BIOLOGY
LIBRARY
G
GENETICS
THE MACMILLAN COMPANY
NEW YORK BOSTON CHICAGO DALLAS
ATLANTA SAN FRANCISCO
MACMILLAN & CO.. LIMITED
LONDON BOMBAY CALCUTTA
MELBOURNE
THE MACMILLAN CO. OF CANADA, Lm
TORONTO
GENETICS
AN INTRODUCTION TO THE STUDY
OF HEREDITY
BY
HERBERT EUGENE WALTER
PROFESSOR OF BIOLOGY, BROWN UNIVERSITY
WITH 92 FIGURES AND DIAGRAMS
REVISED EDITION
THE MACMILLAN COMPANY
1924
All rights reserved
PREFACE TO THE REVISED EDITION
NEARLY ten years have passed since this book first
appeared. The biological Rip van Winkle of today
who, awaking after a decade of somnolence, gazes again
upon the genetic village of Falling Waters, will indeed
need to rub his astonished eyes at the changed scene
that now spreads out before him. Many old familiar
landmarks, such as "unit characters" and "dominance,"
show signs of dilapidation, while strange children,
shouting a medley of outlandish words, "linkage,"
"tetraploidy," and "non-disjunction," for example, are
playing new games on the village green.
Although the remarkable advances in this field of
science are well treated in considerable detail by several
recent text-books, notably those of Castle, Morgan,
Conklin, and Babcock and Clausen, perhaps there still
remains the original need for a more elementary pre-
sentation of the salient points of genetics, not only for
the interested but confused layman, but also for the
initiation of the prospective student who is attracted to
the study of heredity.
To perform this service is the ambitious object of
the present revision.
Three new chapters, XI, XII and XIII, have been
added and the whole book has been thoroughly worked
over and rearranged. Chapter XIII upon Sex De-
termination has practically been written by Professor
S. I. Kornhauser of Denison University and the entire
vii
viii PREFACE TO THE REVISED EDITION
manuscript critically read by Dr. J. W. Wilson of
Brown University.
There are thirty-four new figures and diagrams
which are either original or copied from acknowledged
sources. Mr. C. J. Fish made the drawings for figures
22 and 32. The proof was read by my wife and by my
niece, Miss Dorothy Walter.
I wish to acknowledge the help I have received from
all of these sources as well as from many unnamed
friends who have given valuable suggestions.
H. E. W.
LA JOLUA, CALIF.,
March 18,
PREFACE TO THE FIRST EDITION
THE following pages had their origin in a course
of lectures upon Heredity, given at Brown University
during the winter of 1911-1912, which were amplified
and repeated in part the following summer at Cold
Spring Harbor, Long Island, before the biological
summer school of the Brooklyn Institute of Arts and
Sciences.
An attempt has been made to summarize for the
intelligent, but uninitiated, reader some of the more
recent phases of the questions of heredity which are at
present agitating the biological world. It is hoped
that this summary will not only be of interest to the
general reader, but that it will also be of service in
college courses dealing with evolution and heredity.
The subject of heredity concerns every one, but
many of those who wish to become better informed
regarding it are either too busily engaged or lack the
opportunity to study the matter out for themselves.
The recent literature in this field is already very large,
with every indication that much more is about to
follow, which is a further discouragement to non-
technical readers.
It may not be a thankless task, therefore, out of
the jargon of many tongues to raise a single voice
which shall attempt to tell the tale of heredity. There
may be a certain advantage in having as spokesman
one who is not at present immersed in the arduous
ix
x PREFACE TO THE FIRST EDITION
technical investigations that are making the tale worth
telling. The difficulties in understanding this compli-
cated subject may possibly be realized better by one
who is himself still struggling with them, than by the
seasoned expert who has long since forgotten that such
difficulties exist.
Among others I am particularly indebted to Dr.
C. B. Davenport for many helpful suggestions, to
my colleague, Professor A. D. Mead, for reading the
manuscript critically, to Dr. S. I. Kornhauser who
gave valuable aid in connection 'with the chapter
on the Determination of Sex, and to my wife for
assistance in final preparation for the press.
I wish to thank Professor H. S. Jennings and Dr.
H. H. Goddard, who have given generous permission
to copy certain diagrams, as well as The Outlook
Company and The Macmillan Company for the use
of figures 24 and 66, respectively.
The fact that not all the suggestions which were at
various times offered by my kindly critics have been in-
corporated in the text, absolves them from responsibility
for whatever remains.
H. E. W.
PROVIDENCE, R. I. f
September, 1912
CONTENTS
CHAPTER PAGE
I. INTRODUCTION 1
1. The idea of species 1
2. The triangle of life 3
3. A definition of heredity 6 *-"
4. The maintenance of life 8
5. Somatoplasm and germplasm 12 \)
II. VARIATION 17
1. The most invariable thing in nature 17
2. The universality of variation 18
3. The kinds of variation with respect to their:
a. Nature 19
b. Duplication 20
c. Utility 20
d. Direction in evolution 21
e. Source . 21
f. Normality 22
g. Degree of continuity 22
h. Character 22
i. Relation to an average standard .... 23
j. Heritability 23 <
4. Methods of studying variation
5. Biometry 23
6. Fluctuating variation 24
7. The interpretation of variation curves .... 28
a. Relative variability 28
b. Bimodal curves 29
c. Skew curves 31
8. Graduated and integral variations 33 ^
9. The causes of variation 34 /
a. Darwin's attitude 34 *
b. Lamarck's attitude 34 ^
c. Weismann's attitude 35 tx
d. Bateson's attitude 37
III. HERITABLE DIFFERENCES
1. The mutation idea 38 ^
2. A summary of the mutation theory 41
xi
xii CONTENTS
CHAPTER
3. Lamarck's evening primrose 42
4. Plant mutations found in nature 47
5. Some mutations among animals 48
6. Kinds of mutation 51
7. The origin of mutations 53
8. When mutations occur 55
9. Possible causes of mutation 57
IV. THE INHERITANCE OF ACQUIRED CHARACTERS ... 62
1. Summary of preceding chapters 62
2. The bearing of this chapter upon genetics . . 63
3. The importance of the question 63
4. A historical sketch of opinion 64
5. Confusion in definitions 66
6. Weismann's conception of acquired characters . 67'
7. The distinction between germinal and somatic
characters 67
8. What variations reappear? 69
9. How may germplasm acquire new characters? . 69
10. Weismann's reasons for doubting the inheritance
of acquired characters 71
a. No known mechanism for impressing germ-
plasm with somatic characters ... 71
b. Evidence for the inheritance of acquired
characters inconclusive 74
a. Mutilation 75
b. Environmental effects 76
c. The effects of use or disuse .... 78
d. Transmission of disease 80
e. Immunity and the effect of drugs . . 82
f. Prenatal influences 83
c. The germplasm theory sufficient to account
for the facts of heredity 84
11. The comparative independence of germ and soma 85
12. Acquired characters in the protozoa .... 86
13. The opposition to Weismann 88
14. Various results upon the offspring of parental
acquisitions 90
15. Conclusion 92
V. MENDELISM 93
1. Methods of studying heredity 93
2. The melting pot of cross-breeding 93
3. Johann Gregor Mendel 95
4. Mendel's experiments on garden peas .... 97
5. Some further instances of Mendel's law . . . 101
6. The cardinal principle of segregation .... 103
CONTENTS xiii
CHAPTER
7. Definitions ............ 104
8. The identification of a heterozygote . . . . 105
9. The presence or absence hypothesis .... 106
10. Dihybrids ............ 107
11. The case of the trihybrid ........ 114
12. Summary ........... .117
13. The practical application ........ 119
14. Conclusion ............ 120
VI. THE PURE LINE AND SELECTION
1. Galton's law of regression ........ 121
2. The idea of the pure line ....... 122
3. Johannsen's nineteen beans ........ 124
4. The distinction between a population and a pure
line .............. 127
5. Cases similar to Johannsen's pure lines . . . 132
6. Selection within the pure line ...... 133
a. Vilmorin's wheat ......... 135
b. Clones ............ 135
c. Parthenogenetic progeny ...... 141
d. Homozygous crosses ........ 142
a. Tower's potato beetles ..... 143
6. Drosophila bristles ....... 143
c. Pearl's 200-egg hen ...... 145
d. Castle's hooded rats ...... 145
7. Conclusion ............ 146
VII. THE FACTOR HYPOTHESIS ......... 148
1. The hereditary unit ......... 148
2. Different kinds of genes ........ 149
3. Complementary genes ......... 151
4. Supplementary genes ......... 154
a. Castle's agouti guinea-pigs ...... 154
b. Cuenot's spotted mice ....... 155
c. Miss Durham's intensified mice .... 156
d. Castle's brown-eyed yellow guinea-pigs . . 157
e. Rabbit phenotypes ........ 159
f. The kinds of gray rabbits ...... 162
5. Lethal genes ....... ..... 162
6. Modifying genes and selection .... 166
VIII. BLENDING INHERITANCE .......... 168
1. The relative significance of dominance and seg-
regation ............ 168
2. Imperfect dominance ..... .... 168
3. Delayed dominance .......... 170
xvi CONTENTS
CHAPTER FAGB
XIV. THE APPLICATION TO MAN 296
1. The application of genetics to man . . . .296
2. Modifying factors in the case of man . . . .297
3. Experiments in human heredity 299
a. The Jukes 299
b. The descendants of Jonathan Edwards . . 301
c. The Kallikak family 302
4. Moral and mental characters behave like physical
ones 303
5. The character of human traits 304
6. Hereditary defects 306
7. The control of defects 309
XV. HUMAN CONSERVATION 314
1. How mankind may be improved 314
2. Human assets and liabilities 315
3. More facts needed 317
4. Further application of what we know, necessary 319
5. The restriction of undesirable germplasm . . . 320
a. Control of immigration 322
b. More discriminating marriage laws . . . 324
c. An educated sentiment 325
d. Segregation of defectives 327
e. Drastic measures 329
6. The conservation of desirable germplasm . . . 330
a. By enlarging individual opportunity . . . 331
b. By preventing germinal waste 331
a. Preventable death 331
6. Social hindrances
c. By subsidizing the fit 333
7. Who shall sit in judgment? 334
8. Eugenics not "bluegenics" 335
9. The moral 336
BIBLIOGRAPHY 337
INDEX 941
GENETICS
GENETICS
CHAPTER I
INTRODUCTION
1. THE IDEA OF SPECIES
THE doctors have always disagreed regarding a
definition of species. What determines the exclusive
boundaries that shall isolate from their fellows any
particular group of animals or plants has long been
a mooted question, and still remains so.
The Linnsean concept of a species was that of an
exclusive caste of individuals, inflexibly demarked,
over whose high barriers no nondescript tramps
would dare attempt to climb. When an entomolo-
gist of the old Linnsean school encountered an insect
which did not conform to the morphological tradi-
tions of its fellows, the frequent fate of such a non-
conformist was to perish under the boot-heel rather
than to find sanctuary in the cabinet of the preserved.
Since it was an exception, and a violator of the divine
law of the fixity of species, it deserved to be anni-
hilated! Those were hard days both for heretics and
for variations.
The method of the older school of systematists
1
GENETICS
may be described as one which emphasized differences
and put up barriers that should keep the unlike apart,
at the same time allowing only "birds of a feather"
to flock together. It was a brave and successful
attempt to bring order out of chaos by classifying
the living world, and it served its purpose well until
Darwin's idea of half a century ago, that the origin
of all species is from preceding species, put an en-
tirely new face upon the whole matter. Organisms
of different species were found to be related to one
another, and even man could no longer escape ac-
knowledging his poor animal relations. As a conse-
quence, likenesses rather than differences thereafter
claimed the most attention.
During the reconstruction of pihylogenetic trees,
which seized the imagination and became the prin-
cipal business of post-Darwinian biologists, "connect-
ing links," that is, the crotched sticks in the woodpile
of organisms, which had hitherto been largely dis-
carded, were most eagerly sought after. It was just
these scraggly sticks, that were neither trunk nor
limb-wood but combinations of both, which told the
story of continuity and were indispensable in building
up a reunited whole.
As the analysis of the living world gradually came
to shift from species to individuals, it was shown that
individuals may be regarded simply as aggregates
of imit characters which may combine so variously
that it becomes more and more difficult to maintain
constant barriers of any kind between the groups of
individuals arbitrarily called "species."
INTRODUCTION 3
2. THE TRIANGLE OF LIFE
Accordingly within a generation the center of bio-
logical interest gradually swung from the origin of
species to the origin of the individual. The nineteenth
century was Darwin's century. His monumental work
"On the Origin of Species by Means of Natural
Selection," which appeared in 1859, not only dominated
the biological sciences but also influenced profoundly
many other realms of thought, particularly those of
philosophy and theology.
Now, in the first decades of the twentieth century,
a particular emphasis is being laid upon the study of
heredity. The interpretation of investigations along
this line of research has been made possible through
the cumulative discoveries of many things that were
not known in Darwin's day. Trained students, pa-
tiently and persistently bending over improved micro-
scopes, have untangled the mysteries of the cell, while an
increasing host of investigators, inspired by the
Austrian monk Mendel, have industriously devoted their
energies to breeding animals and plants with an insight
denied to breeders of preceding centuries.
The study of the origin of the individual, which
has grown out of the more general consideration of
the origin of species, forms the subject-matter of
heredity, or, to use the more definitive word of Bate-
son, of genetics.
(It is not with the individual as a whole that
genetics is chiefly concerned, but rather with char-
acteristics of the individual.
4 GENETICS
Three factors acting together determine the char-
acteristics of an individual, namely, environment, re-
sponse, and heritage, as expressed diagrammatically
in Figure 1. It may be said that an individual is the
result of the interaction of these three factors since he
may be modified by changing any one of them. Although
no one factor can possibly be omitted, the student of
genetics places the emphasis upon heritage as the
Heritage
FIG. 1. The triangle of life.
factor of greatest importance. Heritage, or "blood,"
expresses the innate equipment of the individual. JTt Js
what he actually is even before birth. It is his nature.
It is what determines whether he shall be a beast or a
man. Consequently in the diagram (Fig. 1), the tri-
angle of life is represented as resting solidly upon
the side marked "heritage" for its foundation.
Environment and response, although indispensable,
are both factors which are subsequent and secondary.
Environment is what the individual has, for example,
housing, food, friends and enemies or surrounding aids
which may help him and obstacles which he must
INTRODUCTION 5
overcome. It is the particular world into which he
comes, the measure of opportunity given to his par-
ticular heritage.
Response, on the other hand, represents what the
individual does with his heritage and environment. It
is what may be described as the training or educa-
tional factor. Lacking a suitable environment a good
heritage may come to naught like good seed sown upon
stony ground, but it is nevertheless true that the best
environment cannot make up for defective heritage
or develop wheat from tares.
The absence of sufficient response even when the
environment is suitable and the endowment of inherit-
ance is ample will result in an individual who falls short
of his possibilities, while no amount of response or edu-
cation can develop a man out of the heritage of a beast.
Consequently the biologist holds that, although what
an individual has and does is unquestionably of great
importance, particularly to the individual himself,
what he is, is in the long run far more important.
Improved environment and training may better the
generation already born. Improved blood will better
every generation to come. The "triangle of life," when
applied to man, shows that there are theoretically at
least twenty-seven possible kinds of human beings as
shown in Figure . Climbing up this "scale of success'*
is what makes life worth living. It is illuminating
for any one to determine judiciously where he him-
self stands at present or to assign places mentally to
various other people, historical and contemporary, in
this scale.
The left-hand factor does not change throughout
6
GENETICS
Ill
& fl <
n w tf
1.
AAA
2.
AAB
3.
AAC
4.
ABA
5.
ABB
6.
ABC
7.
ACA
8.
ACB
9.
ACC
10.
BAA
11.
BAB
12.
BAG
13.
BBA
14.
BBB
15.
BBC
16.
BCA
17.
BCB
18.
BCC
19.
CAA
20.
CAB
21.
CAC
22.
CBA
23.
CBB
24.
CBC
25.
CCA
26.
CCB
27.
CCC
FIG. 2. The Scale of Success.
A stands for high grade;
B, for mediocrity;
C, for low grade.
life but the other two may.
The sociologist and the
philanthropist are imme-
diately concerned with the
middle column; the edu-
cator and particularly the
parent with the right-hand
column; while the biologist
puts faith in the left-hand
column of heritage. For
example, a child born ACC
is more apt to reach the
top than one born CCC. In
selecting a mate it would be
far wiser to marry ACC
than CAA, since "blood
wiU tell."
What, then, is this
"blood" or heritage? Ex-
actly what is meant by
heredity ?
3. A DEFINITION OP
HEREDITY
The terms heredity and
inheritance come to us from
legal practice. We "in-
herit" the old homestead or
our grandfather's clock.
Moreover, as "heirs to all
the ages" our heredity in-
INTRODUCTION 7
eludes everything that goes to make up civilization,
such as the arts, sciences, literature and traditions.
With this kind of heredity we are not here concerned,
for this is not what is meant by biological heredity.
Professor Castle, in his book on "Heredity in Rela-
tion to Evolution and Animal Breeding," has defined
heredity as "organic resemblance based on descent." '
The son resembles his father because he is a "chip off
the old block." It would be still nearer the truth to say
that the son resembles his father because they are
both chips from the same block, since the actual char-
acters of parents are never transmitted to their off-
spring in the same way that real estate or personal
property is passed on from one generation to another.
When the son is said to have his father's hair and his
mother's complexion it does not mean that paternal
baldness and a vanishing maternal complexion are
the inevitable consequences.
Biological inheritance is more comparable to the
handing down from father to son of some valuable
patent right or manufacturing plant by means of
which the son, in due course of time, may develop an
independent fortune of his own, resembling in char-
acter and extent the parental fortune similarly de-
rived although not identical with it.
So it comes about that "organic resemblance" be-
tween father and son, as well as that which often
appears between nephew and uncle or even more re-
mote relatives, is due not to a direct entail of the
characteristics in question, but to the fact that the
characteristics are "based on descent" from a common
8 GENETICS
source. In other words, an "hereditary character" of
any kind is not an entity or unit which is handed
down from generation to generation, but is rather a
method of reaction of the organism to the constellation
of external environmental factors under which the
organism lives.
To unravel the golden threads of inheritance which
have bound us all together in the past, as well as to
learn how to weave upon the loom of the future, not
only those old patterns in plants and animals and men
which have already proven worth while, but also to
create new organic designs of an excellence hitherto
impossible or undreamed of, is the inspiring task before
the geneticist to-day.
4. THE MAINTENANCE OF LIFE
So far as we know, every living thing on the earth
to-day has arisen from some preceding form of life.
How the first spark of life began will probably
always be a matter of pure speculation. Whether
the beginnings of what is called life came through
space from other worlds on meteoric wings, as Lord
Kelvin has suggested; whether it was spontaneously
generated on the spot out of lifeless components;
or whether life itself was the original condition of
matter, and the one thing that must be explained is
not the origin of life but of the non-living, no one
can say. Leaving aside the first speculation as un-
tenable and the third as irrational, since it jars so
sadly with what astronomers tell us of the probable
INTRODUCTION 9
evolution of worlds, the theory of spontaneous genera-
tion seems to be the last resort to which to turn.
In prescientific days this idea of spontaneous genera-
tion presented no great difficulties to our imaginative
and credulous ancestors. John Milton, with the assur-
ance of an eye-witness, thus described the inorganic
origin of a lion:
"The grassy clods now calved; now half appears
The tawny liorr, pawing to get free
His hinder parts then springs as broke from bonds,
And rampant shakes his brindled mane."
("Paradise Lost/' Book VII, line 543.)
Ovid also in his "Metamorphoses," not to mention a
more familiar instance of special creation, easily suc-
ceeded in creating mankind from the humble stones
tossed by the juggling hands of Deucalion and Pyrrha.
Although under former conditions on the earth
it might have been possible for life to have originated
spontaneously, and although it may yet be possible
to produce life from inorganic materials in the labora-
tory or elsewhere, the exhaustive work of Pasteur,
Tyndall and others effectually demonstrated a genera-
tion ago that to-day living matter always arises from
preceding living matter and this conclusion is gener-
ally accepted as an axiom in genetics.
There are various methods of producing more life,
given a nest-egg of living substance with which to
start. Any organism, whether plant or animal, is
continually transforming inorganic and dead material
into living tissue. Through the process of repair, for
example, an injury to a form as highly developed
10 GENETICS
even as man is frequently made good, if it is not too
extensive and does not involve too highly specialized
tissues, as, for example, in the case of a skin wound.
When the intake of non-living material is in excess
of the outgo, growth results, with the consequence
that more living substance is built up than existed
before. Thus a fragment of a living sponge or a
piece of a begonia leaf is each sufficient to restore a
duplicate of the original organism.
