms, 55. 201
North Equatorial Current, 45 Palmyra, 268
, Northern Circuit " 191 Pamirs, 37, 241, 2 63
Northern Territory, 132, 134 Pampa, 161
'. Northers' of California, ISO Pampero, 43. 159. 164
~ of Texas, etc., 43, 104, 159. 165 Panama, 91, IDS

\
300 CLIMATOLOGY
, Pangaea " 271 Portsmo1!,th (Ohio), 194
Papagayos, 104 Potatoes, 185, 262
Para, 72, 74, 76, 80 .. Po-yang, 200
Paraguay, 103 Prairie, 213, 216
Paramos, 262 Precession of the equinoxes, 269
Parana, 165, 174 . rressure and winds, 2}, et seq.
Paris, 62, 181, 182, 204 - . effect of altitude on, 36
Patagonia, 3, 49, 180, 195, 233, 234, Preston, 17
24.,0,24.,1,24.,'!, Pretoria, 17
Patiala, 125 Proterozoic orogeny, 277
Patna, 124 Provence, 261
Paulsen, A., 203 Ptolemaeus, 267, 286
Peaches, 146 Puebla,106
Peat bogs, 268 Puerto Carranza, 152
, Peat-bog phase', 285 Puge~ Sound, 191
Peking, 168, 182, 199, 20 4 Puna, 162
Pemba,82 Punjab, II2, II3, II4, II6, 124, 126
Penang, 85, 86, 87 Punta Arenas, 177, 204
P~nck, 279, 29 0 Punta del Garda, 156
Perihelion, 77 Punta Tumbez, 152
P,ermo-Carboniferous cHmates,271, 27 2, Pygmies, 2
273
Pernambuco, 47, 100, r 0 3 Qu'Appelle, 218
Persia, 128, 241 Quarternary Climates, 267
Perth (W. Australia), 14, 154, 15 6, 15 8 Quayle, 135, ISS
Petro-Alexandrovsk, 24 2, 24 8 Queensland, 3, 132, 133, 134, 161
Peru, 98, 101, 233, 238, 245, 255, 25 6, - , winds above, 24
262 Quetta, 120, 241 /
Peshawar, II2, 127 Quilpue, 152
Philippine Islands, 94, ~30, 169 Quito, 76, 78, 261
Philippson, A., 155
Phosphates, 246 Rabi crops, 127
Physical Climate, 33 Race, Climate and, 2
Physical features, effect: on climate, 51 Radiani heat, 8
Pigmentation, 2 Radiation, Solar, 7, 253
Pike's Peak, 255, 264 Ragunda, Lake, 282
Pilbarra, 133, 134 Ragusa, 143, 147
- 'low', 133,233 Rainfall, IS et seq.
- I P...fipsj-jJ..tp..... , T~Q,Jqfi, T...Qi1., T"'T.:~." ''_'lll., ?_~?_
1i''';:''''', '!fS,'l.Yl.
Pine Bluff, 174 , effect of altitude on, 39
Plateau climates, 37 - , reliability of, 18, 96, 1I0, 143, 200,
Plummer, F. E., 17 203
Pluvial periods, 280 - . seasonal distribution of, 18, 63, 64
Pluviometric Coefficient, 19 - , types of, 18, 160, 180, 181
Pneumonia, 142, 178 Rains, excessive, 17, 123, 131, 134, 166,
Polar anticyclone, 27, :;t2I, 27 6 173, 183
- bear, 227 Rainy days, 16, 143, 195
- front, 27 et seq., 22J, 27 6 Raised beaches, 284
Pomegranates, 246 Raisins, 146
Pontianak, 84, 85 Rajputana, Il3, 116, 126
Poplar, 212 . Ramah,230
Port Antonio, 91 Ramsay, 273, 289
Port Au Prince, 106 Rangoon, 136
Port Elizabeth, 139, 15 8, 164, 174 Ranns of Cutch, I19
Portillo, 152 Rasputitsa, 210
Portland (Oregon), 204 Rawson, Sir W., 16
- (Australia), 140 Red Bluff, 151, 156
Port Macquarie, 162, 163, 165, 168, 174 Red River (China), 172
Port Nolloth. 248 , Red Wall', the, 287
Port Reyes, 151 Redwoods (of California), 151
~ort St. John, 164
, Re~ • soils, 246
""ort Simpson, 203 Reindeer, 227, 228, 279
INDEX 30r
Relief, Climate and, 5, 227 Santiago, 143, I5 6
Retreat of moIJ.soon, IIO, 126 Santis, 264 .
Rheu.matil;m, 121., 1.<\2 Sao. Francisco, 1.°3·
Rhine Valley, ~70' 253 Saratov, 190
Rhodesia, 97 ' Sarmiento, 248
Rhone Valley, I45, 148; 258 Savanna, 90, 99' 101, 10 3, 144
Rice, 100, 109, .125, 12 7, 1<8,.161, 16 3, Scandinavia, 215
172, I90, ::102, 20 3 Schattenseite, 254
Riga, 61 . Schimper, A. F., 60
Rio de ] aneiro, IOO, 106 Schirmer, H., 24'7
Rio de Oro, 235, 237 Schmidt, R., lOA-
Riss Glaciation, 279 Schuchert, C., 2B9
, R<1!tring Forties ',. I, In Scilly Isles, 143, In, 17;8, 182
Robe, 140, 156 Sclerophy1l6us vl"oodland, 62
Rockhampton, 13 2 Scoto-Icelandic rIse, 4 8, 273
Rock Salt deposits, 244, '},67, 272 Screes, 244 .
