Stutervant - A History of Genetics
Stutervant - A History of Genetics
Stutervant - A History of Genetics
G ENETICS
A. H. STURTEVANT
Thomas Hunt Morgan
Professor of Biology, Emeritus
California Institute of Technology
A History of Genetics
1965, 1967 Alfred H. Sturtevant
2001 (Note from the Publishers and Introduction) Cold Spring Harbor Laboratory
Press and Electronic Scholarly Publishing Project
Published by Cold Spring Harbor Laboratory Press and Electronic Scholarly Publishing Project
Electronic Scholarly Publishing Project
ESP Foundations Reprint Series: Classical Genetics
Series Editor: Robert J. Robbins
Electronic Production: Ana Dos Santos, William Landram
Cold Spring Harbor Laboratory Press
Acquisition Editor: John R. Inglis
Production Editor: Mala Mazzullo
Cover Designers: Robert J. Robbins / Danny deBruin
Front Cover: Species of Drosophila, by Edith M. Wallace, Plate 3, from The North
American Species of Drosophila, by A. H. Sturtevant, 1921, Carnegie Institution of
Washington.
Back Cover: Sex-linked inheritance, Figure 38 from The Theory of the Gene, by
T. H. Morgan, 1926, Yale University Press.
Bibliographical Note: This CSHLP/ESP edition, first published in 2000, is a newly
typeset, unabridged version of A History of Genetics, by A. H. Sturtevant, based on
the second printing of the original 1965 book, published by Harper & Row in their
Modern Perspectives in Biology series. All footnotes were taken from the original
work. The figures have been redrawn for this edition.
Availability: A directory of Cold Spring Harbor Laboratory Press publications is
available at http://www.cshl.org/books/directory.htm. All Electronic Scholarly Publishing Project publications may be obtained online, from http://www.esp.org.
Web site: The full text of this book, along with other supporting materials, including
full-text copies of many of the works discussed by the author, may be obtained at the
books web site: http://www.esp.org/books/sturt/history.
Library of Congress Cataloging-in-Publication Data
Sturtevant, A. H. (Alfred Henry), 1891A history of genetics / A. H. Sturtevant
p. cm.
Previously published: New York: Harper & Row, 1965.
Includes bibliographical references and index.
ISBN 0-87969-607-9 (pbk.: alk. paper)
1. Genetics--History. I. Title
QH428 .S78 2000
576.5'09--dc21
10 9 8 7 6 5 4 3 2 1
Text reprinted with permission from the family of A. H. Sturtevant.
00-052394
C ONTENTS
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
vii
ix
xi
1
9
17
25
33
39
45
51
58
62
67
73
80
87
93
100
107
117
121
126
133
136
139
144
157
168
viii
A H I ST O RY
OF
G EN ET IC S
ROBERT J. ROBBINS
Electronic Scholarly Publishing Project
I NTRODUCTION
The reprinting of this classic book provides students with one of the few
authoritative, analytical works dealing with the early history of genetics.
Those of us who had the privilege of knowing and working with Sturtevant
benefited greatly from hearing first-hand his accounts of that history as he
knew it and, in many instances, experienced it. Fortunately, Sturtevant put it
all together in this book.
In his preface to the book, Sturtevant lists the persons that he knew personally and who were major players in the field, in addition to those who
occupied the famous fly room (Chapter 7) at Columbia University. As a result, much of the history is based on first-hand contacts as well as on a
scholarly and critical review of the literature of genetics and cytology.
Sturtevant was clearly present at the creation of modern genetics, if
dated from 1910 when Morgan commenced work on Drosophila. Of
Morgans three studentsSturtevant, Bridges, and MullerSturtevant was
ideally suited to write the history because of his remarkable memory, his
knowledge of almost all aspects of biology, and a keen analytical ability that
extended not only to his experimental work, but also to tracing the history of
the underlying ideas.
Sturtevant was a gifted writer and also an authority on many of the subjects he covers. While he was a sophomore in college, he deduced the linear
order of the genes. Later, he postulated the existence of inversions and duplications before they were verified cytologically. Sturtevant was especially
interested in how genes produce their effects and, consequently, was the father of a field now called developmental genetics. In this area, his style was
to analyze exceptions to the rule. In so doing, he identified the phenomenon
of position effect, in which the position of a gene (that of the Bar, and
double-Bar, eye mutations) can be shown to affect its function. He identified
the first clear case of a non-autonomously expressed gene, vermilion, mutants of which produce a vermilion, instead of the normal red, eye color.
This was an important exception to the rule that sex-linked mutants behaved
autonomously in gynandromorphs. How this led to the field of biochemical
genetics is explained in Chapter 16.
In the tradition of such biologists as Darwin, Galton, and Bateson and of
many of the early Mendelians, Sturtevant was an ardent evolutionist. He had
a seemingly inexhaustible knowledge of embryology, anatomy, morphology,
and taxonomy that served him well in treating evolutionary concepts historically, as described in Chapter 17. It is a wide-ranging chapter that covers
many topics, including the development of population genetics, the role of
ix
A H I ST O RY
OF
G EN ET IC S
gene mutations in evolution, and, prophetically, the conservation of biochemical pathways in major groups from bacteria to vertebrates. His own
experimental work, typically only briefly referred to, included his work on
interracial and interspecific hybrids in the genus Drosophila, and the demonstration that the genetic content of different species of that genus is
remarkably conserved, whether it be in the X chromosome or in each of the
specific autosomal arms. Sturtevant always had a healthy skepticism, surely
one of the most important qualities of a successful scientist. This is shown
by his doubt of the value of many laboratory experiments in population
genetics, on the basis that they cannot faithfully duplicate what really goes
on in the great out-of-doors.
It may come as a surprise to many students to realize how much opposition there was in some quarters to the early discoveries of the Morgan
school. Sturtevants account of such controversies is a recurrent theme of
this book, as it should be in a historical treatise.
Science has often been advanced by scientists who questioned existing
dogma and found it flawed. Or, conversely, such dogma has probably in
some cases slowed progress for years. Would advances in genetics have been
more rapid had there not been the virtually universal belief that genes were
proteins, or that development of an organism involved cytoplasmic rather
than nuclear heredity? Sturtevant does not waste space speculating about
such issues, but he does discuss several cases in which progress was held
back because of failure to develop a satisfactory terminology and
symbolism.
Sturtevant had a strong social consciousness that comes forth in Chapter
20. There he treats the history of human genetics, stressing the difficulties
and pitfalls that plague studies in this field. He devotes considerable space to
an objective and critical analysis of the so-called nature vs. nurture
question.
In the last chapter, Sturtevant discusses how discoveries in science and
particularly genetics tend to come about. He addresses in his typically analytical way the often-cited dictum: The time has to be ripe for a discovery to
be made and that when that time comes someone is bound to make the discovery. He concludes that this attitude greatly oversimplifies what generally
happens in science.
I believe Sturtevants writing of this book after his retirement was one
more intellectual exercise to stave off boredom. He had reduced his experimental work to an hour or so each day, and it must have been more difficult
to keep up with the expanding literature of the field. His book is clearly a
labor of love and his personality shines through every page.
July 2000
Pasadena, California
E. B. Lewis
A UTHOR S P RE FACE
The publication of Mendel's paper of 1866 is the outstanding event in
the history of genetics; but, as is well known, the paper was overlooked until
1900, when it was found. Its importance was then at once widely recognized.
These facts make the selection of topics for the early chapters of this book
almost automatic. What was the state of knowledge about heredity in Mendel's day; what sort of man was Mendel, and how did he come to make his
discovery; what happened between 1866 and 1900 to account for the different reaction to his results; how did his paper come to be found, and just what
was the immediate reaction?
These questions are discussed in the first four chapters. From that point
on, it has seemed more logical to treat the various topics separately rather
than to follow a more nearly chronological order. The attempt has been, in
each case, to trace the beginnings of a subject and to bring it down to the
work currently being done but not to discuss presently active fields of
work, since these are adequately covered in current books and reviews.
There is no definite terminal date, but work later than about 1950 is generally omitted or is referred to only briefly. In other fields the cutoff date is
even earlier than this.
For Chapters 1 and 3 I have relied largely on secondary sources such as
Sachs (1875), Zirkle (1935), Roberts (1929), and Wilson (1925). For the
period after 1900 I have read or reread much of the original literature and,
for general background, have been fortunate enough to have had some direct
personal contact with many of the people discussed including, among the
early workers, de Vries, Bateson, Johannsen, Wilson, Morgan, McClung,
East, Shull, Castle, Emerson, Davenport, Punnett, Nilsson-Ehle, Goldschmidt, and others. (I have seen Cunot, Baur, Sutton, and Saunders but
never really knew them.)
I am indebted to numerous colleagues who have read part or all of the
manuscript and have made constructive suggestions. Especially to be named
are Drs. N. H. Horowitz, E. B. Lewis, H. L. Roman, C. Stern, G. Hardin, and
C. Fulton. Much of the material has been presented in a series of lectures at
the California Institute of Technology and at the Universities of Washington,
Texas, and Wisconsin; numerous discussions with colleagues at these institutions have been very helpful.
A. H. STURTEVANT
August 1965
Pasadena, California
xi
C HAPTER 1
B EFORE M ENDEL
In discussing the history of a subject it is usual to begin with Aristotle and he forms a convenient starting point for genetics, though the
real beginnings, even of theoretical genetics, go farther back. As a matter
of fact, much of Aristotles discussion of the subject is contained in his
criticism of the earlier views of Hippocrates.
Hippocrates had developed a theory resembling that later proposed
by Darwin, who called it pangenesis. According to this view, each part
of the body produces something (called gemmules by Darwin) which is
then somehow collected in the semen or as we should now say, the
germ cells. These are the material basis of heredity, since they develop
into the characters of the offspring. The view was developed, both by
Hippocrates and by Darwin, largely to explain the supposed inheritance
of acquired characters. Aristotle devoted a long passage to criticism of
this hypothesis, which he discarded for several reasons. He pointed out
that individuals sometimes resemble remote ancestors rather than their
immediate parents (which is in fact one of the arguments used by Darwin
for, rather than against, pangenesis, since Darwin did not suppose that
the gemmules necessarily came to expression in the first generation and
did not suppose, as did Hippocrates, that they were released from the
parts of the body at the moment of copulation). Aristotle also pointed out
that peculiarities of hair and nails, and even of gait and other habits of
movement, may reappear in offspring, and that these are difficult to interpret in terms of a simple form of the hypothesis. Characters not yet
present in an individual may also be inherited such things as gray hair
or type of beard from a young father even before his beard or grayness
develops. More important, he pointed out that the effects of mutilations
or loss of parts, both in animals and in plants, are often not inherited.
Aristotle, like everyone else until much later, accepted the inheritance of
acquired characters; but he was nevertheless aware that there was no
simple one-to-one relation between the presence of a part in parents and
1
A H I ST O RY
OF
G EN ET IC S
its development in their offspring. His general conclusion was that what
is inherited is not characters themselves in any sense but only the potentiality of producing them. Today this sounds self-evident, but at that time
it was an important conclusion, which was not always fully understood,
even by the early Mendelians.
Aristotle was a naturalist and described many kinds of animals
some imaginary, others real and described in surprisingly accurate detail.
He knew about the mule and supposed that other animals were species
hybrids that the giraffe, for example, was a hybrid between the camel
and the leopard. According to him, in the dry country of Libya there are
few places where water is available; therefore many kinds of animals
congregate around the water holes. If they are somewhere near the same
size, and have similar gestation periods, they may cross; this is the basis
for the saying that something new is always coming from Libya.
Some later authorities disregarded Aristotles reasonable limitations
on what forms might be expected to cross, as in the conclusion that the
ostrich is a hybrid between the sparrow and the camel. There is a long
history of such supposed hybrids notably of the crossing between the
viper and the eel, and of the hybrid between the horse and the cow.
Zirkle records accounts of both of these as late as the seventeenth
century.
The knowledge of sex in animals goes far back before the beginnings
of history and was understood quite early even in plants at least in two
important food plants of the Near East, namely, the Smyrna fig and the
date palm, both of which are dioecious (that is to say, have separate male
and female trees). Zirkle shows that a special Near Eastern deity (the
cherub) was supposed to preside over the date pollination, and that representations of this deity can be traced back to about 1000 B.C. There is, in
fact, evidence that male and female trees were grown separately as early
as 2400 B.C.
The condition found in these two trees was definitely related to the
phenomenon of sex in animals, by Aristotle and others, but it was much
later that it was realized that plants in general have a sexual process.
That the higher plants do have sexual reproduction and that the
pollen represents the male element seems to have been first indicated
as an important generalization by Nehemiah Grew in 1676. A sound
experimental basis was first given by Camerarius (1691 to 1694).
From that time on, the view was rather generally accepted, especially
after Linnaeus presented more evidence and lent the prestige of his
name in 1760.
B EFORE M ENDEL
A H I ST O RY
OF
G EN ET IC S
B EFORE M ENDEL
A H I ST O RY
OF
G EN ET IC S
B EFORE M ENDEL
A H I ST O RY
OF
G EN ET IC S
ters to Ngeli, that Mendel himself also repeated this experiment (using
Mirabilis, as had Naudin) and found that a single grain was sufficient. He
did not publish this result, and does not refer to this approach to the
question in his paper.
C HAPTER 2
M ENDEL
Gregor Johann Mendel was born in 1822 in the village of Heinzendorf in northern Moravia then a part of Austria, now in Czechoslovakia, near the Polish border. The area had long been populated by
people of German and Czech ancestry, living side by side and presumably intermarrying. Mendels native tongue was the peculiar Silesian dialect of German; in later life he had to learn to speak Czech. He came of
peasant stock, and only by persistence and hard work was he able to get a
start in education. In 1843 he was admitted as a novice at the Augustinian monastery at Brnn; four years later he became a priest. He took an
examination for a teaching certificate in natural science and failed
(1850). It has been suggested that the examining board was biased because he was a priest or because his scientific views were unorthodox;
the plain fact seems to be that he was inadequately prepared. In order to
remedy this, he spent four terms, between 1851 and 1853, at the University of Vienna, where he studied physics, chemistry, mathematics, zoology, entomology, botany, and paleontology. In the first term he took
work in experimental physics under the famous Doppler and was for a
time, an assistant demonstrator in physics. He also had courses with
Ettinghausen, a mathematician and physicist, and with Redtenbacher, an
organic chemist both productive research men. We may surmise that
this background led to his use of quantitative and experimental methods
in biological work. Another of his professors at Vienna, Unger in botany,
was also an outstanding figure. Unger was one of the important men in
the development of the cell theory; he had demonstrated the antherozoids
of mosses and correctly interpreted them as the male generative cells,
and he had shown (in opposition to Schleiden) that the meristematic cells
of higher plants arise by division. In 1855 Unger published a book on the
anatomy and physiology of plants that is rated by Sachs as the best of its
time; in this book he made the first suggestion that the fluid content of
9
10
A H I ST O RY
OF
G EN ET IC S
animal cells and that of plant cells are essentially similar. Mendel was
thus in contact with at least two first-rate research scientists, and evidence of their influences upon him shows in his major paper.
Mendel returned to Brnn after the summer term of 1853 at Vienna.
At a meeting of the Vienna Zoological-Botanical Society in April, 1854,
his teacher Kollar read a letter from him, in which he discussed the pea
weevil (Bruchus pisi). In the summer of 1854, Mendel grew thirty-four
strains of peas; he tested them for constancy in 1855. In 1856 he began
the series of experiments that led to his paper, which was read to the
Brnn Society for Natural History in 1865 and was published in their
proceedings in 1866. Before discussing this paper and its consequences,
it will be well to describe some later events in Mendels life.
He was interested in honeybees and was an active member of the
local beekeepers society. He attempted to cross strains of bees, apparently without success. It has been suggested by Whiting and by Zirkle
that he probably knew of the work of Dzierzon on bees, and that Dzierzons description of segregation in the drone offspring of the hybrid
queen may have given Mendel the clue that led to his studies of peas. He
is also known to have kept mice, and Iltis and others have suggested that
he may have first worked out his results with them, but hesitated, as a
priest, to publish on mammalian genetics. These suggestions both seem
unlikely to me; there seems no reason to doubt Mendels own statement:
Experience of artificial fertilization, such as is effected with ornamental
plants in order to obtain new variations in color, has led to the experiments which will here be discussed. Perhaps the selection of peas as his
experimental material was due in part to Grtnerss account of the work
of Knight on peas.
Mendel was also interested in meteorology. At least as early as 1859,
he was the Brnn correspondent for Austrian regional reports, and he
continued to make daily records of rainfall, temperature, humidity, and
barometric pressure to the end of his life. He also kept records of sunspots and of the level of ground water as measured by the height of the
water in the monastery well. In 1870 a tornado passed over the monastery, and Mendel published a detailed account of it in the Proceedings of
the Brnn Society. He noted that the spiral motion was clockwise,
whereas the usual direction is counterclockwise. He gave many details,
and attempted a physical interpretation. This paper was stillborn, as was
his earlier one on peas, published in the same journal. According to Iltis,
a catalogue issued in 1917 lists 258 tornadoes observed in Europe but
does not include Mendels account.
M ENDEL
11
In 1868 Mendel was elected abbot of the Brnn monastery. This led
to administrative duties and, beginning in 1875, to a controversy with the
government on taxation of monastery property. It appears that he continued his meteorological and horticultural observations, but his productive
scientific work was finished about 1871. He died January 6, 1884.
Mendel sent a copy of his major paper to Ngeli, together with a letter in which he stated that he was continuing his experiments, using Hieracium, Cirsium, and Geum. Ngeli was professor of botany at Munich
and a major figure of his time in biology. He was also interested in heredity and was actively working on it. He completely failed to appreciate
Mendels work and made some rather pointless criticisms of it in his reply to Mendels letter. He did not refer to it in his publications. He was
greatly interested in Hieracium, however, which fact led to a correspondence with Mendel. Ngelis letters have been lost, but he kept some of
Mendels letters to him. Found among his papers, these were published
by Correns in 1905 (I have used the translation in The Birth of Genetics,
issued in 1950 as a supplement to Volume 35 of Genetics). There are ten
of these letters, written between 1866 and 1873, and they give a picture
of Mendels biological work during the period. Because of Ngelis interest, much of the account has to do with Hieracium, the subject of
Mendels only other published paper in genetics (published in 1870 in
the Proceedings of the Brnn Society for 1869; a translation may be
found in Batesons Mendels Principles, 1909).
The work on Hieracium must have been a great disappointment to
Mendel. He obtained several hybrids by dint of much hard work, and all
of them bred true. It is now known that this occurs because the seeds are
usually produced by apomixis, that is, they are purely maternal in origin
and arise without the intervention of meiosis or fertilization (Raunkir
1903, Ostenfeld 1904). In other words, this was the worst possible choice
of material for the study of segregation and recombination for reasons
that could not be guessed at the time.
It appears from Mendels letters to Ngeli that he was very actively
engaged in genetic studies on several other kinds of plants through 1870.
His experiments (previously mentioned) with single pollen grains of
Mirabilis were repeated in two different years with the same result. He
reports studies on Mirabilis, maize, and stocks. Of these three he says
Their hybrids behave exactly like those of Pisum. The character studied in stocks was hairiness; with respect to flower color in this plant, he
says the experiments had lasted six years and were being continued this
in 1870. He had grown 1500 specimens for the purpose in that year; his
difficulty arose from the mutiplicity of shades that were hard to separate.
12
A H I ST O RY
OF
G EN ET IC S
In Mirabilis he had seen and understood the intermediate color of a heterozygote and had made the appropriate tests to establish this interpretation. He also mentioned experiments with several other plants
Aquilegia, Linaria, Ipomoea, Cheiranthus, Tropaeolum, and Lychnis.
The picture that emerges is of a man very actively and effectively
experimenting, aware of the importance of his discovery, and testing and
extending it on a wide variety of forms. None of these results were published; it is difficult to suppose that his work would have been so completely ignored if he had presented this confirmatory evidence, even
though it was not enough to convince Ngeli.
This, in outline, is the man. I have tried to give an account of him in
order to form a basis for judging his paper how it came about that he
did the work, and what one is to think in view of the analysis by Fisher
that will be discussed. A fuller account of Mendel will be found in the
biography by Iltis.
There are a number of new procedures in Mendels work. He himself
said in the paper, . . . among all the numerous experiments made [by his
predecessors], not one has been carried out to such an extent and in such
a way as to make it possible to determine the number of different forms
under which the offspring of hybrids appear, or to arrange these forms
with certainty according to their separate generations, or definitely to
ascertain their statistical relations. One may agree with Batesons comment on this passage: It is to the clear conception of these three primary
necessities that the whole success of Mendels work is due. So far as I
know this conception was absolutely new in his day.
This was his experimental approach, but it was effective because he
also developed a simple interpretation of the ratios that he obtained and
then carried out direct and convincing experiments to test this hypothesis. The paper must be read to be appreciated. As has often been observed, it is difficult to see how the experiments could have been carried
out more efficiently than they were.
As Fisher (1936) puts it, it is as though Mendel knew the answer before he started, and was producing a demonstration. Fisher has attempted
to reconstruct the experiments as carried out year by year, knowing the
garden space available and the number of years involved.* He concludes
* Fisher's dates are wrong. He gives them as 1857 to 1864, but it is clear from
Mendel's letters to Ngeli that the final year was 1863. Fisher includes the two years of
preliminary testing in the eight years that Mendel says the experiments lasted. I have
interpreted the statement to mean that these two years preceded the eight years of actual
experiments an interpretation also given by Yule (1902). Fisher's interpretation may
be right, but if Yule and I are right there are two more years available and Fisher's
year-by-year reconstruction needs revision. It may also be pointed out that Mendel
M ENDEL
13
that the crosses were carried out in the order in which they are described.
He also points out several other aspects of the work that seem significant.
For example, in testing F2 individuals to distinguish homozygous dominants from heterozygotes, Mendel must have had a much larger number
of seeds illustrating the 3 : 1 ratio than those recorded in F2; but he did
not report these numbers (if he even troubled to count them). Evidently
he felt that larger numbers were of no importance.
The most serious matter discussed by Fisher is that Mendels ratios
are consistently closer to expectation than sampling theory would lead
one to expect. For yellow vs. green seeds, his F2 numbers were
6022 : 2001 a deviation of 5 (from 3 : 1), whereas a deviation of 26 or
more would be expected in half of a large number of trials, each including 8023 seeds. Fisher shows that this same extremely close fit runs
through all Mendels data. He calculates that, taking the whole series, the
chance of getting as close a fit to expectation is only .00007, that is, in
only 1 trial of 14,000 would one expect so close an agreement with
expectation.
If this were all, one might not be too disturbed, for it is possible to
question the logic of the argument that a fit is too close to expectation. If
I report that I tossed 1000 coins and got exactly 500 heads and 500 tails,
a statistician will raise his eyebrows, though this is the most probable
exactly specified result. If I report 480 heads and 520 tails, the statistician will say that is about what one would expect though this result is
less probable than the 500 : 500 one. He will arrive at this by adding the
probabilities for all results between 480 : 520 and 520 : 480, whereas for
the exact agreement he will consider only the probability of 500 : 500
itself. If now I report that I tossed 1000 coins ten times, and got
500 : 500 every time, our statistician will surely conclude that I am lying,
though this is the most probable result thus exactly specified. The argument comes perilously close to saying that no such experiment can be
carried out, since every single exactly specified result has a vanishingly
small probability of occurring.