A process similar to the repair of the begonia leaf
is that employed so effectively in the great groups of
the one-celled animals and plants, the Protozoa and
Protophyta, by means of which their numbers are
maintained. These one-celled organisms usually multi-
ply by fission, that is, by division into halves, and
each half grows to the size of the parent organism
from which it sprang. When two daughter protozoans
are thus formed, they are essentially orphans because
they have no parents, alive or dead. The parental
substance in such a process, along with the regulating
power necessary to reorganization, goes over bodily
into the next generation in the formation of the
daughter-cells, leaving usually no remains whatever
behind. In primitive forms of this description, con-
tinuous life is the natural order, and death, when it
does occur, is, as Weismann has pointed out, acci-
dental and quite outside the plan of nature.
In these cases, it is easy to see the reason for "or-
ganic resemblance" between successive generations.
Parent and offspring are successive manifestations
of the same thmg, just as the begonia plant, restored
INTRODUCTION 11
from a fragment of a begonia leaf, is simply an ex-
tension of the original plant.
Many modifications of the process of multiplication
by fission occur, all of them, however, agreeing in the
fundamental principle that the progeny resemble the
parents because they are pieces of the parents.
Thus the "greening" apple maintains its individu-
ality although coming from thousands of different trees,
because all of these trees through the asexual process
of grafting are continuations of the one original
Rhode Island greening tree grown by Dr. Solomon
Drowne in the town of Foster, nearly a century ago.
Western navel oranges all come, directly or indirectly,
from parts of one tree found near Bahia in Brazil.
Again, certain fresh-water sponges and bryozoans,
quite unlike most of their marine relatives, keep a
foothold from year to year within their particular
shallow fresh-water habitats by isolating well pro-
tected fragments of themselves in the form of geimrndes
and statoblasts. These structures may drop to the
muddy bottom and live in a dormant condition through-
out the icy winter when it would not be possible for
the entire organism to survive near the surface.
In order to meet the conditions imposed by winter,
however, these fragments have become so modified as
temporarily to lose their likeness to the parent genera-
tion, although readily regaining that likeness when
springtime brings the opportunity. The unity of
two succeeding generations, notwithstanding that it
may be interrupted by the temporary interposition of
something apparently different in the form of gemmules
12 GENETICS
or statoblasts, is thus essentially maintained. The
bryozoan colonies of two successive seasons in a fresh-
water pond may be regarded as parts of the same
identical colony, since they present an "organic resem-
blance based on descent," although the sole representa-
tives of the parent colony during midwinter may be the
sparks of life locked up within the statoblasts buried
in the mud.
Similarly, the asexual spores of many plants, such
as molds, mosses and ferns, may be regarded as gem-
mules reduced to the lowest terms, namely, to single
cells. As in the preceding cases so in this instance the
resemblance of the offspring which may arise from these
spores, to the parents which produced them, is due to
the essential material identity of two generations.
These illustrations of heredity in its simplest mani-
festations give the key to "organic resemblance" higher
up in the scale. Sexual reproduction is no less plainly
the direct continuation of life though in this instance
two sporelike fragments out of one generation con-
tribute to form the new individual of the next genera-
tion instead of one fragment. In all cases there is a
material contmmty between succeeding generations.
Offspring become thus an extension of a single parent,
or of two parents, while heredity is simply "organic
resemblance based on descent."
5. SOMATOPLASM AND GERMPLASM
In forms that reproduce sexually there occurs a
differentiation of the body substance into what Weis-
mann terms somatoplasm and germplasm.
INTRODUCTION 13
* Somatoplasm includes the body tissues, that is, the
bulk of the individual, which is fated in the course of
events to complete a life-cycle and die. Germplasm,
on the contrary, is the immortal fragment freighted ^
with the power to duplicate the whole organism and
which, barring accident, is destined to live on and
give rise to new individuals.
Germplasm thus carries potencies for developing both
germplasm and somatoplasm, while somatoplasm, ac-
cording to this conception, has only the power to repair
itself but not to reproduce a new individual. More-
over, germplasm is not freshly formed in each genera-
tion, neither does it arise anew when the individual i
reaches sexual maturity, as it appears to do, but it is
a continuous substance present from the beginning.
Although this theory of the contvnuity of the germ-
plasm has been actually demonstrated in comparatively
few instances, all the facts we know concerning the be-
havior of the germinal substance are consistent with it.
The phrase "life everlasting" is not confined, there-
fore, to the vocabulary of the theologian, and poten-
tial immortality is more than a mystical hope of be-
lieving humanity. They are based upon demonstrable
biological facts,
In many of the Protozoa the entire organism is
possibly comparable to germplasm, but in all forms of
life that are compounded of several cells the germplasm
is probably set aside early in the development of the
individual, and this remains undifferentiated, or in re-
serve, like a savings-bank account put by for a rainy
day, while the somatoplasm is expended in the imme-
GENETICS
diate demands of the tissues that make up the indi-
vidual. In one instance at least, that of the nematode
worm As cans,
according to
Boveri, this
splitting off or
isolation of the
germplasm oc-
curs as early in
thei cleavage of
the fertilized
egg as the six-
teen-cell stage,
when fifteen of
the cells go to
form the soma-
toplasm and the
sixteenth is set
aside as germ-
plasm.
Thus there
results a con-
tinuous stream
of germplasm,
receiving con-
tributions from
other germ-
plasmal streams
at the time of
X
Germplasm \ Somatoplasm
7 ^
FIG. 3. Scheme to illustrate the continuity
of the germplasm. Each triangle repre-
sents an individual made up of germ-
plasm (dotted) and somatoplasm (un-
dotted). The beginning of the life cycle
of each individual is represented at the
apex of the triangle where germplasm and
somatoplasm are both present. As the
individual develops each of these compo-
nent parts increases. In sexual reproduc-
tion the germplasms of two individuals
unite into a common stream to which the
somatoplasm makes no contribution. The
continuity of the germplasm is shown by
the heavy broken line into which run col-
lateral contributions from successive sex-
ual reproductions.
sexual reproduction, as shown diagrammatically in
Figure 3, in which individuals are represented by tri-
INTRODUCTION 15
angles. From this continuous stream of germplasm
there split off at successive intervals complexes of so-
matoplasm, or "individuals," which go so far on the
road of specialization into tissues that the power to
be "born again" is lost, and so after a time they die,
while the germplasm, held in reserve, lives on.
This is what is meant by saying that a father and
son owe their mutual resemblance to the fact that they
are chips off the same block rather than by saying that
the son is a chip off the paternal block. Both somato-
plasms are developments at different intervals from
the same continuous stream of germplasm instead of
one somatoplasm derived from a preceding one. As
a matter of fact the germplasm from which the son
arises is modified by the addition of a maternal contri-
bution, so that father and son in reality hold the same
relation to each other that half-brothers do.
So far as his body or his somatoplasm is concerned,
the son is younger than his father but at the same time
he is older than his father in his germplasm, because this
continuous line of germinal potentiality has a genera-
tion longer span in him than in his parents.
From the point of view of genetics, then, the real
mission of the somatoplasm, which is so marvelously
differentiated into all the various forms that we call
animals and plants, is simply to serve as a temporary
domicile for the immortal germplasm. Thus the parent
becomes as it were the "trustee of the germplasm,"
but not the producer of the offspring, for the soma is
after all only the mechanism through which a fertilized
egg produces in due time another fertilized egg.
16 GENETICS
In the light of these preliminary explanations it is
plain that the hopeful point of attack in the science of
genetics must inevitably be the germplasm which is the
source, or point of departure, in the formation of each
new individual, rather than the somatoplasm, which
represents the end stages of the hereditary processes.
This has not been the method of study in the past.
The resemblances of the visible father and son have
usually been traced instead of the character of their
unseen germplasms. By following this old method, in-
vestigators have often been misled because the visible
or apparent is not always the true index of what lies
behind it. A gray and a white rabbit, for example, may
produce some offspring that are entirely black or
two white-flowering sweet peas when crossed may some-
times produce purple blossoms. Consequently it is a
great fallacy to affirm that always in heredity "like
produces like," since the opposite is quite often the
case.
The new heredity, embodied in the science of genetics,
attempts to go deeper than the surface appearance
of the somatoplasm. It aims to get at the source or ori-
gin of organisms, that is, the germplasm which is the
only connecting thread between succeeding generations
of living forms from the "unbeginning past." It is
concerned not so much with somatoplasm, which repre-
sents what the germplasm has done in the past, as
with the germplasm itself and what it can do in the
future.
CHAPTER II
VARIATION
1. THE MOST INVARIABLE THING IN NATURE
IN the introductory chapter it was shown that "or-
ganic resemblance based on descent," by which is
meant heredity, is due principally to the fact that off-
spring are material continuations of their parents and
consequently may be expected to be like them. The
fact that this is the case in the great majority of in-
stances has given rise to the popular formula, "like
produces like," as a rule of heredity.
But this formula by no means always fits the facts.
Like often produces something apparently unlike.
For instance, two brown-eyed parents may produce a
blue-eyed child, although brown-eyed children are more
usual from such a parentage. It is a common expe-
rience, indeed, for breeders of plants and animals to
meet with continual difficulties in getting organisms to
"breed true."
On the other hand, it is exactly these variations which
so constantly interfere with breeding true that fur-
nish the sole foothold for improvement. If all organ-
isms did breed strictly true, one generation could not
stand on the shoulders of the preceding generation,
and there would be no evolutionary advance.
17
18 GENETICS
The most invariable thing in nature is variation.
This fact is at once the hope and the despair of the
breeder who seeks to hold fast to whatever he has
found that is good and at the same time tries to find
something better. Variation is a veritable Pandora's
box and the chaos that would ensue if it were not con-
fined within certain predictable limits can hardly be
imagined. Obviously the entire subject of variation is
intimately and inevitably bound up with any considera-
tion of genetics, for when the similarities and dissimi-
larities between succeeding generations are clear, then
heredity can be explained.
2. THE UNIVERSALITY OF VARIATION
Much of the variation in nature is patent to the most
casual observer, but it requires a trained eye to see the
universal extent of many minor differences. A flock of
sheep may all look alike to a passing stranger, but not
to the man who tends them. A dozen blue violet plants
from different localities might easily be identified by the
amateur botanist as belonging to the same species when,
to a specialist on the genus Viola, unmistakable differ-
ences would doubtless be clearly apparent.
"Identical twins," for example, constitute so marked
an exception to the universal rule of variational differ-
ence that they challenge the attention at once, yet even
here upon critical examination there appears some de-
gree of variation.
The fact that every attempt at an intimate acquaint-
ance with any group of organisms whatsoever invariably
VARIATION 19
reveals previously unrecognized variations, indicates
that variability is much more widespread in nature
than is commonly believed.
The key to Japanese art, as pointed out by Dr.
Nitobe, consists in being natural and in faithfully
copying nature. It is for this reason that the Jap-
anese artist makes each object that he produces unique,
because nature herself, whom he strives to follow, never
duplicates anything.
The Bertillon system of personal identification
is based upon the constancy of minor variations found
in each individual. Its importance is shown in Figure
4. The faces of the criminals there pictured would
be easily confused by the ordinary observer, but an
examination of their thumb prints shows unmistakable
differences between these three individuals.
On the other hand over-emphasis upon the study and
analysis of variation may tend to obscure the important
fact that parent and offspring in the vast majority of
their characteristics are alike.
3. KINDS OP VARIATION
A brief enumeration of some of the kinds of variation
will reveal their diverse character.
a. With respect to their nature variations may be
morphological, physiological, or psychological. Under
morphological variations are included differences in
shape, size, or pattern as well as differences in number
and relation of constituent parts.
Differences in activity are of a physiological nature.
20 GENETICS
The kea parrot, after the introduction of sheep into
New Zealand, changed from herbivorous to carnivorous
habits and consequently became a pest. Many animals
in captivity are less fertile than when free, while differ-
ent individuals are well known to vary widely with re-
spect to their susceptibility to disease. Nageli, for
example, reports the presence of tubercles in 97 per
cent of the cases in five hundred autopsies, although a
majority of the deaths in question was not due to
tuberculosis at all, a fact which indicates a great
diversity in the resistance of different individuals to the
tubercle bacillus.
Psychological variations in man, such as those which
determine the disposition or mental traits of individuals,
are apparent to every one.
b. With respect to their duplication variations may
be single or multiple. A legless lamb 1 is an example of
a single variation or "sport." Four-leaved clovers, on
the contrary, are multiple for the reason that this
variation, although not common, nevertheless occurs
frequently.
c. With respect to their utility variations may be
useful, indifferent, or harmful to the organism possess-
ing them. Useful variations are of the kind emphasized
by Darwin as being effectively made use of in natural
selection. Indifferent variations, on the other hand,
are those which apparently do not play an important
part in the welfare of their possessor, such, for ex-
ample, as the color of the eyes or of the hair. Finally,
the degree of degeneration in certain organs may be
1 "A Peculiar Legless Lamb." Stockard. Biol. Bull, xiii, p. 288.
VARIATION 21
cited as an illustration of harmful variations. The
amount of closure of the opening from the intestine into
the vermiform appendix in man is an example of a
harmful variation, since the larger the opening, the
greater is the liability to appendicitis.
d. With respect to their direction in evolution varia-
tions may be either definite (orthogenetic) or indefinite
(fortuitous).
Fortuitous or chance variations in all possible direc-
tions furnish the repertory of opportunity, according to
Darwin, from which natural selection picks out those
best adapted to survive in the struggle for existence.
Paleontology furnishes numerous instances of the
former category, such as the series of variations from
a pentadactyl ancestor, all apparently tending in one
direction, which have culminated in the one-toed horse.
The fact that the paleontologist deals historically with
a completed phylogenetic series in which the side lines
lack prominence, while the successful line stands out
with distinctness, makes it easy for him to view succes-
sive variations as orthogenetic, that is, as definitely di-
rected in one course either through intrinsic (Nageli) or
extrinsic (Eimer) causes.
Just as sometimes the individuals of an apparently
continuous series, as shown in a museum collection of
similar insects, may be of very diverse geographical
origin, so genetics is not primarily concerned with re-
semblances from generation to generation but rather
with origins and continuity.
e. With respect to their source, variations may be
somatic or germinal. Somatic, or body variations,
22 GENETICS
arise as modifications due to environmental factors.
They are individual differences which may be quite
transitory in nature, while germinal variations may
arise without regard to the environment, are deep-
seated, and of racial rather than of individual sig-
nificance.
f. With respect to their normality variations may
fall within expected extremes and thus be considered
normal, or they may be outside of reasonable expec-
tations and consequently be reckoned as abnormal,
as in the case of a two-headed calf.
g. With respect to the degree of their continuity
variations may form a continuous series, grading into
each other by intermediate steps, or they may be dis-
continuous in character. An example of continuous
variation is the height of any hundred men one might
chance to meet, which would probably represent all
intermediate grades from the highest among the hun-
dred to the lowest.
On the other hand the number of segments in the
abdomen of a shrimp, for instance, which may be either
eight or nine but cannot be halfway between, illustrates
what is meant by discontinuous variation. The wide-
spread occurrence of this later category of variations
has been pointed out by Bateson in his encyclopedic
volume "On Materials for the Study of Variation."
h. With respect to their character variations may be
quantitative or qualitative. A six-rayed starfish rep-
resents a quantitative variation from the normal num-
ber of five rays, whereas a red variety of a flower may
differ chemically from a blue variety, or a bitter fruit
VARIATION 23
may differ from a sweet fruit in a qualitative way de-
pendent upon the chemical constitution of the fruit in
question.
i. With respect to their relation to an average stand-
ard variations may have a fluctuating distribution
around an arithmetical mean, as when some of the off-
spring have more and some less of the parental char-
acter, or the variations in the progeny may all center
about a new average quite distinct from the parental
standard and consequently come under the head of
mutations.
j. Finally, and most important in the present con-
nection, with respect to heritdbility, variations may
possess the power to reappear in subsequent genera-
tions, or they may lack that power. It is this aspect of
variability which bears most directly upon genetics.
Other possible categories might be mentioned, but
a sufficient number have been cited to show the great
diversity of variations in general.
4. METHODS OF STUDYING VARIATIONS
Roughly stated, there are three ways of studying
variations: first, Darwin's method of observation and
the description of more or less isolated cases; second,
Galton's biometric method of statistical inquiry ; and
third, Mendel's experimental method. The second of
these methods will be considered in this chapter.
5. BIOMETEY
The science of biometry, that is, the application of
statistical methods to biological facts, has been de-
24 GENETICS
\
veloped within recent years. Sir Francis Galton, Dar-
win's distinguished cousin, may be regarded as the
pioneer in this field of research, while Karl Pearson and
his disciples are representatives of the modern school
of biometricians.
Although mathematical analysis of biological data
when not sufficiently ballasted by biological analysis
of the same facts may sometimes lead the investigator
astray, yet often the only way to formulate certain
truths or to analyze data of some kinds is by resort to
statistical methods. Biometricians are quite right in
insisting that it is frequently necessary to go further
than the fact of variation, which may be apparent
from the inspection of an individual case, and to deal
with cumulative evidence as presented through statisti-
cal analysis.
In matters of heredity, however, facts as they occur
in single cases and definite pedigrees seem to offer a
more hopeful line of approach than statistical generali-
zations. It is better to become acquainted with the real
parent than to evolve a hypothetical "mid-parent"
mathematically. In this connection it is well always to
bear in mind the warning of Johannsen, himself a past
master in biometry, when he writes : "3f it Mathematik
nicht als Mathematik treiben wir unsere Studien."
6. FLUCTUATING VARIATION
With respect to any measurable character there
are bound to be deviations from an average condition.
According to the mathematical laws of chance these de-
VARIATION
viations sometimes are plus and sometimes minus, and
consequently they may be termed fluctuating variations.
Pearson gives as a simple illustration of fluctuating
variation the number of ribs present in two sets of
beech-leaves, as shown below. These sets were taken
from two different trees, and each contains twenty-six
leaves.
NUMBER OF RIBS
13
14
15
16
17
18
19
20
TOTAL
First tree . .
Second tree .
3
4
1
9
4
8
7
2
9
4
1
26
26
Total . . .
3
4
10
12
9
9
4
1
It will be seen that, while certain leaves might well
belong to either tree, as, for example, those with sixteen
ribs, the entire group of leaves from either tree is unlike
that of the other tree. In the first instance the number
of ribs fluctuates around eighteen as the commonest
kind ; in the second case, around fifteen. Such a differ-
ence could not easily be detected or expressed by any
other method than the statistical one.
Again, in the case of forty-seven starfishes all of
which were collected from one locality the variation
in the number of rays proved to be, according to
Goldschmidt, an amount indicated graphically in
Figure 5, where the data are arranged in the form of
of a so-called frequency polygon or curve.
From such a polygon certain constants may be com-
puted which conveniently express in a single number,
for purposes of abstract comparison, distinctions that
26 GENETICS
otherwise could be handled only in the most indefinite
way.
List of Constants
- Arithmetical Mean (A.M.) = 4.9
Mode (Af ) = 5
~_ Average Deviation (-A.D.) = .52
Standard Deviation (cr) = .846
. Coefficient of Variability (C. V.) = 1 . 72
Formulae
30
25
20
15
10
A.D.
S = sum
x = deviation of the class from A.M.
f = number in the class
_ = totoZ number
Number of Rays 2T 3 4 5 6 7
FIG. 5. The fluctuating variability of starfish rays. From data by
Goldschmidt.
Thus in this instance the arithmetical mean, ex-
pressed by the hypothetical number 4.915, a number
which of course does not actually occur in nature, is
VARIATION 27
simply the average number of rays in forty-seven star-
fishes selected at random.
The mode which represents the group containing
the largest number of individuals of a kind, namely,
thirty out of forty-seven, is five in this particular
polygon. If all individuals fell within the mode there
would be no variation and the polygon would become a
vertical line.
The average deviation, which is an index of the
amount of variation going on among the starfishes in
question, is .52. In other words, .52 is the average
amount that each individual starfish deviates from the
arithmetical mean of 4.915. Although the one seven-
rayed starfish which happens to be in the lot varies
from the standard of 4.915 to the extent of 2.085
(7 4.915) rays, there are thirty five-rayed starfishes
which vary only .085 c(5 4.915) of a ray, and conse-
quently the average of the entire forty-seven amounts
to .52 of a ray. In another collection of starfishes
where either more seven-rayed or two-rayed specimens
might be present, the average deviation would probably
be greater.
By computing the average deviation, therefore, and
using it as the criterion of variation, a comparison
of the variability of organisms that have been taken
from different localities or subjected to different condi-
tions can be definitely expressed.