'Rocky Mountain Footll~ll' type. of. Scrub,58 .
. rainfall, If)2, 193 ' Sea-breezes, 43, 74, 92, 14 2 , 154, 23 6
Roman Empire, 3, 286, i87 Seathwaite, 18, 189
Rome, 140, 141, 143, 15 6 Secondi,82
Rosario, 164, 174 Selvas, 52, 62, 7 8
Rosemary, 145 " Semipalatinsk, ,f3, 216
R~t\\.am<;',t~, ,,1- S.~=ij:.i._<:. 'LaQ~Qa.'lfs, 289
Rubber, 78, 100 Sen,egal, 44, 102
Run~off, 16, 93, 10 9, II3, <00, 232 Senegambia, 44 .
Russel, R. G., 155 Sequoias, annuiJ,1 rmgs of, 268, 270
Russian Turkestan, 241 'Settlement, Clittl ate and, 73, IOl, 103
Ruwenzori Glaciers, 282 Seville, 143
Rye, 259 Shad~tempetature, 9' .
Shangnai, 40, 168 , 169, 17 1, 197, 199,
Sable Island, z18 200
Sacramento, 140 , 151, 15 6 . Shansi, 198
- Valley, 41 Shantung, 168, 197
Sage, 145 Shaw, Sir N., 16, 52, 290
Sahara, 81, 147, 233, 235, 238, 239 , Shepherd Kings " 3. 28 5
St, Gotthard, z5 6 Shikoku, 281 .
St. Helena, 10 6 Shillong, 123
St. John's, 47.215.218 Shrinagar, 124
St. Lawrence·Valley, 19 1,214 Siberia, 216, 211. 220 ,
St. Louis, 204 Siberian' high " 51, 170 , 198, 199,209,
St. Lucia, 92 211, 21 7
Sagastyr, 223, 23 0 Siam, 130
Sakhalin, 2a1. - , c,\l\\ ~\, >.?:F>
Salinas, 244 Sierra Leone, 81. .
Salisbury, 98, 106 Sierra Nevada (Cahforma), 40, 142, 15 0
Salter, M. de C. S., 39, 20:') - - (Spain), :259
Salt Lake City, 241 Siesta, 141
Salt Lakes, 244 Si-kiang Valley, 170
Salween Valley, 26 3 Silurian climateS, 275
Samarkand, 241, 242 Simla, II3, 136
Samoa, 9I, 92 Simoom, 239 •
Samoyedes, 2z7 Simpson, G. C., 134, 228, 290
Samun, 253 Sind, II2, IIS, JI6, II8
Sanders, E, M., 173 _ 'low', 118, 122, 12 3, 12 4, 125, 128,
San Diego, 150' 24 8 190, 233
San Francisco, 4 1, 140, I'F, ISO, .151, Sion, ]., 134
15 6 Sirocco, 17, 147
San Luis Obispo, 15 6 Sitka, 207, 208, 2 13
Santa Anna de Sobradinh(), 76 Slavery, 3
Santa Annas, 150 , Smaller rains " 74
Santa Cruz, 24 8 Smyrna, 156
Santa Fe (Argentine), 164 Snake River, 244
302 CLIMATOLOGY
Snow blindness, 253 Sylhet, 39
Snowdon, 178 Szechwan, 111
Snowfall, 20, 37, 5Q,:144:182, 194,202,
20 9, 2II, 215, 224, 257 Table Mount.ain, 153
Snowline, 37, 38,257, 282 Tabora,76
Soerabaja, 87 Taiga, 60, 185, 212
SQil, effect on climate, 51 Takla MakaD, 241
- - on vegetation, 52, 186 Tamalpais, Mt., 151
- waste, 109, 143 . TamataYe, 91, 9 2
Solar climate, 33 Tanga, 82
Solar radiation, 7 Ta-pa-shan, 169
Solberg, H., 32 Tapti, River, 12 5
Solomon Islands, 134 Tarim Basin, 24 1, 243, 254
Somaliland, !O2, 128, 135 Tashkent, 242
Sonnblick, 253, 264 Tasmania, 162, 195
Sonnenseite, 254 Tasman Sea, 162
SonaTas, 150 Taylor, G., 7, 13, 3 1, 135, ISS, 228,
South Africa, 23, 97,98,139,153,164 24 6 ,290
;South Australia, 139, 140, 154 Taylor; N., Z0 3
SQuth Downs, 39 Tea, 100, 101, 109, 160, 161, 169
'. Southerly burster " 43, 159, 163 .J:eheran, 241, 24 8
'Southern Circuit " 165, 191 'temperaturt), 8 et seq .