In the present case, however, it appears that in one series of experiprobably used some time and garden space in the later years of this period to carry out
the experiments with beans and hawkweeds and with the several other plants referred to
in the letters to Ngeli.
Fisher also quotes extensively from a paper by Ngeli (1865), and concludes that
it is difficult to suppose that these remarks were not intended to discourage Mendel
personally, without drawing attention to his researches. But this paper of Ngeli's was
published before Mendel's clearly before Ngeli could have known anything about
Mendel's work!
14
A H I ST O RY
OF
G EN ET IC S
M ENDEL
15
Source
Yellow
Green
Total
Dev.
from
3 in 4
Prob.
Error
Dev.
P.E.
Mendel, 1866
Correns, 1900
Tschermak, 1900
Hurst, 1904
Bateson, 1905
Lock, 1905
Darbishire, 1909
Winge, 1924
Total
6,022
1,394
3,580
1,310
11,902
1,438
109,060
19,195
153,902
2,001
453
1,190
445
3,903
514
36,186
6,553
51,245
8,023
1,847
4,770
1,775
15,806
1,952
145,246
25,748
205,147
+ .0024
+ .0189
+ .0021
.0142
+ .0123
.0533
+ .0035
.0180
+ .0008
.0130
.18
.0272
.70
.0169
.12
.0279
.51
.0093
1.32
.0264
2.04
.0030
1.16
.0125
1.44
.0038
.21
16
A H I ST O RY
OF
G EN ET IC S
C HAPTER 3
1866
TO
1900
The period between the publication of Mendels paper and its rescue
in 1900 from oblivion was dominated by the development of the theory
of evolution and its implications. So far as heredity was concerned, it
was largely a period of the production of theories. There were, however,
several real advances which helped to make Mendels results acceptable.
Here we may mention the germ-plasm theory with its emphasis on the
effects of germinal material on the body rather than the reverse; the resulting challenge of the inheritance of acquired characters; the striking
increase of knowledge of the cytological details of fertilization and cell
division; and the increasing emphasis on the importance of discontinuous
variation. This chapter will be concerned chiefly with these topics.
The outstanding figure of the time was August Weismann (1834
1914), who was professor of zoology at Freiburg for many years. From
1862 to 1864 he published several papers on the embryology of Diptera,
and these seem to have led to much of his later theoretical work.
In these flies, the so-called pole cells are set aside in early cleavage
divisions, subsequently to develop into the germ cells. Their early separation from the somatic cells, and their relatively independent development, seem to have suggested the germ-plasm theory to Weismann
(1883), although he was also aware of a similar idea expressed by Nussbaum. According to this scheme, the germ line is the continuous element,
and the successive bodies of higher animals and plants are side branches
budded off from it, generation after generation. This is, of course, only a
way of looking at familiar facts, since Weismann recognized that in the
higher plants and in many animals the visible distinctness of the germ
line only appears late in development and, in fact, that many cells that
will not normally give rise to germ cells still retain the potentiality of
doing so. Nevertheless, the idea was a fruitful one, since it led to an emphasis on the effects of the hereditary material on the soma and to a
17
18
A H I ST O RY
OF
G EN ET IC S
1866 1900
19
ture of the soma and the simplex condition of the germ cells (a relation
deduced by Mendel from his genetic results but not recognized by his
contemporaries). It had, however, been anticipated by Weismann, who
supposed that the function of the polar body divisions in the egg was to
prevent an indefinite accumulation of ancestral hereditary units and predicted that a similar reduction would be found in the formation of the
sperm.
In 1883 there appeared a remarkable essay by Roux, in which he argued that the linear structure of the chromosomes and their point-bypoint division into equal longitudinal halves were such striking and
widespread phenomena that they must have some selective value. This,
he suggested, lay in their effectiveness in assuring that each daughter cell
received the same complement of chromosomal material. He saw this as
a strong argument in favor of identifying the chromosomes as the bearers
of the units of heredity. These units were also here first specified as being arranged in a linear series the visible slender strands of the dividing
chromosomes.
Roux applied these ideas to the cleavage divisions of the fertilized
egg of the frog. He was the father of experimental embryology and had
carried out experiments which he believed had shown that the two cells
arising from the first division are equivalent, but that the second division
leads to differences in the potentialities of the daughter cells.* He therefore concluded, in the 1883 essay, that at the second division the process
of mitosis does not lead to exactly equal complements of hereditary units
in the daughter cells. This was the beginning of the hypothesis that differentiation is due to somatic segregation the sorting out of hereditary
elements at somatic cell divisions.
These ideas were at once adopted by Weismann, who elaborated
them into an intricate theory of heredity and development. According to
this scheme, the chromosomes are the bearers of the hereditary material.
Weismann supposed that each chromosome remains intact in successive
generations, and is simply passed on through the germ line from generation to generation. Since an individual may resemble several different
ancestors in one respect or another, he concluded that each chromosome
carries all the hereditary elements necessary to produce a whole individual. The different chromosomes of an individual may have been derived
from many different ancestral lines, and they therefore differ among
themselves. Each is potentially able to determine the characteristics of a
whole organism, but in the development of a particular part, only one
*
Later experiments have not borne out this conclusion about the second division.
20
A H I ST O RY
OF
G EN ET IC S
chromosome is effective at any given time and place. There is, in a sense,
a competition between the various chromosomes, and the nature of each
characteristic depends on the outcome of this competition at each critical
time and place in the developing embryo. Each chromosome was supposed to be made up of smaller units, and these in turn of still smaller
subunits. These were distributed unequally at somatic divisions, forming
the basis of differentiation.
This theory was elaborated in great detail and was widely known and
discussed, but it was not accepted in detail, because it was so hypothetical and seemed to offer so little basis for experimental testing. There can
be no question of the importance and widespread influence of much of
Weismanns work, but the elaboration of his scheme of heredity and development led to widespread resistance to even the sounder parts of his
interpretations. As Wilson expressed it in 1900:
Weismanns . . . theories . . . have given rise to the most eagerly
contested controversies of the post-Darwinian period, and, whether
they are to stand or fall, have played a most important part in the progress of science. For, aside from the truth or error of his special theories, it has been Weismanns great service to place the keystone
between the work of the evolutionists and that of the cytologists, and
thus to bring the cell-theory and the evolution theory into organic
connection.
1866 1900
21
This dissatisfaction with the then-current views led Bateson, Korschinsky, and de Vries to lay great emphasis on the importance of discontinuous variations. As we can now see, they overemphasized the
distinction between the two kinds of variation; but the immediate result
was to focus attention on sharply separable variations, and these were
more easily susceptible of exact study. Again, it was no accident that de
Vries was one of the discoverers of Mendels paper, and that Bateson
was perhaps the most important of the early advocates of the Mendelian
approach.
During the period in question, a quite different approach to the study
of heredity was developed by Francis Galton. Galton, who was a cousin
of Darwin, had carried out an experiment to test the theory of pangenesis.
He performed extensive blood transfusions between different strains of
rabbits and found no effects on their descendants in either the first or the
second generation. Darwin admitted that he would have expected effects
but felt that his gemmules were not necessarily to be expected in the
blood, since the theory was supposed to apply even to organisms without
a circulatory system. Galton agreed that the experiment was not decisive.
22
A H I ST O RY
OF
G EN ET IC S
Galtons theoretical contribution arose from his feeling for the importance of quantitative study. He felt that almost anything could be
measured. He attempted to develop a quantitative scale for beauty; and
he carried out a study on the effectiveness of prayer, by examining mortality rates for crowned heads (whose subjects prayed for their health),
and by comparing the frequencies of shipwreck for vessels that did and
did not carry missionaries. Like Mendel, he also studied meteorology.
He developed the idea of correlation as a result of tabulating the relationship between the height of parents and that of offspring in human
families. He saw what was needed in geometrical terms and referred the
algebraic problem to the mathematician Dickson, who then produced the
regression coefficient.* Galton used this to give a simple numerical value
for the degree of resemblance between parents and offspring, thus initiating a whole field of study. He tabulated a large series of data on the
colors of pedigreed basset hounds, and based his Law of Ancestral Inheritance on the results. These results showed that, on the average, an
individual inherits of his characteristics from each parent, 116 from each
grandparent, 1 64 from each great-grandparent, and so on. This ingenious
approach was followed by many of his successors, but failed to give the
hoped-for insight into the mechanisms involved. As it happened (through
no fault of Galtons), it led to a long and bitter controversy that wasted
much time and printers ink in the early years of this century (Chapter 9).
The question has often been raised: Would any biologist have appreciated Mendels work if he had seen the paper before 1900? My own
candidate for the most likely person to have understood it is Galton, because of his interest in discontinuous variation, his mathematical turn of
mind, and his acceptance of Weismanns view that the hereditary potentialities of an individual must be halved in each germ cell.
One of the eager controversialists referred to by Wilson was
Haacke, who published a series of anti-Weismann papers between 1893
and 1897. These papers, which have been overlooked by many of the
more recent authorities (but not by all see Correns, 1922), contain the
nearest approach to the Mendelian interpretation before the rediscovery.
They make difficult reading, because the results and conclusions are so
buried in a mass of polemics.
Haacke crossed normal albino mice with waltzing mice that were
colored. In F1 he got only colored normals, and in F2 (which he called the
third generation, the parent strains being considered the first) he rec*
1866 1900
23
24
A H I ST O RY
OF
G EN ET IC S
reason is their use to support the unpopular (and erroneous) idea of the
genetic importance of the centrosomes. Finally, the failure to give actual
counts made the data seem as speculative as the discussion in which they
are imbedded.
mice to study the inheritance of coat colors, instead of cutting off the tails of his
unfortunate mice and those of their children and of their childrens children unto the
twentieth generation.
C HAPTER 4
T HE R EDISCOVERY
To Bateson and to de Vries, the logical approach to the study of
heredity seemed to be the study of variation, which was then to be followed by the study of the transmission of variations. As the event
showed, the effective approach was the reverse of this, since the origin of
variability could begin to be analyzed only after the nature of segregation
and recombination was understood.
By the end of the century both men felt that the time had come to
begin a serious study of the inheritance of discontinuous variations. In
1899 Bateson published an analysis of what was needed, which is remarkable, among other things, for the statement If the parents differ in
several characters, the offspring must be examined statistically, and marshalled, as it is called, in respect to each of those characters separately.
Here was, clearly, a man whose mind was ready to appreciate the Mendelian approach.
The story of the finding of Mendels paper and of the confirming of
his results in 1900 has often been told perhaps most fully by Roberts
(1929).
Mendel had forty reprints of his paper. He sent copies to Ngeli and
to Kerner, professors of botany at Munich and at Innsbrck, respectively,
and both interested in plant hybrids. It is not known what happened to the
other thirty-eight copies; after Kerners death, his copy was found in his
library with the pages uncut.* As was pointed out earlier, Ngeli did reply but did not appreciate the work or refer to it in print. The journal was
perhaps rather obscure, but the Brnn Society had a considerable exchange list, and its Proceedings were sent to more than 120 libraries.
According to Bateson, there were at least two copies in London. Only
* As will appear below, a third reprint was in the library of the Dutch botanist
Beijerinck. I have received, through the kindness of Dr. H. Gloor and Dr. F. Bianchi, a
photostat of the cover and first page of this reprint; there is no indication of how or
when Beijerinck acquired it, or to whom Mendel sent it.
25
26
A H I ST O RY
OF
G EN ET IC S
four printed references to the paper before 1900 are known, however,
other than a listing in the Royal Society Catalogue of scientific papers.
Hoffmann (1869) published an account of experiments with beans, in
which Mendels paper is referred to without any indication of its nature.
Focke (1881) published a rather extensive account of the literature of
plant hybridization, in which he referred to Mendels paper under the
heading Pisum. He failed to appreciate or even to understand the work,
but he did state that Mendel believed that he found constant numerical
relationships between the types a statement that ultimately led to the
paper being found, as will be discussed.
The third reference was by L. H. Bailey (1895), who copied Fockes
statement without having himself seen Mendels paper; this was the
source that led de Vries to Mendel, according to one account. Finally,
Mendel was listed, without comment, as a plant hybridizer by Romanes
in the ninth edition of the Encyclopaedia Britannica (18811895) evidently again following Focke.
Hugo de Vries (18481935) was born in Holland. His university
training was largely in Germany, where he studied plant physiology with
Sachs. In 1871 he became a lecturer at the University of Amsterdam and,
from 1881 until his retirement, was a professor there. His early work was
on local floras, the microorganisms in water supplies, and the turgor of
plant cells. In the latter field, he carried out a beautiful series of quantitative studies of the effects of the concentrations of various salts on
plasmolysis. These results were of importance in the development (by
Arrhenius and vant Hoff) of the ionic theory of the osmotic properties of
solutions of electrolytes. His Intracellular Pangenesis (1889) has been
described in Chapter 3; his work on mutation will be discussed in Chapters 10 and 11.
De Vries published three papers on Mendelism in 1900, one of which
has, for the most part, been overlooked. The first was read by G. Bonnier
before the Paris Academy of Sciences on March 26 and was published in
the Academys Comptes Rendus. A reprint of this paper was received by
Correns on April 21. Another paper by de Vries is dated Amsterdam,
March 19, 1900, and was published in the Revue gnral de botanique,
which was edited by Bonnier. It seems likely, then, that these two French
manuscripts were sent to Bonnier at the same time. The third paper, in
German, was received by the editor of the journal (Berichte der
deutschen botanischen Gesellschaft) in Berlin on March 14 and was
published April 25. These dates are of some interest because the brief
note in the Comptes Rendus, the first to be published, does not mention
Mendel, though it uses some of his terminology. The Revue gnral pa-
T HE R EDISCOVERY
27
per is the one that is rarely cited. It is longer and does mention Mendel
though only on the last page, where is also an added footnote referring to
the Berichte paper and to the papers by Correns and by Tschermak,
which did not appear until May (apparently this paper was published in
July). The reference to Mendel on this page may be translated as follows:
This law is not new. It was stated more than thirty years ago, for a particular case (the garden pea). Gregor Mendel formulated it in a memoir
entitled Versuche ber Pflanzenhybriden in the Proceedings of the
Brnn Society. Mendel has there shown the results not only for monohybrids but also for dihybrids.
This memoir, very beautiful for its time, has been misunderstood
and then forgotten.
In the Berichte paper, the second to be published but evidently the
first to be submitted, there is much the same material as in the longer
French one, but Mendel is mentioned in several places in the text and is
given full credit for his discovery.
In a letter received and published by Roberts, de Vries later stated
that he had worked out the Mendelian scheme for himself, and was then
led to Mendels paper by reading Baileys copy of Fockes reference. In
1954, nineteen years after the death of de Vries, his student and successor Stomps reported that de Vries had told him that he learned of Mendels work through receiving a reprint of the 1866 paper from Beijerinck,
with a letter saying that he might be interested in it. This reprint is still in
the Amsterdam laboratory, as has been stated.
There is a persistent and widespread story to the effect that de Vries
at first intended to suppress any reference to Mendel and changed his
mind only when he found that Correns (or Tschermak) was going to refer
to him. This is based on the failure to refer to Mendel in the Comptes
Rendus paper, the first to be published and, one may add, also in the
Revue gnral paper until the last page that was, at least in part, added
some months later. This view can be maintained only if it is supposed
that the Berichte paper was extensively altered in proof a suggestion
that gets some support from the fact that nine of the twenty-two errata
listed at the end of the volume concern just the pages that would have
had to be altered. These errata are rather minor, but they do make one
wonder if the printer was confused by extensive alterations in the proofs.
A careful comparison of the available dates, however, makes it seem
impossible that such changes could have been a result of a letter from
Correns after he had seen the Comptes Rendus paper, and very unlikely
also that a letter from Tschermak could have been involved. Both of
these men have stated (Roberts) that they learned that de Vries had the
28
A H I ST O RY
OF
G EN ET IC S
T HE R EDISCOVERY
29
Correns reported in detail on his work with peas, in a paper that appeared in May 1900, after he had seen de Vries account. He fully confirmed Mendel* and said that he had observed the same kind of results
with maize; these results were published in full in 1901. He disagreed
with de Vries in that he thought there were cases that did not conform to
the Mendelian scheme. The only one described in any detail, having to
do with the color of the seed coat in peas, seems to have involved the
carrying of a dominant gene for color pattern in a plant which also had a
recesssive that prevented all color, with the result that the F1 did not resemble either parent. This type of situation later became familiar but
seemed then to contradict the Mendelian scheme.
Erich von Tschermak (18711962) was a grandson of Fenzl, under
whom Mendel studied systematic botany and microscopy at Vienna.
Tschermak was trained at Halle, where he received his doctorate in 1895.
His interests were in practical plant breeding, and this led to studies (at
Ghent and Vienna) on the effects of crossing and inbreeding on vegetative vigor, following the work of Darwin. In this connection he made
crosses of peas and raised F2, noting the 3 : 1 ratios and also the 1 : 1 on
back-crossing to the recessive parent. He later wrote (to Roberts) that he
had realized the significance of this result before he found Mendels paper (through the reference by Focke), but since he had reared only two
generations when he published his accounts, he cannot have known that
the recessives bred true or that there were two classes of individuals in F2
that had the dominant character. He published two papers on the subject
in 1900. Of these, the first is much less clearly indicative of a real understanding of the situation than is the second, which was written after he
had seen the de Vries and Correns papers in the Berichte.
William Bateson (18611926) was trained as a zoologist at Cambridge. He was influenced by Sedgwick, by F. Balfour, and by his contemporary, Weldon. The summers of 1883 and 1884 were spent at
Hampton, Virginia, and Beaufort, North Carolina, studying the embryology of Balanoglossus under W. K. Brooks. Bateson has recorded that it
was Brooks who gave him the idea that heredity is a subject worth
studying for itself. In passing, it may be remembered that Brooks also
influenced the history of genetics through the fact that both E. B. Wilson
and T. H. Morgan were trained by him.
* This confirmation included extensive tests over several generations, showing that
extracted homozygotes bred true. This was perhaps the one of Mendels observations
that was hardest to accept at the time. We know that Ngeli balked at it, and that as late
as 1910, Morgan tried to explain it away.
30
A H I ST O RY
OF
G EN ET IC S
Batesons Materials for the Study of Variation (1894) and his outline
of what was needed in the study of heredity (1899) have been discussed
in Chapter 3 and earlier in this chapter. In May, 1900, he read a paper
before the Royal Horticultural Society in London, in which he described
the work of Mendel and its confirmation by de Vries. According to Mrs.
Bateson (1928), he first learned of Mendels work on a train, while going
from Cambridge to London to deliver this paper, and was so impressed
by it that he immediately incorporated it into his lecture.
Bateson at once became the most active proponent of the new approach and developed a very active group of workers at Cambridge, including, in the early years of this century, Saunders, Punnett, Durham,
Marryatt, and others. Mendelian studies were actively pursued in Germany by Correns, in Austria by Tschermak, in France by Cunot, and in
the United States by Castle and Davenport. These workers and others
soon built up a great mass of data and laid the foundations for later developments.
There were two immediate problems: How widespread are the Mendelian phenomena, and what is the interpretation of the so-called compound characters? Another question, that now seems less important,
concerned the generality of the phenomenon of dominance.
Mendels work established his principles for peas and beans; they
were confirmed for peas by Correns and by Tschermak. In his 1900 papers, de Vries showed that the principles applied to about a dozen widely
different genera of seed plants, including one monocotyledon (maize,
which Correns confirmed). There was thus clear evidence for the general
applicability to angiosperms. Correns suggested in 1901 that the principles applied to animals, citing a number of experiments from the earlier
literature that appeared to be Mendelian. That the principles do apply to
animals was definitely shown in 1902, independently by Cunot (mice)
and by Bateson (fowl). It was therefore concluded that the same general
scheme must apply to all higher animals and plants; the later applications
to invertebrates and to lower plants were, when they were made, interesting chiefly because they offered means of studying new kinds of
problems.
The other important question during the early development was that
of the nature of compound characters or, as we should now say,
cases where more than one pair of genes affect the same character.
Mendel reported a cross with beans, using as parents a strain with
white flowers and one with colored flowers. F1 was colored, but the cross
was between quite different species, and these F1 plants were only
slightly fertile. Mendel obtained a total of 31 F2 plants, of which only
T HE R EDISCOVERY
31
one had white flowers. He suggested, tentatively, that there were two
independent dominant genes, A1 and A2 either of which alone would give
colored flowers. The expectation in F2 would then be 15 colored : 1
white; but because of the small number of plants available he did not
urge this interpretation. In 1902, Bateson criticized this suggestion, since
he felt that A1 and A2 would be expected to behave as alleles; that is to
say, he was still thinking in terms of a single gene for each visible character. A similar point of view is found in the 1900 paper by Correns, as
has been pointed out, in his argument that the Mendelian principles could
not be general; the same view later led Cunot to lay no emphasis on his
demonstration of multiple alleles (see Chapter 8).
Several examples of a related sort soon turned up first in the case
of flower color in the sweet pea, where a cross of two whites gave colored flowers in F1, and in F2, 9 colored : 7 white (Bateson and Punnett).
Tests of individual F2 plants showed that this occurred because color
development requires the presence of both of two independent dominants. The F2 ratio is 9 : 3 : 3 : 1, with the last three classes indistinguishable, rather than the first three, as in Mendels beans. With this
cleared up (by Bateson and Punnett), it was not difficult to interpret the
9 : 3 : 4 cases that were soon found.
The case of combs in fowl was puzzling at first. Here the familiar
rose comb was soon shown to be dominant to single, and pea was
also shown to be dominant to single. When rose was crossed to pea, a
new type, called walnut, appeared; this did not seem to be, structurally,
a combination of rose and pea. The relations became clear when it was
found that the F2 ratio is 9 walnut : 3 rose : 3 pea : 1 single (again by
Bateson and Punnett) for Mendel had already shown that this ratio
meant that two independent pairs of genes were segregating.
The occasional occurrences of cases in which the heterozygote is
intermediate, that is, strict dominance is absent, was indicated by Correns
in a footnote added to his 1900 paper in proof. This had been seen and
understood, with the appropriate tests carried out, in Mirabilis by Mendel, according to his letters to Ngeli, and is implied in his account of
flowering time in peas in the 1866 paper. The Mirabilis case and that of
the Andalusian fowl became clear very early in the period after 1900, and
as a result, the phenomenon of dominance was recognized as of only
secondary importance.
New terminology was needed. Many of the now-familiar terms were
introduced by Bateson such as genetics, for the subject itself, and zygote, for the individual that develops from the fertilized egg, as well as
for the fertilized egg itself (which was the older usage). Homozygote,
32
A H I ST O RY
OF
G EN ET IC S
* The confusion existing may be illustrated further by the fact that Haacke (Chapter
3) considered the parental generation as the first and referred to what we now call F1
and F2 as the second and third generations, respectively.