A measure of variability more commonly in use by
biometricians, because of its relation to probable error,
is the standard deviation. This is the square root
of the sum of all the deviations squared and their
28 GENETICS
frequencies divided by n, according to the formula
.. * 7 s (**?)
-\ -T->
in which x represents the deviation of each class from
the arithmetical mean; f, the number of individuals in
each separate class ; 2, the sum of the classes ; and n,
the total number of individuals. 1
In the present instance the standard deviation is
.846, a number that has valuable significance only
when brought into comparison with standard deviations
similarly derived from other groups of starfishes.
Such a variation polygon as the above expresses the
law that the farther any single group is from the
mean of all the groups making up the polygon, the
fewer will be the individuals representing it.
7. THE INTERPRETATION OF VARIATION CURVES
a. Relative Variability
The statistical determination of the relative vari-
ability of two lots of organisms with respect to a cer-
tain character may be illustrated by the case of the
oyster-borer snail, Urosalpinx cinereus, as seen in the
accompanying table on page 29.
The obvious conclusion to be drawn from this table
is that the snails which were unintentionally carried
from the Atlantic coast to California in the transplan-
tation of oysters show more variation in their new
habitat than they did in the old one with respect to the
1 For directions explaining the use of such formulae see Daven-
port's "Statistical Methods."
VARIATION 29
ATLANTIC AND PACIFIC SHELLS COMPARED
NUMBER
PROB-
LOCALITT
OF
A.M.
a
ABLE
SHELLS
ERROR
rWest Shore .
1,001
58.928
2.339
-K0352
Penzance Point
1,002
61.718
2.737
-K0412
Nobska Point .
1,002
61.737
2.152
.0324
Woods
Nobska Point .
1,001
61.944
2.234
.0337
Hole 1
Nobska Point .
496
66.944
2.366
.0507
Barnacle Beach
998
63.932
2.604
.0393
Big Wepecket
.Mid-Wepecket
1,006
500
57.426
57.606
2.052
2.098
H-.0308
.0447
Average for Mass. .
61.066
2.335
.0386
Call- J
Belmont Beds . . .
1,008
59.051
3.023
.0454
f ornia ]
San Francisco Bay . .
520
60.892
3.361
.0703
Average for Cal. . .
59.664
3.138
.0538
Difference ....
.803
particular character measured, namely, the relative size
of the mouth aperture compared with the height of the
entire shell. 1
A further analysis of the data in this particular case
shows that this conclusion is probably biologically in-
correct, a discovery which does not invalidate it, how-
ever, as an illustration of a method of determining
relative variability.
b. Bimodal Polygons
Sometimes two conspicuous modes make their ap-
pearance in a frequency polygon, as Jennings found,
for example, in measuring the body width of a popula-
tion of the protozoan Paramecium (Fig. 6).
^'Variation in Urosalpinx." Walter. Amer. Nat. 1910, Vol.
XLIV, pp. 577-594.
30 GENETICS
It was subsequently found that the two modes in this
polygon were due to the fact that the material in
question was a mixture of two closely related species,
Paramecium aurelia and Paramecium caudatum, the
Number of
Individuals
FIG. 6. The body width of a population of the protozoan Para-
mecium, showing a polygon with two modes. A, Paramecium
aurelia. B, Paramecium caudatum. After Jennings.
individuals of which arranged themselves around their
own mean in each instance.
Although such an explanation does not always turn
out to be the right one, the biometrician is led to suspect
when a two or more moded polygon appears that he is
dealing with a mixture of more than one kind of ma-
terial, each of which fluctuates around its own average.
Heterogeneous material, it should be noted, does
VARIATION
31
not always give a bimodal curve. For example, if Pear-
son's two lots of beech leaves mentioned above are
mixed together, they form a regular series from the
inspection of which no one could infer their double
origin. (See the heavy line in Figure 7.)
12
Number
ofriba
14 15 16 17 18 19 20 21
FIG. 7. The ribs of leaves from two beech trees. When put
together they form a polygon which does not reveal its double
origin. From data by Pearson.
c. Skew Curves
The direction in which variations are tending may
sometimes be determined by the statistical method. As
an illustration of this may be cited the number of ray
florets in 1000 white daisies (Chrysanthemum leucanthe-
mum), 500 of which were collected at random by the
writer from a small patch in a swampy meadow in
northern Vermont, while the other 500 were selected
in the same random manner upon the same day from a
dry hillside pasture hardly more than a stone's throw
GENETICS
distant. Among these two lots of daisies the number
of ray florets varies from twelve to thirty-eight and
their frequency polygons, as shown in Figure 8, form
what are termed "skew curves," because the mode in
12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 88 89
FIG. 8. Variation in the ray florets of the white daisy (Chrysan-
themum leucanthemum) . A, from a swampy meadow. JB,
from a dry hillside pasture near by. Both the curves are
"skew" because in each case there is an admixture of the
other type. The distinction between the two types is due to
heredity rather than to environment.
each case lies considerably to one side of the arith-
metical mean.
It will be seen that lot A from the swampy meadow,
which in spite of the greater fertility of the soil and
the unquestionably greater luxuriance of the plants
themselves, produced heads with fewer florets, fluctu-
VARIATION 33
ates around the number 21, while the dry pasture popu-
lation B, characterized by blossoms which were in
general noticeably smaller, fluctuates around the num-
ber 34. The habitats of the two lots were so near to-
gether, however, that there was probably a considerable
intermixture of the two types, as shown by the tendency
of each polygon to produce a second mode. Thus the A
polygon shows that there is an increasing tendency
or variability in the twenty-one floret type toward the
thirty-four floret type, due probably in this particular
instance to invasion resulting from the proximity of
the B colony.
8. GRADUATED AND INTEGRAL. VARIATIONS
It is comparatively simple to treat statistically
integral variations, illustrations of which have been
given in the case of beech-leaf ribs, starfish rays, and
daisy florets, all of which are characters that can be
readily counted. In the same way any measurable
character, such as the size of snail shells, may fall into
easily limited groups, as, for example, 10 to 11 mm.,
11 to 12 mm., 12 to 13 mm., etc. It is somewhat
more difficult to classify variations when color or
pattern is the character in question, since it then be-
comes necessary to define certain arbitrary limits for
each class of the series within which to group the indi-
vidual variants.
Tower, in his famous researches on potato-beetles,
encountered variations in the pigmentation of the pro-
notum all the way from entire absence of color to com-
34 GENETICS
plete pigmentation but by cutting up this continuous
series of variations into arbitrary groups of equal
extent, it was quite possible to arrange the data so
that they could be statistically treated just as con-
veniently as the integral variations mentioned above.
Groups or classes of this kind are termed graduated
variations.
9. THE CAUSES OF VARIATION
With respect to the causes of variation authoritative
biologists have taken different points of view.
a. Darwin considered variations as axiomatic. An
axiom is self-evident, requiring no explanation. The
absence of variations in organisms rather than the oc-
currence of variations is, from this point of view, the
phenomenon requiring an explanation. Although Dar-
win himself spent some time in pointing out the univer-
sal occurrence of variability, he accepted it as a pri-
mary fact and proceeded from it as a starting point
without attempting to seek its causes.
b. Lamarck and his followers have regarded the
causes of variation either as extrinsic, that is, refer-
able to external factors making up the environment of
the organism, or as intrinsic or physiological, that is,
based upon the efforts which an organism puts forth
to fit into its particular environment successfully. The
causes of variation are to be sought according to the
Lamarckian school, in the "environment" and "re-
sponse" sides of the triangle of life rather than in the
"heritage" side (Fig. 1).
For example, Woltereck, by controlling the single
VARIATION 35
extrinsic factor of food supply, was able to modify
the height of the "head" of the microscopic fresh-
water crustacean, Hyalodaphnia, in the remarkable
manner indicated in Figure 9. When poor food was
supplied, the percentage of the head height to that
of the body averaged hardly forty, while with rich food
it was increased to over ninety.
Similarly Klebs succeeded in changing at will the
35 40 45 50 55 60 65 70 75 80 85 90 95 100 #
Ratio of height of head to length of shell ' *
Fio. 9. Schematic curve of the head height of Hyalodaphnia
under various conditions of nourishment. Adapted from
Woltereck.
number of stamens in the common "live-for-ever,"
Sedum spectabile, by manipulating the environment in
which the plants were kept. Some of his results are
shown in Figure 10. Curve A combines the data for
4260 flowers which were raised in well-fertilized dry
soil under bright light ; curve B represents 4000 flowers
grown in a moist greenhouse under red light ; and
curve C includes 4390 flowers from well-fertilized soil
in moist hotbed conditions under a weak light.
c. Weismawi, on the contrary, believes that the
36 GENETICS
causes of variation, at least of heritable variations,
are intrinsic or inborn in the germplasm. His con-
ception of sexual reproduction is that it is a device for
doubling the possible variations in the offspring by
the mingling of two strains of germplasm (amphimixis).
By far the greater number of observations recorded
go to substantiate this theory.
A
Number of
Flowers
80 1
70
10 98765 10 9876543 10 987654
Number of Stamens
FIG. 10. Variations in the number of stamens in the flowers of
the "live-for-ever" (Sedum spectabile) under various con-
trolled conditions. For detailed description, see text. After
Klebs.
Tower found among his potato-beetles, for example,
that two strains reared in the same environment showed
striking differences in variation, a fact necessarily due
to intrinsic rather than to extrinsic factors. Similar
cases may be recalled by any one.
Nevertheless, heritable variation occurs in the ab-
sence of amphimixis so that, at best, sexual reproduc-
tion furnishes only one of the possible avenues for the
introduction of hereditary variations.
VARIATION 37
d. Lastly, Bateson, whose work "On Materials for
the Study of Variation" already cited is a classic,
takes the agnostic attitude that it is rather futile to
guess at the causes of variation before the facts are
well in hand. He consequently discourages such at-
tempts by saying: "Inquiry into the causes of varia-
tion is, in my judgment, premature."
In conclusion, the words of Darwin written over half
a century ago : "Our ignorance of the laws of variation
is profound," may still be appropriately quoted, not-
withstanding the fact that in biometry we have at least
an excellent analytical method by means of which con-
siderable insight into variation is being gained.
CHAPTER III
HERITABLE DIFFERENCES
1. THE MUTATION IDEA
VARIETY is not only the "spice of life" but it is also
the central necessity in the origin of new kinds of ani-
mals and plants. If there was no variation from gen-
eration to generation then nothing new would appear
which nature could in any way seize upon in order to
escape from conservative monotony and as a result
there would be no possible evolution in any direction.
This deplorable state of affairs we know is contrary to
fact.
There are at least three ways, according to Baur,
by which an organism can become different from its
relatives, viz. 1, modification; , combination; 3,
mutation. Which of these three ways has been followed
in any specific instance can only be determined with
certainty by the test of subsequent breeding, for there
is nothing in the appearance of an animal or plant to
indicate by which of these three paths it has gained
any distinctive variation.
By modifications we understand those widespread
differences which are the result of nurture rather than
nature. They are simply environmental effects upon
the somatoplasm and consequently are, in all probabil-
38
HERITABLE DIFFERENCES 39
ity, transitory so far as their inheritance is concerned.
They are the result of soil rather than seed.
"Combinations" and "mutations" are more deep-
seated. They are conditioned by the germinal nature
of the organism and may, therefore, be passed on as
hereditary.
Combinations are the result of a new deal after a
reshuffling of the cards. Nothing essentially new, which
was not already present in one or the other of the
parental lines, is introduced but a different arrange-
ment or bringing together of old qualities is effected.
This process of variation through hybridization is the
concern of Mendelism and will be considered further on.
Mutations, like Minerva springing full-fledged from)
the head of Jove, are something qualitatively new which'
appear abruptly without transitional steps and that!
breed true from the very first.
A distinctive qualitative character marks mutations,
like the discontinuous differences between such chemical
compounds as carbon monoxide (CO) and carbon
dioxide (CO 2 ), as Bateson has pointed out, but the
leap from one to the other may be so small that it is
difficult to ascertain by inspection whether the differ-
ence is due to mutation or to modification. The test
comes in breeding, since the progeny of a modification,
or "fluctuation" as deVries terms it, will revert to the
old average of the parental generation, while the
progeny of a mutation will vary around a new average
set by the mutation itself.
The series of positions taken by the lower end of a
swinging pendulum illustrate what is meant by these
40 GENETICS
non-heritable fluctuating modifications. They all hold
predictable relations to the average position shown
when the pendulum comes to rest, because whenever
the pendulum is put in motion the various positions
all recur as before. A mutation, on the contrary, is
represented by a change in the point of attachment at
the upper end of the pendulum. It occurs only when
the entire pendulum is unhooked and hung up in a
different place. This new point of attachment must be
chosen arbitrarily and has no such definite relation to
the original attachment as characterizes the variation
in position of the swinging end of the pendulum.
When the attempt is made to arrange a series of suc-
cessive mutations in a curve they do not show a graded
relationship to each other as fluctuations do. The
latter mass around the average standard according to
the laws of chance in much the same way that a hundred
shots by a good marksman may center around a bull's-
eye. Mutations never group in this way. They find no
correspondence even with wild shots at the bull's-eye.
They are shots directed at a different target altogether.
To use the musician's phraseology, a variation elabo-
rated upon an old theme would correspond to a modi-
fication but a mutation would be an entirely new
theme.
Darwin was fully aware of the existence of mutations
or "sports" as he called them, and incidentally gave
time to their consideration, but the great task which
he set out to accomplish in such a masterly manner
was to overthrow the widespread and deep-seated be-
lief of his day in a sudden special creation of distinct
HERITABLE DIFFERENCES 41
species. To this end he marshaled evidence in support of
the gradual transition of one species into another,
emphasizing fluctuating modifications rather than muta-
tions which seemed to him to play a minor role in the
origin of species.
It remained for the Dutch botanist Hugo deVries to
be the first to analyze the character of mutations and
to focus attention upon them. There is something
distinctly suggestive of Darwin's method in the fact
that deVries worked in silence for twenty years before
he gave the world the "Mutationstheorie" with which
his name will be forever connected.
2. A SUMMARY OF THE MUTATION THEORY
The main features of the mutation theory of deVries
may be indicated as follows:
a. New species arise abruptly regardless of environ-
ment without transitional forms, and at present they
are not known to arise in any other way.
6. New forms arise as unusual deviations from the
parent form, which itself remains unchanged although
it may repeatedly give rise to similar deviations.
c. New mutations are, from the first, constant, that
is, they produce their like. They do not become
gradually evolved as the result of natural selection
although natural selection may act upon them after
they appear.
d. Among mutations there may occur forms char-
acterized by the addition of something new, progres-
sive elementary species, as well as forms lacking
42 GENETICS
something present in the parental type, regressive
varieties.
e. The same mutation may arise simultaneously in
many individuals instead of as a single "sport."
f. Mutations do not vary around an arithmetical
mean with respect to the parent form, as is the case
with fluctuating variations, but each fluctuates around
a new average of its own, thus forming a discontinuous
series with the parent form.
g. Mutations may occur in all directions, that is,
they are not necessarily definite or orthogenetic.
h. Mutations probably appear periodically.
i. Every mutation means two possible species where
one existed before.
j. Useless or insignificant fluctuating variations are
not necessarily the material from which natural selec-
tion must sift out new species.
k. Natural selection is not a causative agent in
evolutionary advance but is simply a sieve which picks
out successful survivors from mutations.
3. LAMARCK'S EVENING PRIMROSE
Perhaps the most widely known plant mutations are
the progeny of Lamarck's evening primrose, (Enothera
lamarckiana, because it was these plants that led
deVries to formulate his mutation theory.
It is believed by botanists in general that this plant
is a native of the southern United States, although,
so far as is known, it is now extinct as a wild species
in America, and native specimens are included in but
few American herbaria.
It was exported to London as a garden plant about
HERITABLE DIFFERENCES 43
1860, and thence it spread to the continent, where,
escaping from gardens, it became wild in at least
one locality near Hilversum, a few miles from Amster-
dam. Here, in an abandoned potato field, it fell under
the seeing eye of Hugo deVries in 1885, and now both
botanist and primrose are famous.
DeVries found among these escaped plants not only
0. lamarckiana, but also two other kinds of mutants,
0. brevistylis, characterized by short-styled flowers,
and 0. Icevifolia, which has smooth leaves. These two
were entirely new species hitherto unknown at the great
botanical clearing-houses of Paris, Leyden, and the
Kew Gardens.
Since the seeds of the (Enofhera are produced by
self-fertilized flowers, deVries felt safe in regarding
these plants as mutants rather than hybrids, and he
continued to study them with especial care. Trans-
planting the mutants along with representatives of 0.
lamarckicma to his private gardens in Amsterdam,
where it was possible to maintain them in normal
healthy condition, deVries was able to follow their indi-
vidual histories with certainty.
The wild mutants Icevifolia and brevistylis did not
reappear under cultivation but he found that, out of
54,343 plants of the species 0. lamarckiantt grown as
descendants from nine original plants during eight
years, there appeared 837 mutants comprising seven
different elementary species, all of which, with the ex-
ception of 0. scintttlans, bred true. See table on the
next page.
Some explanatory comment on this table may be
of value.
44
GENETICS
MUTANTS or CENOTHERA LAMARCKIANA
<
95
fc
OQ
>.
<j
M
GENERATION
u
1
B
w
H
|
<
J
H]
^
TOTAL
2
O
9
1
3
<5
1
H
X
<!
1
1
I
1886-7
9
II
1888-9
15,000
5
5
III
1890-1
1
10,000
3
3
IV
1895
1
15
176
8
14,000
60
73
1
V
1896
25
135
20
8,000
49
142
6
VI
1897
11
29
3
1,800
9
5
1
VII
1898
9
3,000
11
VIII
1899
5
1
1,700
21
1
= T =
IT 5
"350
T
53,509
W
~229"
== 1 =
54,343
The seeds in each generation were self-fertilized
lamarMana.
The mutant gig as occurred once, in 1895. From
the seeds of this one plant were produced 450 true
gigas offspring in the first year, and the strain con-
tinues to breed true.
Albida was first noted in 1895, but deVries remem-
bered having seen it before and dismissing it as
pathological. Because of its poverty in chlorophyll it
is a mutant which probably would not maintain itself
successfully in nature, but it breeds constant under
cultivation.
Oblonga always bred true with the exception of
throwing an albida in 1895 and a single example of
rubrinervis in 1899.
Of rubrinervis over 2000 invariably bred true, while
nanella bred true in over 20,000 offspring, with but
three exceptions when oblonga characters appeared.
HERITABLE DIFFERENCES 45
Lata, since it produces only female flowers and so
cannot be self-fertilized, had constantly to be crossed
back with the parent lamarckiana, when it produced
from 15 to 20 per cent lata and 80 to 85 per cent
lamarckiana.
Finally, scmtillans which appeared at three separate
times proved constant only in its inconstancy because
it invariably produces a heterogeneous progeny. The
1895 plant gave 53 per cent lamarckiana, 35 per cent
scintillans, 10 per cent oblonga, and 1 per cent lata.
One of the 1896 plants gave 51 per cent lamarckiana,
39 per cent scmtillans, 8 per cent oblonga, 1 per cent
lata, and 1 per cent nanetta, while another 1896 plant
gave only 8 per cent lamarckiana, but 69 per cent
scmtillans, 21 per cent oblonga, and 2 per cent of
nanella and lata together.
These different kinds of evening primroses are dis-
tinguished from each other by features which are un-
mistakable even to the uninitiated. The old-time sys-
tematist would undoubtedly have regarded them as dis-
tinct species.
DeVries distinguishes four categories among the
CEnothera mutants, the first three of which are
quite likely to maintain themselves in nature. They
are:
1. Progressive species, (gig as, rubrinervis) , due to
the addition of certain characteristics ;
2. Retrogressive varieties, (nanella, Iceviflora, bre-
vistylis), characterized by the loss of some-
thing that was present in the parent form;
46 GENETICS
3. Inconstant species, (scintillans and lamarckiana
itself), that do not always breed true but pro-
duce mutants, and
4. Degressive species, (lata, albida), which are de-
fective in some way and are incapable of main-
taining themselves in nature.
DeVries' experiments and observations have been
repeated on a large scale and extended, notably by
MacDougal in the New York Botanical Gardens, by
Shull at the Carnegie Institution for Experimental
Evolution, Cold Spring Harbor, Long Island, and by
Gates in England, and his conclusions have been con-
firmed in all essential points. The mutability of 0.
lamarckiana is as unmistakable and as diverse in
America and England as it is in Holland.