.South Georgia, 225, 230 - anomaly, 12, 108, 177"
- Orkneys, 225, 230 - , acc~~~la+~d,ll),~,~l
Spitzbergen, 222, 223, 226, 220 - , effects of altitude on, 36
Spokane, 193 - in the sun,S, 223, 254, 262
Spruce, 186, 213 - , means of, 9, 10
, Squall line " 29 - , sensible, 13, 96, 120, 143, 159, 209
Squantum tillite, 272 - , wet bulIJ, 13
Stanovoi, 51 Tenasserim, 108, 123
, Steering line', 29 Tertiary cliI11ates, 277. 27 8
Steppes, 3, 61, 67, 139, 149, 179, 186, Teutonic peoples, 3
190, 196, 211, 212, 213, 216, 233, Texas, 167
285, 286, 287 Thayetmyo, 123
Stevens, A., 32 'Thermal belt', 260
Stockholm, 216 Thorshavn, 218
Stockwell, 269 Thunder, 84, 96, 150, 166, 183, 190
, Stowed' winds, 83 Thyme, 145
Straits Settlements, 86 Tian Shan, 40 , 2 6 3
Sub-arctic forest, 61 Tibesti Higlllands, 240, 25 6
Sub-tropical Climates, 62, 138 et seq. Tibet, 250, 262, 263
Sucre, 106 Ticino Valley, 253
Sugar, 16, 98, 161, 169, 171, 190 Tientsin, 197
Sulairnan Mountains, 110 Tierra del Fuego, 43
Sumatra, 85, 123 Tierra caliente, 100
, Sumatras " 84 1 - [ria, 78, 98, 101

Sungari, River, 178 - templada, 98, 100


Sunlight, 4, 103 Tigris, River, 24 1
Sunshine, 21, 143, 183, 196 Timbuktu, 19,95
Sunspots, 270 Timor, 85
Sunstroke, 1.15 Tinnevelly, 122
Supan, 53, 54 Titicaca, uke, changes of level, 281
Surat, 122 Tobacco, 23, 100, 101, 161, 17 1,24 6
Surazos, 103 Tobolsk, 209, 210, 2Il
Swakopmund, 236, 237 Toce Valley, 253
Swanland, 140, 154 Toeal,85
Swatow typhoon, 94, 173 Tokio, 182, 201, 202
Sweden, 216 Tondano,85
Swertzow, 40 Tongking, 173
Switzerland, 260 et seq. -,.Gulf of, 131, 17 2
Sycamore, 185 Tornado (U.S.A.), 183. 194
\ Sydney, IS, 158, 160, 162, 163, 174 - (W. Africa), 9 6
INDEX
Toronto; 218 Verkhoyansk, za9, 210, 216, 218
Torquay, SI Vestmanno, n8
Townsville, 132 Vicksburg, 165, 174.
Trade, climate and, 1 Victoria, 140, 162
Trade winds, 1, 6, 27, 71, 73, 77, 90 - (B.C.), 192, 193.
94, 101, 132, 138, 148, IS8, 169 Victoria Nyanza, 50, 102
Trajan, 287 . Vienna, r82, 183,204
Transhumance, 40, '14S, 245, 258 Vine, 146, 147, 153, 154,246,253, 259
Transvaal, 17, 97 . - , past limits of, 268 .
Tree ferns, 80, 162 Visher, S. S., 52, 270, 289
Tree growth, limit of, 55, 212 Vladivostok, 51, 1I0, 217, 218
Trewartha, G. T., 7
Triassic climates, 267 . Wadis, 240, 243, 281
Trinidad, 91, 92 Walfish Bay, 235, 236, 237, 2.f8
Tripoli, 141, 142 Walker, 290
Tromso, 215 Wallen, A., 16
Trondhjem, 218 '
Wallis, B. C., 19, 31, 135
Tropical climates, 64, 90 et seq. .Walnut, 153, 161
Tsanpo Valley, 263 Ward, R de C., 7, 20, 31, 155, 173:-
Tulip tree, 161 .
20 3,247
Tundra, 220, 223, 226 Warm temperate climates, 65
Tunis, 140, 146, 246 Warrnambool, 140
tung-ting, Lake, 200 Warsaw, 42, 182, r84, 189, 204
Turan, 242 Washington (D.C.), 47, 197
Turfan, 242 \ Washington, 150, 192, 204
Turkestan, 241, 242 Weather and climate, 6
Tycho Brahe, 267 Wegener, A., 271, 289
Tyndall's theory, 270' Wellington (N.Z.), 174, 196
Tyrol,257 Welsh Mountains, 39
Typhoons, 84, 92, 131, 158, 172, 200 Western Australia, 18, 132, 133, 134,
l4 0 , I4S, 23 2 ,235, 237
Ubac, 254 Western Ghats, 40, 108, 122, 130, 133
Uganda, 104 West Indies, 19, 94, lOS
United Provinces, 1I2, 121, 124, 126, , West Monsoon. " 82
12 7 Wet bulb temperature, 13
Unstead, J. F., I I Wheat, ro, 16, i8, 21, 33, 51, 101, 153,
Upernivik, 230 154, 169, 186, 199, 21 7, 232, 233,
Upper Air Currents, 24, 126, 133, 134,
24 6, 26 3
149,281 'White Australia '. 3, 14
Upsala, 218 'White Man's Grave " 81
Urals, 170 White Nile, 102
Urga, 248 Wilcannia, 174-
Uruguay, 163, 164 Willow, 212, 227
Urumtsi, 218 Willy Willies, 134
Uspallata, 264 Windhoek,237
Utrecht, 189 'Windjammers', 1
Windmills, 23
Valais, 256 Wind roses, 24
Valdivia, 152, 204 Wine harvest, records of dates of, 268
Valencia, 43, 181, 182,204 Winnipeg, 15, 214, 218
Valley winds, 142, 252 Wisconsin glaciation, 279
Valparaiso, 152, IS6, 158 Woeikof, A., 217. 247
Van, Lake, 244 Woolly rhinoceros, 279
Vancouver, 193, 204 Wright, W. B., 289
Vard6, 209, 2r5, 230 Wiirm glaciation. 279
Varve clays, 270, 282 Wyndham, 1~2, 133, 13 6
Vegetation and climate,s, 52, 54, 58,
98, 103, 10 9, 144, r45, 161, r69,
184, r85, 200, 212, 223, 224, 226, Xerophilous vegetation, 59, 61,99, 145.