C HAPTER 5
G ENES
AND
C HROMOSOMES
For many years the standard authority on the chromosomes was Wilsons The Cell in Development and Inheritance. The second edition of
this work was published in 1900; it gives a full account of the state of
knowledge and of current theories about chromosomes at the time of the
discovery of Mendels paper.
The constancy of chromosome number for a species was known, and
it was known that this number was usually even, equal numbers coming
from the egg and from the sperm. It was known that each chromosome
divides longitudinally at each somatic division, and that this division is
initiated by an equal division of each visible granule along the length of
the chromosome. It was also known that the reduction in chromosome
number is accomplished by the last two divisions before the production
of the mature gametes (in animals) or gametophytes (in plants). Further,
it was generally supposed that the chromosomes are the bearers of the
essential hereditary material.
There were, however, a number of things, now part of common biological knowledge, that were not known. It was generally supposed that,
when the chromosomes reappear at the end of the resting stage, they first
do so as a single continuous thread, or spireme, which then breaks into
the number of chromosomes characteristic for the particular species. It
had been postulated by Rabl and by Boveri that the chromosomes do not
lose their individuality at the close of division, but persist in the reticulum of the resting nucleus. Although Weismann adopted this view, Wilson felt that it was far from proved. The details of chromosome reduction
at meiosis were not at all clear, chiefly because the two-by-two pairing in
meiotic prophase was not recognized. The whole idea of definite pairs of
chromosomes was missing. It was not recognized that there are different
kinds of chromosomes in a single cell. In short, one chromosome was
33
34
A H I ST O RY
OF
G EN ET IC S
G ENES
AND
C HROMOSOMES
35
(though small) embryo. Boveri found that if the three or four cells produced by the first cleavage of dispermic eggs were separated, they would
sometimes develop into normal embryos but never would all of the
cells from any one egg do so. He showed that this result could not be
explained by the number of chromosomes present, since in cells that
failed to develop this was often greater than the haploid number and a
single complete haploid set was already known, from other experiments,
to be enough for normal development. The results can only be explained
on the assumption that the chromosomes differ in their effects on development, and that a cell will not give rise to a normal embryo unless it has
at least one complete haploid set of chromosomes.
The paper has a footnote, in which Boveri points out that he, like
Weismann, had previously supposed that the chromosomes of an individual were equivalent one to another. This view he now finds untenable,
and therefore Weismanns interpretation of chromosome reduction at
meiosis must now be revised. This same footnote contains a statement
that may be freely translated thus: I shall consider in another place these
and related problems, such as the connection with the results of the botanists on the behavior of hybrids and their offspring. Before this further
discussion appeared, the whole matter was clearly analyzed by Sutton,
but there can be no doubt that Boveri was near to the solution.
In 1900 Correns had already raised the question of where the Mendelian segregation occurs and had discussed it in several papers, the fullest
account being in 1902. He knew, from his experiments with maize hybrids,
that the embryo and the endosperm of a given seed are alike. It follows that
the three maternal nuclei involved in the double fertilization are alike, as
are the two from the pollen. He concluded that segregation is accomplished
at latest by the time the megaspore is produced, and that it occurs in the
anther at some time before the final division that produces the two sperm
nuclei in a single pollen grain. On the other hand, the earliest time it can be
supposed to occur is after the sexual organs are formed, as both the ovaries
and the anthers of heterozygous plants produce both kinds of gametes.
Since the ratio of dominant to recessive gametes is so very close to 1 : 1,
he concluded that segregation must come very late in the development in
order to avoid chance (or selective) differences in multiplication of the
products. He was, of course, also aware that the last two sporophyte divisions (that is, the meiotic ones) are of a different type and lead to haploid
nuclei. He therefore concluded that segregation occurs in these two divisions in the ovule.
In the anther he recognized that the same reasoning would lead to the
view that the meiotic divisions were also those that produce segregation.
36
A H I ST O RY
OF
G EN ET IC S
This should mean that the pollen grains of a heterozygote were of two
kinds, in equal numbers. He tested this deduction. He found two flower
colors in Epilobium that gave Mendelian results and were associated
with differences in the color of the cell sap of the pollen grains. He examined the pollen of the heterozygote and found it to be uniform in
color. An exactly similar result was found in one of the poppies. He concluded that segregation had not yet occurred, and that it must therefore
occur in the first pollen division, which separates the tube nucleus from
the generative nucleus the latter then dividing to produce the two
sperm nuclei. Strasburger made the suggestion (which we now know to
be correct) that the color of the pollen grains is determined by the composition of the plant that produced them, not by the gene content they
have just acquired. Correns admitted this possibility but argued, reasonably, that this was a special hypothesis made up to save another hypothesis, and he preferred to avoid such a procedure.
In this same paper, Correns discussed the relation of the genes to the
chromosomes. He supposed that the genes were carried by the chromosomes, and he drew a diagram that looks very much like the beads-on-astring type that later became familiar. One string was labelled A, B, C,
and so on; on another, closely apposed string a was placed opposite A, b
opposite B, and so on. But this was supposed to represent a single mitotic
chromosome, which divided in the plane of the paper at each division,
except at the time of segregation, when it divided at a right angle to the
paper to yield strands ABC and abc. He also figured what looks now like
crossing over, supposing that the pairs of genes could rotate about the
long axis of the chromosome, so that ABc and abC (or AbC and aBc, and
other combinations) could be produced. This was to explain independent
assortment (Mendels second law).
While this scheme related the genes to the chromosomes, it was
wrong in many critical points. There was no explanation of how a single
mitotic chromosome came to have maternal and paternal halves; the segregation division was not identified with meiosis; and independent assortment was not related to independent segregation of nonhomologous
pairs of chromosomes.
Cannon pointed out in 1902 that there is a close parallelism between
Mendelian segregation and chromosome reduction and concluded that
this is because the genes are in the chromosomes. Like Correns, he seems
to have thought that all the paternal chromosomes went to one pole at
meiosis and all the maternal ones to the other; he offered no explanation
for independent assortment. De Vries, in 1903, also discussed these
G ENES
AND
C HROMOSOMES
37
questions, and he seems likewise to have supposed that the paternal and
maternal chromosomes were separated as groups at meiosis. He accounted for independent assortment by postulating that, at meiosis, the
members of individual gene pairs could be freely exchanged between
homologous chromosomes (as Correns had supposed) and so would segregate at random.
Guyer, in 1902 and 1903, also understood the situation. The 1902
paper did not mention Mendel, and both were largely concerned with the
cytology of sterile hybrids. Guyer did, however, understand that random
assortment between different pairs of chromosomes would give independent assortment of genes, although both Wilson and Sutton thought
that he had missed this latter point.
Thus there were several people who were close to the correct interpretation at this time, but the first clear and detailed formulation was that
of Sutton. W. S. Sutton (18771916) was a student of McClung, whose
work on the sex chromosomes will be discussed later. Sutton was working with him at the time that McClung first suggested the relation of the
X chromosome to sex determination (1901), but Sutton was a graduate
student under Wilson at Columbia University when he wrote his two important papers (1902 and 1903). He never finished his graduate work, but
did later receive an M.D. degree, and became a practicing surgeon [see
biography by McKusick (1960) for more details].
As was pointed out above, the first of Suttons papers contained the
earliest detailed demonstration that the somatic chromosomes (of a
grasshopper) occur in definite distinguishably different pairs of like
chromosomes. He knew of the earlier work of Montgomery on pairing
and of Boveris paper (also published in 1902) on dispermic eggs. The
paper closed with the statement: I may finally call attention to the probability that the association of paternal and maternal chromosomes in
pairs and their subsequent separation during the reducing division . . .
may constitute the physical basis of the Mendelian law of heredity.
The 1903 paper contains a full elaboration of this hypothesis, including the view that the different pairs of chromosomes orient at random on the meiotic spindles,* thus accounting for the independent
segregation of separate pairs of genes seen by Mendel. He suggested,
following Fick and Montgomery, that in those cases where both meiotic
divisions had been described as longitudinal, the initial pairing had been
side by side rather than end to end, as he supposed it to be in insects. The
* The cytological demonstration of the random assortment of different pairs of
chromosomes was made by Carothers in 1913.
38
A H I ST O RY
OF
G EN ET IC S
paper contains a discussion and criticism of the 1902 accounts by Cannon and by Guyer, referred to previously.
With this paper, this phase of the history is finished. The conclusions
were not at once generally accepted, but they could not be disregarded
and stand today as essentially correct. At last, cytology and genetics were
brought into intimate relation, and results in each field began to have
strong effects on the other.
C HAPTER 6
L INKAGE
With the work of Sutton, the relation of the chromosomes to segregation and to independent assortment became clear. But there was a difficulty, already recognized in 1903 by Sutton and by de Vries: It must be
supposed that there are more separately Mendelizing genes than there are
chromosomes in the cells. That is to say, there are not enough chromosomes to make it possible to identify each gene with one whole chromosome. De Vries pointed out that this difficulty could be avoided by
supposing that genes were freely exchanged between homologous chromosomes at meiosis the process that he appealed to as an explanation
of independent assortment. Boveri made a similar suggestion in 1904.
The real solution showed that the principle of independent assortment is
not as regularly applicable as was then thought. The discovery and analysis of linkage, increasing knowledge of the nature and behavior of the sex
chromosomes, and more detailed cytological study of the meiotic prophases finally led to a resolution of the difficulty.
The first report of linkage was that of Correns (1900). He crossed
two strains of Matthiola (stocks), one of which had anthocyanin in the
petals and seeds, and also had hoary leaves and stems; the other had
white flowers and seeds, and smooth leaves and stems. The F1 had
colored flowers and seeds and was hoary. In F2 he expected to find many
types, resulting from independent segregation of three pairs of genes, but
actually recovered only the two parental combinations, in the ratio 3 : 1.
He suggested that the flower color and seed color might be due to the
same pair of genes, but interpreted the relation between color and hoariness as being due to the absence of recombination between two different
pairs of genes. He knew of the existence of smooth strains with colored
flowers and of hoary strains with white flowers, which confirmed the
view that there were two pairs of genes. Later work by Tschermak and by
Saunders has shown that the genetic situation is very complex, there being at least four (and probably more) genes in which the various known
39
40
A H I ST O RY
OF
G EN ET IC S
Bateson and Punnett did not present this algebraic formulation nor specifically
include the 1 : 1 case in the series, but this seems to be the simplest way of stating their
scheme.
Bateson, an embryologist by training, was impressed by the circumstance that
sometimes the two cells arising from the cleavage of a fertilized egg give rise to the
right and left sides, respectively, of the embryo. The mirror-image symmetry of these
L INKAGE
41
42
A H I ST O RY
OF
G EN ET IC S
L INKAGE
43
44
A H I ST O RY
OF
G EN ET IC S
* One of the genes involved in the case described by Lock concerned date of
flowering in peas and did not lead to clearly separable classes. While his data indicated
linkage, they were not amenable to exact analysis.
C HAPTER 7
T HE F LY R OOM
When Daniel Coit Gilman became the first president of Johns Hopkins University in 1875, he assembled a remarkable group of scholars to
supervise the graduate work there. Among these were two biologists: W.
K. Brooks, who had studied with L. Agassiz, and H. Newell Martin, a
student of Michael Foster and T. H. Huxley. These two trained a whole
generation of outstanding zoologists; among them Edmund Beecher Wilson and Thomas Hunt Morgan were of especial importance in the history
of genetics.
E. B. Wilson (18561939) took his Ph.D. at Johns Hopkins in 1881,
with a thesis on the embryology of the colonial coelenterate Renilla. He
then went to Europe, where he worked at Cambridge, at Leipzig under
Leuckart, and at Naples. The Naples station, where he returned several
times later, greatly influenced him and led to lasting friendships with
such men as Dohrn, Herbst, Driesch, and especially Boveri. In 1885 he
became the first professor of biology at the newly opened college at Bryn
Mawr. In 1891 he became professor of zoology at Columbia University
and remained there for the rest of his life. He also spent many summers
working at the Marine Biological Laboratory at Woods Hole, Massachusetts. Wilsons early work was largely in embryology, at first descriptive,
and later of an experimental nature; his interest here was chiefly in the
analysis of the gradual limitation of the potentialities of the cells of the
developing embryo and the extent to which formative stuffs were involved in development. The first edition of his great The Cell in Development and Inheritance appeared in 1896, the second in 1900, and the
third (really a new, much larger, and wholly rewritten book) in 1925.
This was the standard work for many years and exerted a very great influence on biology. Wilsons own studies on chromosomes began about
1905, with the work on the sex chromosomes referred to in Chapter 6,
and led to a series of detailed accounts that are models of accuracy and
clarity of expression.
45
46
A H I ST O RY
OF
G EN ET IC S
T HE F LY R OOM
47
48
A H I ST O RY
OF
G EN ET IC S
middle of July, 1912, the tests carried out by Bridges and Sturtevant had
shown that this linkage group (the second) included not only black,
curved, and vestigial, but five additional mutant types. In the same
month, we also found two additional types linked to pink, thus beginning
the study of the third linkage group (Sturtevant, 1913). The fourth and
last linkage group was found by Muller in 1914.
Stevens (1908) described the chromosomes of the female of Drosophila melanogaster (under the name D. ampelophila) as they are now
known, but she found the male difficult to study, and interpreted her figures as meaning that there was a rather small X attached to an autosome,
and no Y. This interpretation was followed in early genetic literature on
the species, until the work of Bridges (at first on XXY females in 1914),
and then of Metz (at first on other species of the genus, also in 1914) established the relations now known. Bridges insisted from the first, and
rightly, that the Y is J shaped and longer than the rod-shaped X; but the
rest of the group was at first unwilling to accept this, since in other animals
(even in other species of Drosophila) the Y was known to be absent,
smaller than the X or equal to it, but never larger. A corollary of the earlier
interpretation was that there were four, rather than three, pairs of autosomes, with one of them having some sort of relation to the X.
Bridges cytological work grew out of his studies of nondisjunction.
In the first paper on the sex-linkage of white eyes, Morgan reported a
few white sons from the original mutant male, which he supposed represented further mutation; there can be no doubt that they were due to nondisjunction.
Further examples kept appearing, and in 1913 Bridges published an
extensive genetic analysis of the phenomenon, giving it the name nondisjunction. Further studies led to no satisfactory causal interpretation
until he looked at the chromosomes and saw that females that gave exceptional offspring were XXY in composition (1914). As Bridges understood, this was really a proof of the chromosome theory and made it
inconceivable that the relation between genes and chromosomes was
merely some kind of accidental parallelism especially after the publication of his detailed account in 1916, as the first paper in volume 1 of
the newly founded journal Genetics.
A further consequence of the cytological work of Bridges and of Metz
was that it became clear that D. melanogaster had three pairs of autosomes two large and one small corresponding to the three autosomal
linkage groups, of which two were also large and one was small.
By 1915 the work with Drosophila had progressed to the point where
the group at Columbia was ready to try to interpret the whole field of
T HE F LY R OOM
49
It should be added that by his third edition (1926) Johannsen accepted the chromosome interpretation, and that Bateson thus closed the
review (from which the quotation just cited is taken): . . . not even the
most skeptical of readers can go through the Drosophila work unmoved
by a sense of admiration for the zeal and penetration with which it has
been conducted, and for the great extension of genetic knowledge to
which it has led greater far than has been made in any one line of work
since Mendels own experiments.
Not all critics were as generous, nor did they always receive soft answers. In short, there were a good many polemical papers; and there
surely would have been more if the work had not had the whole-hearted
support of Wilson, who had the respect and admiration of all zoologists,
making him an invaluable ally.
With the publication of the Mechanism and of Bridges 1916 paper,
this part of the story closes. There was still much exciting and fundamental work to be done with Drosophila, and the Columbia laboratory
was still the center for such work, but it had become a question of how
the chromosome mechanism worked, not of whether it could be demonstrated to be the true mechanism.
There was a give-and-take atmosphere in the fly room. As each new
result or new idea came along, it was discussed freely by the group. The
published accounts do not always indicate the sources of ideas. It was
often not only impossible to say, but was felt to be unimportant, who first
had an idea.* A few examples come to mind. The original chromosome
* There are, in the later literature, some examples of a concern about priority in the
development of ideas in the early period, but at the time such a concern never inhibited
free and open discussion.
50
A H I ST O RY
OF
G EN ET IC S
C HAPTER 8
D EVELOPMENT OF
D ROSOPHILA W ORK
One of the striking things about the early Drosophila results is that
the ratios obtained were, by the standards of the time, very poor. With
other material it was expected that deviations from the theoretical Mendelian ratios would be small, but with Drosophila such ratios as 3 : 1 or
1 : 1 were rarely closely approximated. This was recognized as being due
to considerable differences in the relative mortalities of the various
classes in larval and pupal stages, before counts were made. In fact the
culture methods were poor, so that these viability differences were very
marked, and they still are a source of difficulty.
There has, however, been a steady improvement in the technique of
handling the material most of the early work here being due to Bridges,
who brought improved optical equipment and temperature control devices into use, and who had the largest share in the development of improved and standardized culture media.
Bridges was also responsible for the finding and analyzing of a great
many new mutants. This undertaking, which was the source of usable
material for the studies, was participated in by everyone in the laboratory, but Bridges had the best eye for new types and contributed many
more of them than did the rest of us. He also had the skill and patience
required to build up useful combinations of mutant genes, and many of
his multiple strains are still among the most useful, twenty-five years
after his death.
Aside from these matters of routine technique, a series of new
genetic relations was worked out, and new genetic techniques were developed which made it possible to attack problems that were not
recognized or were not approachable before. But one of the first needs
was a convenient symbolism; the development of such a system was, in
fact, connected with a theoretical question.
51
52
A H I ST O RY
OF
G EN ET IC S
D EVELOPMENT
OF
D ROSOPHILA W ORK
53
54
A H I ST O RY
OF
G EN ET IC S
D EVELOPMENT
OF
D ROSOPHILA W ORK
55
56
A H I ST O RY
OF
G EN ET IC S
strand diagrams as long as they were adequate to explain the data. With
these results of Andersons, and the confirmatory results from triploid
females, published at the same time by Bridges and Anderson, the 4strand interpretation was firmly established.
Crossing over may occasionally occur at mitotic divisions, as was
shown by Stern (1936) for Drosophila. Here again it occurs after the
chromosomes have undergone a split, and the crossing over is between
two (nonsister) strands of the four that are present. The result may be the
production of sister cells homozygous for genes that were heterozygous
in the original cell.* If these cells then undergo further division and differentiation, there arise adjacent twin spots which may show the phenotypes of recessive genes that were originally heterozygous, one in each
homologous chromosome.
The production of twin spots was used by Demerec to study the effects of recessive lethal genes that had become homozygous in small
areas of a heterozygous individual. He found that some of these are cell
lethal under these conditions, while others are viable in at least some
tissues either because their normal alleles are not essential for the reactions going on in these tissues or because the necessary substances can
somehow be supplied from the rest of the animal.
More recently, it has been found that somatic crossing over occurs
regularly in some fungi; this has proved useful for linkage studies, especially in Aspergillus (Pontecorvo and co-workers) and in yeast (Roman
and co-workers). Somewhat similar phenomena also occur in bacteria in
connection with transformation and transduction and are now under intensive study for their bearing on the mechanism of crossing over.
Bridges followed his genetic and cytological studies of nondisjunction of X with a similar account of nondisjunction of the small IV-chromosomes (1921), and with genetic demonstrations of sectional
duplications, deficiencies, and translocations involving the longer chromosomes; these had to wait for the discovery of the properties of the
salivary gland chromosomes before they could be fully analyzed (see
Chapter 12). Also in 1921 came his studies on triploidy, which will be
discussed later (Chapter 13).
Another type of chromosome modification that was worked out step
by step may be described here, since it furnished part of the working
methods for the analysis of mutation and other phenomena. This is the
* That is, somatic segregation occurs. The process is much too rare and too erratic in
its incidence to furnish support for Batesons hypothesis of somatic segregation
(Chapter 6), though it sometimes occurs in the germ line and leads to the production of
crossover gametes.
D EVELOPMENT
OF
D ROSOPHILA W ORK
57
C HAPTER 9
G ENETICS OF
C ONTINUOUS V ARIATION
Galtons use of correlation methods for the analysis of parent-offspring resemblances, resulting in the Law of Ancestral Inheritance,
was described in Chapter 3. This approach was followed by Weldon and
by Pearson, the latter, in particular, contributing more sophisticated
mathematical techniques, as did Fisher and others more recently.
Karl Pearson was a mathematical physicist by training, and wrote a
book, The Grammar of Science (1892), that had a great deal of influence.
According to Pearson, science is simply descriptionby which he meant
description in quantitative termsand he was very suspicious of the idea
of causality. He was an extreme philosophical idealist; he was also convinced that there is nothing to which the scientific method cannot be
applied.
Weldon was a contemporary of Bateson at Cambridge, and the two
were at first close friends but later became bitter enemies. According to
Batesons letters, this enmity must have begun about 1890; by 1895 it
appeared publicly in a controversy about the origin of the cultivated
races of Cineraria, in the course of which each accused the other of deliberately misrepresenting the published statements of an authority that
each cited in favor of his own view.
This personal quarrel, which came to involve Pearson as an ally of
Weldon, seems to have been a chief reason for the anti-Mendelian stand of
both Weldon and Pearson. Bateson felt that they were trying to strangle the
new development, and he resisted vigorously. The controversy was pursued in debates and published works for years, and certainly delayed the
utilization of the powerful methods of statistics in much of genetics.
In 1902 Yule suggested that the Law of Ancestral Inheritance
could be thought of as due to the operation of the Mendelian principles in
a random-breeding population. In 1904 Pearson disputed this conclusion
58
G ENETICS
OF
59
60
A H I ST O RY
OF
G EN ET IC S
forms, but was inclined to minimize the effects of selection, and to agree
with de Vries that it could produce nothing really new.
In his original paper, Mendel had suggested the beginnings of the
theory of multiple genes and had recognized (in connection with flowering time in peas) the confusing effects of environmental differences; but
it was not at once evident that continuously varying characters could be
studied effectively by Mendelian methods.
In 1902 Bateson pointed out that it should be expected that many
genes would influence such a character as stature, since it is so obviously
dependent on many diverse and separately varying elements. This point
of view was implied by Morgan in 1903 (Evolution and Adaptation, p.
277), and by Pearson in 1904. It was developed by Nilsson-Ehle in 1908
and 1909, especially as a result of his experiments with wheat and oats.
Nilsson-Ehle applied the idea of numerous separately Mendelizing pairs
of genes plus the confusing effects of environmental differences to
quantitative characters such as size, length of awns, winter-hardiness,
and so forth, but his actual detailed analyses concerned somewhat simpler cases, such as color and presence or absence of ligules.
Lock and Castle had both reported crosses in which quantitative differences supposedly gave no increased variability in F2, but in 1910 East
and R. A. Emerson, separately, reported cases in several different plants
in which the F2 was much more variable than the F1 and interpreted them
as being due to the segregation of several pairs of genes. Their joint
paper (1913) on maize is a classic in the field and marks the bringing of
the inheritance of quantitative characters into the general scheme of
Mendelism.