The critics of deVries, however, regard (Enothera
lamarckiana as a hybrid to begin with, from which
different strains have simply been bred out. Both
Bateson and Lotsy have called attention to the pres-
ence of deformed or defective germ cells in (Enothera
lamarckiana as evidence of its hybridity, and Davis,
by crossing 0. franciscana and O. biennis, has pro-
duced a hybrid (Enothera, which he has christened
(Enothera neo-lamarckiana because it not only resem-
bles 0. lamarckiana but behaves like it in producing
mutations. He consequently proposes "dissolution of
hybrids" as a substitution for mutation in explaining
the phenomena that deVries has described.
It is somewhat questionable whether this classical
plant, which has added at least a five-foot shelf to
HERITABLE DIFFERENCES 47
the biological literature of the last thirty-five years,
is after all the most fortunate organism for demon-
strating mutation since its "mutations" may represent
simply combinations becoming isolated from something
already present as the result of past hybridization.
In either case the new form would breed true and behave
like a true mutation.
4. PLANT MUTATIONS FOUND IN NATURE
The oldest known authenticated case of a plant
mutation is the oft cited instance of the "fringed
celandine," Chelidonium laciniatwm, which made its
appearance in the garden of the Heidelberg apothe-
cary Sprenger in 1590 among plants of the "greater
celandine," Chelidonium majus. The fringed celan-
dine bred true at once and is now a widespread and
well-known species.
The purple beech has appeared historically as a
mutant among ordinary beeches upon at least three
occasions in widely separated localities, and it has
always given rise to a constant progeny.
The "Shirley poppy," notable for its remarkable
range of color, which was discovered in 1882 by Rev.
W. Wilks, originated from a single plant of the small
red poppy, Papaver rlioeas, which is commonly found
in English cornfields.
The first double petunia was found in 1855 in a
private garden in Lyons. (Ziegler.) Other instances
are known of double flowers among roses, azaleas,
stocks, carnations, primroses, etc., arising from single
48 GENETICS
flowering plants, the seeds of which in turn produce
double flowers.
The giant primrose is a mutation from a normal
strain of known pedigree. (Keeble.)
"Mutations in certain pericarp color patterns of
maize are so common that a wide range of variability
results. Selection is able from such material to iso-
late types relatively stable but very diverse in appear-
ance." (Emerson and Hayes.)
That plant mutations may occur in nature and per-
sist successfully without isolation or external selection
is shown, for instance, by Schaffner 1 who reports an
unusual white verbena growing wild in Ohio over about
a square mile of territory along with the typical pur-
plish blue Verbena stricta without transitional forms.
Hayes discovered a tobacco mutant in which the
average number of leaves produced was 70 instead of
20, and Cockerell found a single red mutant plant of
the sunflower, Helianthus lenticuLaris coronatus, which
has bred true. The list of similar plant mutations
could be almost indefinitely extended.
5. SOME MUTATIONS AMONG ANIMALS
In 1791 a Massachusetts farmer, by name Seth
Wright, found in his flock of sheep a male lamb with
long, sagging back and short, bent legs resembling
somewhat a German dachshund. With unusual fore-
sight he carefully brought up this strange lamb be-
cause it was an animal that could not jump fences.
'Ohio Naturalist. Dec., 1906.
HERITABLE DIFFERENCES 49
It occurred to this hard-headed Yankee that it would
be much easier to get together a flock of short, bow-
legged sheep, unable to negotiate anything but a low
hurdle, than to labor hard at building high fences.
So it came about that this mutating lamb, in the hands
of a man who appreciated labor-saving devices, became
the ancestor of the Ancon breed of sheep. Later on
this breed gave place in public favor to another mu-
tant, the Merino, which produces a superior grade of
wool.
Some mutations, however, that may be selected and
maintained by man are unlikely to succeed in nature
when left to themselves. Albino animals, for example,
are so handicapped by defective eye-sight that they
have a hard struggle in the wild condition. Albino
rats set free by Dr. Hatai a few years ago upon Goose
Island, a small uninhabited bit of land in Long Island
Sound, all succumbed to the native rats in a short
time.
Hornless cattle suffer fewer injuries from one an-
other than horned cattle. It has consequently become
quite a general practice among farmers to "dehorn"
their stock surgically. It is an obvious advantage to
have cattle born hornless, and many breeds having
this character are now established. In 1889 a mutant
among horned stock appeared at Atchison, Kansas,
in the form of a hornless Hereford. From this mutant
has descended the well-established race of polled Here-
ford cattle, constituting a bovine aristocracy with
registry books and blue blood all their own.
Taillessness in cats, dogs and poultry, as well as
50 GENETICS
hairlessness in cattle, dogs, mice and horses, are fur-
ther instances of mutations.
Davenport, 1 writing of his experiments with poultry,
says: "During the past four years I have handled
and described over 10,000 poultry of known ancestry.
Of striking new characters I have observed many,
some incompatible with normal existence; others in
no way unfitting the individual for continued life.
In the egg unhatched I have obtained Siamese twins,
pug jaws, and chicks with thigh bones absent. There
have been reared chicks with toes grown together by
a web, without toenails or with two toenails to a toe;
with five, six, seven, or three toes; with one wing or
both lacking; with two pairs of spurs; without oil-
gland or tail ; with neck devoid of feathers ; with cere-
bral hernia and a great crest; with feather shaft re-
curved, with barbs twisted and dichotomously branched
or lacking altogether. Of comb alone I have a score
of forms. All of these characters have been offered
to me without the least effort or conscious selection on
my part, and each appeared in the first generation as
well-developed peculiarities, and in so far as their
inheritance was witnessed, each refused to blend when
mated with a dissimilar form."
Bateson (1894), in his "Materials for the Study of
Variation," gives a detailed list of 886 cases of "dis-
continuous variations" among animals, many of which
doubtless belong to the category of mutations, al-
1 Davenport, C. B., 1909. "Inheritance of Characteristics in
Domestic Fowl." Carnegie Institution of Washington, Publica-
tion No. 131.
HERITABLE DIFFERENCES 51
though several may be "combinations" or must be
placed even in the non-inheritable class of "freaks."
The chief reason why definite examples of mutation
are so infrequently noted and recorded is because
the attention of the investigator has generally been
directed, not to them, but to gradual fluctuating varia-
tions which, according to Darwin's conception, furnish
the material for the operation of natural selection.
Mutations are doubtless much more common than has
been generally supposed, and it is likely that they will
receive more attention in the future than they have in
the past.
No stock when bred on a large scale breeds abso-
lutely true for all specific characters. Gerould re-
ports that in his butterflies (Colias), he found blue-
green instead of yellow-green eyes, uncoiled instead of
coiled tongue, the absence of orthodox wing spots,
one proleg less in the caterpillar, etc. Drosophila
is a famous example of many deviations from type
which have been revealed upon persistent and careful
scrutiny.
6. KINDS OF MUTATION
Multiple or aggregate mutations are those germinal
upsets that affect many parts of an organism instead
of a single part. This type is of frequent occurrence
and is in contrast to a single gene mutation which in-
volves only an hereditary unit that determines a single
somatic feature. For example, Babcock describes a
new walnut, Juglans quercina, which appeared inde-
52 GENETICS
pendently in four different widely separated localities
in California. This, like deVries' evening primrose,
was an aggregate mutation, for differences appeared
in size, shape, color and texture of leaves ; size, form
and color of flower-parts; color of bark, habit of
growth, etc. That this was a true mutant and not a
hybrid between the oak and the walnut was indicated
by negative results in cross-pollinating experiments.
Similar aggregate mutations have been reported for
cotton, tomato, tobacco and other organisms.
Another phenomenon that probably indicates com-
mon ancestral germplasm among species, at present
apparently independent of each other, is the occurrence
of parallel mutations. The North African ostrich
(Struthio camelus) and the South Australian ostrich
(S. australis), although separated from each other for
long geological time, show, according to Duerden,
similar mutations in size, length of neck and legs, skin-
color and bald-head as well as in size and shape of the
egg and the character of its surface, whether pitted or
ivory-smooth.
A long list of parallel mutations in Drosophila
melanogaster and D. virilis has been described by Metz,
and similarly, Sturtevant reiports mutations in D.
funebris that are likewise parallel to those of D.
melanogaster, in which the occurrence of mutations
has probably been more carefully studied than in any
other animal.
Sumner with the deer-mouse, Peromyscus, has found
albinism, spotting and red-eyed yellow, all mutations
known to occur in other mice.
HERITABLE DIFFERENCES 53
Reverse mutations have also been repeatedly ob-
served. This is something resembling the unscrambling
of an egg. Morgan and Bridges obtained, for exam-
ple, normal red-eyed flies from white-eyed mutants and
May, also with ubiquitous Drosophila, got back nor-
mal-eyed individuals from bar-eyed mutants.
The frequent occurrence of recurrent mutations,
that is, the reappearance of the same mutations, sug-
gests that the cause underlying these irregular heredi-
tary changes is something continuous and definite even
if we are at present unable always to put our finger
upon it. The evening primroses have repeatedly shown
the same mutations in widely different localities and
under the eyes of different investigators. Morgan says
of his famous banana flies, "One of the first mutants
that appeared, viz., white eyes, has appeared anew in
our cultures about three times, in cultures known to
be free from it before and not contaminated. The same
mutant has been found by several other observers.
The eye color vermilion has appeared at least six
times; the wing character called rudimentary, five
times ; cut wing has been found four times," etc.
7. THE ORIGIN OF MUTATIONS
Mutations may be gametic, zygotic or somatic in
their origin. There seems to be no reason why muta-
tion may not occur at any stage in the life-cycle of an
organism. In the first place, it may be gametic in
origin if the onset is in the germ-cell before or during
the maturation changes that prepare it for union with
54 GENETICS
another germ-cell (See Chap. X). In this instance
its effect may be profound and patent upon the entire
development of the individual, although if it chances
to be relegated to an abortive polar cell during meiosis
or to an unmated spermatozoon it will be entirely lost
at once. There are no doubt many such "mute inglo-
rious mutations" (Muller) that never see the light
of day.
It is furthermore obvious that a gametic mutation
usually enters the organism concerned singly, that is,
from one parent only, and if recessive 'in character
will fail to put its appearance in the somatoplasm
until some subsequent generation when two hybrids
from the new stock each chance to contribute the re-
cessive mutant character in question to the formation
of a new individual.
The appearance of such mutants, therefore, unless
dominant, must come two or more generations after
the mutation has taken place. The time when a
gametic mutation is initiated, consequently, and when
it manifests itself are by no means necessarily the same.
This fact needs to be kept in mind in considering the
evidences from experiments for the determining causes
of mutations.
Perhaps the reason why mutations are more fre-
quently reported in self-fertilizing (autogamous)
plants than in cross-fertilizing (heterogamous) ani-
mals is because in self-fertilizing organisms the inbreed-
ing necessary to bring about the doubling of a single
character so that it will come into expression is more
likely to occur.
vw -
HERITABLE DIFFERENCES 65
Secondly, the mutation may occur in the fertilized
egg. This is zygotic mutation. In this case the change
is evident at once in the resulting individual since the
developing individual is the unfolding of what is pres-
ent in the zygote. Such a mutation, for example, oc-
curring after fertilization and not as the result of a
combination or cross, is reported in tobacco by Hayes
and Beinhart. 1
Thirdly, in contrast to the two kinds of mutations
just described which, are distinctly germinal in origin,
there may be somatic mutations which fall directly
upon some individual somatic cell or tissue arising out
of the original germplasm and produce in turn such
abnormalities as "bud variations," chimaeras and the
like. In such a case all the cells and tissues arising
from the mutant somatic cell will express the mutation
and no others. Lehmann, 1920, proposes the term
metaclonosis for hereditary somatic modifications, re-
serving the term mutation for solely those instances
that involve a change in the genes.
8. WHEN MUTATIONS_DCCUR
It has been suggested by Standfuss that species
may go through the same kind of a life-cycle that indi-
viduals do, only taking infinitely more time to do it.
As shown in Figure 11, they are born of other species
and enter the prodigious growth period of infancy and
youth, both of which are characterized by much fluctua-
tion. With maturity they gradually become compara-
tively stable until the reproductive period is reached,
Science XXXIX, No. 992, p. 34.
56
GENETICS
when they throw off their progeny, as on a tangent.
They finally pass into the excessively differentiated
period of old age, from which there is no recall, al-
though they approach in many features the infantile
condition, and end in death or extinction. This cycle is
repeatedly illustrated by phylogenetic lines of fossil
forms which have
long since become
extinct.
Beecher has
pointed out that,
in paleontolog-
ical times just
before they be-
c a m e extinct,
species often un-
derwent extreme
Fio. 11. Diagram of the relation of re- ...
production to the life-cycle. specialization in
the form of fan-
tastic shapes, an excessive number of spines or elabo-
rate sculpturings on the shells as seen among the
ammonites, belemnites, and trilobites, or of gigantic
size as in the dinosaurs, plesiosaurs, and theromorphs.
All of these facts indicate a species-cycle in which these
abnormal features were the unmistakable signs of old
age.
The reproductive period of a species when mutants
are being thrown off, as of an individual, may extend
over a considerable period of the whole cycle, or it
may be confined to a relatively small segment. It is
possible that in the evening primrose deVries may
HERITABLE DIFFERENCES 57
have caught a plant passing through the crucial period
of species-reproduction.
Another reason why so few mutations have as yet
been seen may be because the majority of organisms
are not, during the short span of human observation,
in the reproductive part of their cycles. When it is
remembered that accurate observation with this object
in view has extended over only a brief period, insig-
nificant in comparison with the vast geologic stretches
of time concerned in species-building, the marvel is
that so much, rather than so little, has been seen.
9. POSSIBLE CAUSES OF MUTATION
There are at least three avenues of approach to the
analysis of mutation: (1) Anatomical, depending upon
observation of its occurrence in nature and under
control; (2) Genetical, consisting of the experimental
breeding of test cases, and (3) Cytolqgical, or the
microscopic examination of the germplasm. It is this
latter method that furnishes perhaps the most hope of
gaining some insight into the fundamental causes under-
lying the phenomena of mutation.
No doubt the conclusions in this paragraph could
be better presented after the consideration of the re-
maining chapters of the book, particularly the section
on the cellular basis of heredity (Chaps. X, XI and
XIII), but some discussion, nevertheless, seems desir-
able at this point, even if it may be necessary to return
and reread it later.
Babcock and Clausen have classified mutations from
58 GENETICS
the cytological standpoint into two groups, viz.,
chromosomal aberrations and factor mutations. Chro-
mosomal aberrations are accidents or irregularities
occurring in the nuclear make-up of the germ-
cells. These aberrations may be of various kinds and
probably take place during meiosis when the germ-
cells are going through the preparatory process of
reduction of the chromosomes which precedes the for-
mation of the fertilized egg.
For example in the unpairing of homologous chro-
mosomes after synapsis it is conceivable that the pro-
cess may not be clean-cut and complete but that a piece
of one chromosome may adhere to its mate thus
changing its size and composition. Or again, a frag-
ment of a chromosome, during the complicated elimi-
nation performances accompanying the marriage cere-
mony of germ-cells, may be shuffled out and lost, thus
creating a deficient chromosome. Such accidents to the
germ-cells would be reflected in all the subsequent
mitotic divisions of the somatic cells derived there-
from and a mutation would be the result. At any rate
an examination of the nuclear structure of mutants
frequently reveals chromosomal irregularities so that
an unmistakable relation between the two phenomena
undoubtedly exists.
Another irregularity that occurs is an unequal
migration of the chromosomes to the poles of a germ-
cell during the reduction division, which, of course,
results in a cell progeny of mature gametes having
a number of chromosomes unlike the number in the
normal gametes. This appears to be the reason for
HERITABLE DIFFERENCES 59
the mutation, (Enothera lata, which has 15 chromo-
somes instead of 14, the typical number for 0.
lamarckiana from which it sprang. What occurs in
the formation of this mutation is that for some reason
0. lamarckiana during reduction division instead of
dividing as usual into 7-7 makes the unequal division
of 6-8, a phenomenon known as non-dis junction
(Bridges). When this 8-chromosome gamete joins
with a normal 7-chromosome gamete the new mutant
number of 15, characteristic of O. lata, is the result.
Gates and others in their extensive cytological
studies on (Enothera mutants, have found not only 15
chromosomes instead of 14 but also, associated with
various other mutations, the abnormal numbers of 16,
20, 22, 23, 24, 27, 28, 29, and 30.
(Enothera gigas is a mutant in which 28 chromo-
somes, or twice the normal number, appear and,
moreover, these chromosomes represent actually twice
the original amount of chromatin material. "Gigas"
mutants have been found in various other forms, such
as the tomato (Winkler), the jimson weed or Datura
(Blakeslee and Belling), Primula (Gregory) and
Narcissus (Stomps), and they are always character-
ized by a doubling of the chromosomes. This condition
is termed tetraploidy because it shows four times the
gametic number of chromosomes.
When a normal diploid Datura is crossed with a tet-
raploid gigas individual, a triploid mutant results
with a different constellation of somatic characters so
that the best of evidence is now at hand that one
category of mutations, at least, that of chromosomal
60 GENETICS
aberrations, is dependent upon, or associated with,
abnormal quantitative differences in the chromosomes.
The other category of mutations, factor mutations,
is qualitative and concerns the character of hereditary
units or genes rather than quantitative groups of these
genes as they are assembled in the chromosomes.
Whatever it is that causes the character of a gene
to change in quality, with the resultant expression in
the somatoplasm, is still apparently beyond the pale
of scientific proof. Some investigators find satisfac-
tion in assigning external environmental causes to the
solution of the problem while others prefer to conceal
their ignorance under the blanket of "internal causes,"
whatever these may be. At least it is reasonable to
say when a new variety appears suddenly in a bottle
full of flies or in a field of plants in the same environ-
ment with all of its unmodified fellows, that mutation
can arise somehow without outside interference.
The wild jungle fowl presents a large and useful
series of mutations which have cropped out in poultry
under the spell of domestication while the goose, on
the contrary, although domesticated for an equally
long period, has remained practically the same. The
nature of the plastic hen must be different from that
of the more conservative goose.
Meanwhile the secret of the real causes of muta-
tions remains a challenge to every geneticist and suc-
cess surely awaits some clever workman who knows
how to use skilfully the indispensable tools of obser-
vation and experimentation.
The bearing of the whole matter of mutation upon
HERITABLE DIFFERENCES 61
heredity lies in the fact that, contrary to Darwin's
belief, it is apparently mutations, and not fluctuations
or "modifications,'* that make up heritable variations.
If this supposition proves to be true, mutations furnish
the essential material in the study of heredity. Conse-
quently, whatever knowledge we may gain of them has
a direct relation to the entire problem of genetics.
CHAPTER IV
THE INHERITANCE OF ACQUIRED
CHARACTERS
1. SUMMARY OF PRECEDING CHAPTERS
HEREDITARY resemblance is due to the derivation of
offspring from the same stock as the parent, and
successive generations, therefore, are simply periodic
expressions of the same continuous stream of germ-
plasm.
Perfect inheritance, or uniformity of generations,
does not exist, since variations always occur in suc-
cessive generations. It is upon these variations that
evolution depends. Without them there would be no
change of type and consequently no possibility of
evolutionary advance.
Some variations are fluctuating or continuous in
character and may be detected and analyzed by sta-
tistical methods, while others are mutations, or dis-
continuous variations, representing qualitative differ-
ences which do not lend themselves readily to statistical
analysis.
Mutations are more common than was formerly
believed, and since they are germinal rather than
somatic in character, they play an important role in
heredity.
62
ACQUIRED CHARACTERS 63
2. THE BEARING OF THIS CHAPTER UPON GENETICS
Only those variations which reappear in succeeding
generations have to do with heredity. Hence it be-
comes important to inquire as to what kind of varia-
tions actually reappear. Can variations that are not
inborn, but which are acquired during the lifetime of
the individual, be inherited? Does the experience of
the parent become a direct part of the child's heritage,
or can the environment of the one enter in any way
into the heredity of the other? Can changes wrought
in the somatoplasm be so impressed upon the germ-
plasm as to change it in such a way that it, in turn,
will give rise to similarly modified somatoplasm in the
next generation? To use Shakespeare's antithesis, can
nurture as well as nature be transmitted? As Conklin
says : "Few questions have been discussed so fully and
so fruitlessly as this."
In answering these questions we are of course con-
cerned solely with biological inheritance and not at all
with those extra-biological accumulations in the way
of arts, literature, tradition, invention, and the like
which constitute civilization and which make us the
"heirs of the ages." Such benefits are entailed upon
us much in the same way as property is "inherited,"
but they form no part of the personal biological
heritage into which we are now inquiring.