258,262 244, 745 , 258, 2.62
Venezue1a, roo, 103 Xerophytrc adaptatlOns, 144, 146, 226,
Vera Cruz, 47 245,25 8
CLIMATOLOGY
Yakutsk, 62, 218 Yunnan, 172, 245
Yang-tse-Kiang, I68, ~69, I7I, I99/
262 .: Zambesi,90
Yemen, I28 Zanzibar, 82, I29
Yenesei, River, I70, 210 Zikawei, 204
Yezo, 201 Zomba, 95
'.Yoldia Sea', 283 Zonda, 159
Yukon, River, 2IO Zone of maximum precipitation, 40,
Yuma, 248 25 8
,
METHUEN'S
'

GENERAL LITERATURE
A SELECTION OF

MESSRS. METHUEN'S
PUBLICATIONS
This Catalogue contains only a selection of the more important
. books published by Messrs. Methuen. A complete catalogue of
their publications may be obtained on application.

ARMSTRONG (Anthony) (' A. A.' THE LIVERY OF EVE


of Punch) THE SUBSTANCE OF A DREAM
WARRIORS AT EASE Each 3S. 6d. net.
WARRIORS STILL AT EASS AN ECHO OF THE SPHERES
PERCIVAL AND I lOS. 6d. net.
PERCIVAL AT PLAY
BALFOUR (Sir Graham)
ME AND FRANCES
THE LIFE OF ROBERT LoUIS
How TO DO IT
STEVENSON
Each 3'. 6d. net.
LIVESTOCK IN BARRACKS
Twenty-first Editiun. lOS. 6d. net.
Illustrated by E. H. SHEPARD.
BARKER (Ernest)
6s. net.
NATIONAL CHARACTER
Two LEGS AND FOUR
10$. 6d. net.
Illustrated by RENE BULL.
GREEK POLITICAL THEORY II
5s. net.
14$. net.
CHURCH. STATE AND STUDY
BAGENAL (Hope) and WOOD (Alex.) los. 6d. net.
PLANNING FOR GOOD ACOUSTICS ~
Illustrated. £1 IS. nero BELLoe (HIlaiN)
PARIS 8s. 6d. net.
BAIKIE (James) THE PYRENEES 8s. 6d. net.
EGYPTIAN ANTIQUITIES IN THE NILE MARIE ANTOINETTE 18,. net.
VALLEY. Illustrated. £1 IS. net. A HISTORY OF ENGLA.'lD
In 5 Vols. Vol•. I. II and III
BAIN (F. W.) Each ISS. net.
A DIGIT OF THE MOON ON NOmiNG
THE DESCENT OF THE SUN HILLS AND THE SEA
A HEIFER OF THE DAWN ON SOMETHING
IN THE GREAT GOD'S HAm THIS AND THAT AND THE OrnER
A DRAUGHT OF THE BLUE ON
AN ESSENCE OF THE DUSK FIRST AND LAST
AN INCARNATION OF THE SNOW ON EVERYmING
A MINE OF FAULTS ON ANYnuNG
THE ASHES OF A GOD EMMANUEL BURDEN
BUBBLES OF THE FOAM A PICKED CoMPANY
A SYRUP OF THE BEES Each 3$. 6d. net.
Messrs. Methuen's Publications:. 3
BffiMINGHAM (Geor~e A.) DUGDALE (E. T. !'I.)
A WAYFARER IN HUNGARY GERMAN DIPLOMATIC DOCUMENTS,
Illustrated. 8s. 6d. net. 1871-1914.
SPILLIKINS : ESSAYS 3s. 6d. net. Selected from the Documents pub-
SHIPS AND SEALING-WAX: ESSAYS [ished by the German Foreign
3S. 6d. net. Office. In 4 vol.. Vol. I; 1871-
90. Vol. II, 1891-8. Vol. III,
BUDGE (Sir E. A. Wallis) 1898-1910. Vol. IV, 191I-1914.
A HISTORY OF ETHIOPIA: NUBIA Each £1 IS. net.
AND ABYSSINIA
Illustrated. 2 vols. £3 13S. 6d. net. EDWARDES (Tickner)
THE LORE OF THE HONEY·BEB
CHESTERTON (G. K.) Illustrated. 7S. 6d. and 3S. 6d. net.
COME TO THINK OF IT • • • BEEKEBPING FOR ALL
6s. net. lIlustrated. 3$. 6d. net.
G.K.C. AS M.C. ,5.
6d. net. THE BEE-MASTER OF WARRILOW
THE BALLAD OF THE WHITE HORSE Illustrated. 7S. 6d. net.
3$. 6d. net. BEE-KEEPING Do's AND DONT'S
Also Illustrated by ROBERT 25. 6d. net.
AUSTIN. 12S. 6d. net.
CHARLES DICKENS
EINSTEIN (Albert)
GENERALLY SPEAKING
RELATIVITY: THE SPECIAL AND
ALL THINGS CoNSIDERED GENERAL THEORY 5S. net.
TREMENDOUS TRIFLES
SIDELIGHTS ON RELATIVITY
FANCIES VERSUS FADS
ALARMS AND DISCURSIONS
3'. ·6d. net.
THE MEANING OF RELATIVITY
A MISCELLANY OF MEN
THE USES OF DIVERSITY 5'. net.
THE OUTLINI! OF SANITY
THE BROWNIAN MOVEMENT
5S, net.
THE FLYING INN
Each 3$. 6d. net.