There is another method of studying the inheritance of quantitative
characters, namely, the use of selection. This, of course, has a long history; we are here concerned with the relation between it and Mendelian
genetics. De Vries was inclined to minimize the effects of selection, and
argued that it could produce nothing new. Johanssen had a similar view;
but this was so contrary to the point of view of Darwin, Weismann, and
the whole generation that followed Darwin, that it was not generally accepted. There followed a series of selection experiments by numerous
workers. Those of Johanssen have been discussed, and the circumstance
of their being done with self-fertilizing homozygous strains has been
pointed out.
The most extensive and most widely discussed selection experiment
of the time was that which Castle carried out with rats, beginning in
1914. The hooded type here is white with a colored area on the head
and usually a narrow, colored line on the back. Castle selected for an in-
G ENETICS
OF
61
crease in the colored area in his plus line, and for a decrease in his
minus line. Selection was effective and ultimately yielded individuals
far beyond the limits of the variability of the original series. Both extremes, when crossed to self-colored rats, gave self (that is, uniformly
colored) in F1, and 3 self to 1 hooded in F2. Castle argued that this
pointed to a gradual change in the hooded gene, rather than to an accumulation of modifying genes a conclusion not accepted by most
geneticists. Considerable discussion resulted some of it (including one
paper of mine) rather heated.
Finally Castle crossed the two selected lines, separately, to wild type
and examined the grades of the extracted hooded F2 rats. In each case
these were less extreme than their hooded grandparents, that is, more like
the original, unselected strain. After two more crosses to wild type and
extractions, the two lines were nearly identical, and by 1919 Castle concluded that most of the effects of selection had been due to sorting out
and accumulation of modifying genes at loci other than that of the
hooded gene, although it seemed probable that a small part of the result
was due to minor changes in the hooded gene.
With this result there came to be general agreement that selection operates chiefly through the sorting out of modifiers already present, especially after the demonstration in Drosophila (by Payne, Sturtevant, and
others) that such modifiers could be located on chromosome maps by using marker genes. With the realization that modifiers are numerous, it
came to be recognized that they are often linked to each other. It is to be
expected also that plus modifiers will be linked to minus ones sometimes,
thus greatly complicating the analysis. Since practical breeding of domestic animals and cultivated plants depends to a large extent on selection,
knowledge about the nature of its action has been of importance in agriculture and horticulture; though perhaps this importance has been more in
the understanding of the principles than in the practice of the art of selection.* This art has, however, profited much from the application of statistical methods that grew out of the work discussed in this chapter. Here the
work of Fisher has been especially important.
Finally, an understanding of the way selection works has been of the
first importance in the application of genetics to the problems of evolution
(Chapter 17).
* The understanding of the principles may, of course, lead to practical results. As an
example, the increased vigor of hybrids was observed long ago by Klreuter, and even
earlier, in the particular case of the mule; but it was the application of Mendelian
methods by Shull (1908) and Jones (1917) that led to the most important agricultural
uses, beginning with hybrid corn.
C HAPTER 10
O ENOTHERA
As was pointed out in Chapter 3, there was a growing interest in discontinuous variations in the 1890s. In 1901 there appeared the first volume of de Vries monumental Die Mutationstheorie, in which he
developed the idea that evolution occurs through discrete steps (saltations or mutations) rather than by gradual changes accumulated by
selection. This conclusion rested on a vast amount of data concerning
many kinds of plants but was based more especially on the work of de
Vries on the evening primrose, Oenothera Lamarckiana.
The members of this genus are American in origin, but several of
them have escaped from cultivation in Europe, and grow in sandy or
disturbed soil there, as they do in much of the United States and Canada.
De Vries found a patch of Lamarckiana growing in an abandoned field at
Hilversum in Holland, and noticed that two variant types were present.
He brought all three types into his garden and found that the typical form
produced a series of mutant types, generation after generation. Many of
these new types bred true, and most of them differed from the parental
form in a whole series of relatively slight respects. It is now known that
this is because they differ from the parental form in many genes, and that
Lamarckiana is a very unusual and special kind of multiple heterozygote.
But to de Vries these new forms were essentially new species, and their
sudden occurrence meant that selection had little or nothing to do with
the origin of new species that differed from their parents in numerous
ways. This was the mutation theory in its original form; it is ironic that
few of the original mutations observed by de Vries in Oenothera would
now be called mutations.
It seems likely that the properties of these new types were largely
responsible for the emphasis placed by many geneticists on the multiplicity of phenotypic effects of single gene changes. It became the custom to emphasize the cases where such multiple effects occur though
surely if geneticists had approached their material without preconceived
62
O ENOTHERA
63
ideas, the striking thing would have been the relative scarcity of obvious
multiple effects of single-gene substitutions.
It soon became evident that the genetic behavior of Oenothera is unusual. The short-styled type (brevistylis) of Lamarckiana was one of the
Mendelian characters listed by de Vries in 1900, but it gradually became
a puzzle in itself, since nothing else in the plant behaved in so orthodox a
way.
The first examples of the new types to be explained were gigas, a
tetraploid with 28 chromosomes instead of the usual 14, and lata, a trisomic with 15 chromosomes (Lutz, 1907, 1909). These led to a whole
series of observations and experiments with other organisms, but they
left unexplained the majority of the Oenothera mutant types, since these
were found to have the 14 chromosomes of typical Lamarckiana.
The behavior of these 14-chromosome types when crossed to
Lamarckiana, and the results of crosses between various distinct wild
forms (biennis, muricata, and so on) were puzzling, sometimes giving
twin hybrids (that is, two distinct types in F1 from true-breeding parents), usually giving different results from reciprocal crosses, and usually
producing hybrids that bred true. Bateson, and later Davis, suggested that
Lamarckiana is really a hybrid but this suggestion, while probably correct, did little to explain its anomalous behavior. Meanwhile de Vries
published many data that seemed to show regularities but resisted all attempts at a systematic analysis.
The solution of the problems was really begun by Renner in a remarkable series of papers that were long disregarded, even by those of us
who were actively trying to relate the published data to a scheme consistent with what was known elsewhere. This neglect of Renners work was
undoubtedly due to his use of a system of terminology which was and is
very convenient for Oenothera but makes the papers unintelligible unless
the special terminology is first learned. When it is learned, the papers are
found to be written in a very clear and logical style.
The series of papers began in 1913 with one on fertilization and early
embryology; it showed that a suggestion of Goldschmidts (merogony)
was incorrect. This was followed in 1914 and 1917 by a study of the embryos and seeds from Lamarckiana after self-pollination and after crossing with other species. These studies showed that Lamarckiana is a
permanent heterozygote between two complexes called gaudens and
velans. After self-fertilization about half of the seeds contain inviable
embryos. Half of these die at an early stage, and the other half at a later
one. Renner concluded that these inviable seeds represented the gaudens-
64
A H I ST O RY
OF
G EN ET IC S
O ENOTHERA
65
66
A H I ST O RY
OF
G EN ET IC S
C HAPTER 11
M UTATION
Early studies on variation were concerned with phenomena that were
in part due to recombination, and the origin of new types could be studied effectively only when the question could be formulated in terms of
the origin of new genes. The first question was: Do new genes in fact
arise, or is all genetic variability due to recombination of preexisting
genes? This question was seriously discussed though the alternative to
mutation seems to be an initial divine creation of all existing genes.
It is true that most strains studied are of hybrid derivation or may
be so and that recessives may remain undetected for many generations.
The result is that, while one may argue that the origin of new genes is a
logical necessity, it is not so easy to specify that a given gene must be of
recent origin. Perhaps the clearest evidence in the early studies came
from the sweet pea, which was known to be descended entirely from a
single small collection of wild material (from Sicily) and to have been
propagated for many generations, mostly by self-pollination, before new
types appeared. The history of the various color types was known, and
left no doubt that particular genes had arisen suddenly in cultivated material. But most of the newly arisen forms in such cases are recessive to the
original type, and, on the long-accepted presence and absence hypothesis, represent merely losses of genes not the acquisition of new ones.
As late as 1914, Bateson discussed the possibility that evolution had
come about solely through the loss of genes, followed by recombination.
In diploid organisms there are only two classes of genes whose mutability can be studied simply and effectively newly arising dominants,
and sex-linked genes, where in the heterozygous sex each individual is
effectively haploid. Both were used in early work, and special methods
were developed fairly quickly for the study of autosomal recessives, especially in self-fertilizing plants.
In the 1890s de Vries studied the genetic behavior of variegation in
the flowers of Antirrhinum, the snapdragon. Here the flowers are white
67
68
A H I ST O RY
OF
G EN ET IC S
or yellow, with red stripes, the variegation being recessive to the self red
condition. On a variegated plant, the size of the red areas is variable, and
sometimes includes an entire branch. De Vries showed (1901, Die Mutationstheorie) that the flowers on these red branches behave like the red
flowers on a wholly red F1, that is, on self-pollination they give 3 red : 1
variegated. The results were not given a Mendelian interpretation because, at the time the experiments were done, de Vries did not know
about Mendelism. He did, however, suggest that some kind of segregation (Spaltung) was occurring, so that self color was somehow split off
from the variegated element; he knew, and emphasized the point, that no
true whites were produced.
Correns (1909, 1910) studied similar cases in the four oclock,
Mirabilis, the most detailed account concerning variegated leaves. He
gave a Mendelian interpretation but thought of segregation rather than of
mutation though, like de Vries, he found that the change was in one
direction only: from variegated (pale green and dark green) to self dark
green. His main emphasis was on the change in somatic tissues from a
homozygous variegated condition to a heterozygous (self/variegated)
one. Neither de Vries nor Correns seems to have been clear that the
variegation itself, within individual flowers or leaves, could most easily
be interpreted as being due to the occurrence, late in development, of the
same change responsible for large sectors, which could then be transmitted. This view, and the interpretation of the whole series of events as recurrent gene mutations, was developed by R. A. Emerson (1914) as a
result of his studies on the variegated pericarp of the calico type of
maize. He also pointed out what was evident in his results and in those of
de Vries and of Correns (when finally mutation was thought to cause
variegation), namely, that the mutation regularly occurred in only one of
the two genes in a single homozygous cell.
The regular and frequent occurrence of mutations in such cases was
made clear by these papers, but there was, and still remains, a reluctance
to generalize from them about the mutation process, since the usual frequency is so very much less than they report and is for that reason
much more difficult to study.
The early work on Drosophila, especially that of Bridges, furnished
several instances of the occurrence of new dominant genes, and many
new sex-linked recessives, in pedigreed material where the event could
be analyzed in some detail. These examples confirmed the conclusion
that mutation occurs in a single gene in a single cell, and that it can occur
at any stage of development. In fact, Muller concluded that the frequency
M UT AT ION
69
per cell is probably the same for each stage in the germ line.
These results were merely qualitative, since the frequencies were too
low for a quantitative study and were also strongly influenced by the personal equation of the observer. What was needed was an objective index,
and one that would recognize a class of mutations that was frequent
enough to give significant numerical values. Both of these requirements
were met in the system devised by Muller for the study of newly arisen
sex-linked lethals in Drosophila.
As first used by Muller and Altenburg (1919), the method depended
on the study of the sex-ratios from individual females that were heterozygous for sex-linked marker genes. This technique made possible
an objective and unambiguous determination of the frequency of sexlinked lethals, but it was very laborious, since rather detailed counts had
to be made for each tested chromosome. There was a suggestion of an
increase in the rate of occurrence of new lethals with an increase in temperature; this was later confirmed with improved techniques. It was also
evident that the rate per unit time could not be as high in man as in
Drosophila, for the approximate 1 : 1 human sex-ratio would then be
quite impossible.
Muller then improved the technique by using the ClB chromosome
that he found in his experiments. This is an X chromosome that carries a
crossover reducer (later shown to be an inversion), a lethal, and the
dominant mutant gene Bar. As is well known now, this chromosome
made it possible to detect new sex-linked lethals without anesthetizing
the flies or making counts merely by rapid examination of individual
culture bottles. In this way one could test many more chromosomes than
possible ever before and thus could obtain adequate data on lethal frequencies.
The first great advance through the use of this technique was the
demonstration of the mutagenic action of X rays (Muller, 1927).
X rays were reported in 1895 by Roentgen, and, in the next year,
radiation burns caused by them were reported by L. G. Stevens. There
followed a rapidly increasing volume of literature on the biological effects of X rays and (beginning in 1900) of radium rays. Radiation sickness, described in 1897 by Seguy and Quenisset, and by Walsh, was
characterized by heart symptoms, headaches, and insomnia. The first
successful therapeutic use seems to have been by Steenbeck, who in 1900
reported the destruction of a small tumor in the nose of a patient. Radium
burns were reported in this same year by Walkoff and by Griesel a result confirmed in 1901 by Becquerel and Curie, who deliberately ex-
70
A H I ST O RY
OF
G EN ET IC S
M UT AT ION
71
was mature, each shoot was separately self-pollinated and the seeds were
planted. Any induced recessive segregated in the ratio 3 : 1 and could be
identified as induced, since it would not appear from other shoots of the
same plant. In all, Stadler found forty-eight mutations in distinct seedling
characters from irradiated seeds, and none from a large control series.
Some of these seeds had been given X rays, others had been exposed to
radium the latter proving to be also mutagenic.
These results of Muller and of Stadler at last opened the way to an
experimental attack on problems having to do with mutation, and they
and others began an extensive study of these problems a study that is
still being very actively prosecuted.
Muller, in his first account, concluded that high-energy radiation is
dangerous not only to the exposed individual, but to his descendants as
well. This conclusion has come to be generally understood and accepted,
and is now of much public concern in connection with the medical and
dental use of X rays, and the worldwide distribution of radioactive isotopes produced by nuclear fission.
In this same paper, Muller made two tentative suggestions that were
soon shown, by Muller himself and by others, to be incorrect: first, that
spontaneous mutations might be largely or wholly due to normal background radiation, and second, that the relation between dosage and mutation rate might be exponential rather than linear.
Stadler, in his second paper in 1928, reported (in an abstract about
half a page long) three major discoveries made with barley:
1. Seeds soaked in water to initiate germination gave almost eight times
the mutation rate of dormant seeds.
2. Mutation rate was independent of the temperature at the time of irradiation.
3. The relation of mutation rate to total dosage was linear doubling
the dose doubled the mutation rate.
In 1929, Hanson and Heys studied the induction of lethals in Drosophila by radium. They interposed lead shields of different thicknesses
and recorded the ionization (using an ionization chamber) in each treatment. The curves for ionization and for mutation rate were superposable,
leading to the conclusion that ionization is responsible for the mutations
and that the relation is a simple, direct one a one-hit phenomenon. In
1930, Oliver confirmed this conclusion by showing that varying the dose
by varying the duration of exposure to a constant X ray source also gave
a linear curve relating dosage to mutation rate. It was apparent that projecting this curve to a zero dosage did not give zero mutations, and Mul-
72
A H I ST O RY
OF
G EN ET IC S
ler and Mott-Smith showed in this same year that the amount of
background radiation is far less than would be required to produce the
normal spontaneous frequency of mutations.
It came to be very generally accepted that total ionization is all that
need be considered in connection with radiation-induced mutations, at
least within a strain. Stadlers dormancy results should have suggested
that it was dangerous to argue from sperm (which are essentially dormant) to actively growing tissues; but it was not until rather recently that
it was shown (by Lning, Russell, and others) that there are differences
in dose response in different tissues and developmental stages in Drosophila and in mice. These studies, now being actively prosecuted, fall outside the scope of this book.
The induction of mutations by chemical means was attempted by
many people over a long period, but until 1941 there were no clear and
convincing positive results. In that year Auerbach and Robson obtained
clear evidence that mustard gas is mutagenic. They were led to study it
because pathologists had found much similarity between the appearance
and behavior of skin burns induced by mustard gas and by irradiation.
The mutation effect was observed in a (British) government-supported
war project, and the information was classified and could not be published until 1946. With its publication began the present active and productive study of chemical mutagens another development outside the
scope of this book.
C HAPTER 12
C YTOLOGICAL M APS
AND THE C YTOLOGY
OF C ROSSING O VER
Rearrangements of the chromosomal material were first detected in
Drosophila by genetic methods. Deficiencies were reported by Bridges
(1917) and by Mohr (1919), duplication by Bridges (1919), translocation
by Bridges (1923), and inversion by Sturtevant (1926). These were all of
spontaneous occurrence, and none of them were cytologically identifiable by the methods then available.
The first successful cytological attempt to analyze the chromosomes
from a genetic point of view, in terms of units less than a whole chromosome, was made by Belling. By 1924, Belling and Blakeslee reported a
series of extra-chromosome types in Datura. In this paper they developed
the idea that the meiotic pairing of chromosomes, even at a late stage
(diakinesis), could be used to determine the homologies of separate arms.
They described secondary trisomics, in which the extra chromosome
was made up of two like arms of a normal chromosome, presumably
having arisen by a somatic division that was normal except that the centromere had divided transversely to the long axis of the chromosome instead of parallel to that axis. These types gave evidence as to the
phenotypic effects of the separate arms. Unfortunately, however, there
were few mutant genes available in the plant, and the more critical earlier
stages (pachytene) of meiosis were not favorable for study in Datura.
A different attempt to get cytological information on the genetic
composition of individual chromosomes was developed for maize by
McClintock, Randolph, Longley, and others. Here it was possible to
study the pairing of homologous chromosomes at pachytene, when they
were longer and showed more recognizable detail than at the later stages
73
74
A H I ST O RY
OF
G EN ET IC S
C YTOLOGICAL M APS
AND T HE
C YT OLOGY
OF
C ROSSING O VER 75
76
A H I ST O RY
OF
G EN ET IC S
In the same month (December, 1933) there appeared Painters account of the salivary gland chromosomes of Drosophila melanogaster,
which he showed were quite workable, if one studied old larvae nearly
ready to pupate. In this paper he presented a drawing of the euchromatic
part of the X, with over 150 bands, and with 13 corresponding points
shown, that had been determined both cytologically and genetically from
a long deletion, seven translocations, and two inversions (one of the latter being the familiar ClB). Here at last was a detailed correspondence in
sequence between the crossover map and cytologically visible landmarks, and a technique that was clearly capable of refinement to give the
precise loci of genes in terms of recognizable bands. Instead of two or
three landmarks per chromosome (the ends and centromeres), there were
now hundreds, and there soon came to be thousands for the whole
complex.
There followed a series of studies in several laboratories, which
rapidly gave more and more detailed cytological maps of all the chromosomes, both of melanogaster and of other species. In the case of melanogaster, where the available genetic data were much more extensive, the
detailed studies of Bridges were especially useful, and his drawings of
the salivary gland chromosomes of that species (1935, 1938, 1939) are
still the standards.
In 1935 Bridges recognized 725 bands for the X chromosome, 1320
for the second, 1450 for the third, and 45 for the fourth. In 1938 the
number for the X was raised to 1024, and in 1939, the number for the
right limb of the second was raised from 660 to 1136. He recognized that
even these numbers did not exhaust the potential resolving power of the
method. Bridges also developed a convenient system for designating
each band a system that is still in use.
An early result of the salivary gland studies was the discovery of
repeats by Bridges (1935) in Drosophila and by Metz (1938) in Sciara.
These repeats were shown to be either direct or reverse in their orientation with respect to each other, and to be either adjacent or separated
by other regions. Their origin is not altogether clear, but their frequent
occurrence is, as Bridges pointed out, of considerable evolutionary interest, since they furnish extra genes that are presumably not needed by the
organism, and that may be of importance in making possible the origin of
new genes with new functions.
Another result of these studies, recently found to be of great interest,
is that of the puffing of certain regions. It was shown by Metz (1938)
that certain regions of the salivary gland chromosomes undergo a reversible process in which the bands swell and show a much looser struc-
C YTOLOGICAL M APS
AND T HE
C YT OLOGY
OF
C ROSSING O VER 77
78
A H I ST O RY
OF
G EN ET IC S
same level. In the later forms of these models, he related the phenomenon to the production of new daughter chromatids an idea that has been
involved in many of the more recent interpretations.
The most ambitious attempt at a general scheme is that of Darlington, embodied in a long series of papers and first developed in detail in
his book of 1932. This scheme was very generally accepted, and for a
time came to be considered the very backbone of cytogenetics. It depends on Darlingtons precocity theory, which he sums up as follows:
Meiosis differs from mitosis in the nucleus entering prophase before the
chromosomes divide instead of after they divide.
According to this scheme there is a tendency for chromosomes and
their constituent parts to form pairs of like elements at the beginning of
prophase. If chromosome division has already occurred (as at mitosis),
this affinity is satisfied by the fact that daughter chromatids are still
closely apposed; at meiosis it leads to a conjugation between homologues. In the latter case, when the conjugated chromosomes divide there
are four apposed strands, and the attractive force is supposed to be satisfied when two elements are apposed. Therefore there occurs a separation
(reductional) into two double bodies, each made up of a pair of sister
elements. If, now, there has been an exchange (that is, a crossover), there
will be a chiasma corresponding to it, since only in this way can each
part undergo a reductional separation. These chiasmata hold the structure
together and ensure that the orientation at the metaphase of the first
meiotic division will lead to the passage of two chromatids to each pole.
This scheme was elaborated in great detail, and gave a satisfying
geometrical picture, which was correlated with the genetic results by
many workers. To many of us, it came to be accepted as basic (see, for
example, Sturtevant and Beadle, 1939). But there were skeptics from the
beginning. Belling was very critical of much of the scheme, as were Sax
and others. It was soon apparent that, in some forms, the chromosomes
are visibly double at the time they first conjugate; the view that the initial
separation is always reductional at each level was questioned as being an
unsupported hypothesis. It was pointed out that quite regular first meiotic
segregation occurs without any accompanying crossing over in the
Drosophila male and, sometimes, also in the female. Some of the supporting observations themselves were questioned notably the quantitative agreement between observed frequencies of crossing over and
counted numbers of chiasmata. Here the fact is that the counting of chiasmata can be carried out in a really convincing and unambiguous manner in only a few very favorable objects, and these unfortunately do not
C YTOLOGICAL M APS
AND T HE
C YT OLOGY
OF
C ROSSING O VER 79
* I should like here to enter a protest against the current use, especially by students
of bacteria and bacteriophage, of the word chromosome as synonymous with linkage
group. A chromosome is a body that is visible under the light microscope, contains
both DNA and other material, and has a whole series of reasonably well-understood
properties. The bodies so designated in bacteria and in bacteriophage are very much
smaller, seem to be wholly DNA, and lack many of the properties of true
chromosomes. They do agree in containing the genes and in being subject to
recombination. No one can question the importance of the studies being made about
them but it seems essential to avoid confusion by using a different term; genophore,
suggested by Ris, seems appropriate and desirable.
C HAPTER 13
S EX D ETERMINATION
Theories of the determination of sex were already numerous in
Aristotles time, and he discussed many of them. His own view was that
there is, in each embryo, a sort of contest between the male and female
potentialities, and the question of which prevails, that is, the frequencies
of the two sexes, may be influenced by many factors, such as the age of
the parents, the direction of the wind, and so forth. This idea of a competition between opposing influences has been involved in most theories,
down to the present; the current form is described by the term genic
balance.