3. THE IMPORTANCE OF THE QUESTION
This inquiry concerning the inheritance of acquired
characters, which Professor Brooks has called "the
64 GENETICS
interminable question," is not simply an academic
matter. Its solution is of vital importance from sev-
eral viewpoints. For breeders, who are trying to
maintain or improve particular strains of animals or
plants ; for physicians, who, in fighting disease, are
honestly seeking to substitute an ounce of prevention
for a pound of cure ; for sociologists and philanthro-
pists, who have at heart the permanent bettering of
human conditions; for educators, who cherish hopes
that their life-work of unfolding the youthful mind
may prove cumulative and lasting rather than tran
sitory; for religious workers, who want their faith
strengthened that conquests in character-building may
outreach the individual and so enrich the race; for
parents, who entertain hopes that their own efforts
may give their children a better biological start in
life, for all these and many more, it is important to
know the answer to the question: Can acquired
characters be inherited?
4. AN HISTORICAL SKETCH OF OPINION
That the personal accumulations of a lifetime are
heritable was generally believed throughout the credu-
lous ages. A century ago Lamarck made this idea
the corner-stone of his theory of evolution. It was all
very simple. The reason evolution occurs in nature
is because individual acquirements are being continually
added to the onflowing stream of living forms. This
cumulation of characters indeed is evolution. How
else can the present stage of adaptation of organisms
ACQUIRED CHARACTERS 65
to their several niches in nature be explained save by
seeing in it the final results of generations of gradu-
ally inherited adaptations?
Darwin also believed in the inheritance of acquired
characters, although he differed from Lamarck with
respect to how such characters are acquired.
Francis Galton in 1875 was one of the first to ex-
press skepticism regarding this generally accepted
belief, but the man who, in a masterly manner, focused
the growing doubt, and who did more than any other
to inspire thought and investigation upon the subject,
was August Weismann, who conspicuously bore the
torch of genetics between 1880 and 1890. Weismann
made the issue so clear that the heritability of acquired
characters became the parting of the ways which
divided biologists into the two camps of Neo-Lamarck-
ians who affirm, and Neo-Danmnians who deny, such
inheritance. His conclusions, which are the natural
outgrowth of the theory of the "continuity of the
germplasm," were based, however, upon logical rather
than upon experimental grounds.
Comparative anatomists and paleontologists, who
are accustomed to work from results back to their
causes, are frequently inclined to look favorably upon
the inheritance of acquired characters while, on the
other hand, geneticists and embryologists, representing
the two lines of study which furnish the most imme-
diate approach to this problem, are well-nigh agreed
that acquired characters are not inherited. Experi-
ment from cause to result is undoubtedly the best cri-
terion for if the question could be decided by a vote
66 GENETICS
or by an expression of opinion, the result would be
doubtful, since each column contains the names of men
whose scientific accomplishments entitle them to a
respectful hearing. But just what are the facts of
the case?
5. CONFUSION IN DEFINITIONS
The source of much of the lack of agreement in
this controversy lies in the definition of what con-
stitutes an "acquired character." One is reminded
of the two knights who fought so bitterly over the
color of a shield, one maintaining that it was red,
the other that it was black. So they hacked away
at each other, as all good knights should do in the
defense of the truth, until they both fell -down dead
beside the shield which was black on one side and
red on the other.
Of course actual characters are never inherited, but
only the determiners or potentialities which regulate
the way in which the organism reacts to its environ-
ment with respect to the characters in question. Reid
has pointed out that in one sense every adult charac-
ter is "acquired" because it has no expression at first.
For instance, there is no beard on the face of a male
infant, but one will presumably be "acquired" later
on in the life-cycle due to a heritable and not to an
environmental cause.
It is plain that every new character which repre-
sents a forward evolutionary step must have been
"acquired," or added, sometime and somewhere, else
it would not be present, as there is evidence that it is.
ACQUIRED CHARACTERS 67
Perhaps the question, as Montgomery has suggested,
ought to be changed to read: "What kinds of acquired
characters are inherited?" It is obvious that discus-
sion is futile until a common denominator in the shape
of a definition of acquired characters shall be accepted.
6. WEISMANN'S CONCEPTION OF ACQUIRED
CHARACTERS
Weismann defines an acquired character as any
somatic modification that does not have its origm in
the germplasm.
Of course those somatic modifications which are
phases of the developing individual, such as the
acquisition of a deeper voice at puberty or the substi-
tution of the permanent dentition for the milk-teeth,
are somatic variations which have their rise and con-
trol in the germplasm and consequently cannot prop-
erly be included under the head of acquired characters.
Examples of acquired characters in the Weisman-
nian sense are mutilations, the effects of environment,
the results of function as in the use or disuse of certain
organs, and such diseases as may be due either to in-
vading bacteria or to the neglect or abuse of the bodily
mechanism.
7. THE DISTINCTION BETWEEN GERMINAL AND
SOMATIC CHARACTERS
Redfield has thrown light on the classification of
the characters which make up the individual by quot-
ing the familiar lines:
68 GENETICS
"Some are born great,
Some achieve greatness,
Some have greatness thrust upon them."
"Born" characters are constitutional, having their
origin in the germplasm itself. They are never
Weismannian acquired characters and may be illus-
trated by eye-color, mental disposition, or facial fea-
tures. Lightning calculators and musical prodigies
may have their gifts developed and enlarged, but the
fact that their talent is nevertheless an unmistakable
gift, and not an acquisition, remains true.
"Achieved" characters are functional and are
gained by exercise. Many things are achieved, how-
ever, which are not acquired characters, as, for in-
stance, wealth, reputation, or an education. Not any
of these are biological characters, and therefore we
are not concerned with them in this connection, al-
though in the case of education it should be noticed
that the mental exercise necessary to bring about a
trained mind, if not the subject matter of the educa-
tion itself, is distinctly an acquired character of the
"achieved" type.
"Thrust" characters are the results of environment.
They are acquired without functional activity on the
part of the organism and usually in spite of anything
the organism can do to prevent. Sometimes these
characters are thrust upon the individuals before birth,
as in the case of blindness caused by parental gonor-
rhoea or tuberculosis arising from uterine infection,
in which case they are termed congenital characters.
Congenital or prenatal characters, however, are in
ACQUIRED CHARACTERS 69
no way the same as germinal characters, for they fall
just as truly into the category of acquired variations
as do those which make their appearance in later life.
8. WHAT VARIATIONS REAPPEAR?
Returning now to Montgomery's question, "What
kinds of acquired characters are inherited?'* it is
apparent that only the "born" ones can be, which have
their roots in the germplasm whence the new individual
arises, and that "achievements" and "thrusts," in
order to reappear in the succeeding generation, can
do so only by first becoming incorporated in the germ-
plasm.
Any character that is not acquired must have been
present in the germplasm from which the organism
arose, as there is no transfer of characters between
organisms except through the germ-cells. Thus it is
evident that the only inherited acquisitions are those
which, either primarily or .secondarily, bring about
variation in the germplasm. Such temporary acquisi-
tions as a coat of tan or a display of freckles do not
impress the germplasm, for when the cause that incites
their appearance is removed, they soon vanish.
9. How MAY GERMPLASM ACQUIRE NEW CHARACTERS?
In addition to mutation considered in the last chap-
ter, various sorts of rearrangement in the germplasm
may present something different.
First may be mentioned the "amphimixis" of Weis-
mann, that is, the mixture of two nearly related strains
70 GENETICS
of germplasm in sexual reproduction within a species,
and secondly, the mixture of two more remotely related
strains resulting in hybridization. In either case the
strain of germplasm undergoes a shake-up that may
result at least in new combinations of characters, if
not in the production of entirely new characters. This
recombination of characters in amphimixis and
hybridization will receive further attention in a later
chapter.
The fact that successive parthenogenetic genera-
tions, in which amphimixis does not of course occur,
may show a larger degree of variability than sexually
produced generations, indicates that amphimixis in
itself is by no means sufficient to account for all kinds
of variations.
It is conceivable that the external factors that act
upon the germplasm may be grouped into three classes :
first, external factors acting upon the somatoplasm
and then through the agency of the somatoplasm af-
fecting the germplasm ("somatic induction" of Detto
or "pangenesis" of Darwin) ; second, external factors
acting directly upon the germplasm and the somato-
plasm at the same time ("parallel induction" of
Detto) ; and third, external factors acting upon the
germplasm without necessarily at the same time hav-
ing any effect upon the somatoplasm.
Many instances of direct influence of external fac-
tors upon germplasm are known in biological litera-
ture, and these have led to some of the misunderstand-
ings concerning the "interminable question" of the in-
heritance of acquired characters. Sitkowski fed the
caterpillars of the moth Tineola biselliella with an
ACQUIRED CHARACTERS 71
aniline dye (Sudan red III), obtaining therefrom, in-
stead of normal whitish ones, moths that laid col-
ored eggs, and these in turn hatched into caterpillars
still tinged with the color of the red dye. Riddle, with
guinea-pigs, and Gage, with poultry, obtained quite
similar results. This case of apparent parallel induc-
tion, however, is not a matter of inheritance at all,
since the germinal substance itself was not involved,
but of animals who got their red color directly from
external sources while they were eggs within the moth-
er's body.
10. WEISMANN'S REASONS FOE DOUBTING THE INHER-
ITANCE OF ACQUIRED CHARACTERS
Weismann's reasons for questioning the popularly
accepted view that acquired characters are inherited
may be briefly stated as follows :
First, there is no known mechanism whereby somatic
characters may be transferred to the germ-cells.
Second, the evidence that such a transfer actually
does occur is inconclusive and unsatisfactory.
Third, the theory of the continuity of the germ-
plasm is sufficient to account for the facts of heredity
without assuming the inheritance of acquired somatic
characters.
Let us examine these three statements a little more
closely.
A. NO KNOWN MECHANISM FOR IMPRESSING THE
GERMPLASM WITH SOMATIC ACQUISITIONS
Each germ-cell remains an independent unit and
does not participate in the activities of the body but
72 GENETICS
lies within the body like a commensal or parasite. It
is hard to see, therefore, how a germ-cell can be
changed except in a general nutritive way which is
quite different from a change in character of any
hereditary significance.
The somatoplasm is something that has traveled
out from the original fundamental germplasm along
the paths of differentiation and elaboration. The more
complex the body-cells become, that is, the more suc-
cessive modifications they undergo, the more difficult
it is for these somatic cells to return to their original
primitive germinal estate.
In many lower forms of life where cell elaboration
is not so great, a part lost by amputation is often
regenerated, but this process is not possible in higher
forms where the parts represent cell complexes too
hopelessly differentiated to begin anew the unfolding
sequences of their elaboration. This difficulty was
a very real one in the mind of that famous nocturnal
inquirer Nicodemus when he asked: "How can a man
be born when he is old? Can he enter a second time
into his mother's womb and be born?"
Not only the development of the race which we call
evolution, but also the determination of the individual
in heredity, is a chain of onward-moving sequences like
the succession of events in history. It is hard to see
how recent events can influence preceding events. It is
hard to see how the water that has gone over the dam
can return and affect the flow of the river upstream in
any direct way. It is likewise hard to see how differ-
entiated somatoplasm, which represents the end stage
ACQUIRED CHARACTERS 73
of a successive series of modifications, can make any
definite impress upon the original germplasmal sources
from which it arose.
Darwin felt this difficulty and presented with apolo-
gies his provisional hypothesis of pangenesis in which
he assumed that every bodily part sends contributions
to the germ-cells in the form of "gemmules." These
gemmules, or hypothetical somatic delegates, then
reconstruct in the germ-cells the characters of the
entire body, including acquired modifications as well
as all others, and thus there is no reason why acquired
characters cannot readily be transmitted. Unfortu-
nately there is no tangible basis in fact for this
delightfully simple explanation to rest upon. It is a
theory assuming that all parental somatic cells take
part in the formation of the new individual, hence it
was called "pangenesis," or origin from all.
Nothing we have subsequently learned of minute
cell structure favors this hypothesis, while many facts
go quite against it. Moreover, it is directly opposed
to the theory of the continuity of germplasm so con-
vincingly set forth later on by Weismann. Darwin
indeed advanced it only in the most tentative way,
being entirely ready to see it abandoned at any time
for something better. It at least performed one valu-
able service to science, namely, that of demonstrating
how far investigators were from an adequate concep-
tion of any means by which somatic modifications might
become incorporated in the germ-cells.
We must acknowledge, however, with Lloyd Morgan
that the fact that a mechanism for the transfer of
74 GENETICS
somatic characters to the germ-cells has not been dis-
covered, is not proof that such a mechanism does net
exist. It may simply be beyond our present powers of
penetration.
B. EVIDENCE FOR TRANSMISSION OF ACQUIRED
CHARACTERS INCONCLUSIVE
The evidence for the inheritance of acquired charac-
ters was, for a long time, taken for granted. This
theory was the most obvious explanation of many facts
and so was accepted without question. An obvious
interpretation, however, is not always the correct one.
The sun appears to go around the earth, but astrono-
mers assure us that it does not.
When Weismann began to sift the evidence for the
inheritance of acquired characters, he found that it
was largely based upon opinion rather than fact, much
like the popular belief with regard to the causation of
warts by handling toads.
The supposed evidence for the inheritance of ac-
quired characters falls chiefly into the following cate-
gories :
a. Mutilations;
b. Environmental effects ; .
c. The effects of use or disuse; .
d. The transmission of disease ;
e. Immunity;
f. Prenatal influences.
ACQUIRED CHARACTERS 75
a. Mutilations
It is fortunate that the sons of warriors do not
inherit their fathers' honorable scars of battle, else
we would now be a race of cripples.
The feet of Chinese women of certain classes have
for centuries been mutilated into deformity by ban-
daging, without the mutilation in any way becoming
an inherited character. The same result is also true
of tattooing and of circumcision, the latter a mutila-
tion practised from ancient times by the Jews and
certain other Eastern peoples. The progressive degen-
eration or crippling of the little toe in man has been
explained as the inheritance of the cramping effect
of shoes upon generations of shoe wearers, but, as
Wiedersheim has pointed out, the fact that Egyptian
mummies show the same crippling of the little toe is
unfavorable to this hypothesis, for no ancient Egyptian
could ever be accused of wearing shoes or of having
had shoe-wearing ancestors. Sheep and horses with
docked tails as well as dogs with trimmed ears never
produce young having the parental mutilation.
Weismann's classic experiment with mice, an experi-
ment subsequently confirmed by others, is additional
negative evidence upon this same point. What Weis-
mann did was to breed mice whose tails had been cut
off short at birth. He continued this decaudalization
through twenty-two generations with absolutely no
effect upon the tail-length of the new-born mice.
One may see in the catacombs of the Zoologisches
Institut at Freiburg, filed carefully away on shelves,
76 GENETICS
as a "document," long rows of labeled bottles contain-
ing the fifteen hundred and ninety-two martyrs to
science which made up the twenty-two generations of
mice in this famous experiment.
Blaringhem, it is true, obtained mutations which
bred true from latent buds that were forced into devel-
opment following mutilation of normal buds, but
Griffon has shown that similar mutations occur with-
out preceding mutilations so that this, as Shull points
out, is simply a case of segregation of biotypes already
present in the mutilated parent.
Conklin has hit the nail upon the head with respect
to mutilations by saying: "Wooden legs are not in-
herited, but wooden heads are."
b. Environmental Effects
Trees deformed by prevailing winds, like the willows
that line the canals in Belgium and Holland, or storm-
crippled trees along the exposed seacoast are not
known to produce a modified progeny when their ad-
verse environmental conditions are removed. Simi-
larly, the persistent sunburn of Englishmen long resi-
dent in India does not reappear in their children born
in England.
Sumner kept mice in a constant but abnormally high
temperature of 26 C. with the result that the ears,
tail, and feet grew noticeably larger than in control
animals kept in ordinary lower temperatures, while at
the same time the general hairiness of the body de-
creased. It should be remembered, however, that mice
ACQUIRED CHARACTERS 77
are mammals which pass through an extended uterine
existence, so that it is easy to see how the offspring
in this case were subjected to the same excessive tern-
perature as the parents for a period sufficient to amply
account for their subsequent variation when removed
to a normal environment.
Zederbaur finds that the wayside weed Capsella,
which in the course of many years has gradually crept
along the roadsides up into an Alpine habitat and
there "acquired" Alpine characters, upon being trans-
planted to the lowlands retains its Alpine modifications.
Although this case has been cited as an authentic in-
stance of the inheritance of acquired characters, is it
not possible that the conquest of the Alps by CapseUa
has been due, in the course of time, not to the inherit-
ance of acquired characters at all, but to a gradual
natural selection of just those germinal variations
which best fitted it to cope with Alpine conditions
until, finally, a strain of germplasm producing somato-
plasm suitable to Alpine conditions has been isolated
in the form of an elementary species derived from the
original type? If this is what has happened, of course
such germplasm would give rise to Alpine plants
whether individual plants grew to maturity near the
snow-line or in the warm valleys at a lower altitude.
Kammerer, by reducing the water supply, succeeded
in transforming Salamandra maeulosa, a salamander
normally producing about seventy eggs which, when
hatched in water, become gill-breathing tadpoles, into
a salamander producing only two to seven young
which are born alive without gills and are able to live
78 GENETICS
out of water entirely, in damp situations. These
land-adapted offspring, moreover, when supplied with
abundant water, produce in turn tadpoles which spend
days only, instead of months, in the water undergoing
their metamorphosis, thus showing an apparent inherit-
ance of an acquired character.
It should be pointed out, however, that in these cases
the gill-breathing forms in each instance represent a
case of arrested development. Axolotl is simply a
larval form of Amblystoma which, under normal con-
ditions of an abundant water environment and high
temperature, gets no further in its metamorphosis than
the tadpole stage, when it produces eggs and sperms
and finishes its life story. A change in environment
simply permits the life-cycle to go on further. Chang-
ing from gill-breathing to lung-breathing is not, there-
fore, an acquired character, but a purely germinal
character that may be either blocked or released by
changing conditions in the environment. The phe-
nomenon is termed neotony.
c. The Effects of Use or Disuse
The callosities on the end of a violinist's left-hand
fingers are acquired by use, but they are not inherited.
There are callosities on the knees of the wart-hog,
Phacochcerus, which are also apparently the result of
use, for these animals kneel as they root for a living in
the African forests, and have done so for untold gen-
erations. It has been noticed that young wart-hogs
as soon as they are born possess the callosities, so that
ACQUIRED CHARACTERS 79
this instance looks like one of inheritance of a charac-
ter acquired through use or exercise.
The skin on the soles of human feet is thicker than
the skin elsewhere, and by use it becomes still thicker.
This is apparently another instance of the same sort.
The writer has observed, however, that a cross sec-
tion through the foot of a "mud puppy," Necturus
maculatus, shows a much thickened sole. Necturus,
it should be noted, is a very primitive salamander
living always under water and never using the soles
of its feet in any way to bear its weight, nor is it
reasonable to suppose that it ever had any ancestors
who did so, for the hands and feet of the Amphibia
are the most primitive and ancient hands and feet to
be found in the animal kingdom without any known
ancestral types. The thickening of the skin on the
sole of the mud puppy's feet must be due, therefore,
to germinal determiners and is in no way an acqui-
sition through use. The same may also be true of the
wart-hog's knees and of human soles.
The strong arm, the skilled hand, and the trained
ear are not inherited. They have always to be re-
acquired in each succeeding generation just as surely
as the ability to walk, or to read and write.
Herbert Spencer has defined instinct as "inherited
habit." But surely those instincts which determine a
single isolated action during the lifetime of the indi-
vidual, such as the spinning of a peculiar cocoon, can-
not be the result of habit, since habits are formed only
through repeated action.
Dr. Hodge, who succeeded in hatching tame quail
80 GENETICS
chicks out of "wild" eggs, asks the pertinent ques-
tion: "How can a fear hatch out of an egg?" The
habit of wildness, particularly with precocial chicks
like quails, may, under an inciting environment, be
very soon established but it is difficult to see how cau-
tion, gained by the experience of the parents, can find
its way into the fertilized egg. If, then, some instincts
require a different explanation from that of "inherited
habit," may it not be likely that all instincts do? Is
it not better to assume that the structure of the germ-
plasm determines a particular response to a particular
stimulus regardless of whether in the past the ances-
tors have made a similar response to a similar stimulus ?
d. Transmission of Disease
If acquired diseases were heritable we would all have
been dead long ago. When a son, whose father died
of pneumonia, succumbs himself to pneumonia after
an interval of years there may be no more causal
or hereditary connection between the two events than
when a second house burns down on the same site where
a former house went up in flames.