A GLEAMING COHORT 2S. 6d. net. EISLER (Robert)
WINE, WATER AND SONG THE MESSIAH JESUS AND JOHN THE
IS. 6d. net. BAPTIST: according to Flavius
Josephus' recently rediscovered
CLUTTON-BROCK (A.) 'Capture of Jerusalem' and
WHAT IS THE KINGDOM OF HEAVEN? the other Jewish and Christian
EsSAYS ON AliT sources. Translated by A. HAG-
SH.'J()?SPEARE'S HAMLET GERTY KRAPPE.
Each 5S. net. Illustrated. Demy 8vo. £2:as. net.
ESSAYS ON BOOKS 3S. 6d. net.
MOI!B EssAYS ON BOOKS
FIELD (G. C.)
ESSAYS ON LIFE
'MORAL THEORY 6s. net.
EsSAYS ON RELIGION
PLATO AND HIS CONTEMPORARIES
ESSAYS ON LITERATURE AND LIFE
125. 6d. net.
MOI!B EssAYS ON RELIGION
Each 6$. net.
SHELLEY, THE MAN AND THE POET FYLEMAN (Rose)
Illustrated. 7S. 6d. net. FAIRIES AND CHIMNEYS
THE FAIRY GREEN
CRAWLEY (Ernest) THE FAIRY FLUTE
THE MYSTIC ROSE. Re,,;sed and THE RAINBOW CAT
Enlarged by THEODOllE BESTER- EIGHT LITTLE PLAYS FOR CHILDREN '
MAN. 2 vols. £1 lOS. net. FORTY GoOD"NIGHT TALES
STUDIES OF SAVAGES AND SEX FAIRIES AND FRIENDS
Edited by THEODORE BESTERMAN. THE ADVENTURE CLUB
lOS. 6d. net. FORTY GOOD-MORNING TALES
DRESS, DRINKS AND DRUMS. SEVEN LITTLE PLAYS FOR CHILDREN
Edited by THEODol!B BESTER- TWENTY TEA-TIME TALES
MAN. 10$. 6d. net. Each 3'. 6d. net.
4 -Messrs. Methuen's Publications
FYLEMAN: (Rose)-continued HALL (H. R.)
THE DOLLS" HOUSE THE ANCIENT HISTORY OF THE
Illustrated by MARGARET TEMPEST. NEAR EAST
SS, net. Illustrated. £1 IS. net.
GAY Go Up THE CIVILIZATION OF GREECE IN
Illustrated by DEelE MERWIN. THE BRONZE AGE
5s. net. Illustrated. £1 lOS. net.
THE ROSE FYr.EMAN FAIRY BOOK A SEASON'S WORK AT UR OF THE
Illustrated by HILDA MILLER. CHALDEES. Illustrated. £1 5s. net.
lOS. 6d. net. HEATON (Rose Henniker)
A GARLAND OF ROSE'S: COLLECTED THE PERFECT HOSTESS
POEMS Decorated by A. E. TAYLOR
Illustrated by RENE BULL.
8s. 6d. net.
'S.
6d. n,t. Gift Edition, £1 IS. net.
HERBERT (A. P.)
GIBBON (Edward) HONEYEUBBLE & Co. 6s. net.
THE DECLINE AND FALL OF THE MISLEADING CASES IN THE COMMON
ROMAN EMPIRE LAW. With an Introduction by
With Notes, Appendixes and Maps, LoRD HEWART. SS. net.
by J. B. BURY. Illustrated. 7 vols. MORE MISLEADING CASES. 5'. net.
Demy 8uo. I5s. net each volume. WISDOM FOR THE WISE
Also, unillustrated. Crown 8vo. Illustrated by GEORGE MORROW.
7S. 6d. net each volume. 5s. net.
THE WHEREFORE AND THR WHY
GLADSTONE (Mary) (Mrs. Drew) Illustrated by GEORGE MORROW.
HER DCARCE:S AND LETTERS 3s. 6d. net.
Illustrated. £1 IS. net. THE BOMBER GIPSy 3S. 6d. net.
GLOVER (T. R.) THE SECRET BAITLE 3S. 6d. net.
VIRGIL TA..'ITIVY TOWERS 2S. 6d. net.
THE CONFLICT OF RELIGIONS IN THE r1oLDSWORTIl (Sir W. S.)
EARLY ROMAN EMPIRE A HISTORY OF ENGLISH LAW
POETS AND PURITANS Nine Volumes. £1 5$. net each.
Each lOS. 6d. net. HUDSON (W. H.)
FROM PERICLES TO PHILIP A SHEPHERD'S LIFE
6d. net.
I2S.
Illustrated. lOS. 6d. net.
GRAHAM (Harry) Also unillustrated.. 3s. 6d. net.
THE WORLD WE LAUGH IN: More IlUTTON (Edward)
Deportmental Ditries CITIES OF SICILY
Illustrated by '·FISH'. Eighth Illustrated. lOS. 6d. net.
Edition. 5S. net. MILAN AND LOMBARDY
STRAINED RELATIONS THE CITIES OF ROMAGNA AND THB
Illustrated by H. STUART MENZIES MARCHES
and HENDY 6s net. SIENA AND SOUnlERN TUSCANY
THE WORLD'S WORKERS NAPLES AND SOUTHERN ITALY
Illustrated by , FOUGASSE '. Illustrated. Each 8,. 6d. net.
SS. net.. A WAYFARER IN UNKNOWN Tus-
ADAM'S ApPLES
CANY
Illustrated by JOHN REYNOLDS.
THE CITIES OF SPAIN
5s. net. THE CITIES OF UMBRIA
GRAHAME (Kenneth) COUNTRY WALKS ABOUT FLORENCE
THE WIND IN THE WILLOWS ROME
207th Thousand. 7S. 6d. net. FLORENCE AND NORTHERN TUSCANY
Also illustrated by WYNDHAM VENICE AND VENETIA
PAYNE. 7S. 6d. net. Illustrated. Each 7S. 6d. net.