The existence of males and females, in approximately equal numbers, continued to intrigue both philosophers and biologists. Thomson in
1908 wrote: The number of speculations as to the nature of sex has been
well-nigh doubled since Drelincourt, in the eighteenth century, brought
together 262 groundless hypotheses, and since Blumenbach caustically
remarked that nothing was more certain than that Drelincourts own
theory formed the 263rd. Subsequent investigators have long ago added
Blumenbachs theory to the list.
The discovery of the sex chromosomes and the demonstration of
their relation to sex determination, have been described in Chapters 6 and
7. One result of the nondisjunction studies was not pointed out: the Xbearing sperm is not in itself female determining, since it may produce a
male if the egg carries no X, that is, sex is determined by the composition
of the zygote. This conclusion was confirmed and extended by studies of
gynandromorphs. Morgan and Bridges (1919) showed, by a detailed
study of a large series of these, that the separate parts of the body of
Drosophila are largely independent in their determination, and that the
sex of each part is due to its chromosome composition.
Another relation established by Bridges in his nondisjunction work is
that diploid individuals of Drosophila with the composition XXY are
80
S EX D ET ERMINATION
81
normal and fertile females, and that those with a single X and no Y (XO)
are normal males in appearance though they are sterile. That is to say,
the Y is not the primary sex-determining agent.
A much fuller analysis resulted from the study of triploids by
Bridges (1921). These results were based on a remarkable series of parallel genetical and cytological studies, and furnished convincing proof
that sex in Drosophila is due to a balance between the number of Xs
(which have a net female-producing effect) and the number of sets of
autosomes (which have a net male-producing effect). Thus, addition of
an X to the normal male composition produces a female, while addition
of a set of autosomes to the female composition produces an intersex.
This conclusion has been fully confirmed by the later finding of a few
additional types from the tetraploid females.
Bridges interpreted these results on the basis of the genic balance
idea that he had deduced early in 1921 from his studies on haplo-IV individuals, which have a recognizable phenotype that differs from the wild
type in a number of respects. Bridges pointed out that there is evidence
for the existence of numerous genes affecting a given character some
in one way, some in another. Each individual represents the resultant of a
particular balance between these variously acting genes. It is unlikely
that any given chromosome, or section of a chromosome, will have a set
of genes with the same net effect as the whole complement; therefore it is
to be expected that duplications or deficiencies will cause changes in this
balance and will alter the phenotype usually to the detriment of the
individual, and often with even a lethal effect.
In the case of sex, this interpretation means that there are genes with
male-producing effects, and others with female-producing ones; and that
the former predominate in the autosomes, the latter in the X. This is, evidently, a form of the competition idea of Aristotle, which had been previously elaborated by Weismann, and by Goldschmidt especially for sex
determination. But Bridges based it on more direct experimental evidence and proceeded to use it as a working hypothesis to suggest further
experimental approaches. His studies of the effects of different dosages
of the small fourth chromosome on the phenotype of intersexes were designed to test the net effect of the genes in this chromosome on the sex of
the individual. The results were inconclusive, but the method was used
successfully by Dobzhansky and Schultz (1931, 1934) in studying the
effects of various fragments of the X. When these fragments were added
to intersexes (that is, 2X + 3A + fragment of X), they found that the
inert (that is, heterochromatic) region was without effect, but that each
82
A H I ST O RY
OF
G EN ET IC S
of the very diverse euchromatic duplications shifted the degree of intersexuality toward femaleness. That is to say, not only does the X as a
whole have a net female effect, but the numerous tested portions of it
also have such an effect.
Similar tests with duplications for various autosomal segments have
given no such clear-cut result, but evidence of another kind indicates that
there are several autosomal genes that affect the sex of the individual.
Several autosomal mutant genes change diploid females into male-like
intersexes (Sturtevant, 1920, for Drosophila simulans; Lebedeff, 1934,
and Newby, 1942, for D. virilis; L. V. Morgan, 1943; Sturtevant, 1945;
Gowen, 1948, for D. melanogaster). Four of these genes are recessive, so
that the supposition is that their wild-type alleles probably influence development in the female direction, rather than in the male direction as the
whole set of autosomes does. They do, however, serve to indicate that
there are several (many?) autosomal loci concerned.
In 1946 I found that intersexes are produced in hybrids between
Drosophila repleta and D. neorepleta, The analysis indicated that neorepleta carries a dominant autosomal gene which so conditions the eggs
before meiosis that two repleta Xs are insufficient to produce the full
female phenotype and intersexes result though two neorepleta Xs, or
one X from each species, produce normal fertile females. Here is evidence for an autosomal gene with the expected male effects; whether or
not the unusual maternal effect is generally present in other Drosophila
species remains to be determined. It may be noted that this particular
gene would not be detected by the usual technique for testing the effects
of duplications or deficiencies on triploid intersexes. This gene, since it
operates before meiosis and fertilization, cannot be responsible for all the
autosomal effects observed in the triploid experiments.
The earlier interpretation of the sex chromosomes of Drosophila,
giving the male the formula XO, meant that the male was haploid for
this chromosome, and this was consistent with the facts of sex-linkage.
When the work of Bridges and of Metz established that the normal male
is XY, it became necessary to suppose that the Y lacked dominant alleles
of the sex-linked genes. The study of the XO (exceptional) males by
Bridges (1916) showed that, though phenotypically normal, they were
sterile. It was shown by Stern (1929) that fertility is dependent on the
presence of both arms of the Y, and also (Stern, 1926) that the Y normally carries the wild-type allele of the sex-linked mutant bobbed, so it
is not quite inactive genetically. Finally the cytological studies of Heitz
and of Painter in 1933 showed that the Y is heterochromatic, like the
S EX D ET ERMINATION
83
84
A H I ST O RY
OF
G EN ET IC S
S EX D ET ERMINATION
85
86
A H I ST O RY
OF
G EN ET IC S
honeybee, but seems not to hold (at least in unmodified form) for some
of the other Hymenoptera.
Still another type of sex determination is to be found in the mosses
and liverworts, where the haploid generation is sexual. It was shown by
the Marchals (1906, 1907) that, in certain mosses, regeneration from the
diploid sporophyte (which reproduces by asexual spores) led to the production of sexual gametophytes. These were diploid, and were shown to
be hermaphroditic, although the normal haploid gametophytes in these
species are either male or female, as are homozygous diploids.
The conclusions suggested were confirmed by the cytological work
of Allen (1917) on the liverwort Sphaerocarpos. He showed that the
sporophyte has an unequal pair of chromosomes, of which the larger
(called X) is present in the (haploid) female gametophytes, while the
smaller (Y) is present in the male gametophytes.
Baltzer (1914) has shown that in the marine worm Bonellia sex is
determined by the environment, and not by genetic means. This form has
perhaps the most extreme sexual dimorphism known, the male being a
minute degenerate creature that lives as a parasite in the female. Baltzer
has shown that the larvae are not differentiated sexually. Those that settle
on the sea bottom develop into females, while those that settle on the
proboscis of a female develop into males.
There are many special devices in connection with several of the different types of sex determination, but even a simple catalogue of the
more interesting ones would be out of proportion here.
It has sometimes been felt that the determination of sex offered the
best opportunity for the study of the manner of action of genes, and the
results described here have contributed largely to our understanding in
this field; it now appears, however, that it will be more profitable to
study simpler situations, and it is to these that attention is now more
often directed (Chapter 16).
C HAPTER 14
P OSITION E FFECT
In the absence of radiation, one mutation appeared in Drosophila
with sufficient frequency to be worth detailed study: namely the case of
Bar eye. Although the analysis was successful, the situation turned out to
be too special to serve as a basis for any general picture of mutation, but
it did lead to the discovery of the position effect, which has played a
large part in later developments.
Bar is a sex-linked dominant that reduces the size of the eye. It was
studied intensively by Zeleny and his students, especially with respect to
the effects of temperature on the size of the eye, as measured by counting
the numbers of facets. In the course of these experiments it was noticed
by May (1917) that Bar stocks occasionally revert to wild type. This
phenomenon was studied by Zeleny (1919, 1920, 1921), who found that
about 1 in 1600 offspring from a Bar stock carries a wild-type allele (B+).
He concluded that the event occurs in females, late in the development of
their eggs. He also found that there is a more extreme type produced, a
type that he called ultra-Bar, which I later gave the name double-Bar. He
showed that double-Bar stocks also revert to wild type and also may give
rise to Bar.
Zelenys evidence indicated that these mutations occurred in females
near the time of meiosis, and consequently Morgan and I were led to an
investigation of whether the mutations had any relation to crossing over.
Our result was clear: 6 reversions were obtained, and all were crossovers
between marker genes (forked and fused) lying on opposite sides of the
locus of Bar and less than 3 units from each other. That is to say, all 6
reversions were in a class (the crossovers) that included less than 3 percent of the population (Sturtevant and Morgan, 1923).
I then carried out more extensive tests (Sturtevant, 1925) that confirmed this result, not only for reversion of Bar but for the production of
double-Bar and for other changes, such as the production of Bar from
double-Bar/wild-type heterozygotes.
87
88
A H I ST O RY
OF
G EN ET IC S
P OSITION E FFECT
89
discovered by Lewis (1945). The two mutants Star (dominant) and asteroid (recessive) have very similar phenotypes and lie adjacent to each
other. Lewis studied the heterozygote (Star/asteroid) and recovered from
it both the wild type and the double mutant, Star asteroid. If we compare
the two kinds of double heterozygotes the cis type (Star asteroid/wild
type) and the trans type (Star/asteroid) it is clear that the trans type
differs more decidedly from the wild-type phenotype than does the cis
type. This case is complicated by the dominance of Star, but more recent
examples have shown that the principle illustrated here is a general one
in such cases: the cis heterozygote (carrying a normal unmutated chromosome) is more nearly wild type in phenotype than is the trans (Bar is
an exception to this rule).
Situations like this were soon found in which the dominance complication was absent. The first of these was reported by Green and Green in
1949, for the mutant lozenge in Drosophila. Oliver (1940) had shown
that females heterozygous for two independently arisen lozenge mutants,
called glossy and spectacled (that is, glossy/spectacled), had the typical
phenotype of the lozenge series but gave some wild-type chromosomes
that were always crossovers for outside markers. Oliver failed to detect
the contrary crossover and was therefore in doubt as to the significance
of the result. Green and Green used glossy (lzg) and two new independently arisen types (lzBS and lz46). They were able to recover from each of
the double heterozygotes of the trans type, both the wild type and the
double mutant; all these events were again associated with crossing over
between outside markers. These results showed the sequence in the
chromosome map to be lzBS lz46, lzg . When the six possible heterozygotes
for two mutants were made up, it was found that all three cis types were
wild type in phenotype, whereas all three trans types (lzBS/lz46, lzBS/lzg,
lz46/lzg) were lozenge in appearance.
The same kind of result was soon demonstrated for several other
series of independently arisen alleles for vermilion and for beadex by
Green, for white and for bithorax by Lewis, and for several other series
by other investigators. The case of white was particularly unexpected, for
this had long been the type case of multiple allelism, and it became clear
that the then- current hypothesis must be revised.
Multiple alleles had been supposed to represent changes in a single
original gene, and there were two criteria for their recognition: they occupied the same locus in the chromosome and were not separable by crossing
over; and their heterozygote (trans type) was mutant with respect to their
common recessive phenotype, since neither carried the wild-type allele of
the other. With the discoveries noted above, these two criteria were shown
90
A H I ST O RY
OF
G EN ET IC S
P OSITION E FFECT
91
92
A H I ST O RY
OF
G EN ET IC S
the most detailed and illuminating studies were by Demerec (1940, 1941)
and by Demerec and Slizynska (1937). These studies showed that there is
a spreading effect. If heterochromatin is represented by the symbol H
and if a series of wild-type alleles A, B, C, and so forth be brought next
to it in the sequence HABC, then the suppression of gene activity proceeds from H; A is inactivated first, then B, then C, and so on. Tissue
with inactivation of A alone or of both A and B may occur but not with A
active and B inactive. It appears that there is no skipping of genes, that is,
there seem to be no genes immune to the effect. (A supposed exception
to this rule reported for the fourth chromosome has been found to be
based on an incorrect map of the fourth chromosome.)
The most likely interpretation is that there is a progressive inhibition
of the production of gene products but not of gene replication; that is, in
modern terms, RNA is not produced, but DNA replication does occur. One
possible interpretation is that the timing of the DNA replication is retarded
as it seems to be in heterochromatin; these matters are, however, beyond
the scope of this book.
It is, in fact, premature to formulate any definitive scheme for the Vtype position effects, since several facts remain to be further analyzed:
the effect of removal from heterochromatin upon genes normally in or
near it (possibly an inhibition of suppressors normally present ?);* the
striking effects of temperature and of the number of Y chromosomes present (both reported in 1933 by Gowen and Gay); the occurrence of dominant V-type effects, and numerous other unexplained relations. These are
now under active study in several laboratories, and there can be no doubt
that the V-type effects will contribute largely to future ideas about the
nature of gene action in development and differentiation.
C HAPTER 15
93
94
A H I ST O RY
OF
G EN ET IC S
TABLE 2
Current
Designation
Jansky
System
Moss
System
Agglutinogens
on Cells
Agglutinins
in Serum
IV
None
II
II
III
III
AB
IV
A, B
None
G ENETICS
AND I MMUNOLOGY
95
served frequencies, in populations with very different absolute frequencies. According to this scheme, the four groups have the following
genotypes: O is OO, A is AA or AO, B is BB or BO, AB is always AB.
There was a serious difficulty here, as Bernstein recognized: On this
scheme AB and O can never be related as parent and offspring, yet the
published pedigrees included many examples of such relationships.
In 1929 Snyder presented a summary of the published records for the
offspring of one of the critical matings that between O and AB parents.
If these are divided into two groups, those published before and those
after Bernsteins paper, the totals are as shown in Table 3.
TABLE 3. OFFSPRING RECORDED FROM M ATINGS BETWEEN O AND AB
Published Records
AB
Number
of
Papers
27
80
59
24
18
228
234
96
A H I ST O RY
OF
G EN ET IC S
the effective serum components antibodies, which may cause agglutination (as in the ABO system), hemolysis, or other reactions are produced only as a result of the previous introduction of the corresponding
antigen (agglutinogen in the ABO system) into the blood of an individual
that does not itself produce it.
The extensive studies of blood antigens other than A and B in man,
and those carried out on other animals, have most often utilized such induced antibodies.
If human blood cells are injected into a rabbit, a series of antibodies
that will react with any human red cells will be produced in the rabbits
serum. Landsteiner and Levine (1927) carried out this experiment, using
a series of rabbits, each injected with cells from one of a series of human
donors. They then treated each antiserum with cells from several different donors separately. This procedure resulted in the absorption of all the
general antibodies against all human red cells, but in some cases the
treatment left antibodies that would still react with the cells from other
individuals. They were able to show that there were two such reactive
substances in the cells, which they called M and N. They found three
types of individuals M, N, and MN. These substances have been found
to be dependent on a pair of allelic dominant genes: no inactive allele,
corresponding to the O of the ABO system, is known.* This paper marks
the beginning of the study of specific induced antibodies to human blood
cells, which has been greatly extended since then.
A modification of this technique was used by Landsteiner and
Wiener (1940). They injected red cells from a Rhesus monkey into
guinea pigs, and carried out absorption with red cells from different human subjects. The resulting absorbed antisera were, in some cases, reactive with cells from other human subjects. They recognized two types of
individuals, Rh positive and Rh negative, the symbol being derived from
the name of the original donor species.
The clinical importance of the Rh antigens arises from the fact that
human subjects who are Rh negative may develop Rh antibodies which
lead to transfusion reactions, if the subjects have previously received
transfusions from an Rh-positive donor (Wiener and Peters, 1940) or,
more often, if an Rh-negative mother has had an Rh-positive child
(Levine and Stetson, 1939). In the latter case, it is evident that Rh antigens from the fetus enter the maternal circulation, and there induce the
formation of anti-Rh in the serum. This antibody may, in turn, enter the
* This series of alleles has since been extended by the discovery of antibodies to S, and to
other properties. Many alleles are now known.
G ENETICS
AND I MMUNOLOGY
97
98
A H I ST O RY
OF
G EN ET IC S
G ENETICS
AND I MMUNOLOGY
99
vincing, but they were fully confirmed on a large scale by Loeb and
Wright (1927), using a series of long-inbred lines of guinea pigs.
This general approach has since been carried much farther in mice,
especially by Snell, who has succeeded in genetically isolating a series of
genes concerned with graft compatibility, and in locating these on the
linkage maps.
One of the most striking results in this field was an outgrowth of the
extensive work on the red-cell antigens of cattle. The existence of a great
individual diversity in cattle antigens was shown by Todd and R. G.
White in 1910, but the genetic analysis was begun by Ferguson (1941)
and extended by a group including Irwin, Ferguson, Stormont, and
Owen. In 1945, Owen reported on a pair of twins, one of which had a
Guernsey sire, the other a Hereford sire. He showed that each of these
twins had antigens that could only have come from the sire of the other
and, similarly, had two kinds of red cells. That is to say, each had a kind
of cell proper to its own genetic composition and another kind proper to
that of its twin. This was interpreted as being due to the anastomosis of
blood vessels in the placenta, already shown by Lillie (1916) and others
to result in the passage of hormones from male embryos to their female
co-twins. The new results indicated the reciprocal passage of cells ancestral to red cells; since the cells characteristic of a twin were found to
persist into adult life, it followed that these foreign erythropoietic cells
persisted and reproduced in the recipient animals.
The consequences of this finding were far reaching, and have led to
developments in the study of the immunological basis of individual
specificity, acquired tolerance to foreign tissue, and other topics (Owen,
Medawar, and their co-workers).
C HAPTER 16
B IOCHEMICAL G ENETICS
There have been two chief biochemical approaches to the study of
genetics through the biochemical study of the effects of gene substitutions, and through a direct attack on the chemical nature of the genetic
material itself. Both approaches have been highly successful in recent
years, but both went through a rather long period of slow development
that was often rather discouraging.
The study of the biochemical effects of genes may be dated from the
work of Garrod on alkaptonuria in man. In 1902, he concluded that this
condition is an inherited one and that it is due to an alternative pathway in
the metabolism of nitrogenous materials, leading to the excretion of homogentisic acid rather than its further degradation product, urea in the
urine. He consulted with Bateson on the genetic question, and in 1902 the
latter discussed the case, giving what appears to be the first suggestion of
gene action in terms of ferments (enzymes) and of a recessive owing its
properties to the absence of a particular ferment. Later, Garrod (1908)
discussed the case in much more detail and concluded that the condition is
due to the blocking of a particular enzymatically controlled metabolic reaction, leading to an accumulation and excretion of the substrate (homogentisic acid) normally degraded by this reaction. He deduced the probable
earlier compounds produced in this degradation and administered them to
alkaptonuric patients. They were degraded to homogentisic acid, thus
showing that the mechanism was interrupted at that one point but was still
capable of functioning up to that point.
He also drew similar conclusions concerning some other less fully
analyzed biochemical defects in man such as albinism, cystinuria, and
porphyrinuria.
This work was early recognized as of great importance in the study
of the chemistry of metabolism; it illustrated the principles of blocking of
specific metabolic pathways, and the resultant accumulation of intermediate products, in intact living organisms. In 1909 the results were dis100
101
102
A H I ST O RY
OF
G EN ET IC S
103
This type of argument has since been applied to many systems and has
proved to be one of the most powerful methods in the analysis, through
the study of mutant types, of metabolic syntheses in intact organisms.
The study of this system was actively carried on by Ephrussi and coworkers (Khouvine and others), by Beadle and his co-workers (Tatum
and others), and by others in several laboratories. These studies, culminating in the identification of the v+ substance as kynurenine (Butenandt,
Weidel, and Becker, 1940) have been described many times for example by Ephrussi (1942).
The use of Drosophila for this kind of work was stimulated by the
existence of a large number of mutant types, whose interaction, it was
hoped, could be useful in future studies; but with the discovery that the
great majority of these are autonomous it became evident that the transplantation technique was not useful in their analysis.
This realization led Beadle and Tatum to look for a more favorable
object. They chose the fungus Neurospora. The techniques for handling
this material had been worked out by Dodge and by Lindegren, who had
also used it for interesting genetic studies. It had also been shown by
Robbins that Neurospora can be grown on a relatively simple synthetic
minimal medium, containing a carbon source, small amounts of inorganic salts, and a single complex organic compound (biotin). On this
medium the plant manufactures for itself the other organic substances
that it needs amino acids, vitamins, and other, still more complex, substances such as proteins. Beadle and Tatum reasoned that it should be
possible to detect and to study any mutants affecting the ability to synthesize such substances, providing they were reparable. That is to say, if
the substance not produced by a mutant could be supplied in the medium
and then utilized, growth should occur, and genetic and biochemical
studies could be carried out.
The plant has other advantages for this purpose. It is haploid, so that
dominance and recessiveness do not complicate the analysis; it usually
reproduces asexually by the production of numerous conidiospores,
which make it possible to work with very large numbers of individuals of
identical genetic composition and also simplify the detection of large
numbers of mutant individuals; sexual reproduction can occur, making
possible a detailed genetic analysis of the mutants produced.
In their first paper on the subject, Beadle and Tatum (1941) reported
the recovery of three mutant strains after X-ray treatment. These grew
normally on a complete medium made by adding malt extract and yeast
extract to the minimal medium described earlier but were (unlike the
wild type from which they came) unable to grow on the minimal medium
104
A H I ST O RY
OF
G EN ET IC S
In the next year he suggested that it is the nucleic acid component that is
responsible for heredity.
The existence of two kinds of nucleic acid was shown by Kossel in his
105
studies of material derived from thymus and from yeast cells. These two
types, long known as thymus nucleic acid and yeast nucleic acid, are
now known as deoxyribose nucleic acid (DNA) and ribose nucleic acid
(RNA), respectively. Ascoli (1900) and Levene (1903) showed that both
contain adenine, cytosine, and guanine, while the thymine of DNA is replaced by uracil in RNA. Levene also established the presence of deoxyribose and ribose, respectively, as the sugars present.
The early analyses of nucleic acids suggested that the four bases
were, in each case, present in equimolar amounts, and the tetranucleotide interpretation was developed. According to this view, an essential
component of the nucleic acid molecule is a unit composed of one of
each of the four bases. Chargaff and his co-workers showed (1950 and
later) that in DNA the bases need not be present in equimolar amounts,
but that the amount of cytosine does equal that of guanidine, and that of
adenine is equal to that of thymine.
With the development of a stain specific for DNA (Feulgen and
Rossenbeck, 1924) and, later, of enzymatic methods for distinguishing
DNA and RNA, it became possible to study the distribution of both substances in the cell. Feulgen (1937) showed that, in most cells, almost all
the DNA is in the nucleus.