Many diseases, like tuberculosis, have their imme-
diate cause in invading pathogenic bacteria. Bacteria
themselves cannot be inherited for the reason that it
is not possible for them to become an integral part of
the fertilized egg and thus cross the "hereditary
bridge" which joins two generations. A general pre-
disposition to bacterial disease, that is, a lack of re-
sistance to bacterial invasion due to defectiveness in
ACQUIRED CHARACTERS 81
physical or physiological equipment, may be present
as a combination of characters in the germplasm, or
an individual, as the result of disease, may "ac-
quire" a generally weakened germplasm and so pro-
duce a progeny exhibiting general liability to disease;
but it is doubtful if such a condition can properly be
termed the inheritance of an acquired character, since
the particular definite disease in question is not de-
monstrably heritable.
When alcoholism "runs in a family," its reappear-
ance in the son is probably due to the fact that he is
derived from the same weak strain of germplasm as
his father. The fact that the father succumbed to
the alcohol habit is not the determining cause of
drunkenness in the son. The same thing that caused
the father to become an alcoholic, namely, weak germ-
plasm, and not the resulting drunkenness in the parent,
is the causal factor for alcoholism in the son.
At the same time it is entirely probable that heredi-
tary alcoholism may in some cases arise through
"parallel induction," that is to say, acquired alco-
holism may end in the simultaneous poisoning and
consequent modification of both the somatoplasm and
germplasm of the parent, with the result that the germ-
plasm has less resistance to alcoholism in a succeeding
generation. The offspring are consequently more
likely to succumb to the disease. This, however, is
not the inheritance of an acquired character or of a
definite somatic modification.
When a man of the present generation has rheu-
matic gout, it is a severe stretch both of patriotism
82 GENETICS
and of the powers of heredity to trace the origin of
the affliction back to a revolutionary ancestor who
acquired sciatic rheumatism by sleeping on the ground
at Valley Forge, yet this is quite as direct as many
alleged instances of the inheritance of disease.
In the majority of instances, apparent cases of the
inheritance of disease are merely instances of reinfec-
tion. This reinfection of the offspring may occur very
early in embryonic life, even in the egg, in the case of
pebrine in silkworms (Pasteur) and in the tick which
transfers the protozoan parasite causing Texas fever.
Or it may happen after birth, provided the offspring
are exposed to the same environment as that in which
the parent acquired the disease, but in any case reinfec-
tion is not heredity.
e. Immwrdty and the Effect of Drugs
Ehrlich subjected mice to increasing doses of ricin
until they became immune to doses which are ordi-
narily fatal. When these ricin-immune mice were bred
to non-immune mates the offspring in turn showed some
degree of immunity if the immunized parent was a
female but not if the immunized parent was a male.
In other words, the immunity was transferred through
the female only, where the blood of the mother is for
a considerable period during foetal life in intimate rela-
tion with the blood of the offspring. Even here, just
as in the lifetime of an immunized individual, the im-
munity tended to fade out after a short time.
As a matter of fact many of the instances that have
ACQUIRED CHARACTERS 83
been advanced to show the inheritance of acquired
characters are simply transient hold-over somatic
effects that have gained no permanent grip upon the
hereditary stream of germplasm, and which conse-
quently soon fade away.
In a similar way the gradual acclimatization of the
mold, Penicillium, to a salt solution of a density suffi-
cient to cause its death if placed in it at once, has been
effected, and the resulting spores have produced molds
that are able to survive in the concentrated solution.
Here, of course, the spores have been acclimatized as
well as the parent plant and it was to be expected that
these spores would develop into molds habituated to
the increased saline environment. This, however, is
pseudo-heredity, for no permanent method of response
has been established.
/. Prenatal Influences
Perhaps the most illogical and at the same time the
most widespread of all types of supposed transmis-
sion of acquired characters are the so-called "maternal
impressions." The prevalence of this superstition has
caused expectant mothers untold needless misery.
Popenoe and Johnson, after an excellent and ex-
tended discussion of the matter, conclude as follows :
"To recapitulate, the facts are
(1) That there is, before birth, no connection be-
tween the mother and child, by which impressions on
the mother's mind or body could be transmitted to the
child's mind or body.
84 GENETICS
(2) That in most cases the marks or defects whose
origin is attributed to maternal impression, must
necessarily have been complete long before the incident
occurred which the mother, after the child's birth,
ascribes as the cause.
(3) That these phenomena usually do not occur
when they are, and by hypothesis ought to be, expected.
The explanations are found after the event, and that
is regarded as causation which is really coincidence.
It is easily understandable that any event which
makes such an impression on the mother as to affect
her health, might so disturb the normal functioning of
her body that her child would be badly nourished, or
even poisoned. Such facts undoubtedly form the basis
on which the airy fabric of prenatal culture was reared
by those who lived before the days of scientific biology."
C. THE GEEMPLASM THEORY SUFFICIENT TO ACCOUNT
FOE THE FACTS OF HEREDITY
Weismann holds that the theory of the continuity
of the germplasm, already considered in a previous
chapter, is sufficient in itself to account for the facts
of heredity. Hence it is quite unnecessary to fall back
upon the inheritance of acquired characters as an
explanation, since this theory is at least difficult, if
not impossible, of satisfactory proof.
To prove the inheritance of acquired characters,
according to Weismann three things are necessary:
first, a particular somatic character must be called
ACQUIRED CHARACTERS 85
forth by a known external cause; second, it must be
something new or different from what was already
exhibited before, and not be simply the reawakening
of a latent germinal character; and third, the same
particular character must reappear in succeeding
generations in the absence of the original external
cause which brought forth the character in question.
As yet these conditions have not been convincingly
met in the evidence which has been brought forward
in support of the inheritance of acquired characters.
11. THE COMPARATIVE INDEPENDENCE OF GERM
AND SOMA
The fact that the germ is only a pilgrim stranger
passing through the homeless land of the soma is well
brought out by the critical ovarian transplantation
experiments of Castle and Phillips upon guinea-pigs.
The ovaries of an albino guinea-pig were removed
and those of a black guinea-pig were grafted in their
place. After recovery from the operation the animal
was mated with an albino male three times before
pneumonia unfortunately put an end to this famous
experiment. The resulting offspring were all black,
as shown in Figure 12. Ordinarily when albinos are
crossed they produce only albinos. It is obvious that
the pneumonia victim was not the mother of the six
black offspring although she bore them. "The conclu-
sion is forced upon us," to quote Babcock and Clausen's
comments on the case, "that the egg-cell during its
growth does not change in germinal constitution. Its
growth is like the growth of a parasite or of a wholly
86
GENETICS
independent organism ; what it takes up serves as food ;
this is not incorporated merely in the growing organ-
ism, it is made over into the same kind of living sub-
stance as composes the assimilating organism."
FIG. 12. Diagram of ovarian transplantation experiment to
show the influence of somatoplasm upon germplasm. Black
is dominant over albino. The ovaries from a black guinea-pig
were engrafted into a female albino whose ovaries had been
removed. Upon recovery this female was crossed three times
with an albino male. All the progeny were black. Data from
Castle and Phillips.
12. ACQUIRED CHARACTERS IN THE PROTOZOA
Although the problem of the inheritance of acquired
characters is much better defined among the higher ani-
ACQUIRED CHARACTERS 87
mals where the distinction between the soma and the
germ is more sharply cut than among the lower ani-
mals and plants, yet, as Jennings points out, one meets
FIG. 13. The behavior of an "acquired character," a spiny pro-
jection at one end of the body, in the case of Parameciwm.
The original individual is represented in the center and its
offspring, which arise by fission, are in successive circles. In
the fifth generation only one out of 32 shows the spine.
Data from Jennings.
the same difficulties in the protozoa as in the metazoa.
The difficulty in the inheritance of acquired charac-
ters is not so much in separating germ and soma as in
the mechanism of cell-division. There seems to be no
way in which an acquisition located at one end of a
88 GENETICS
cell can overleap the barrier of cell division and appear
at the other end after mitosis.
In his cultures Jennings found a Paramecium with
an abnormal spine at one end. This acquisition was
handed on for five generations before it disappeared
but never in any generation did more than one of the
offspring have the spine. In other words, it did not
become hereditary although it continually reappeared
in one individual in every generation. The reason for
this will be apparent upon referring to Figure 13. The
fission-half bearing the spine holds the same relation
to the spineless half as soma to germ and there is here
no mechanism for the transmission from one half to
the other. Simple transmission, like the persistence of
the spine for five generations of Paramecium is not
heredity. In order that a character shall be really
inherited, that is, shall appear in more than one of the
progeny and so affect the race, it must be produced
anew in each generation from a germinal determiner.
This is just as true for the protozoa as it is for the
higher organisms.
13. THE OPPOSITION TO WEISMANN
The opponents of Weismann point out, as a weak
place in his argument, the assumption that the germ-
plasm is so insulated from the somatoplasm as not to
be influenced by it. Weismann assumes, of course,
that the germplasm is isolated from the somatoplasm
very early in the development of the fertilized egg into
an individual, and that when once isolated it thereafter
ACQUIRED CHARACTERS 89
takes no active part in, nor is in any way affected by,
the vicissitudes through which the somatoplasm, or the
body itself, passes. The somatoplasm is thus merely
a carrier of the germplasm and unable to affect the
character of it any more than a rubber hot-water bag,
although capable of assuming a variety of shapes, can
affect the character of the water within it.
In opposition to this view it is urged that every
organism is a physiological as well as a morphological
unity, and that cells entirely insulated within such a
unity would be a physiological miracle.
There is abundant evidence that germ-cells, or
rather the hormones in the sexual organs producing the
germ-cells, do affect the somatoplasm under particular
conditions, as, for example, in cases of castration when
those somatic features called "secondary sexual char-
acters" undergo profound modification.
Even here, however, it must be pointed out that it
is not the germ-cells themselves that are directly re-
sponsible for the modifications which occur, but rather
the hormones of the interstitial gonadal cells. A most
serious fly in the Weismannian ointment is due to the
results of certain recent experiments by Guyer and
Smith. 1 These ingenious experimenters injected into
fowls the freshly removed lenses of rabbits' eyes that
had been pulped up in Ringer's solution. The fowls
developed an "anti-body" which tended to dissolve
and disintegrate the rabbit lenses. When serum
from these fowls was in turn injected into pregnant
rabbits the mother was unaffected but nine out of sixty-
*Jour. Exp. Zool III, 1920.
90 GENETICS
one surviving young were born with degenerate eyes.
The affected young have carried the defect even in
the male line through eight generations without the
injection of any more serum containing the lens anti-
body. "The degenerating eyes are themselves, directly
or indirectly, originating anti-bodies in the blood serum
of their bearers which in turn affect the germ-cells."
If these conclusions are substantiated, the cardinal
principle of the inheritance of acquired characters is
conceded. The end is not yet !
14. VARIOUS RESULTS UPON OFFSPRING OF PARENTAL
ACQUISITIONS
In Diagram 14 an attempt is made to visualize the
various results of parental acquisitions, both somatic
and germinal, upon the generations following.
It will be noted that Case I, where the soma of the
parent is represented as determining the soma of the
offspring, is contrary to fact for in sexual reproduc-
tion the offspring arises from the undifferentiated
germplasm of its parents.
The usual result of a somatic modification is shown
in Case II.
Pangenesis, Case III, postulates a reversal of the
universal process of differentiation in that it demands
a return of the elaborated soma with the modifications
it has acquired during the course of its elaboration,
to the primitive condition of the germ.
In Case IV the apparent inheritance of acquired
characters is due not to the fact that the parental soma
ACQUIRED CHARACTERS
91
92 GENETICS
was modified, but because at the same time and in the
same way that the parental soma was taking on a
modification, the germ was likewise modified. This, to
use the drug clerk's phraseology, is "something just as
good" as the inheritance of acquired characters but
it is not the Weismannian brand.
Finally, Case V shows a true mutation which occurs
in the parental germplasm but does not appear to the
light of day until the offspring develops.
15. CONCLUSION
But even granting that the somatoplasm affects the
germ-cells, the inheritance of acquired characters is
by no means thereby established.
In order to do this, the precise acquired character
in question, which indirectly exercised its influence
upon the germ, must be redeveloped, and, although
the germplasm might conceivably receive an influence
from the somatoplasm and be affected by it in a gen-
eral way, it is a different matter entirely to develop
anew the replica of the character itself which is sup-
posed to have been acquired.
It will be seen in subsequent pages, under the dis-
cussion of data furnished by experimental breeding,
that the weight of probability is decidedly against
the time-honored belief in the inheritance of acquired
characters.
CHAPTER V
MENDELISM
1. METHODS OF STUDYING HEREDITY
MODERN studies in heredity have been pursued princi-
pally in three directions : first, by microscopical ex-
amination of the germ-cells; second,, by statistical
consideration of data bearing upon heredity ; and
third, by experimental breeding of animals and plants.
In the present chapter attention will be directed to a
consideration of experimental breeding with reference
to hybridization, that is, breeding from unlike parents,
a process which Jennings characterizes by the expres-
sive phrase, "the melting-pot of cross-breeding."
2. THE MELTING-POT OF CROSS-BREEDING
Hybridization, or cross-breeding, as analyzed by
^Galton (1888), results in one of three kinds of inherit-
ance, namely, blending, alternative, or particulate.
Of these, blending inheritance may be called ,the
typical ''melting-pot" in which contributions from the
two parents fuse into something intermediate and dif-
ferent from that which was present in either parent.
Galton illustrated this process by the inheritance of
human stature in which a tall and a short parent pro-
93
GENETICS
duce offspring intermediate in height. A more thorough
consideration of this type of inheritance will be pre-
sented in Chapter VIII.
By the method of alternative inheritance the pa-
rental contributions do not melt upon union, but re-
tain their individuality, reappearing intact in the off-
Charaettristica
Of parental
Blending
Alternative
Particulate
may produce
FIG. 15. Three kinds of inheritance described by Galton, when
applied to a single pair of characters.
spring. In inheritance of human eye-color, fo
ample, the offspring usually have eyes colored like -tKtV
of one of the parents when the parental eye-co!0/ 5
unlike in the two cases, rather than eyes intermedia
in color between those of both parents.
Particulate inheritance results when the offspring
present a mosaic of the parental characters, that is,
when parts of both the maternal and paternal charac-
ters reappear in the offspring without losing their iden-
MENDELISM 95
titles by blending or without excluding one another.
Piebald races of mice arising from parents with solid
but different colors have been cited as illustrations of
this sort of inheritance, although it will be seen later
in connection with the "factor hypothesis" that another
interpretation of this phenomenon is not only possible
but probable.
The distinctions between these three categories of
inheritance are diagrammatically represented in Figure
15.
3. JOHANN GREGOR MENDEL
Our understanding of the working of inheritance
in hybridization we owe largely to the unpretentious
studies of an Austrian monk, Johann Gregor Mendel,
who, although a contemporary of Darwin, was prob-
ably unknown to him. Bateson says of Mendel: "Un-
troubled by any itch to make potatoes larger or bread
heaper he set himself in the quiet of a cloister garden
D find out the laws of hybridity, and so struck a mine of
uth, inexhaustible in brilliancy and profit." For
Mit years Mendel carried on original experiments by
Deeding peas and then sent the results of his work to a
j jrmer teacher, the celebrated Karl Nageli, of the Uni-
- vversity of Vienna. At the time Nageli's head was full of
other matters, so that he failed to see the significance of
lis old pupil's efforts. However, in 1866 Mendel's
results appeared in the Transactions of the Natural
'History Society of Briinn, 1 an obscure publication
1 Verhandlungen naturf. Verein in Briinn. Abhandl. IV, 1865
r (which appeared in 1866).
96 GENETICS
that reached hardly more than a local public. Here
Mendel's investigations were buried, so to speak, because
the time was not ripe for a general appreciation or
evaluation of his work.
At that time neither the chromosome theory nor the
germplasm theory had been formulated. Moreover,
much of our present knowledge of cell structure and
behavior was not even in existence. Weismann had not
yet led out the biological children of Israel through the
wilderness upon that notable pilgrimage of fruitful
controversy which occupied the last two decades of the
nineteenth century, and the attention of the entire
thinking world was being monopolized by the newly
published epoch-making work of Charles Darwin.
Mendel died in 1884, and his work slumbered on
until it was independently discovered, almost simul-
taneously, by three botanists whose researches had
been leading up to conclusions very much like his own.
These three men were deVries of Holland, von Tscher
mak of Austria, and Correns of Germany. Their con
tributions were published only a few months apart c
1900 and were closely followed by important papers
from Bateson in England, Cuenot in France and.,
Davenport and Castle in America, extending Mendelisr^
to animals, with a rapidly increasing number from ot
biologists the world over. To-day the literature uj
this subject has grown to be very large, and the end
by no means yet in sight.
Castle has well said: "Mendel had an analyti
mind of the first order which enabled him to plan an
carry through successfully the most original and in
structive series of studies in heredity ever executed."
MENDELISM
97
l t. MENDEL'S EXPERIMENTS ON GARDEN PEAS
JVhat Mendel did was to hybridize certain varieties
garden peas and keep an exact record of all the
igeny, in itself a simple process but one that had
;ver been faithfully carried out by any one.
"To Mendel's foresight in arranging the conditions
his work, as much as to his astuteness in interpreting
e data, is due his remarkable success." (Morgan.)
Before examining Mendel's results it may be well to
late the difference between normal and artificial self-
:ilization. Self-fertilization occurs when from the
>llen and ovule of the same flower are derived the two
imetes which uniting produce a zygote that develops
[to the seed and subsequently into the adult plant of
next generation. In artificially crossing normally
[f-fertilized flowers it is necessary to carefully re-
ive the stamens from one flower while its pollen is
ill immature, and later, at the proper time, to transfer
it ripe pollen from another flower.
[Mendel's cross-breeding experiments on peas showed
;ain numerical relations among the progeny which
re rise to what has come to be rather indefinitely
as "Mendel's law." This law may be temporarily
ftrmulated as follows :
B*Vhen parents that are unlike with respect to any
acter are crossed, the progeny of the first genera-
will apparently be like one of the parents with
to the character in question. The parent
nresses its character upon the offspring in this
manner is called the dominant. When, however, the
Jiybrid offspring of this first generation are in turn
ch
98 GENETICS
crossed with each other, they will produce a mixefl
progeny, 25 per cent of which will be like the dominanB
grandparent, 25 per cent like the other grandparenl
and 50 per cent like the parents resembling the doi
nant grandparent.
An illustration will serve to make plain the manm
in which this law works out.
Mendel found that when peas of a tall varietj
were artificially crossed with those of a dwarf variety
all the resulting offspring were tall like the first parenj
It made no difference which parent was selected as tl
tall one. The result was the same in either cas
showing that the character of tallness is independei
of the character of sex.
When these tall cross-bred offspring were subsj
quently crossed with each other, or allowed to p
duce offspring by self-fertilization which amounts
the same thing, 787 plants of the tall variety and 277?
of the dwarf kind were obtained, making approximated
the proportion of 3 to 1.
On further breeding the dwarf peas thus derivdH
proved to be pure, producing only dwarf peas, whih*
the tall ones were of two kinds, one third of thai
"pure," breeding true like their tall grandparent, and
two thirds of them "hybrid," giving in turn the PWJ
portion of three tall to one dwarf like their pareiM
These crosses may be expressed as follows :
Tall, T, X dwarf, t, = tall, T(t).
That is, tallness crossed with dwarfness equals tallness
with the dwarf character present but latent.
MENDELISM
99
Mendel termed the character, which became apparent
in such a hybrid, in this case tallness, the dominant,
and the latent character which receded from view, in
this instance dwarfness, the recessive.
The members of such a Mendelian pair are termed
allelomorphs.
When now the hybrids, T(t), were crossed together,
the result algebraically expressed was as follows:
T + t (all possible egg characters)
T -j- t (all possible sperm characters)
TT + Tt
Tt + tt
TT + 2T(t) + tt
That is, one of the four possible cases was dwarf, 1 1,
in character and the other three were apparently tall,
although only one out of
the three was pure tall,
TT, while the remaining
two were tall with the
dwarf character latent,
T(t).
The same thing may be
expressed more graph-
ically by the checkerboard
plan, which Punnett sug-
gested (Fig. 16). Each
Male Gametes ^
L f
W3 1 -
t
<
- tT
it
FIG. 16. Diagram to illustrate
theoretically the formation of
the four possible zygotes in
the second filial generation of
a monohybrid.
square of the checker-
board represents a zygote
which, having received a gamete from each of the two
parents, may develop into a possible offspring. The
character of the gametes of the parents is shown out-
100
GENETICS
side of these squares, while the arrows represent the
parental source from which the offspring have received
their hereditary composition.
The essential feature of Mendel's law is briefly this :
hereditary characters are visually independent units
which segregate out upon crossing, regardless of tem-
porary dominance.