4\180 unillustrated. 3S. 6d. net.
Se. also Milne (A. A.)
HADFIELD (J. A.)
I INGE (W. R.), D.O., Dean of St.
Paul's
CHRISTIAN MYSTICISM
PsYCHOLOGY AND MORALS (The Bampton Lectures of 1399).
Eighth Edition. Crown 8vo. 6s. net. Sixth Edition. 7" 6d. net.
Messrs. Methuen's Publications. 5
KIPLING (Rudyard) THE SURVIVAL·OF M4.N
BARRACK-RoOM BALLADS 7S. 6<1. net.
255th Thousand. RAYMOND lOS. 6d. net.
THE SEVE.'1 SEAS RAYMOND REvISED 6s. net.
186th Thousand. . MODERN PROBLEMS 3S. 6d. net.
THE FIVE NATIONS REASON AND BELIEF 3s. 6d. net.
143rd Thousand. THE SUBSTANCE OF FAITH
DEPARTMENTAl, DITTIFS 2$. nel.
"7th Thousand. RELATIVITY IS. net.
THE YEARS BETWEEN CONVICTION OF SURVIYAL 2S. net.
95th Thousand.
Four Editions of these ifamo us LUCAS (E. V.)
volumes of poems are now pu~ THE LIFE OF' CHARLES LAA{II
Iished, viz. :- 2 Vol.. £1 IS. net.
Crown 800. Buckram, 7s. '6d. net. THE COLYINS AND THEIR FRIENDS
.Fcap. 8uo. Cloth', 6s. net. £1 IS. net.
L,ath", 'S. 6d. net. VERMEER THE MAGICAL 5S. net.
Service Edition. Two volumes A WANDERER IN ROME
each book. Square Fcap. 8i10. A WANDERER IN HOLLAND
3S. net each volume. A WANDERER IN LONDON
A KIPLI>1G ANTHOLOGY-VERSI' LONDON REVISITED (Revised)
Sixth Edition. A WANDERER IN PARIS
Leather 7S. 6d. net, A WANDERER IN FLORENCE
Cloth 6,. net and 3'. 6d. net. A WANDERER IN VENICE
TWENTY POEMS FROM RUDYARD Each lOS. 6d. net.
KIPLING A WANDERER AMONG PICTVRES
486th Thousand. IS. net. 8s. 6d. net.
A CHOICE OF SONGS E. V. LUCAS'S LONDON {.I net.
Second Edition. 25. net. INTRODUCING LONDON
INTRODUCING PARIS
LAISTNER (M. L. W.) Each 2S. 6d. net.
ThOUGHT AND LETTERS IN WESTERN THE OPEN ROAD ,6s. net.
EUROPE (A.D. 500-900). ISS. net Also, illustrated by CLAVDE A.
LAMB (Charles and Mary) SHEPPERSON, A.R.W.S.
THE COMPLETE WORKS lOS. 6d. net.
Edited by E. V. LUCAS. Six Also, India Paper.
Volumes. With Frontispieces. Leather, '$. 6d. net.
6s. net each. THE JOY OF LIFE
SELECTED LETTERS 6s. net. Leather EditiOn, '$. 6d. net.
Edited by G. T. CLAPTON. Also, India Paper,
3S. 6d. net. Leat{ler, '$. 6d. net.
THE CHARLES LAMB DAY BOOK TilE GENTLEST AnT 3$. 6d. net.
Compiled by E. V. LUCAS. And THE SECOND POST 3$. 6d. net.
65. net. Also together in one volunle.
LANKEST.ER (Sir Rayl . 7s. 6d. net.
SCIENCE FROM AN EASY CHAIR FIRESIDE AND SUNSHINE
SCIENCE FROM AN EASY CHAIR. CHARACTER AND COMEDY
Second Series GOOD COMPANY e-
DIVERSIONS OF A NATURALIST ONE DAY AND ANOTHER
GREAT AND SMALL THINGS OLD LAMPs FOR NEW
Illustrated. Each ,s. 6d. net LOITERER'S HARVEST
SECRETS OF EARTH AND SEA LUCK OF THE YEAR
Illustrated. 8s. 6d. net. EVENTS AND EMBROIDERIES
A FRONDED ISLE
LINDRUM (Walter) A ROVER I WOULD BE
BILLIARDS. Illustrated. 6s. net. GIVING AND RECEIVING
LODGE (Sir Oliver) HER INFINITE VARIETY
MAN AND THE UNIYERSE ENCOUNTERS AND DIVERSIONS
7S. 6d. net ~nd 3S. 6d. net. Each 3$. 6d. net.
6 • ~ Me!lsrs. Methuen's Publications
LUCAS (E; V.l-'continued MAETERLINCK (Maurice)
CLOUD AND SILVER THE BLUE BIRD 6s. net.
A BOSWELL OF BAGHDAD Also, illustrated by F. CAYLEY
'TWIXT EAGLE AND DOVB ROBINSON. lOS. 6d. net.
TfIE PHANTOM JOURNAL OUR ETERNITY 6•• net.
ZIGZAGS IN FRANCE THE UNKNOWN GUEST 65. ne!.
TURNING THINGS OVEl! POEMS 55. net.
TRAVELLER'S LUCK THE WRACK OF THE STORlll
Each 00. net. 6,. net.
FRENCH LEAVES THIl BURGOMASTER OF STILEMONDE
Illustrated. 55. net. 55. net.