Another approach grew out of the studies of Griffith (1928) on the
bacterium Pneumococcus. It had been found previously that this organism
exists in a number of serologically distinct types (Neufeld and Haendel,
1910; Dochez and Gillespie, 1913, and others, later). The type specificity
is due to differences in the mucopolysaccharides in the capsular envelopes
of the bacteria, and in culture of the strains in vitro it may be lost, leading
to rough strains that have lost both their type specificity and their pathogenicity. Griffith (1928) used a rough strain that had been derived from a
Type II strain, which he injected into mice along with killed individuals of
Type III. The mice died, and from their bodies Griffith recovered virulent
Type III strains of Pneumococcus. This was the first example of transformation, as the phenomenon came to be called. Dawson and Sia (1931) succeeded in getting it in vitro instead of making the mixture in a mouse. The
studies were continued, in an effort to isolate the transforming principle
from the killed virulent strain, and in 1944 it was shown by Avery, MacLeod, and McCarty that this agent is in fact DNA though the specificity
conferred by it and inherited by the descendants of the transformed cells is
due to a polysaccharide.
It had been generally supposed that the degree of specificity evidently present in hereditary material could only reside in proteins. Partly
because of the tetranucleotide interpretation, it had seemed that the nu-
106
A H I ST O RY
OF
G EN ET IC S
cleic acids were too simple for this degree of specific diversity. With the
breakdown of the tetranucleotide theory, and even more because of the
Pneumococcus result, attention turned to DNA. The spectacular results in
this field in recent years are not within the scope of this book, since they
have been described many times, and are still being very actively developed. I can only mention the solution of the structure of DNA by Watson
and Crick (1953) and the work on the RNA code by Nirenberg, Ochoa,
and many others.
C HAPTER 17
108
A H I ST O RY
OF
G EN ET IC S
Here was the essence of the basic formula of population genetics, though
it was derived by a longhand method and was not stated in simple algebraic form. The result, without selection, was also derived by Pearson
(1904), that is, for the case where p = q, where p = the frequency of A
genes, q = the frequency of a, and p + q = 1.
The generalization that the stable frequency of genotypes is
p2AA : 2pqAa : q2aa was made by Hardy and by Weinberg, independently, in 1908. Both knew of Pearsons result. To Hardy, who was a
mathematician, the generalized result seemed so self-evident that he
commented: . . . I should have expected the very simple point which I
wish to make to have been familiar to biologists. That it was not familiar is shown by the fact that it had been seriously suggested that dominant genes would automatically increase in frequency in mixed
populations.
The Hardy-Weinberg formula is strictly valid only if several conditions are fulfilled:
1. The population must be large enough so that sampling errors can be
disregarded. As Hardy pointed out, any chance deviations in the values of p and q will be as stable in succeeding generations as were
those of the preceding one.
2. There must be no mutation, since change of A to a or of a to A will
alter the values of p and q.
3. There must be no selective mating.
4. There must be no selection, that is, A and a must have no differential
effect on the reproductive capacity of individuals bearing them.
(In the requirements 2 and 4, the wording given is not strictly correct, since it is possible that balanced mutation and selection rates may
exist in which case there will be no net changes in the frequencies of A
and a.)
These are rather stringent requirements, and it may be doubted if
they are ever fully met; nevertheless, they are often approximated closely
enough to make the formula useful in analysis of populations.
The further developments in this field have depended on the algebraic analysis of the effects of deviations from the four requirements.
This development dates from Haldanes (1924 and later) analysis of the
effects of selection. He determined the number of generations required to
alter gene frequencies, as related to the intensity of selection. This was
worked out for dominant and for recessive genes, both in haploid and in
diploid organisms, and for sex-linked as well as for autosomal genes.
This analysis has been followed by detailed studies on the algebraic
AND
E VOLUTION
109
110
A H I ST O RY
OF
G EN ET IC S
AND
E VOLUTION
111
112
A H I ST O RY
OF
G EN ET IC S
nads of the hybrid blocked at almost any point most often just before meiosis; meiosis may be abnormal or essentially normal; the
gametes may be normal in structure and be functional but may give
rise to some or many abnormal or sterile individuals in the next generation. In other words, there are many different kinds of mechanisms
that prevent or hinder interspecific exchange of genes. The problem
is, how do these arise?
There is probably no single general answer to this question, but
there is an answer applicable to many cases, namely, polyploidy.
The study of polyploidy may be dated from the work of Boveri
and others in the 1880s and 1890s on the two races of Ascaris
megalocephala univalens, with one pair of chromosomes in the
germ line, and bivalens, with two pairs. There is some question about
this being a simple case of polyploidy. The first unambiguous cases
seem to be those reported by the Marchals (1906, 1907) in mosses
(described in Chapter 13) and since studied in great detail by F. von
Wettstein, and in Oenothera by Lutz, also in 1907 (see Chapter 10).
Other more or less probable examples followed (Strasburger, Tischler, and others) together with cytological studies of meiosis in the
triploids produced by crosses between diploids and tetraploids. The
results were confusing and contradictory, until the analysis by Winge
(1917), which began to clarify the situation.
Winge made a detailed study of the available data on chromosome
numbers in plants and found that in many groups there was a basic
number, with various multiples of this number represented in different
species. He pointed out that, in a hybrid between two species, it
sometimes happens that some (or all) chromosomes are sufficiently
different so that they do not pair at meiosis, so that the resulting gametes do not all contain a single complete set of chromosomes, and
partial, or essentially complete, sterility results. Now, if the chromosomes of such a hybrid are doubled (by chromosome division without
an accompanying cell division), each chromosome will now have an
exact mate, and meiosis can be expected to be normal with a restoration of fertility, as had already been shown by Federley (1913). This
process was demonstrated by Clausen and Goodspeed (1925) and
Clausen (1928), in Nicotiana. N. tabacum (24 pairs of chromosomes)
was crossed to N. glutinosa (12 pairs). The hybrid, with 36 chromosomes, was sterile and at meiosis showed only a few paired chromosomes. One hybrid individual was fertile, however, and was found to
be a tetraploid, with 72 chromosomes that formed 36 bivalents at
AND
E VOLUTION
113
meiosis; this plant was fully fertile and bred true to type.
The terminology which was suggested by Kihara and Ono (1926) has
been generally accepted and has helped to clarify the relations. They
suggested that:
Under polyploidy we must distinguish two phenomena, namely,
autopolyploidy and allopolyploidy. Autopolyploidy signifies the doubling of the same chromosome set; allopolyploidy the multiplication
of different chromosome sets brought together by hybridization.
114
A H I ST O RY
OF
G EN ET IC S
AND
E VOLUTION
115
This conclusion was based on solid facts derived from extensive species crosses in the genus Gossypium (cotton). I have elsewhere (Sturtevant, 1948) given my reasons for an alternative interpretation of these
facts, based on the polyploid nature of cotton. The more recent comparative biochemical data also favor the idea of the great stability of genetic
systems, since they show essential identity of some of the gene-controlled basic biochemical pathways in bacteria, fungi, and vertebrates.
It is true, however, that in many, probably most, loci there exist series of isoalleles (Stern and Schaeffer, 1943), which carry on the function
116
A H I ST O RY
OF
G EN ET IC S
C HAPTER 18
P ROTOZOA
Weismanns germ-plasm theory was based on the differentiation of
germinal and somatic cell lines in many-celled organisms. In unicellular
forms there is evidently no such cellular differentiation, and Weismann
surmised that these organisms are potentially immortal a view already
expressed by Ehrenberg in 1838. This conclusion was questioned as a
result of the growing knowledge of the occurrence and nature of conjugation first noted in the ciliate Infusoria.
Conjugation was observed by Leeuwenhoek (1695) and by O. F.
Mller (1786) but was usually misinterpreted until the work of
Btschli (1873 and later) and Engelmann (1876). The ciliates have
two types of nuclei macronucleus and micronucleus which were
earlier interpreted as ovary and testis, respectively. Btschli and
Engelmann made out their nature and found that the macronucleus
was lost at the time of conjugation. They developed the view that
these animals are not potentially immortal but have a definite life
cycle that requires conjugation to renew the vitality of the line (which
becomes senescent and dies out in the absence of conjugation). This
speculation was based on observation of the structural changes.
The cytological details of conjugation were worked out independently by Maupas and by R. Hertwig in 1889. They described the degeneration of the macronucleus, the exchange of micronuclei with the
fusion of those from the two individuals, and the reconstitution of the
new macronucleus from a division product of the biparentally derived
new micronucleus. With these results, it became possible to relate the
process of conjugation to the better-known sexual reproduction of
higher forms.
Maupas had already (1888) published his elaborate studies on the
relation of conjugation to senescence and rejuvenescence and had fur-
117
118
A H I ST O RY
OF
G EN ET IC S
P ROTOZOA
119
when paired with extreme males, and as males when paired with extreme
females.
Similar systems were found in other algae and were studied in detail
by Hartmann and his students. One of the most striking cases, that of
Chlamydomonas, was studied by Moewus, who developed an elaborate
biochemical and genetic interpretation. This work at one time figured
largely in the literature of genetics, though there were doubters almost
from the beginning. It has now been very generally concluded that the
experimental results of Moewus are wholly unreliable, and that the data
were largely made up to fit his theories rather than based on observations. This is, in fact, now considered to be one of the very few scientific
scandals in biology (see Ryan, 1955, Science, 122:470).
There is no doubt about the reality of the phenomenon of relative
sexuality, and it is clearly something that needs study and analysis; but
the papers of Moewus have unfortunately given it unpleasant associations.
There is one special process that occurs in Paramecium aurelia
and rarely or not at all even in other species of the genus that has
played a large part in the genetical study of the ciliates. This process,
called endomixis (or, now, autogamy) was described by Woodruff and
Erdmann in 1914; their account of the cytological details was corrected
and brought into better agreement with later genetic results by Diller in
1934. In this process, the macronucleus degenerates and is replaced by a
division product of the micronucleus. The micronucleus itself undergoes
meiosis, and one of the resulting haploid nuclei divides to give two identical nuclei a process similar to what happens at conjugation. But at
conjugation one of these two genetically identical nuclei migrates to the
other individual, and there fuses with the nonmigratory product in that
individual, thus producing in each ex-conjugant a diploid biparental micronucleus, from which the whole nuclear complement of its descendants
arises by division. In autogamy, the two identical sister micronuclei fuse,
and again the whole nuclear complement of the descendants arises by
division from this uniparental diploid nucleus. It follows from this that,
after autogamy has occurred, a line of P. aurelia individuals is genetically homozygous until conjugation (or mutation) occurs.
One result of the more recent studies is the demonstration that the
phenotype of a ciliate is dependent on its macronucleus, which was derived from a micronucleus and does not survive conjugation or autogamy. The germ-plasm theory is valid here; the macronucleus is somatic,
the micronucleus is germinal.
120
A H I ST O RY
OF
G EN ET IC S
The unusual properties of the ciliates have been exploited by Sonneborn and his co-workers (Kimball, Preer, Beale, Siegel, and others see
Beale, 1954, for a summary) to produce a coherent body of knowledge
about the genetics of the ciliates that is important in many current developments but is beyond the scope of this book.
C HAPTER 19
M ATERNAL E FFECTS
Aristotle wrote at length on the relative effects of the male and female parents on the properties of their offspring his discussion is now
interesting as an example of the Aristotelian method rather than as a real
contribution. With the increase in knowledge of development there
gradually appeared the idea of preformation, according to which the fertilized egg contains the parts of the developing individual in miniature
development consisting of the unfolding of these parts in a manner similar to the development of a flower from a bud. In its extreme form, this
led to the conclusion that each egg also contains miniature representatives of the eggs of all potential future descendants.
With the development of clearer ideas about fertilization, two
schools emerged: the ovists who thought the preformed parts were
contained in the unfertilized egg and were merely activated by the sperm,
and the spermists who thought of the sperm as a complete animalcule
that was merely nourished by the egg.
C. F. Wolff (1759) initiated a reaction from this view. Wolff thought
of the fertilized egg as a relatively homogeneous structure, from which
the parts of the embryo developed de novo. This view, known as epigenesis, was more in accord with the direct observations of embryologists and avoided some of the absurdities to which the preformationists
had been led. It came to dominate the thinking of embryologists. It was
also more easily reconciled with the cell theory and with the experimental results of the hybridizers.
The older views implied inequalities between the parents in the determination of the properties of their offspring; for example, that the form
was determined by the mother, the color by the father. I have, in fact, encountered such views currently held by a few amateur plant breeders.
Klreuter (17611766) seems to have been the first to carry out systematic reciprocal crosses, and to have concluded that the two parents
121
122
A H I ST O RY
OF
G EN ET IC S
contributed equally to the characteristics of their offspring. This conclusion was confirmed by most of the plant hybridizers who followed him
(see Chapter 1), but was long resisted by zoologists.
It may be supposed that the zoologists were at first influenced by
the often-discussed differences between the mule and the hinny the
results of reciprocal crosses between the horse and the ass. Hinnies
are rare, and I have never been able to find a satisfactory account of
them. The only supposed hinny I have ever seen impressed me as being merely a small mule and the smaller size is the only generally
recognized peculiarity. This may be due to their having a smaller
mother; it is also probable that they are usually the offspring of small
and inferior individuals of both parent species, since they are usually
accidental in origin.
In later times zoologists were undoubtedly influenced by the study of
hybrid embryos of marine animals, especially species hybrids in sea urchins. Here the effects of foreign sperm are sometimes not at once apparent, and the hybrid embryo begins its development according to the
maternal plan. Often the later embryos from reciprocal crosses are not
distinguishable (Boveri, 1892, 1903; Driesch, 1896, 1898). These authors
concluded that up to a certain point the development is controlled entirely by the cytoplasm of the egg though Boveri later realized that this
cytoplasmic specificity might be under the control of the chromosomes
of the mother. Others, however, were led to conclude that the general
ground plan of development is not under chromosomal determination. As
Loeb (1916, 1919) expressed it, the embryo in the rough is determined
by the cytoplasm alone, or as Conklin (1918) stated, we are vertebrates
because our mothers were vertebrates and produced eggs of the vertebrate pattern; but the color of our skin and hair and eyes, our sex, stature
and mental peculiarities were determined by the sperm as well as by the
egg from which we came.
The suggestion that the maternal cytoplasm in such cases may be
determined by the chromosomal genes of the mother received experimental support from the work of Toyama (1912) on the color of the
embryonic serosa in the silkworm, and more clearly in the case of the
snail Limnaea (Boycott and Diver, 1923; Sturtevant, 1923). In this
form the shell may be coiled either dextrally or sinistrally, the two
types being exact mirror images of each other. It turned out that the
difference is due to a single pair of genes, with dextral dominant; but
the direction of coiling is determined not by the constitution of the
individual but by that of its mother. Thus, for example, a heterozygous individual, mated as a female (the animals are hermaphroditic)
M AT ERNAL E FFECT S
123
to another from a pure sinistral line, will produce only dextral offspring, even though half of these individuals do not carry the gene for
dextral coiling. The fate of the egg is thus determined, before polarbody formation and fertilization, by the genes of the mother. The nature of the coiling is visibly determined early in development; the two
types can be distinguished by the pattern of cleavage at the second
division following fertilization; but the mirror image relation persists
throughout the life of the animal.
The somewhat similar case of the color of the sap in pollen grains
studied by Correns has already been described in Chapter 5. Many
other instances are known; for example, the gene affecting sex in
Drosophila neorepleta, described in Chapter 13. In a few of them, the
maternal nature of the hybrids is due to a failure of some or all of the
paternally derived chromosomes to persist in the foreign cytoplasm
(Baltzer, 1909; Godlewski, 1911).
There is another group of maternally inherited characteristics that
is different in kind, namely, certain chloroplast defects in plants. In
many plants there are ordinary chromosomal genes that affect the
green pigment, giving regular Mendelian results, with white or pale
green seedlings segregating in the usual ratios. One of these, in the
snapdragon, has already been described (Chapter 8) as the first clearly
demonstrated lethal gene (Baur, 1907, 1908).
In many plants there are strains in which the leaves and stems are
variegated with respect to chlorophyll color; some of these behave
differently. The first unambiguous case of maternal inheritance in
such strains was reported by Correns (1909) in Mirabilis. In one strain
of this plant the leaves are irregularly mottled dark green and yellowish white, the difference being in the color of the individual plastids. At the boundaries between the two areas there are some cells that
contain both kinds of plastids. The pattern is so irregular that some
branches are wholly green and others are wholly white. Correns used
flowers on such uniform branches and found that seeds from those on
wholly green branches gave green offspring only, regardless of the
source of the pollen used; those from wholly white branches gave
white seedlings only, again without regard to the source of the pollen
used; flowers on variegated branches gave seeds that produced green,
variegated, or white seedlings whether selfed or pollinated by wholly
green plants. Evidently, then, the plastids act as though they or their
precursors were self-reproducing bodies, with their properties unaffected by the chromosomal genes.
124
A H I ST O RY
OF
G EN ET IC S
M AT ERNAL E FFECT S
125
C HAPTER 20
T HE G ENETICS OF M AN
Man is, in many ways, very unsuitable as an object for the study of
genetics. Families are too small for dependable determination of ratios,
desired test matings cannot be made, and study of more than a very few
generations for any particular purpose is not often possible. The social
implications of human genetics are so great, however, that the subject
must be investigated; and there are some real advantages in the material.
For no other organism do we have such detailed and extensive information on anatomy, development, biochemistry, physiology, pathology,
evolution, and population statistics. These advantages have, in fact, led to
important advances in basic genetics through the study of human material, notably in connection with the blood groups (see Chapter 15) and
the biochemistry of hemoglobin variants.
The systematic study of the genetics of man began before the Mendelian era with the work of Francis Galton, beginning in 1865; the two
best-known accounts of his work are the books Hereditary Genius (1869)
and Natural Inheritance (1889).
Following 1900 there accumulated a body of information concerning
the Mendelian inheritance of a large series of aberrant conditions in man,
beginning with Farabees account of brachydactyly (short fingers) in
1905.
In 1902, Garrod and Bateson suggested that alkaptonuria is due to a
single recessive gene, but the evidence did not seem conclusive until
Garrod reported additional families in 1908. This case illustrates one of
the difficulties in the study of the genetics of man, namely, the difficulty
of finding an adequate number of critical families.
It is important that suspected cases of Mendelian inheritance in man
should be recorded, so that they may be checked by other workers and, if
valid, may be incorporated in studies of possible linkage and of anthropological questions. There is an unfortunate tendency, however, to accept
126
T HE G ENET ICS
OF
M AN
127
128
A H I ST O RY
OF
G EN ET IC S
T HE G ENET ICS
OF
M AN
129
biologists including some who were more sophisticated than Davenport. Two examples follow:
Bateson (1912, Biological Fact and the Structure of Society, Herbert Spencer lecture, Oxford):
How hard it is to realize the polymorphism of man! Think of the varieties which the word denotes, merely in its application to one small
society such as ours, and of the natural genetic distinctions which differentiate us into types and strains acrobats, actors, artists, clergy,
farmers, labourers, lawyers, mechanics, musicians, poets, sailors, men
of science, servants, soldiers, and tradesmen. Think of the diversity of
their experience of life. How few of these could have changed parts
with each other. Many of these types are, even in present conditions,
almost differentiated into distinct strains . . . I never cease to marvel
that the more divergent castes of civilized humanity are capable of
interbreeding and of producing fertile offspring from their crosses.
Nothing but this paradoxical fact prevents us from regarding many
classes even of Englishmen as distinct species in the full sense of the
term.
130
A H I ST O RY
OF
G EN ET IC S
T HE G ENET ICS
OF
M AN
131
of this attitude have not usually been scientifically trained; the terrible example is Hitler, of course, but he was preceded by many pseudoscientific writers (such as Gobineau, Houston Chamberlain, and
Madison Grant), most of whom would have been horrified by Hitlers
methods. There have, however, been biologists with some background
in genetics who have leaned in this direction. Since racism is a dirty
word, it is perhaps kinder (and certainly more agreeable to the writer)
not to name them.
Galton was one of the first to suggest the possibility of the genetic improvement of human populations; he introduced the word
eugenics to designate this field of study and planning. There are two
approaches here, which have been described as negative and positive. The first proposes to decrease or eliminate the more extreme
inherited defects physical and mental and the second proposes to
increase the number of better individuals, and thereby to make possible the production of still better ones. Both approaches, especially the
positive one, are based on the obvious success of animal and plant
breeders in improving the populations with which they work.
It is estimated that something like 4 percent of human infants
have tangible defects that can be detected in infancy some of them
very serious and others much less so, and some of them remediable
and others not. It is also estimated that perhaps about half of these are
largely genetic in origin. If it were possible to eliminate these by preventing their birth, this would obviously be a great advantage to society, in economic and, especially, in humanitarian terms.
In the early days of Mendelism, there were many people who felt
that this objective could be rather simply achieved, but with increased
knowledge this hope has been somewhat dimmed. The easiest class of
defects to eliminate should be the dominant, but it has turned out that the
more serious of these are apt not to appear until the normal reproductive
age has largely passed (the typical example here is Huntingtons chorea).
Presumably those that appear earlier in life have, for the most part, been
eliminated by natural selection. Any appreciable decrease in the incidence of recessive defects would depend on the identification of heterozygous carriers which is not usually possible. There has also come
to be a growing realization that, in some cases, heterozygosis for a particular gene may (at least under certain conditions) confer an advantage
even when homozygosis is very disadvantageous. The best-known example here is sickle-cell anemia in man. Homozygosis for this gene causes
the serious defect from which the name is derived; but it was shown by
132
A H I ST O RY
OF
G EN ET IC S
C HAPTER 21
G ENERAL R EMARKS
There is a widespread view that scientific discoveries are more or
less inevitable, and that it makes little difference whether or not a particular individual makes a discovery at a given time: if the time is not
ripe for it, it will not be understood and will have little or no effect on
future events; if the time is ripe, then someone else will soon make the
discovery anyhow.
The history of Mendelism is one of the often-cited examples here. According to this interpretation, Mendels paper was not understood in 1866
because the time was not ripe; in 1900, when the time was ripe, the principles were discovered independently by three different people. To me, this
account seems greatly oversimplified though it must be admitted that the
development of the subject would probably have been much the same,
even to the dates, if Mendels paper had never been written.
It is true that the paper was ahead of its time, but it was not difficult
to understand, and it seems unlikely that it would have remained unappreciated for so long if it had appeared in a less obscure journal, or if
Mendel himself had published the further cases that he reported in his
letters to Ngeli. It must be remembered that Ngelis failure to appreciate the paper in 1866 can be matched by Pearsons failure in 1904. Both
were outstanding men, and both were actively studying heredity, but to
both of them Mendels results appeared as trivial cases involving a few
superficial characters, obviously neither useful nor illuminating for any
general theory of inheritance. It does not follow that no biologist was
likely to have appreciated the paper if he had seen it before 1900 I have
suggested above that Galton, for one, would very likely have done so.
As for the simultaneous discovery in 1900, I have pointed out in
Chapter 4 that it seems likely that the independent discovery was the
finding of Mendels paper, and that the actual working out of the principles without knowledge of Mendels work was only accomplished
once by Correns and even here it is not possible to be certain how
133
134
A H I ST O RY
OF
G EN ET IC S
G ENERAL R EMARKS
135
A PPENDIX A
C HRONOLOGY
circa 323 B.C.