Mendel carried on further experiments with garden
peas, using other characters. He obtained practically
the same result as in the instance already given, for
the actual progeny in the second generation of the
cross-bred offspring figured up, as seen in the table
below, very nearly to the expected theoretical ratio of
3 to 1.
CHARACTER
NUMBER OF
DOMINANTS
NUMBER OF
RECESSIVES
RATIO
Form of seed . .
Color of seed coat .
Length of stem . .
Color of flowers
Position of flowers
Form of pods .
Color of unripe pods
5474 smooth
6022 yellow
787 tall
705 colored
651 axial
882 inflated
428 green
1850 wrinkled
2001 green
277 dwarf
224 white
207 terminal
299 constricted
152 yellow
2.96 to
3.01 to
2.84 to
3.15 to
3.14 to
2.95 to
2.82 to
Total
14949
5010
2.98 to 1
These results have been confirmed by other investi-
gators, for example the yellow-green seed-color cross
has been repeated by Correns, Tschermak, Hurst, Bate-
son, Lock, Darbishire and White, with results totalling
195,477 in the second generation of which number
146,802 were yellow and 48,675 were green. This is a
proportion of 3,016 to 1.
MENDELISM 1(H
5. SOME FURTHER INSTANCES OF "MENDEL'S LAW"
Since the rediscovery of Mendel's law the ratio of
3 to 1 in the second hybrid generation has been found
by a number of different investigators to be constant in
a large array of characters observed both in animals
and plants of diverse kinds when these are cross-bred
with reference to the characters in question.
Botanists have an advantage perhaps in this matter,
as they deal with forms which usually produce a large
number of offspring from a single cross, a very desir-
able condition in estimating ratios. On the other hand,
they are handicapped by being unable usually to obtain
more than one generation in a year, while zoologists
may secure from animals like rabbits and mice several
generations in a year, although ordinarily the number
of progeny is much smaller and the ratios obtained
have a larger chance of error than is the case with the
more numerous plant offspring.
Semi-microscopic animals, as, for example, the pom-
ace fly, Drosophila, which produces a large progeny
every two weeks or so, may combine the general ad-
vantages mentioned for the two groups of organisms
indicated above, but they have the disadvantage of
being so small that the detection of their distinctive
phenotypic characters is attended with considerable
technical difficulty.
What the modern experimenter in genetics desires is
an organism, first, which possesses conspicuous distinc-
tive somatic characters, and, second, that will come to
sexual maturity early and breed either in captivity or
103
G'ENETICS
under cultivation both numerously and frequently with
a minimum of trouble and expense.
The following table, compiled chiefly from Bateson l
and Baur, 2 might easily be much extended. It shows
ORGANISM
AUTHOR
Q
DOMINANT
RECESSIVE
Nettles
Correns
'03
Serrated leaves
Smooth-margined
leaves
Sunflower
Shull
'08
Branched habit
Unbranched habit
Cotton
Balls
'07
Colored lint
White lint
Snapdragon
Baur
'10
Red flowers
Non-red flowers
Wheat
Biffen
'05
Susceptibility
[mmunity to rust
to rust
Tomato
Price and
'08
Two-celled fruit
Many-celled fruit
Drinkard
Maize
deVries
'00
Round, starchy
Wrinkled, sugary
kernel
kernel
Silkworm
Toyama
'06
Yellow cocoon
White cocoon
Cattle
Spillman
'06
Hornlessness
Horns
Pomace fly
Morgan
'10
Red eyes
White eyes
Horses
Bateson
'07
Trotting habit
Pacing habit
Land snail
Lang
'09
Jnbanded shell
Banded shell
Mice
Darbishire
'02
formal habit
Waltzing habit
Guinea-pig
Castle
'03
Short hair
Angora hair
Canaries
Bateson and
'02
Crest
Plain head
Saunders
Poultry
Man
Davenport
Farrabee
'06
'05
:lumplessness
Brachydactyly
Long tail
formal joints
Barley
von Tschermak
'01
Beardlessness
Beardedness
Salamander
(Ambly-
stoma)
Haecker
'08
Dark color
Light color
from what diverse sources confirmatory evidence of the
truth of Mendel's law has been derived within the
first ten years of observation and experiment after its
rediscovery.
1 "Mendel's Principles of Heredity," 1909.
a "Einfiihrung in die experimented Vererbungslehre," 1911.
MENDELISM 103
6. THE CARDINAL, PRINCIPLE OF SEGREGATION
The essential thing which Mendel demonstrated was
the fact that, in certain cases at least, the determiners
of heredity derived from diverse parental sources may
unite in a common stream of germplasm from which,
in subsequent generations, they may segregate out ap-
parently unmodified by having been intimately asso-
ciated with eadh other. This law of segregation, or
"independent assortment" as Morgan prefers to call it,
depends upon the conception that the individual is
made up of a bundle of unit characters. It may be
illustrated by the separate flowers picked from a garden
which, after being made into a nosegay, may be taken
apart and rearranged without in any way disturbing
the identity of the separate blossoms.
The general formula of segregation that covers
all cases of organisms cross-bred with respect to a
single character, that is, monohybricfo, is given in
Figure 17. ,
The parents of a hybrid are usually referred to as
the parental generation (P). The hybrid generation
formed by crossing diverse characters 'in parents is
designated as the first filial generation (Fj). The
offspring of F x are F 2 , and so on.
Incidentally this diagram hints how it is possible
to derive a pure strain from an impure (hybrid) source,
a fact of immediate interest not only to breeders of ani-
mals and plants but also to breeders of men.
Such "extracted" recessives or dominants will be en-
tirely free of the hybrid impurity.
104 GENETICS
7. DEFINITIONS
A character which is present in the offspring in
double quantity because it was present in both parents
is said by Bateson to be homozygous, while an or-'
ganism which is homozygous with respect to any char-
D (Dominant)
DD
ZD&)
RR
\
D
2 Dp)
^
DD DD DD. 2D(R) RR RR RR
FIG. 17. General Mendelian formula for a monohybrid.
,->
acter is called a homozygote so far as that particular
character is concerned (DD or RR.)
In contrast to the homozygous condition, an organ-
ism is said to be heterozygous when it derives the deter-
miner of a character from only one parent. Such
an organism is described as a heterozygote with respect
to the character in question (DR).
Organisms that appear to be alike, regardless of
their germinal constitution, are said by Johannsen to
be identical phenotyjpically (DD and DR), while or-
ganisms having identical germinal determiners- are
said to be genotypically alike (DD and DD or RR
and RR).
MENDELISM 105
The word "genotype" was suggested by Jbhannsen
in honor of Darwin and his theory of pangenesis, al-
though there are certain objections to its use in this
connection for the reason that systematists have
already appropriated it in a different sense. As here
used it signifies "the fundamental hereditary constitu-
tion or combination of genes of an organism" (Shull).
8. THE IDENTIFICATION OF A HETEROZYGOTE
"Homozygote" and "heterozygote" are terms then
descriptive solely of the genotypical constitution of
organisms, and, as has been said, it is not always pos-
sible to distinguish one from the other by inspection.
The only sure way to identify a heterozygote is by
breeding to a recessive and observing the kind of off-
spring produced.
Peas of the formulae TT and T(t) 9 for example,
both look alike, since a single determiner for the tall
character, T, is sufficient to produce complete tallness.
When, however, these two kinds of tall peas are each
bred to a recessive dwarf pea, of the formula tt, the
progeny will differ distinctly in the two cases as fol-
lows :
Case I. T + T X t + t = 100 per cent T(t).
Case II. T t X * + * = 50 per cent T(t) + $0 per cent tt. -
That is, if the dominant to be tested is homozygous
(Case I), the entire progeny will exhibit the dominant
character, but if the dominant to be tested is heterozy-
gous (Case II), then only one half of the progeny will
show the character in question.
106 GENETICS
Sometimes when dominance is not pronounced it is
possible to distinguish the heterozygote dominant from
the homozygote dominant. Correns has described an
excellent instance of this type. When plants of a
white-flowering race of the four-o'clock, MirabUis
jalapa, are crossed with those of a red-flowering race,
all the offspring in the first filial generation, unlike
either parent, exhibit rose-colored flowers. When, how-
ever, these rose-colored flowers are crossed with each
other, they produce red, rose, and white in the Men-
delian ratio of 1:2:1; that is, three colored to one
white. The red-flowering race thus proves to be
homozygous and the rose-flowering race heterozygous.
Here color dominates the absence of color, or white,
but the degree of the color depends upon whether the
dose of pigment is double, from both parents, or single,
from only one parent.
9. THE "PRESENCE OR ABSENCE" HYPOTHESIS
In place of Mendel's conception that every dominant
character is paired with a recessive alternative or
allelomorph^ there has been proposed the presence or
absence hypothesis which was first suggested by Correns
but later logically worked out by others, particularly by
Hurst, Bateson, and Shull. According to this inter-
pretation, a determiner for any character either is,
or is not, present. When it is present in two parents,
then the offspring receive a double, or duplex, "dose,"
to use Hurst's word, of the determiner. When it is
present in only one parent, then the offspring have
MENDELISM 107
a single, or simplex, dose of the character. When it
is present in neither parent, it follows that it will not
appear in the offspring. In this case the offspring
are said to fie nulivplex with respect to the character
in question. Take the case of tall and dwarf peas,
the determiner for t alines s when present produces tall
peas, even if it comes from only one parent, but if this
determiner for tallness is absent from both parents,
the offspring are nulliplex, that is, the absence of tall-
ness results and only dwarf peas are produced.
The difference between the presence or absence
theory and the dominant or recessive theory of allelo-
morphs is that in the former case the "recessive" char-
acter has no existence at all, while in the latter" in-
stance it is present, but in a latent condition.
The reasons for and against the presence or absence
interpretation may be more suitably considered later.
10. DIHYBRIDS
So far reference has been made exclusively to mono-
hybrids, any two of which are supposed to be similar
except with respect to a single unit character. Mono-
hybrids are comparatively simple, but when two or-
ganisms are crossed which differ from each other with
respect to two different unit characters, the situation
becomes more complicated.
Mendel solved the problem of dihybrids by crossing
wrinkled-green peas with smooth-yellow peas. He found
that smoothness, S, is dominant over wrmkledness, W,
and that yellow coloi^ F, is dominant over green, G,
\
108 GENETICS
or, as it would be stated according to the presence or
absence theory, smoothness is a positive character
which fills out the seed-coat to plumpness while its
absence leaves a wrinkled coat, and yellowness is a
positive character due to a fading of the green which
causes the yellow to be apparent. In the absence of
this green-fading factor, or determiner, the green of
course appears.
If smooth-yellow, SY, and wrinkled-green, WG, are
crossed, all the offspring are smooth-yellow, but
they carry concealed the recessive determiners for
wrinkledness and greenness according to the formula
S(W)lf(G). When the determiners of these cross-
breds segregate out during the maturation of the germ-
cells, they may recombine so as to form four possible
double gametes, namely, smooth-yellow, 5F, and
wrinkled-green, WG, which are exactly like the grand-
parental determiners from w_hich they arose, and in
addition, two entirely new combinations, smooth-green,
SG, and wrinkled-yellow, WY.
Since the male and the female cross-breds are each
furnished with these four possible gametic combina-
tions, the possible number of zygotes formed by their
union will be sixteen (4X4=16). That is, the mono-
hybrid proportion of 3 to 1 in dihybrid combinations
is squared, (3+l) 2 =16.
It of course does not follow that the offspring in
dihybrid crosses will always be sixteen in number, or
that they will always conform strictly to the theoreti-
cal expectation of (3+1 ) 2 . The offspring obtained
undoubtedly obey the laws of chance, but the greater
MENDELISM 109
the number of offspring, the nearer they come to fall-
ing into the expected grouping.
The sixteen possible zygotes resulting from a dihybrid
cross will give rise to sixteen possible kinds of indi-
viduals which in turn, as will be demonstrated directly,
present four kinds of phenotypic and nine kinds _o
genotypic constitutions.
A dihybrid mating, using the same symbols em-
ployed in the case just described, would be expressed
algebraically as follows:-***
SG+ WY+ /SF+ WG = all the possible egg gametes
SG+ WY+ SY+ WG = all the possible sperm gametes
80SG+ 80WY+ SGSY+ SGWG
SGWY + WYWY + WYSY+ WYWG
SGSY -f WYSY +SYSY+ SYWG
SO WO + WYWG + SYWG+WGWQ
SGSG+2SGWY+28G8Y+2SGWG+WYWY+2WYSY-\-2WYWG+SYSY+28YWG-}-WGWG
\+~* V~
The second and the ninth items in this result are
alike; by combining them the revised result reads:
>gOSG + 4SGWf+ 2SGSY+ 2SGWG+WTWY ^T
There are then these nine different combinations
of germinal characters or nine different genotypes
in any dihybrid cross. By placing the recessive char-
acters in parentheses whenever the corresponding
dominant is present, to indicate that the dominant
causes the former to recede from view, these nine geno-
types may be combined into four phenotypes as shown
in the table at the top of page 110.
From this analysis it may be said that the Mendelian
ratio for a typical dihybrid is phenotypically
that for a moriohybrid, as we have
110
GENETICS
Phenotypes
Genotypes
9SY
SYSY
3 SO
SGSG
3WY
WYWY
already seen, is phenotypically 3:1. This expected
ratio corresponds essentially with the actual results
SG WY SY WG
* * \ I
SG
WY ,
SY .
WG
&
SG
SG
i /7\
SG
o, *
o 1 ar
WY
WY
V
SY
WG,
'
WY
sp
WY
SY
WG
SY
.\SY
SY
4 SY
SG
lyY <<
SY '
WG
\WG
. WG
WG
WG
FIG. 18. Diagram to illustrate the possible combinations arising
in the second filial generation (F 2 ) following a cross between
yellow-smooth, YS, and green-wrinkled, G^V, peas.
which Mendel obtained in crossing smooth-yellow and
wrinkled-green peas.
Figure 18 presents a graphic representation of the
different combinations resulting from a dihybrid cross
following the checkerboard plan used in Figure 16
to illustrate monohybrids.
The nine genotypes and four phenotypes which
MENDELISM
111
result from a dihybrid cross are shown in the following
table:
Number in
Each Class
GENOTYPE
Number of
Squares in
Fig. 18
PHENOTYPE
Number in
Each Class
1
8Y8Y
11
8Y
9
2
(W)YSY
7:10
2
S(G)SY
3 -9
4
8(G)(W)Y
2 & 12 15
1
SGSG
1
8G
3
2
SG(W)G
13 -4
I
WYWY
6
WY
3
2
WYW(G)
8-14
1
WGWG
16
WG
1
16
16
Another illustration of dihybridism is shown in Fig-
ures 19 and 20 based upon data furnished by the
Davenports. 1 In the matings given here, dark or
pigmented hair, represented by the solid black circles,
is dominant over light-colored, that is, unpigmented or
slightly pigmented hair, symbolized by the open circles,
while curly hair is dominant over straight, represented
by crooked and straight lines respectively in the dia-
gram. In other words, the presence of pigment is
dominant over the absence of pigment, while the factor
that causes curliness is dominant over the absence of
this factor, with respect to human hair.
1 "Heredity of Eye-color in Man," Science, N. S. 26, p. 589,
1907; "Heredity of Hair Form in Man," Amer. Nat. 42, p. 341,
1908. Davenport, C. B. and G. C.
GENETICS
When a homozygous individual with dark curly
hair crosses with a homozygous individual with light
\\r\ f7 / / s ...
x V //--/ /-" / .-
X \ v AA' / .'7 s
O = Light
B <= Curly
= Straight
FIG. 19. The heredity of human hair according to data by C. B.
and G. C. Davenport. The arcs represent the somatoplasms
of four individuals. Within the arcs are the gametes formed
by these individuals. The dominant character is placed on
the outside of the arc where it will be visible.
straight hair, all the offspring have dark curly hair.
The dark curly-haired individuals of this second
generation, however, are heterozygous with respect to
MENDELISM
113
each of these two hair characters. When any two
individuals having this particular genotypic composi-
tion mate, therefore, they may produce any one of four
possible phenotypes dark curly, dark straight, light
curly or light straight
haired individuals.
These four phenotypes
in turn will present nine
different genotypic
combinations out of
sixteen possible cases,
as shown in Figure 20.
Figure 19 further-
more serves to make
clear, first, the distinc-
tion between somato-
jplasm and germplasm ;
second, the maturation
of germ-cells ; third,
the segregation of
gametes ; and fourth,
jthe formation of zy-
gotes in sexual repro-
duction.
The cells of the so-
in each
class
Genotype
Phenotype
in each
class
4
(*^\
Dark curly
9
2
2
/^\
1
1
Dark straight
3
2
1
Light curly
3
2
(^\
1
Light straight
1
16
16
FIG. 20. Diagrams showing the
possible genotypic and pheno-
typic combinations resulting
when two heterozygous individu-
als with dark curly hair mate.
Symbols are the same as in
Figure 19.
matoplasm are represented as making up the arcs within
which are inclosed the germ-cells after their reduction
through maturation, which results in giving to each
germ-cell half the number of determiners that are
present in the somatic cells.
It will be remembered that when two gametes, or
114 GENETICS
mature germ-cells, unite, they form a zygote having the
proper number of determiners normal to the species
in question instead of double that number. Symbols for
dominant characters in the diagram are placed on the
outside of the somatic arcs, because these are the char-
acters that are visible or phenotypic, while the non-
apparent recessives are placed on the inside out of
sight.
11. THE CASE OF THE TRIHYBRID
Mendel went even further and computed the possi-
bilities which would result when two parents were
crossed differing from each other with respect to three
unit characters. He found that the results actually
obtained by breeding closely approximated the theo-
retical expectation.
This expectation in the case of a trihybrid cross is
that the cross-breds resulting will all exhibit the three
dominant characters, while their genotypic constitution
will include six factors, namely, these three dominant
characters plus their corresponding recessives or "ab-
sences."
Cross-breds of the first generation will, therefore,
have eight possible kinds of triple gametes and when
interbred may form a possible range of sixty-four
(8X8) different zygotes, which corresponds to a
monohybrid raised to the third power (3-f-l) 3 . These
sixty-four zygotes group together in eight different
phenotypes and twenty-seven different genotypes as
shown on page 116.
The trihybrid cross with its resulting combinations
MENDELISM
115
is well illustrated by Castle's work on guinea-pigs
which confirms the Mendelian hypothesis on an extensive
scale. In Figure 21 dominant characters are repre-
? t
rSP-*-
rsp-
RSP RsP RSp Rsp rSP rsP rSp rap
I * 1 1 1 t 1
RSP
RSP
RsP
RSP
RSp
RSP
Rsp
RSP
rSP
RSP
rsP
RSP
rSp
RSP
rsp
RSP
RSP
RsP
RsP
RsP
RSp
RsP
Rsp
RsP
rSP
RsP
rsP
RsP
rSp
RsP
rsp
RsP
RSP
RSp
RsP
RSp
RSp
RSp
Rsp
RSp
rSP
RSp
rsP
RSp
rSp
RSp
rsp
RSp
RSP
Rsp
RsP
Rsp
RSp
Rsp
Rsp
Rsp
rSP
Rsp
rsP
Rsp
rSp
Rsp
rsp
Rsp
RSP
rSP
RsP
rSP
RSp
rSP
Rsp
rSP
rSP
rSP
rsP
rSP
rSp
rSP
rsp
rSP
RSP
rsP
RsP
rsP
RSp
rsP
Rsp
rsP
rSP
rsP
rsP
rsP
rSp
rsP
rsp
rsP
RSP
rSp
RsP
rSp
RSp
rSp
Rsp
rSp
rSP
rSp
rsP
rSp
rSp
rSp
rsp
rSp
RSP
rsp
RsP
rsp
RSp
rsp
Rsp
rsp
rSP
rsp
rsP
rsp
rSp
rsp
rsp
rsp.
FIG. 21. Diagram showing the possible combinations in a guinea-
pig trihybrid of the F 2 generation. R, resetted coat; r, non-
rosetted coat (absence of R) ; S, short hair; s, angora hair
(absence of S) ; P, pigmented; p, albino (absence of pig-
ment). The eight possible triple gametes of each parent are
placed in the upper and left hand margins respectively.
Each of the sixty-four squares represents a possible zygote
or fertilized egg, having received a triple gamete from each
parent.
sented by capital letters, while recessives or absences
are indicated by corresponding small letters.