ROVING EAST AND ROVING WEST THE BETROTHAL 6,. net.
55. net. MOUNTAIN PATHS 61. net.
Mr. Punch's COUNTY SONGS THE GREAT SECRET 78. 6d. net.
Illustrated by E. Ii. SHEPARD. THE CLOUD TIL~T LIFTED and THE
105. 6d. net. POWER OF THE DEAD 78. 6d. net.
'THE MORE I SEE OF MEN •• .' MARY MAGDALENE 21. net.
Our OF A CLEAR SKY MARLOWE (Christopher)
IF DOGS COULD WRITE THE WORKS. In 6 volumes.
, •• AND SUCH SMALL DEER' General Editor, R. H. Case.
Each 3'. 6d. net. THE LIFE OF MARLOWE, by C. F.
THE PEKINESE NATIONAL ANTHEM TUCKER BROOKE, and DIDO.
Illustrated by PERSIS KIRMSE. Edited by the Same. 88. 6d. net.
IS. net. TAMBURLAINE, I AND II. Edited by
See also Lamb (Charles). U. M. Ellis-Fermor. lOS. 6d. net.
THE JEW OF MALTA and THE
LYND (Robert)
MASSACRE AT PARIS
IT'S A FINE WORLD Si. net; Edited by H. S. BENNEIT.
THE GREEN MAN
lOS. 6d. net.
THE PLEASURES OF IGNORANCE MASEFffiLD (John)
T:m; GOLDFISH
ON THE SPANISH MAIN 8s. 6d. net.
THE LITTLE ANGEL
THE BLUE LION
A SAILOR'S GARLAND 3S.
SEA LIFE IN NELSON'S TIME
net. "d.
THE PEAL OF BELLS
T:m; MONEy-Box
'5. 6d. net.
T:m; ORANGE TREE METHUEN (Sir A.)
Each 35. 6d. nct. AN ANTHOLOGY OF MODERN VERSE
232nd Thousand.
McDOUGALL (William) SmKESPEARE TO HARDY: An
AN INTRODUCfION TO SOCIAL
Anthology of English Lyrics
PSYCHOLOGY lOS. 6d. net.
28th Thousand.
NATIONAL WELFARE AND NATIONAL
Each, Cloth, 6s. net.
DECAY 6s. net. Leather, 75. 6d. net.
AN OUTLINE OF PSYCHOLOGY MILNE (A. A.)
lOS. 6d. net. By WAY OF INTRODUCTION
AN OUTLINE OF ABNORMAL PSYCHO- 6s. net.
LOGY ISS. net. TOAD OF TOAD HALL
BODY AND MIND 121. 6d. net. A Play founded 00 Kenneth
CHARACTER AND THE CONDUCT OF Grahame's • The Wind in the
LIFE lOS. 6d. net. Willows,' SS. net and 2S. 6d. net.
MODERN MATERIALISM AND EMERG- THOSE WERE THE DAYS: Collected
ENT EVOLUTION 75. 6d. net. Stories 'S.
6d. 11'01.
ETHICS AND SOME MODERN WORLD NOT THAT IT MATTERS
~ROBLEMS 7S. 6d. net. IF I MAY
THE SUNNY SIDE
MALLET (Sir C. E.) THE RED HOUSE MYSTERY
A HISTORY OF THE UNIVERSITY OF ONCE A WEEK
OXFORD THE HOLIDAY ROUND
In 3 vols. Each £1 IS. net. THE DAY'S PLAY
Messrs. Methuen's Publications 7

MILNE (A. A.)-continued NEUBURGER (AJ!>ert) •


MR. PIM PASSES By THE TECHNICAL ARTS AND SCIENCES
Each 3S. 6d. net. OF THE ANCIENTS
WHEN WE WERE VERY YOUNG Translated by H. L. BROSE.
21Ith Thousand. Illustrated. £2 2'. net.
WINNIE-THE-POOH OMAN (Sir Charles)
I I 8th Thousand. A HISTORY OF THE ART OF WAR IN
Now WE ARE SIX THE MIDDLE AGES, A.D. 378-1485
I I 9th Thousand. 2 vol.. Illustrated. £1 16s. net.
THE HOUSE AT POOH CoRNEll STUDIES IN THE NAPOLEONIC WARS
[osth Thousand. 8s. 6d. net.
Each illustrated by E. H. SHEPARD.
'S. 6d. net. Leather, lOS. 6d net. PERRY (W. J.)
THE CHRISTOPHER ROBIN STORY THE ORIGIN OF MAGIC AND
BOOK. Third Edition. RELIGION
Illustrated by E. H. SHEPARD. THE GROWTH OF CIVILIZATION
SS. net. Each 6s. net.
THE CHRISTOPHER ROBIN BIRTH- THE CHILDREN OF THE SUN
DAY BOOK £1 IS. net.
Illustrated by E. H. SHEpARD.
3S. 6d. net. PETRIE (Sir Flinders)
MILNE (A. A.) and FRASER-SIM- A HISTORY OF EGYPT
SON (H.) In 6 Volumes.
FOURTEEN SONGS FROM' WHEN WE Vol. I. FROM THE 1ST TO THE
"WERE VERY YOUNG' XVITH DYNASTY
Thirteenth Editidit. 'S. 6d. net. Eleventh EdiIion, Revised.
TEDDY BEAR AND OTHER SONGS [2S. net.
FROM 'WHEN WE WERE VERY Vol. II. THEXVIITHANDXVIIITH
YOUNG' 6d. net. ,s. DYNASTIES
THE KING'S BREAKFAST Seventh Edition, Revised. 9'. net.