1676
1677
1716
1759
17611766
18231846
1853
1859
1866
1868
1871
1875
1881
18821885
1883
18831889
18881889
1889
1894
1900
1901
136
C HRONOLOGY
1902
1903
1904
1905
1906
1907
19071908
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1919
1921
137
138
A H IS TOR Y OF G E NE TIC S
1922
1924
1925
1926
1927
1928
1930
1932
1933
1935
1937
1940
1941
1944
1945
1946
A PPENDIX B
I NTELLECTUAL
P EDIGREES
The accompanying diagrams are presented with some hesitation,
since they are certainly greatly oversimplified. In general, the attempt has
been to show teacher-student relationships and to neglect the interactions
between contemporaries and of students on their teachers both of
which are evidently often important. They also neglect the effects of
brief contacts and of influences through reading, rather than direct personal relations. Finally, they are very incomplete even for what has been
attempted, and I am sure there are inaccuracies.
There nevertheless seems to be some advantage in attempting to give
a picture of the various interrelations and a sense of the continuity of personal influences even in an incomplete and imperfect form.
The information has been derived from many sources: biographies
and other published accounts, my own personal knowledge, and the help
and advice of numerous friends about their own backgrounds and about
others with whom they were familiar.
The diagrams are self-explanatory for the most part. It will be seen
that the same name often occurs in several; this is because of an effort to
keep them in a relatively simple form. In one case (p. 141, top) there is a
series of names enclosed in parentheses; these represent men who came
as postdoctoral students, already trained elsewhere, to the California Institute of Technology where they were influenced by the group shown.
Many names that might well have been included are omitted in order
to simplify the diagrams; I have had to be quite arbitrary about this.
139
140
A H IS TOR Y OF G E NE TIC S
Buffon
Lamarck
Cuvier
Purkinje
de Candolle
Ngeli
Sachs
Haberlandt
Pfeffer
de Vries
Correns
Johannsen
von Wettstein
Winge
J. Loeb
Westergaard
Kant
Fichte
Schelling
Harrison
Kielmayer
Dllinger
Oken
Huxley
Agassiz
Martin
Brooks
Morgan
Wilson
Bridges
Metz
Carson
Cuvier
Conklin
Gowen Plough
Schultz
Schrader Lancefield
Cooper
Ryan
I NTELLECTUAL P EDIGREES
Morgan
Bridges
Sturtevant
Anderson
S. Emerson
Poulson
E. B. Lewis
Novitski
Lindegren
141
Dobzhansky
Cuvier
Schultz
( Beadle
Rhoades
Ephrussi
Stern
Delbrck
Owen )
Wagner
Leuckart
Mark
Gerould
Cole
Weismann
Davenport
Petrunkewitsch
Khn
G. H. Shull
Castle
Jennings
Caspari
Bauer
Zeleny
Dunn
Little
Irwin
Wright
Charles
Stormont
Owen
Russell
Sonneborn
Preer
Beale
Rudolphi
von Klliker
J. Mller
Gegenbaur
Beerman
Nanney
Siegel
Haeckel
O. Hertwig
Roux
Driesch
Baltzer
Hadorn
R. Hertwig
Boveri
Nachtsheim
Strasburger
Goldschmidt
Painter
Lang
Philiptschenko
Spemann
Morgan
Dobzhansky
142
A H IS TOR Y OF G E NE TIC S
Whitman
Marsh
Wheeler
Williston
Wilson
McClung
Robertson
Carothers
Sutton
Foster
Sedgwick
Balfour
Brooks
Bateson
Hopkins
Punnett
Saunders
Winge
Vavilov
Haldane
Newton
Westergaard
Darlington
Torrey
A. Gray
Bessey
East
Jones Brink
Giles
Sax
Rick Swanson
R. A. Emerson
Stadler
Anderson
Roman
Laughnan
Beadle Rhoades
S. Emerson
Demerec
McClintock
I NTELLECTUAL P EDIGREES
143
J. Mller
Henle
Virchow
Schwann
Haeckel
Btschli
R. Hertwig
Helmholtz
Koch
Ehrlich
Schaudinn
Hartman
Belar
Morgan
Goldschmidt
Stern
Nachtsheim
Seiler
Agassiz
Whitman
Lillie
Brooks
Martin
Oliver
E. B. Lewis
R. Hertwig
Petrunkewitsch
Boveri
Williston
Harrison
Muller
Weismann
Patterson
Painter
Glass
Stone
M. Green
Crow
Wilson
Schrader
Morgan
Tyler
Wagner
R. Emerson
Beadle
Peterson
Ryan
Horowitz
Tatum
Lederberg
Doerman
Srb
Fred
B IBLIOGRAPHY
The list of publications that follows does not include all the references given in the text, but it is designed to cover the most important
papers in the mainstream of genetics, plus some accounts that are themselves useful sources of literature citations. A fuller listing might have
been useful to the serious student, but that seemed to be out of proportion
with the rest of this book. The selection made is, unfortunately, somewhat arbitrary.
Allen, C. E. 1917. A chromosome difference correlated with sex in Sphaerocarpos. Science, 46: 466467.
Allison, A. C. 1954. Protection afforded by sickle-cell trait against subtertian
malarial infection. Brit. Med. J. (4857): 290294.
Anderson, E. G. 1925. Crossing over in a case of attached X chromosomes in
Drosophila melanogaster. Genetics, 10: 403417.
Aristotle. Historia Animalium. Translated by DArcy W. Thompson. 1910.
Clarendon Press, Oxford (vol. 4, The Works of Aristotle). Original work,
about 323 B.C.
Aristotle. De Generatione Animalium. Translated by A. Platt. 1912. Clarendon
Press, Oxford (vol. 5, The Works of Aristotle). Original work, about 323
B.C.
Auerbach, C., and J. M. Robson. 1946. Chemical production of mutations. Nature, 157: 302.
Avery, O. T., C. M. MacLeod, and M. McCarty. 1944. Studies on the chemical
nature of the substance inducing transformation of pneumococcal types. J.
Exper. Med. 79: 137158.
Balbiani, E. G. 1881. Sur la structure du noyau des cellules salivaires chez les
larves de Chironomus. Zool. Anz., 4: 637641, 662666.
Baltzer, F. 1909. ber die Entwicklung der Echiniden-bastarde. Zeits. ind. Abst.
Vererb., 5.
Bateson, B. 1928. William Bateson, Naturalist. Cambridge Press, Cambridge.
473 pp.
Bateson, W. 1894. Materials for the Study of Variation. The Macmillan Co.,
New York. 598 pp.
144
B IBLIOGRAPHY
145
146
A H IS TOR Y OF G E NE TIC S
B IBLIOGRAPHY
147
148
A H IS TOR Y OF G E NE TIC S
B IBLIOGRAPHY
149
Fisher, R. A. 1928. The possible modification of the response of the wild type to
recurrent mutations. Amer. Nat., 62: 115126.
Fisher, R. A. 1930. The Genetical Theory of Natural Selection. Clarendon Press,
Oxford. 272 pp.
Fisher, R. A. 1936. Has Mendels work been rediscovered? Ann. Sci. 1: 115
137.
Focke, W. O. 1881. Die Pflanzenmischlinge. 570 pp. Borntraeger, Berlin.
Gabriel, M. L., and S. Fogel. 1955. Great Experiments in Biology. Prentice-Hall,
Englewood Cliffs, N.J. 317 pp.
Galton, F. 1869. Hereditary Genius. Macmillan & Co., London. (Second edition
1892.)
Galton, F. 1883. Inquiries into Human Faculty. Macmillan & Co., London.
Galton, F. 1889. Natural Inheritance. Macmillan & Co., London. 259 pp.
Galton, F. 1908. Memories of My Life. Methuen & Co., London. 339 pp.
Garrod, A. E. 1908. Inborn errors of metabolism. Lancet, 2: 17, 7379, 142
148, 214220. Also separately published as a book, 1909. Oxford
University Press, London. 168 pp.
Glass, B. 1947. Maupertuis and the beginnings of genetics. Quart. Rev. Biol., 22:
196210.
Goldschmidt, R. 1912. Erblichkeitsstudien an Schmetterlingen, I. Zeits. ind.
Abst. Vererb., 7: 162.
Goldschmidt, R. 1933. Lymantria. Bibliogr. Genet., 11: 1186.
Gowen, J. W., and E. H. Gay. 1933. Eversporting as a function of the Y-chromosome in Drosophila Melanogaster. Proc. Nat. Acad. Sci., 19: 122126.
Gowen, J. W., and E. H. Gay. 1933. Effect of temperature on eversporting eye
color in Drosophila melanogaster. Science, 77: 312.
Green, M. M., and K. C. Green. 1949. Crossing-over between alleles at the lozenge locus in Drosophila melanogaster. Proc. Nat. Acad. Sci. 35: 586591.
Griffith, F. 1928. The significance of Pneumococcal types. J. Hyg., 27: 113159.
Guyer, M. F. 1902. Some notes on hybridism, variation, and irregularities in the
division of the germ-cell. Science, 15: 530531.
Guyer, M. F. 1903. The germ-cell and the results of Mendel. Cincinnati LancetClinic. 2 pp.
Haacke, W. 1893. Die Trger der Vererbung. Biol. Centralbl., 13: 525542.
Haldane, J. B. S. 1924. A mathematical theory of natural and artificial selection.
Proc. Cambridge Phil. Soc., 23: 1941, 158163, 363372, 607615, 838
844.
Hanson, F. B., and F. Heys. 1929. An analysis of the effects of the different rays
of radium in producing lethal mutations in Drosophila. Amer. Nat. 63: 201
213.
150
A H IS TOR Y OF G E NE TIC S
Harland, S. C. 1936. The genetical conception of the species. Biol. Rev., 11: 83
112.
Heitz, E. 1933. ber totale und partielle somatische Heteropycnose, sowie
strukturelle Geschlechtschromosomen bei Drosophila funebris. Zeits. Zellf.
mikr. Anat. 19: 720742.
Heitz, E., and H. Bauer. 1933. Beweise fr die Chromosomennatur der
Kernschleifen in den Knuelkernen von Bibio hortulanus. Zeits. Zellf. mikr.
Anat. 17: 6782.
Henking, H. 1891. ber Spermatogenese bei Pyrrhocoris apterus. Zeits.
wissensch. Zool., 51: 685736.
Hershey, A. D. 1946. Spontaneous mutations in bacterial viruses. C. Spr. Harb.
Sympos. Quant. Biol., 11: 6777.
Hirszfeld, L., and H. 1919. Serological differences between the blood of different races. Lancet 1919 (2): 675679.
Iltis, H. 1924. Gregor Johann Mendel, Leben, Werk und Wirkung. Springer,
Berlin. Translated, E. and C. Paul, 1932 (title of translation, Life of Mendel).
W. W. Norton & Company, New York. 336 pp.
Imai, Y. 1928. A consideration of variegation. Genetics, 13: 544562.
Irwin, M. R., and L. J. Cole. 1936. Immunogenetic studies of species and of species hybrids in doves. J. Exper. Zool., 73: 85108.
Janssens, F. A. 1909. La thorie de la chiasmatypie. La Cellule, 25: 389411.
Jennings, H. S. 1929. Genetics of the Protozoa. Bibliogr. Genet., 5: 105330.
Johannsen, W. 1909. Elemente der exakten Erblichkeitslehre. Fischer, Jena. 516
pp.
Kihara, H., and T. Ono. 1926. Chromomenzahlen und systematische
Gruppierung der Rumex-Arten. Zts. Zellf. mikr. Anat., 4: 475481.
Lancefield, D. E. 1922. Linkage relations of the sex-linked characters in Drosophila obscura. Genetics, 7: 335384.
Landsteiner, K. 1900. Zur Kenntnis der antifermentativen, lytischen und agglutinativen Wirkungen des Blutserums und der Lymphe. Zbl. Bakt., 27: 357
362.
Landsteiner, K., and P. Levine. 1927. Further observation on individual differences of human blood. Proc. Soc. Exper. Biol. Med., 24: 941942.
Landsteiner, K., and A. S. Wiener. 1940. An agglutinable factor in human blood
recognized by immune sera for rhesus blood. Proc. Soc. Exper. Biol. Med.,
43: 223224.
Lewis, E. B. 1945. The relation of repeats to position effects in Drosophila
melanogaster. Genetics, 30: 137166.
Lewis, E. B. 1950. The phenomenon of position effect. Advances in Genetics, 3:
73115.
B IBLIOGRAPHY
151
152
A H IS TOR Y OF G E NE TIC S
B IBLIOGRAPHY
153
Oliver, C. P. 1930. The effect of varying the duration of X-ray treatment upon
the frequency of mutation. Science, 71: 4446.
Owen, R. D. 1945. Immunogenetic consequence of vascular anastomosis between bovine twins. Science, 102: 400401.
Painter, T. S. 1933. A new method for the study of chromosome rearrangements
and the plotting of chromosome maps. Science, 78: 585586.
Painter, T. S., and H. J. Muller. 1929. Parallel cytology and genetics of induced
translocations and deletions in Drosophila. J. Hered., 20: 287298.
Pavan, C., and M. E. Breuer. 1952. Polytene chromosomes in different tissues of
Rhynchosciara. J. Hered., 43: 151157.
Payne, F. 1918. An experiment to test the nature of the variations on which
selection acts. Indiana Univ. Studies, 5: 145.
Peters, J. A. 1959. Classic Papers in Genetics. Prentice-Hall, Englewood Cliffs,
N.J. 282 pp.
Renner, O. 1917. Die tauben Samen der Oenotheren. Ber. deutsch. botan. Gesellsch., 34: 858869.
Renner, O. 1919. Zur Biologie und Morphologie der mnnlichen Haplonten
einiger Oenotheren. Zeits. Bot., 11: 305380.
Renner, O. 1921. Heterogametie im weiblichen Geschlecht und Embryosackentwicklung bei den Oenotheren. Zeits. Bot., 13: 609621.
Renner, O. 1924. Die Scheckung der Oenotherenbastarde. Biol. Zentralbl., 44:
309336.
Renner, O. 1925. Untersuchungen ber die faktorielle Konstitution einiger komplexheterozygotischen Oenotheren. Biblioth. Genet., 9, 168 pp.
Rhoades, M. M. 1943. Genic induction of an inherited cytoplasmic difference.
Proc. Nat. Acad. Sci., 29: 327329.
Roberts, H. F. 1929. Plant Hybridization before Mendel. Princeton University
Press, Princeton, N.J. 374 pp.
Robertson, W. R. B. 1916. Chromosome studies. I. J. Morph., 27: 179331.
Roux, W. 1883. ber die Bedeutung der Kerntheilungsfiguren. Engelmann,
Leipzig, 19 pp.
Sachs, J. 1875. History of Botany. Translated by H. E. F. Garnsey 1890. Oxford
University Press, Oxford, 568 pp.
Schultz, J. 1936. Variegation in Drosophila and the inert chromosome regions.
Proc. Nat. Acad. Sci., 22: 2733.
Seiler, J. 1913. Das Verhaltung der Geschlechtschromosomen bei Lepidopteren.
Zool. Anz., 41: 246251.
Smith, T., and F. L. Kilborne. 1893. Investigations into the nature, causation and
prevention of southern cattle fever. U. S. Bur. Anim. Indust. Bull. 1. 301 pp.
154
A H IS TOR Y OF G E NE TIC S
B IBLIOGRAPHY
155
Sturtevant, A. H. 1948. The evolution and function of genes. Amer. Sci., 36:
225236. Reprinted in Science in Progress 6: 250256 (1949) and in Smithsonian Report for 1948, pp. 293304.
Sturtevant, A. H. 1959. Thomas Hunt Morgan, 18661945. Biograph. memoirs
Nat. Acad. Sci., 33: 283325.
Sturtevant, A. H., and G. W. Beadle. 1939. An Introduction to Genetics. Saunders, Philadelphia. 391 pp.
Sturtevant, A. H., and E. Novitski. 1941. The homologies of the chromosome
elements in Drosophila. Genetics, 26: 517541.
Sturtevant, A. H., and C. C. Tan. 1937. The comparative genetics of Drosophila
pseudoobscura and D. melanogaster. J. Genet., 34: 415432.
Sutton, W. S. 1902. On the morphology of the chromosome group in Brachystola magna. Biol. Bull., 4: 2439.
Sutton, W. S. 1903. The chromosomes in heredity. Biol. Bull., 4: 231251.
Todd, C. 1930. Cellular individuality in the higher animals, with special reference to the individuality of the red blood corpuscle. Proc. Roy. Soc., 106 B,
2044.
Tschermak, E. 1900. Ueber knstliche Kreuzung bei Pisum sativum. Zeits.
landw. Versuchsw. Oesterr., 3, Heft 5.
Tschermak, E. 1900. ber knstliche Kreuzung bei Pisum sativum. Ber. deutsch.
botan. Gesellsch., 18: 232239. Translation. 1950. The Birth of Genetics.
Suppl. Genetics. 4247.
Ullerich, F.-H. 1963. Geschlechtschromosomen und Geschlechtsbestimmung bei
einigen Calliphorinen. Chromosoma, 14: 45110.
Vries, H. de. 1889. Intracellulare Pangenesis. 149 pp. Fischer, Jena. (Translation by C. S. Gager, 1910. Open Court, Chicago.)
Vries, H. de. 1900. Sur la loi de disjonction des hybrides. C. R. Acad. Sci.
(Paris), 130: 845847. Translated, 1950, in The Birth of Genetics. Suppl.,
Genetics, 3032.
Vries, H. de. 1900. Sur les units des caractres spcifiques et leur application a
ltude des hybrides. Rev. gnr. botan., 12: 257271.
Vries, H. de. 1900. Das Spaltungsgesetz der Bastarde. Ber. deutsch. botan. Gesellsch., 18: 8390.
Vries, H. de. 1901. Die Mutationstheorie. Vol. 1. Veit, Leipzig. 648 pp.
Warmke, H. E., and A. F. Blakeslee. 1939. Sex mechanism in polyploids of
Melandrium. Science, 89: 391392.
Watson, J. D., and F. H. C. Crick. 1953. Genetic implications of the structure of
deoxyribonucleic acid. Nature, 171: 964.
Weismann, A. 18911892. Essays on Heredity. Translated by A. E. Shipley, S.
Schnland, and others. Oxford University Press, 2 vols., 471; 226 pp.
156
A H IS TOR Y OF G E NE TIC S
INDEX
Ascoli, 105
Aspergillus, 56
Ass, 122
Assortment, independent, 37
Attached X, 55, 70
Auerbach, C., 72
Autogamy, 119
Autopolyploidy, 113
Avery, A. G., 113
Avery, O. T. (18771955), 104, 105, 138
Bach family, 128
Background radiation, 72
Bacteria, 56, 79, 104, 115, 135
Bacteriophage, 79, 90, 104, 135
Baehr, W. von, 42
Bailey, L. H. (18581954), 26
Balanoglossus, 29
Balbiani, E. G., 75
Balfour, F. M. (18511882), 29, 142
Baltzer, F. (b. 1884), 86, 111, 123, 141
Bancroft, F. W. (18711923), 70
Barley, 70
Basset hounds, 22
Bateson, B., 30
Bateson, W. (18611926), 12, 15, 21, 25,
2932, 40, 41, 49, 52, 53, 56, 5860,
63, 64, 67, 100, 101, 107, 126, 129,
132, 134, 136, 137, 142
Batrachoseps, 42
Bauer, H. (b. 1904), 75, 138, 141
Baur, E. (18751933), 54, 124, 137
Beadle, G. W. (b. 1903), 55, 57, 78, 102
104, 135, 138, 141, 142, 143
Beale, G. H., 120, 141
Beans, 30, 59
157
158
A H ISTORY
OF
G ENETICS
Caldwell, 18
Caliban, 128
Calkins, G. N. (18691943), 118
Cambridge group, 30
Camerarius, R. J. (16651721), 2
Candolle, A. P. de (17781841), 140
Canaries, 42
Cannon, W. A. (b. 1870), 36
Carcinogenesis, 70
Carothers, E. E. (b. 1882), 37, 142
Carson, H. L. (b. 1914), 140
Caspari, E. W. (b. 1909), 102, 141
Castle, W. E. (18671962), 30, 43, 47,
54, 60, 61, 107, 135, 137, 141
Catcheside, D. G. (b. 1907), 66, 91
Cattle, blood groups, 99
Cell, division, 17
lethal, 56
Centromere, 74
Centrosomes, 23
Chamberlain, H., 131
Chargaff, E. (b. 1905), 105
Charles, D. R. (19091955), 141
Chemical induction of mutation, 72
Cherub, 2
Chetverikov, S. S. (18801959), 110
Chiasma, 7779
Chiasmatype, 42, 44
Chino, M., 114
Chironomus, 75, 83
Chlamydomonas, 119
Chlorella, 125
Chlorophyll, 123
Chloroplasts, 123125
Chromocenter, 75
Chromosome, elements, 115
fragmentation, 70
individuality, 33
map, 47, 53
number, 33
pachytene, 73
pairing, 33, 34, 37
rearrangements, 74
rings, 65
Chromosomes, 1820, 3339, 7379,
Ciliates, 117120
Cineraria, 58
Cistron, 90
Cis-type heterozygote, 89
Clausen, R. E. (18911956), 112
ClB chromosome, 69, 70, 76
I NDEX
Cleland, R. E. (b. 1892), 65, 138
Coiling of snail shells, 122, 123
Colchicine, 113
Cole, L. J. (18771948), 97, 141
Color blindness, 127
Columbia University, 45, 46
Combs of fowl, 31, 53
Compatibility, tissue, 99
Complementation, 90
Complexes, 63, 65
Compound character, 30
Conidiospores, 103
Conjugation, 119
Conklin, E. G. (18631952), 122, 140
Cooper, K. W. (b. 1912), 140
Correlation, 58
coefficient, 22
Correns, C. (18641933), 11, 15, 18, 22,
2630, 31, 3537, 39, 40, 68, 123,
133, 136, 137, 140
Cotton, see Gossypium
Coupling, 40
Creighton, H. B. (b. 1909), 74
Crew, F. A. E. (b. 1888), 114
Crick, F. H. C. (b. 1916), 106
Crossing over, 36, 44, 54, 55, 7779
mitotic, 56
Crow, J. F. (b. 1916), 143
Cunot, L. (18661951), 30, 31, 5254,
101, 134, 137
Curie, P. (18591906), 69
Cuvier, G. L. C. (17691832), 140, 141
Cytological map, 74, 76
Cytology, 135
Cytoplasm, egg, 122
Cytosine, 105
Darbishire, A. D. (18791915), 15, 59
Darlington, C. D. (b. 1903), 78, 129, 142
Darwin, C. (18091882), 1, 35, 7, 20,
21, 28, 29, 60, 107, 111, 136
Date palm, 2
Datura, 65, 73
Davenport, C. B. (18661944), 30, 59,
128, 129, 141
Davis, B. M. (18711957), 63
Dawson, M. H. (18961945), 105
De Bary, H. E. (18311880), 7
Decastello, A., 93
Deficiency, 56, 73, 81
Delbrck, M. (b. 1906), 141
159
160
A H ISTORY
OF
G ENETICS
Fowl, 30, 42
blood groups, 97
Freeman, F. N. (b. 1880), 130
Frieben, 70
Frog, 19
Frolowa, S., 75
Fungi, 115
Galton, F. (18221911), 21, 22, 58, 59,
107, 126, 128, 130, 131, 133
Garrod, A. E. (18571936), 100, 126,
134, 137
Grtner, C. F. (17721850), 3, 57, 10
Gay, H. (b. 1918), 91, 92
Gegenbaur, C. (18261903), 141
Gemmules, 1
Gene, 32, 90
frequency, 108, 109
Genetic load, 111
Genetics, 31
Genic balance, 80, 81
Genophore, 79
Genotype, 59
Germ-plasm theory, 17, 117, 119
Gerould, J. H. (18681961), 141
Gibor, A. (b. 1925), 125
Gigas, 63
Giles, N. H. (b. 1915), 142
Gillespie, 105
Gilman, D. C. (18311908), 45
Giraffe, 2
Glass, B. (b. 1906), 6
Gloor, H., 25
Gobineau, 131
Godlewski, E., 123
Goldschmidt, R. (18781958), 63, 81,
84, 90, 116, 137, 141, 143
Goodspeed, T. H. (b. 1887), 112
Gortner, R. A. (18851942), 101
Gossypium, 115
Gowen, J. W. (b. 1893), 82, 91, 92, 140
Grafts, 98
Granick, S. (b. 1909), 125
Grant, M., 131
Grasshopper, 37, 114
Gray, A. (18101888), 142
Green, K., 89
Green, M. M. (b. 1916), 89, 143
Grell, R. F. (b. 1914), 79
Grew, N. (16411712), 2, 136
I NDEX
Griesel, 69
Griffen, A. B. (b. 1915), 88
Griffith, F. (18771941), 105, 138
Grouse locusts, 114
Guanine, 105
Guinea pig, 99
Guppy, 83
Guyer, M. F. (18741959), 37
Gynandromorphs, 80
Gypsy moth, see Lymantria
Haacke, W., 2224, 32
Haberlandt, G. (18541945), 140
Habrobracon, 85
Hadorn, E. (b. 1902), 141
Haeckel, E. (18341919), 141, 143
Haendel, 105
Haldane, J. B. S. (18921964), 97, 108,
109, 132, 138, 142
Hanson, F. B. (18861945), 71
Haplo-IV, 81
Haploidy, 35
Hardy, G. H. (18771947), 108, 137
Harland, S. C. (b. 1891), 115
Harrison, R. G. (18701959), 46, 140,
143
Hartmann, M. (18761962), 118, 119,
143
Heitz, E. (18921965), 75, 82, 138
Helmholtz, H. L. F. (18211894), 143
Hemiptera, 41
Hemoglobin, 126, 127
Henking, H., 41
Henle, J. (18091885), 143
Herbert, W. (b. 1778), 5, 6
Herbst, C., 45
Hershey, A. D. (b. 1908), 104, 138
Herskowitz, 1. H. (b. 1920), 140
Hertwig, O. (18491922), 18, 136, 141
Hertwig, R. (18501937), 117, 118, 141,
143
Heterochromatin, 75, 81, 83, 91, 92
Heterozygote, 32
intermediate, 12, 31
Heys, F. M. (b. 1906), 71
Hieracium, 11
Hinny, 122
Hippocrates (fl. about 400 B.C.), 1
Hirszfeld, H., 94
Hirszfeld, L. (18841954), 94, 95, 98,
137
161
162
A H ISTORY
Kerner, A. (18311878), 25
Khouvine, Y., 103
Kielmayer, K. F. (17681844), 140
Kihara. H. (b. 1893), 112
Kimball, R. F. (b. 1915), 120
Knight, T. A. (17591838), 10
Koch, H. H. R. (18431910), 143
Koernicke, M., 70
Kollar, 10
Klliker, R. A. von (18171905), 141
Klreuter, J. G. (17331806), 2, 57,
121, 136
Korschinsky, S. (18611900), 21
Kossel, A. (18531927), 104
Kossikov, 88
Khn, A., 141
Kynurenine, 103
Lamarck, J. B. (17441829), 140
Lamy, R., 114
Lancefield, D. E. (b. 1893), 114, 140
Lancefield, R. C. (b. 1895), 114
Landsteiner, K. (18681943), 93, 96,
136, 138
Laughnan, J. R. (b. 1919), 142
Lebedeff, G. A. (18941964), 82
Lebistes, 83
Lecoq, H., 5, 6
Lederberg, J. (b. 1925), 104, 143
Lee family, 128, 130
Leeuwenhoek, G. van (16321723), 7,
117, 136
Lethal, 54, 56, 69, 110
Lethals, balanced, 64
Leuckart, R. (18231898), 45, 141
Levene, P. A. T. (18691940), 105, 137
Levine, P. (b. 1900), 96, 97, 138
Lewis, E. B. (b. 1918), 8890, 138, 141,
143
Li, C. C. (b. 1912), 109
Lillie, F. R. (18701947), 99, 143
Limnaea, 122
Lindegren, C. C. (b. 1896), 103
Linear curve, 71
Linkage, 3944, 47
Linkage group, 79
Linnaeus, C. (17071778), 2
Little, C. C. (b. 1888), 53, 54, 98, 137,
141
Liverworts, 86
Lock, R. H. (18791915), 15, 43, 60, 137
OF
G ENETICS
Locus, 90
Loeb, J. (18591924), 70, 122, 140
Loeb, L. (18691959), 99, 138
Loew, H. (18071879), 43
Longley, A. E. (b. 1893), 73
Lning, K. G., 72
Lutz, A. M., 63, 112, 137
Lymantria, 84, 116
Lynch, C. J. (b. 1882), 47
McCarty, M. (b. 1911), 104, 105, 138
McClintock, B. (b. 1902), 73, 74, 142
McClung, C. E. (18701946), 37, 41,
136, 142
MacDougall, D. T. (18651958), 70
McKusick, V. A. (b. 1921), 37
MacLeod, C. M. (b. 1909), 104, 105, 138
Macronucleus, 117119
Mainx, R, 83
Maize, 2, 11, 2830, 35, 60, 68, 73, 74,
124, 135
Malaria, 132
Male-sterile maize, 124
Malogolowkin, C, 125
Malpighian tubes, 75
Mammals, melanins, 101
sex determination, 83
Man, 42, 126132
biochemical defects, 100
blood groups, 9396, 126, 127
eye color, 127
hair form and color, 127
sex determination, 83
Marchal, E. and E., 86, 112, 137
Marine Biological Laboratory, 45, 46
Mark, E. L. (18471946), 141
Marryatt, D. C. E, 30, 42
Marsh, O. C. (18311899), 142
Martin, H. N. (18481896), 45, 140, 143
Maternal effect, 82, 121125
Mather, C. (16631728), 2, 136
Mating types, 118
Matlock, P., 127
Matthiola, 11, 39, 41
Maupas, E. (18441916), 117, 136
Maupertuis, P. L. M. (16981750), 6
Mavor, 1. W. (b. 1883), 70
May, H. G, 87
Mechanism of Mendelian Heredity, 49
Medawar, P. B. (b. 1904), 99
Megaselia, 83
I NDEX
Megaspore competition, 64
Meigen, J. W. (17631845), 43
Meiosis, 18, 3335, 39, 42, 78, 112
Meisenheimer, J., 102
Melandrium, 83
Melanins, 101
Mendel, G. J. (18221884), 37, 917,
1820, 22, 2527, 29, 3032, 37, 49,
52, 60, 85, 111, 133, 134, 136
Mental properties, 127131
Merogony, 63
Meteorology, 10
Metz, C. W. (b. 1889), 48, 75, 76, 82,
114, 115, 140
Meves, F. (18681923), 85
Mice, 10, 18, 23, 5254, 72, 91, 98, 99
sex determination, 83
Michaelis, P. (b. 1900), 124
Micronucleus, 117119
Miescher, F. (18441895), 104, 136
Minimal medium, 103, 104
Mirabilis, 8, 11, 31, 68, 123
Mitchell, H. K. (b. 1913), 104
Mitchell, M. B. (b. 1911), 104, 124
Mitochondria, 124125
Mitosis, 18, 78
MN blood groups, 96
Modifying genes, 61
Moewus, R., 119
Mohr, O. L. (b. 1886), 73, 128
Montgomery, T. H. (18731912), 34, 37,
41
Morgan, L. V. (18701952), 55, 82, 138
Morgan, T. H. (18661945), 29, 4250,
5254, 60, 70, 80, 87, 101, 135, 137,
140, 141, 143
Moriwaki, D. (b. 1906), 114
Mosses, 86, 112, 124
Moths, 84
Mott-Smith, L. M. (b. 1902), 72
Mourant, A. E. (b. 1904), 97
Mucor, 118
Mule, 2, 122
Muller, H. J. (18901967), 46, 4850,
54, 57, 6872, 74, 75, 88, 90, 91, 113,
130, 137, 138, 140, 143
Mller, J. P. (18011858), 141, 143
Mller, O. F. (17301784), 117
Multiple alleles, 89
Mustard gas, 72
Mutants, 51
163
164
A H ISTORY
Ovists, 121
Owen, R. D. (b. 1915), 99, 138, 141
Painter, T. S. (b. 1889), 7476, 82, 138,
141, 143
Pangenesis, 1, 6, 20, 21
intracellular, 20
Pangens, 20
Panshin, I. B., 91
Paramecium, 117120, 135
Paramecium aurelia, 118, 119
Paramecium bursaria, 118, 125
Patterson, J. T. (18781960), 91, 143
Pavan, C. (b. 1919), 77
Payne, F. (b. 1881), 61, 140
Pearson, K. (18571936), 5860, 108,
133
Peas, 1016, 26, 2830, 44, 60
Pelargonium, 124
Pellia, 75
Penrose, L. S. (b. 1898), 128
Perthes, G., 70
Peters, H. R., 96
Petrunkewitsch, A. (18751964), 141,
143
Pfeffer, W. (18451920), 28, 140
Phenotype, 59
Phenylketonuria, 128
Philiptschenko, L., 141
Phormia, 83
Phylogeny, 57
Pigeons, 98
Pisum, see Peas
Plastid precursors, 124
Plastids, 123, 124
Platypoecilus, 83
Plough, H. H. (b. 1892), 55, 140
Plunkett, C. R. (b. 1892), 57
Pneumococcus, 105, 106
Pollen, germination, 3
grains, 35, 36
lethal, 41, 64
Polyploids, 83, 86, 112, 113
Polysaccharide, 105
Polyspermy, 34
Pontecorvo, G. (b. 1907), 56, 140
Poppy, 36
Population genetics, 107116
Position effect, 8792
S-type, 88, 90
V-type, 88, 90
OF
G ENETICS
Poulson, D. F. (b. 1910), 125, 141
Precocity theory, 78
Preer, J. R. (b. 1918), 120, 141
Preformation, 121
Presence and absence, 52, 67
Pringsheim, N. (18231894), 7
Protozoa, 117120
Pseudoalleles, 53, 90
Puffing, 76
Punnett, R. C. (18751967), 30, 31, 40,
137, 142
Pure lines, 59
Purkinje, J. E. (17871869), 140
Pyrrhocoris, 41, 83
Quantitative characters, 60
Quantitative study, 22
Quenisset, 69
Rabbits, 47, 53
Rabl, K. (18531917), 33
Race, R. R. (b. 1907), 97
Racial differences, 130
Racism, 131
Radiation sickness, 69
Radium, 69, 71
Randolph, L. F. (b. 1894), 73
Random drift, 110
Rats, hooded, 60, 61
Raunkir, C., 11
Raynor, G. H., 42, 137
Recombination, 25
Recon, 90
Red hair, 127
Redtenbacher, J. (18101870), 9
Reduction, 34
chromosome, 33
Reduplication, 40, 41
Regression coefficient, 22
Rejuvenescence, 117
Relative sexuality, 118, 119
Renner, O. (18831960), 6366, 124,
134, 137
Reparable mutant, 102, 103
Repeats, 76
Repulsion, 40
Reversion, 4, 87
Rh blood groups, 96, 97
Rhesus monkey, 96
Rhoades, M. M. (b. 1903), 123, 141, 142
I NDEX
Rhynchosciara, 77
Ribose nucleic acid, see RNA
Rick, C. M. (b. 1915), 142
Right and left handedness, 127
Ris, H. (b. 1914), 79
RNA, 92, 105, 106
Robbins, W. J. (b. 1890), 103
Roberts, H. F. (18701937), 4, 5, 25, 27
Robertson, W. R. B. (18811941), 77,
115, 142
Robson, J. M. (b.1900), 72
Rodents, 135
Roentgen, W. K. von (18451923), 69
Rollins, 70
Roman, H. L. (b. 1914), 56, 142
Romanes, G. J. (18481894), 26
Rossenbeck, 105
Roux, W. (18501924), 19, 136, 141
Rudkin, G. T. (b. 1917), 77
Rudolphi, C. A. (17711832), 141
Russell, L. B. (b. 1918), 83, 91
Russell, W. L. (b. 1910), 72, 141
Ruud, L., 128
Ryan, F. J. (19161963), 119, 140, 143
Sachs, J. (18321897), 9, 26, 28, 140
Salivary gland chromosomes, 56, 7476
Saltation, 62
Sanger, F. (b. 1918), 97
Saunders, E. R. (b. 1865), 30, 39, 142
Sax, K. (b. 1892), 78, 142
Schaeffer, E. W., 115
Schaudinn, F. (18711906), 143
Schelling, F. W. J. (17751854), 140
Schleiden, M. J. (18041881), 9
Schrader, F. (18911962), 85, 140, 143
Schultz, J. (b. 1904), 81, 91, 140, 141
Schwann, T. (18101882), 143
Sciara, 76
Scipio family, 128
Scott-Moncrieff, R., 101
Sea urchin, 34, 122
Sedgwick, A. (18541913), 29, 142
Segregation, 25
Mendelian, 35
somatic, 19
Seguy, 69
Seiler, J. (b. 1886), 42, 143
Selection, 6062, 107, 108, 110
Selective mating, 108
165
Self-sterility, 66
Senescence, 117
Sex, chromosomes, 39, 41
determination, 8086
linkage, 42, 43
in plants, 2
Shakespeare, W. (15641616), 128, 132
Shull, G. H. (18741954), 61, 70, 141
Sickle-cell anemia, 131
Sidorov, B. P., 91, 92
Siegel, R. W. (b. 1924), 120, 125, 141
Silkworm, 122
Site, 90
Slizynska, H., 92
Smith, T. (18591934), 125
Smyrna fig, 2
Snapdragon, see Antirrhinum
Snell, G. D. (b. 1903), 99
Snyder, L. H. (b. 1901), 95
Somatic segregation, 41, 56
Sonneborn, T. M. (b. 1905), 118, 120,
138, 141
Species, 111
hybrids, 2, 122
Spemann, H. (18691941), 141
Sperm, 7
nuclei, 35
Spermists, 121
Sphaerocarpos, 86
Spireme, 33, 34
Sports, 4, 21
Spreading effect, 92
Srb, A. M. (b. 1917), 143
Stadler, L. J. (18961954), 7072, 142
Stanford University, 104
Statistical methods, 59
Statistics, 135
Stature, 60, 127
Steenbeck, 69
Steiner, E. (b. 1919), 66
Sterility, interspecific, 111
Stern, C. (b. 1902), 56, 74, 82, 92, 115,
141, 143
Stetson, R. E., 96
Stevens, L. G., 69
Stevens, N. M. (18611912), 41, 42, 48,
75, 137
Stocks, see Matthiola
Stomps, T. J., 27
Stone, W. S. (b. 1907), 140, 143
Stormont, C. (b. 1916), 99, 141
166
A H ISTORY
OF
G ENETICS
Twin hybrids, 63
Twin spots, 56, 101
Twins, 130
dizygotic, 130
identical, 130
monozygotic, 130
Tyler, A. (b. 1906), 143
Tyzzer, E. E. (b. 1875), 98, 137
Vail, A. M., 70
Van't Hoff, J. H. (18521911), 26
Variability, 59
Variation, 5, 25
continuous, 4, 5860
discontinuous, 4, 17, 21, 62
Variegation, 68, 91, 92, 124
Vavilov, N. I. (18871943), 143
Vertebrates, 115
Viability, 51
Vigor, hybrid, 61
vegetative, 29
Virchow, L. C. (18211902), 143
Vries, H. de (18481935), 5, 20, 21, 25,
2629, 36, 39, 43, 60, 62, 63, 67, 68,
107, 124, 136, 140
Ullerich, F.-H., 83
Unequal crossing over, 88
Unger, F. (18001870), 910
Unit character, 32
Uracil, 105
Urine, biochemistry of, 127
I NDEX
Wheat, 60
Wheeler, W. M. (18651937), 142
Wheldale, M., see Onslow, M. W.
White, R. G., 99
Whitefly, see Aleurodes
Whiting, A. R. (b. 1892), 85
Whiting, P. W. (b. 1887), 10, 85
Whitman, C. O. (18431910), 142, 143
Whitney, D. D. (b. 1878), 125
Wichura, M. E. (18171866), 5, 6
Wiener, A.S. (b. 1907), 96, 97, 138
Wild type, 53
Williston, S. W. (18521918), 142, 143
Wilson, E. B. (18561939), 20, 29, 33,
37, 41, 45, 49, 77, 104106, 137, 140,
142, 143
Winge, O. (18861964), 15, 83, 104,
112, 137, 138, 140, 142
Winiwarter, H. von, 34
Wolff, C. F. (17331794), 121, 136
Woodruff, L. L. (18791947), 118, 119
Woodworth, C. W. (18691940), 43
Wright, S. (b. 1889), 99, 101, 109, 138,
141
Xenia, 28
X chromosome, 41
X rays, 69, 70, 71, 74, 91, 111
Y chromosome, 48, 8183
Yeast, 56, 104, 124
Yeast nucleic acid, see RNA
Yule G. U. (18711951), 12, 58, 59, 107
Zeleny, C. (18781939), 87, 140, 141
Zirkle, C. (b. 1895), 2, 10
Zygote, 31
167
A FTERWORD
R EMEMBERING S TURTEVANT *
Alfred Henry Sturtevant (18911970) was the youngest of six children of Alfred Henry and Harriet (Morse) Sturtevant. His grandfather,
Julian Sturtevant, was a Yale graduate, a Congregational minister, and
one of the founders and later president of Illinois College in Jacksonville,
Illinois. Sturtevants father taught mathematics for a while at that college, but later took up farming, first in Illinois and later in southern Alabama, where the family moved when Sturtevant was seven years old.
Sturtevant went to a one-room country school and later to a public high
school in Mobile.
At the age of 17, Sturtevant entered Columbia University, where his
brother Edgar, who was 16 years older, was teaching at Barnard College.
Edgar and his wife took the young Sturtevant into their family, and
Alfred lived with them while attending the University. Edgar was a
scholar who later became a professor of linguistics at Yale and an
authority on the Hittite language. Sturtevant said that he learned the aims
and standards of scholarship and research from Edgar. It was a great
pleasure for Sturtevant when he and Edgar were awarded honorary degrees at the same Yale commencement many years later. Also present at
the ceremony were Sturtevants nephew, Julian (Edgars son), Professor
(now Emeritus) of Organic Chemistry at Yale, and Sturtevants elder
son, William, then a graduate student in Yales department of anthropology and now curator of anthropology at the Smithsonian Institution in
Washington.
Sturtevant said that he became interested in genetics as the result of
tabulating the pedigrees of his fathers horses. He continued this interest
at Columbia and also collected data on his own pedigree. At Edgars
suggestion he went to the library and read some books on heredity, with
* Reprinted from Lewis, E. B., 1995. Remembering Sturtevant, Genetics, 141:12271230. Used with permission.
168
169
170
A FT ERWORD
ments. Afternoons were spent in the biology library checking on any incoming journals, few of which in any phase of biology he did not at least
dip into. The pace of science was not so frenetic as it is nowadays, so
there was time for extended afternoon tea sessions at which Sturtevant
might bring up a paper he had read that afternoon and that had attracted
his attention. These sessions were very stimulating for the graduate students in genetics and embryology who usually attended them; among the
faculty in genetics, Schultz, Emerson, and Dobzhansky were likely to be
present in addition to Sturtevant, and in embryology, Albert Tyler, who
was working on the biochemistry of fertilization. Although a rift had developed between Sturtevant and Dobzhansky, there was no sign of it in
front of the graduate students.
Sturtevant taught the undergraduate course in genetics at Caltech for
many years. From time to time he also gave a course for undergraduates
in entomology, complete with a field laboratory session. His lectures on
topics in advanced genetics were scholarly reviews of special areas of
genetics, often dealing with organisms with bizarre genetics, such as the
protozoa. His lectures were especially valuable because he covered areas
of research not ongoing at Caltech. The elementary course in genetics
that Sturtevant taught was based on a textbook that he and George Beadle wrote (1939). It was not so widely used as perhaps it should have
been, probably because it was considered too difficult for the average
student. It was tailored for Caltech students, and the problems especially
were a challenge, even for Caltech undergraduates.
Sturtevant and Beadle planned to revise the textbook, but the pressure of other work and the rapidity of developments that followed the
discovery of the role of DNA prevented the revision. Sturtevant also
liked to point out that both he and Beadle found after writing the book
that each had used the term gene differently. For example, the white
gene to Sturtevant was the specific white mutant, but to Beadle it represented the constellation of white alleles including the wild-type allele.
Sturtevant facetiously blamed their inability to get out a second edition
on this difference in thinking about the gene. Characteristically, he would
ask each geneticist whom he met how he or she used the term, and he
then promptly catalogued such persons according to whether they
thought of the gene the way he did or the way Beadle did. The person
asked did not, of course, need to worry about his answer being in good
company in either case.
Sturtevant read widely and kept abreast of many topics of current
interest, especially politics. He would, for example, read the Sunday New
York Times and the Manchester Guardian Weekly virtually from cover to
171
172
A FT ERWORD
173
174
A FT ERWORD
Lewis, E. B., 1976 Sturtevant, Alfred Henry, pp. 133138 in Dictionary of Scientific Biography, vol. 13, edited by C. C. Gillispie. Charles Scribners
Sons, New York.
Provine, W. B., 1991 Alfred Henry Sturtevant and crosses between Drosophila
melanogaster and Drosophila simulans. Genetics 129: 15.
Roe, A., 1953 The Making of a Scientist. Dodd, Mead & Co., New York.
Snell, G. D., and S. Reed, 1993 William Ernest Castle, pioneer mammalian
geneticist. Genetics 133: 751753.
Sturtevant, A. H., 1910 On the inheritance of color in the American harness
horse. Biol. Bull. 19: 204216.
Sturtevant, A. H., 1961 Genetics and Evolution: Selected Papers of A. H.
Sturtevant, edited by E. B. Lewis with a foreword by G. W. Beadle. W. H.
Freeman, San Francisco.
Sturtevant, A. H., 1965a A History of Genetics. Harper & Row, New York.
Sturtevant, A. H., 1965b The fly room. Am. Sci. 53: 303307.
Sturtevant, A. H., and G. W. Beadle, 1939 An Introduction to Genetics. W. B.
Saunders Co., Philadelphia.