Third Edition. 3" 6d. net. Vol. Ill. XIXTH TO XXXTH
SONGS FROM ' Now WE ARE SIX' DYNASTIES
Second Edition. 7s. 6d. net. Third Edition. 12S. net.
MORE 'VERY YOUNG' SONGS Vol. IV. EGYPT UNDER THE
'S. 6d. net. PTOLEMAIC DYNASTY
THE HUMS OF POOH 7s. 6d. net. By EoWYN BEVAN. ISS. net.J
Words by A. A. MILNE. Vol. V. EGYPT UNDER ROMAN RULE
Music by H. FRASER-SIMSON. By]. G. MILNE.
Decorations by E. H. SHEPARD. Third Edition, Revised. I2S. net.
Vol. VI?GYPT IN THE MIDDLE
MORTON (H. V.) AGES
THE HEART OF LONDON By STANLEY LANE POOLII.
35th Thousand. 3S. 6d. net. Fourth Edition. lOS. net.
Also with Scissor Cuts by L.
HUMMEL. 6,. net. PONSONBY OF SHULBREDE
THE SPELL OF LoNDON (Lord)
20th Thousand. ENGLISH DIARIES £1 IS. net.
THE NIGHTS OF LONDON MORE ENGLISH DIARIES
13th Thousand. Each 3s. 6d. net. US. 6d. net.
IN SEARCH OF ENGLAND SCOTTISH AND IRISH DIARIES
74th Thousand. lOS. bJ. net.
THE CALL OF ENGLAND
RUTTER (Frank)
31St Thousand.
EL GRECO. Illustrated. £1 100. net.
IN SEARCH OF SCOTLAND
96th Thousand SELLAR (W. C.) and YEATMAN
IN SEARCH OF IRELAND (R.J.)
40th Thousand. 1066 AND ALL THAT
IN SEARCH OF LONDON Jllustrated by JOHN REYNOLDS.
Each illustrated. 'S. 6d. net. 50th Thousand, S', net.
8 .Messrs. Methuen's Publications
STEVENSON (R. ,L.) UNDERHILL (Evelyn)
THE LllTI1llcl Edited by Sir SID- MYSTICISM. Revised Edition. 151. net.
NEY COLYIN. 4 Vols. Each 6,. I14L THB LIFE OF THB SPIRIT AND THE
STOW (G. W.) and BLEEK (Doro~
LIFE OF TO-DAY ,s. 6d. net.
thea F.)
ROCK P AlNTINGS IN SOUTH AFRICA
MAN AND THE SUPERNATURAL

CONCERNING THE INNER LIFE ~


,s. 6d. mt.
Illustrated. £,2 2$. net.
2$. mt.
SURTEES (R. S.) THE HOUSE OF THE SOUL 2$. net.
HANDLEY CROSS
MR. SPONGE'S ~PORTING TOUR VARDON {Harry}
ASK MAMMA Bow TO PLAy GOLF
MR. FACEY ROMFORD'S HOUNDS Illustrated. Nimteenth Edition.
PLAIN OR RINGLETs? St. net.
HILLINGDON HALL
FAch illustrated. ". 6d. net. WELD-BLUNDELL (Dom B.)
]ORROCKS'S JAUNTS AND JOLLITIBS SELF-DISCIPLINE AND HOLINESS
HAWBUCK GRANGE S" net.
Each illustrated. 6,. 114t.
TAYLOR (A. E.) WILDE (Oscar)
PLATO: THE MAN AND BIS WORK THE WORKS
[,1 Ii. 114t. In 16 Vol.. Each 6$. 6d. net.
PLATO: TIMlEUS AND CRITIA~ I. LoRD ARTHUR SAVILE'S CRIME
61. ng&. AND TIlE PORTRAIT OF MR. W. H.
ELE.\{ENTS OF MBTAPRYSlC'l II. THE DUCHFSS OF PADUA
12$. 6d. n.t, Ill. POEMS
IV. LADY WINDERMERE'S FAN
TILDEN (William T.) V. A WOMAN OF No IMPORTANCE
THE ART 'OF LAWN TENNIS VI. AN IDEAL HUSBAND
Revised Edition VII. THE IMPORTANCE OF BEING
SINGLES AND DOllBLES EARNEST
Each illustrated. 6s. net. VIII. A HOUSE OF POMEGRANATES
TIm COMMON SENSE OF LAWN IX. INTENTIONS
TENNIS X. DE PROFUNDiS AND PRISON
MATCH PLAY AND THE SPIN OF THE LETTERS
BALL XI. ESSAYS
Each illustrated. 51. net. XII. SALOME, A FLORENTINE
TRAGEDY, and LA SAINTB
TILESTON (Mary W.) COURTISANE .
DAILY STRENGTH FOR DMLY NEEDS XIV. SELEcrED PROSE OF OSCAR
Thirty-fourth Edition. 3$. 6d. mt. WILDE
India Paper. Leather. 6s. net. XV. ART AND DECORATION
TRAPP (Oswald Graf) XVI. FOR LoVE OF THE KING
THE ARMOURY OF THE CASTLE OF 5$. net.
CHURBURG XVII. VERA, OR THE N!.RILISTS
Translated by J. G. MANN.
Richly illustrated. WILLIAMSON (G. C.)
Limited to 40<> copies. THE BOOK OF FAMILLE ROSE
[,4 14'. Od. net. Richly Illustrated. £8 8r. ml.

Methuen & Co. Ltd., 36 Essex Street, London, W.C.2

You might also like