When a smooth, or non-rosetted (r), short-haired
(S), pigmented (P) guinea-pig is crossed with a
116
GENETICS
Number in
each class
GENOTYPE
PHENOTTPE
Number in
each class
1
SS PP RR
SPR
Short, pigmented, resetted
27
2
2
4
SS Pp RR
Ss PP RR
Ss Pp RR
2
SS PP Rr
4
SS Pp Rr
4
8
Ss PP Rr
Ss Pp Rr
1
SS pp RR
SpR
Short, albino, resetted
9
9
2
2
Ss pp RR
SS pp Rr
4
Ss pp Rr
1
ss PP RR
sPR
Angora, pigmented, resetted
2
ss Pp RR
2
ss PP Rr
4
ss Pp Rr
1
SS PP rr
SPr
Short, pigmented, non-rosetted
9
2
SS Pp rr
2
4
Ss PP rr
Ss Pp rr
1
ss pp RR
spR
Angora, albino, resetted
3
2
ss pp Rr
1
SS pp rr
Spr
Short, albino, non-rosetted
3
2
Ss pp rr
1
ss PP rr
sPr
Angora, pigmented, non-resetted
3
1
2
ss Pp rr
1
ss pp rr
spr
Angora, albino, non-rosetted
64
64
MENDELISM 117
resetted (#), long-haired (s), albino (p) guinea-pig,
all the offspring appear to be of one phenotypic
constitution, namely, resetted, short-haired, and pig-
mented (RSP). Their genotypic constitution is rep-
resented by the formula RrSsPp. These six factors
may form eight possible triple gametes, as follows :
RSP, RsP, RSp, Rsp, rSP, rSp, rsP, rsp. When two
germ-cells each made up of these eight triple gametes
unite in sexual reproduction, they will give rise to sixty-
four (8X8) possible zygotes as displayed in Figure 21.
An analysis of Figure 21 shows among the offspring
eight different phenotypes in the ratio of 27 :9 :9 :9 :3 :
3:3:1 and 27 different genotypes in the proportions
indicated on page 116. The order of the three pairs
of symbols is changed from that in Figure 21 to
emphasize the fact that with independent unit char-
acters the order is immaterial.
Sketches, drawn from photographs in Castle's "Ge-
netics and Eugenics," of the eight phenotypically differ-
ent guinea-pigs here described are shown in Figure 22.
12. SUMMARY
Three principles are concerned in Mendel's law:
/ independent unit characters, dominance, and segrega-
tion.
a. Independent Unit Characters. An organism,
although acting together as a physiological and mor-
phological whole, may be regarded from the point of
view of heredity as consisting of a large number of
independent heritable unit characters.
118
GENETICS
SPR
SpR
sPR
SPr
spR
Spr
sPr
spr
FIG. 22. The eight phenotypically different kinds of guinea-pigs
in the F a generation of a trihybrid. S = short hair ; s = long
hair or angora ; P = pigmented coat ; p = non-pigmented coat
or albino ; R = rough or resetted coat ; r = smooth coat.
Drawn from Castle's photographs by C. J. Fish.
MENDELISM
b. Dominance. In every individual there are two
determiners for every unit character, one derived from
each parent. If this pair is different, i. <?., if the zygote
is a heterozygote, one dominates the other and deter-
mines the apparent character of the organism.
The alternative recessive characters, although they
may be present in the germplasm, are unable to be-
come manifest in the somatoplasm so long as the domi-
nant characters are present. When, however, the domi-
nant determiner is absent, and the recessive is dupli-
cated, the recessive character becomes manifest.
c. Segregation. The determiners of unit characters,
although they may be intimately associated together in
the individual, during the complicated process of ma-
turation that always precedes the formation of a new
individual, separate or segregate out as if independent
of each other and thus are enabled to unite into new
combinations.
13. THE PRACTICAL APPLICATION
Although the ratios for more than a trihybrid were
computed by Mendel, the experimental test was not
carried out by him, since it involves such large and com-
plicated proportions.
In the case of four differing unit characters in the
parental generation, the offspring of the quadruple
hybrids derived from such an ancestry would include
256 or (3+1) 4 possibilities instead of 64 or (3+1) 3 ,
as in the case of trihybrids. When ten differing char-
acters are combined in the parental generation,
120 GENETICS
there would result over a million possible kinds of
offspring among the hybrids of the second filial
generation, (3+l) 10 =l,048,576.
From the foregoing it is apparent that in practical
breeding the only hope lies in dealing with not more
than one or two characters at a time. Since unit
characters usually behave independently of each other,
one may breed for a single character until it is segre-
gated out in a homozygous, that is pure, condition,
and then in the same way obtain a second character, a
third, and so on.
This is not difficult if the character sought is a
recessive for, in that case it is already homozygous or
pure and consequently appears. When a character is
dominant it takes longer to determine whether the
individual is heterozygous (hybrid) or homozygous
(pure).
14. CONCLUSION
The Mendelian method is an attempt to analyze the
behavior of a particular characteristic in heredity
rather than to get at the lump performance of the in-
dividual as a whole. Herein lies the scientific control of
heredity which the trinity of Mendelian principles
namely, independent unit characters, segregation, and
dominance, has placed in human hands. Following this
method there can be obtained in a few generations of
properly directed crosses, combinations of characters
united in one strain that formerly were never obtained
at all or were only hit upon by the merest chance at
long intervals.
CHAPTER VI
THE PURE LINE AND SELECTION
1. GALTON'S LAW OF REGRESSION
GALTON was one of the first * to attempt to express
mathematically the relationship between parents and
offspring by means_ojF__ir^atmg_^tatisjically a single
unit character. According to Galton, a mathematical
expression of the relationship between two generations
should serve as a corner-stone of heredity.
What Galton did was to take human stature as a
unit character in comparing 204 English parents and
their 928 adult offspring, because human stature is
not complicated by environmental influences and is
consequently, a purely hereditary matter.
The results of his measurements expressed in inches
are shown in Figure 23 in which the circles connected by
the diagonal line represent the graded parental heights,
while the arrowpoints indicate the average heights of
the offspring in each group.
This illustrates Galton's Law of Regression or the
tendency in successive generations toward mediocrity.
The law may be stated as follows :
Average parents tend to produce average children;
minus parents tend to produce minus children; plus
1 "Hereditary Genius," 1869.
121
122 GENETICS
parents tend to produce plus children ; but the progeny
of extreme parents, whether plus or minus, inherit the
parental peculiarities in a less marked degree than the
latter were manifested in the parents themselves.
FIG. 23. Scheme to illustrate Gallon's law of regression. The
circles represent graded groups of parental height while the
arrowpoints indicate the average heights attained by the re-
spective offspring. The offspring of undersized parents are
taller, and of oversized parents are shorter than their respec-
tive parents. Based on data from Galton.
2. THE IDEA OF THE PURE LINE
It was Galton's law of regression that suggested
to the Danish botanist Johannsen a possible means of
controlling heredity. In his mind arose the question
whether it would not be possible by continually breeding
THE PURE LINE AND SELECTION
from plus parents, granting that plus parents produce
plus offspring and making allowance for some regres-
sion to type, to shove over the offspring more and more
into the plus territory and so to establish a plus race.
To test this hypothesis, Johannsen selected beans,
Phaseolus, with which to experiment, since this group
of plants is self-fertilizing, prolific, and easily measur-
able. Somewhat to his surprise, the beans refused to
shove over as much as expected. That is, big beans
did not yield principally big offspring, nor little beans
little offspring, according to the expectation, although
they each produced offspring that varied in the manner
of fluctuating variability around an average unlike the
parental type. This gave Johannsen the idea that he
was using mixed material, so he next isolated the prog-
eny of single beans, which, being self -fertilized for many
generations, each constituted unmistakably a single
hereditary line. In this way nineteen beans, now fa-
mous, became the known ancestors of Johannsen's
original nineteen "pure lines," a further study of which
has led the way to some of the most brilliant biological
discoveries of recent years.
A pure line has been defined by Johannsen as "the
descendants from a single homozygous organism ex-
clusively propagating by self-fertilization," and more
briefly by Jennings as "all the progeny of a single
self-fertilized individual."
It should be pointed out, however, that this technical
idea of a "pure line" is not at all the same as that which
the breeder has in mind when he uses the same term.
The nearer individuals can be bred to conform to an
GENETICS
arbitrary standard agreed upon, the better they illus-
trate the stock-breeder's idea of a pure line. For ex-
ample, in "The Standard of Perfection," a book pub-
lished by the American Poultry Association, there are
recognized 42 breeds and 121 varieties of chickens.
To belong to any particular breed in this gallinaceous
Blue Book the chicken must look the part regardless of
its germinal derivation.
To the biologist, on the contrary, the pure line is
like an imaginary mathematical concept depending en-
tirely upon similarity of the determining hereditary
complex. The biologist's pure line is genotypic. The
stock-breeder's is phenotypic, a difference of definition
which has given rise to considerable confusion.
In a certain general way it will be seen that the pure
line stands over against mutation, since it is concerned
with the conservative maintenance of type while muta-
tion attempts to change it.
The inevitable monotony of a pure line may be con-
siderably masked by individual somatic modification.
DeVries has said paradoxically, "The pure line is com-
pletely constant and extremely variable." That is, it
is "completely constant" except for mutations, and
it is "extremely variable" in the somatic development
that may be attained by separate individuals.
3. JOHANNSEN'S NINETEEN BEANS
To return to experiments with beans, Johannsen
found out that the progeny of every one of his pure
lines varied around its own mean, which was different
THE PURE LINE AND SELECTION 185
in each of the nineteen instances. When, however, ex-
tremes from any pure line series were selected and bred
from, the results showed complete regression away from
the extreme condition of the parent bean back to the
Average of_
all progeny"
Weight of parent seed \
10203040606070 10 20 80 40 BO 60 70 10203040506070 10203040606070 10208040606070
Pure line number >U VU XV XVIII
FIG. 24. The result of selection in four pure lines of beans. The
vertical columns, representing the average progeny from
different sized parents all derived from the same pure line,
contain groups nearer alike than the horizontal columns, rep-
resenting progeny from the same sized parents, but different
pure lines. All the numbers indicate centigrams. Data from
Johannsen.
type of the entire pure line in question. That is,
selection within a pure line is absolutely without effect
in modifying a particular character in the offspring of
the line in question.
This is illustrated in Figure 24 in which the results
of selecting for size in the year 1902 is shown for four
126
GENETICS
pure lines only. The average for each pure line is
given at the top of its column. When, for example,
beans weighing 60 eg. were selected from pure lines II,
VII, and XV, the average weights of their progeny
were 56.5, 48.2, and 45.0 eg. respectively, which in
each instance is nearer to the average for the pure
line than to the weight of the parental seed.
It will be seen at once that the averages in the vertical
columns are nearer alike than the averages in the hori-
zontal columns. In other words, the beans bred true
to their pure line rather than to their fluctuating
parent.
As a further example of this law, take the result
of selection for six years in pure line I as shown in
the accompanying table and in Figure 25.
HARVEST YEAR
MEAN WEIGHT OF
SELECTED PARENT SEED
MEAN WEIGHT or
OFFSPRING
Minus
Plus
From Minus
Parent
From Plus
Parent
1902
1903
1904
1905
1906
1907
60
55
50
43
46
56
70
80
87
73
84
81
63.15
75.19
54.59
63.55
74.38
69.07
64.85
70.88
56.68
63.64
73.00
67.66
It is evident, for instance, that in 1907 the smallest
beans, weighing an average of 56 eg., gave an average
progeny weighing 69.07 eg., while the largest ones
for the same year, weighing an average of 81 eg., pro-
duced nearly the same average in their progeny as did
the smallest beans, that is, 67.66 eg.
THE PURE LINE AND SELECTION 127
Incidentally all the progeny from both large and
small parents averaged notably less in 1904 than all
the progeny from large and small parents in 1906,
a result due to a "poor year" when certain factors
of environment were unfavorable. Such unfavorable
conditions, however, are known to influence in no way
the hereditary qualities of the beans. Thus it appears
that, although the progeny of a pure line present
plenty of variations of the fluctuating type, due prob-
ably to environmental differences in nutrition, moisture,
etc., such variations are quite ineffectual so far as
inheritance is concerned, and it makes no difference
whether the largest or the smallest beans within a pure
line are selected from which to breed, the result will be
the same, in that there is a complete return to medioc-
rity or type with no "inheritance" of the parental modi-
fication. As a matter of fact in 1903, 1906 and 1907
the lighter parents gave heavier progeny than the
heavier parents.
It will be seen at once that here is a discovery of
far-reaching importance which may require us to
reconstruct certain cherished ideas about the part
played in the evolution of species, as well as in heredity,
by natural selection.
4. THE DISTINCTION BETWEEN A POPULATION AND
A PURE LINE
A mixture of pure lines has been called a population
(Johannsen).
It is not possible to distinguish by inspection a group
128
GENETICS
THE PURE LINE AND SELECTION 129
of individuals composing a pure line from a group
making up a population, since both may be phenotypi-
cally alike. Fluctuations about the average occur in
both cases with no appreciable difference in character,
although such fluctuations, when they occur within a
pure line, are simply somatic differences caused in
general probably by modifications in nutrition or some
other external factor of environment, while fluctua-
tions in a population include not only modifications of
this transient nature, but also permanent hereditary
differences due to germinal differences in the various
pure lines of which the population is composed.
Johannsen has made the distinction between pure
lines and populations clear by the following figure
(Fig. 26), in which five pure lines of beans are com-
bined artificially to form a population.
The beans which make up the pure lines noted in
this figure are represented inclosed within inverted
test tubes. The beans in any single tube are all of
one size. Tubes vertically superimposed upon each
other also contain only beans of one size.
Thus it is seen that what may be a rare size of
bean in one line, for instance that in the left-hand
tube of pure line 3, may be identical with the com-
monest size in another line, as pure line 2. The five
pure lines represented in Figure 26 are combined in
a population at the bottom of the figure. In this
population array the five pure lines are hidden.
Hence, while selection within a pure line has no he-
reditary influence, it is evident that selection within a
population may shift or move over the type of the
130
GENETICS
progeny obtained, in the direction of the selection simply
by isolating out a pure line of one type. Thus beans
Pure Line chosen from the
extreme 1 e f t-
hand test tube in
the population
cited would be-
long only to pure
line 2, while those
taken from the
extreme right-
hand test tube
could belong only
to pure line 3.
Galton's "law
of regression,"
namely, that
minus parents
give minus off-
spring and plus
parents plus off-
spring, with a
tendency to re-
version from gen-
eration to gener-
ation, depends
FIG. 26. Diagrams showing five pure lines s i m ply upon a
and a population formed by their union. r t j r
The beans of each pure line are repre- partial but not
sented as assorted into inverted test tubes . i
making a curve of fluctuating variability. <
Test tubes containing beans of the same tion of pure lines
weight are placed in the same vertical f ,
row. After Johannsen. out of a popula-
tion.
THE PURE LINE AND SELECTION 181
From this distinction between pure lines and popula-
tions it is clear why breeders in selecting for a particu-
lar character out of their stock need to keep on select-
ing continually in order to maintain a certain standard.
As soon as they cease this vigilance, there is a "reversion
to type" or, as they say, "the strain runs out," which
means that the pure lines become lost in the mixed popu-
lation which inevitably results as soon as selective iso-
lation of the pure line ceases.
Such reversion must always be the case in dealing
with a population made up of a mixture of pure lines,
for only by the isolation of pure lines can the constancy
of a character be maintained. When, however, a pure
line is once isolated, then all the members of it, large
as well as small, are equally efficient in maintaining the
pure line in question, regardless of their phenotypical
constitutions.
Conceding that natural history and common usage as
well as the older theories of heredity are concerned with
phenotypic constitution of organisms, we are now
coming to see more clearly than before that heredity
must always be a case of similarity in origin, that is,
in germinal composition, and that similarity in appear-
ance by no means always indicates similarity in origin
or true relationship.
The assumption that similarity in appearance does
indicate relationship has been made the foundation of
many conclusions in comparative anatomy and phy-
logeny, but to the modern student of genetics who places
his faith in things as they are, rather than in things as
they seem to be, conclusions based upon phenotypical
GENETICS
distinctions alone have in them a large source of error
which must be taken into account.
In a museum of heredity, should such a collection
ever be assembled, the specimens would not be arranged
phenotypically as they are in an ordinary museum
where things that look alike are placed together as if
in bonds of relationship, but they would be arranged
historically from a genetic point of view to show their
true origin one from another.
5. CASES SIMILAR TO JOHANNSEN'S PURE LINES
Although, according to Johannsen, pure lines are
"the progeny of a single self-fertilized individual," it
is plain that in at least three other possible cases some-
thing quite similar to "pure lines" may be obtained.
These are clones, partheno genetic progeny and homo-
zygous crosses. "In principle pure lines, partheno-
genetic reproduction and vegetative propagation are
concerned with nearly the same situation" (Morgan).
First, in asexual reproduction where the progeny
are simply the result of continued fission of the original
individual, a pure line may be said to continue from
generation to generation because it is a germinally un-
changed sequence of individuals. Such an asexual
progeny is termed a clone (Webber). Shull's definition
of a clone is "a group of individuals of like genotypic
constitution, traceable through asexual reproductions
to a single ancestral zygote, or else perpetually
asexual."
Second, in cases of parthenogenesis, the progeny
THE PURE LINE AND SELECTION 133
arising from a single female individual without the
customary maturation of the germ-cells which accom-
panies sexual reproduction, constitute a pure line or an
unmixed strain because as in clones there has been no
segregation nor addition of outside germplasm.
Third, in homozygous crosses when two organisms
identical in their germinal determiners inbreed, their
progeny will form a pure line just as truly as two
parents that are united in a single hermaphroditic in-
dividual produce a pure line progeny as the result of
self-fertilization.
In the case of clones and parthenogenesis it should be
pointed out that the "pure line" is assured only so
long as asexual reproduction continues. It is quite
possible for an organism, even heterozygotic in composi-
tion, to continue to breed true or to produce an ap-
parently pure line so long as asexual methods are em-
ployed. As soon as such an organism, however, changes
to the sexual method of reproduction, segregation of
characters may occur and different combinations result.
A pure line, therefore, implies freedom from admixture
of different germplasm rather than any necessary
equality or likeness of individuals.
The different kinds of "pure lines" are diagram-
matically represented in Figure 27.
6. SELECTION WITHIN A PURE LINE
The basic idea of the pure line concept is that every
member of any pure line is genetically identical with
every other member of the same fraternity, therefore,
134
GENETICS
THE PURE LINE AND SELECTION 155
any differences found between individuals of a pure line
are entirely somatic and not hereditary.
The importance of the problem of pure line selection
for any general consideration of the mechanism of evo-
lution is at once apparent. There have been many re-
cent investigations besides those of Johannsen to test
the result of selection within the four kinds of "pure
lines." Some of these investigations are enumerated in
the table on pages 136 and 137.
It is apparent in the first section of the following
table that the pure line sensu stricto, that is, the pure
line of Johannsen, must be studied with plants alone,
since among animals only certain highly specialized
parasitic worms, which do not lend themselves readily to
selection experiments, produce offspring by means of
self-fertilization. The work of the other authors
upon plants, mentioned in the first group of the table, is
in entire agreement with the work of Johannsen.
The noteworthy contribution of L. de Vilmorin con-
sists in a detailed comparison of preserved specimens
of certain pure lines of wheat which were isolated in
France about 1840, with their lineal descendants of
to-day. In spite of continuous selection for better-
ment within these self-fertilized strains during more than
60 years, their constancy has been maintained.
B. CLONES
With respect to selection within a clone there is an
apparent conflict of results.
136 GENETICS
THE RESULTS OF SELECTION WITHIN A PURE LINE
Kind of
pure line
Author
Organism
Character
selected
Result
Johannsen, '03
Beans
Size
No effect
K
Barley
Mealiness of
kernel
Progeny
Nilsson
Wheat
Various
of a
Oats
characters
c<
single
Barley
self-
Surface and
fertil-
Pearl
Oats
Yield per acre
ized
Fruwirth, '17
Lentils
indi-
Peas
it
vidual
Soy beans
Lupines
L. de Vilmorin
Wheat
Wolf, '09
Bacteria
Pigment pro-
duction
Barber, '07
Wiuslow and
Walker, '09
it it
Meader, '19
Form, fer-
mentative
reaction,
virulence
East, '09-'10
Potato
11 it
Vogler,'14
Garlic
(( tl
Stout, '15
Coleus
Color pattern
Effective
Mendiola, '19
Lemna
Size and shape
of frond,
speed of
budding
No effect
Jennings, '08
" '16
Paramecium
Difflugia
Size
Six shell
{( U
Clones
characters
Effective
Calkins and
Gregory, '13
Paramecium
Size, rate of
fission, etc.
{<
Jollos, '13
Resistance to
arsenical
poisoning
No effect
Stocking, '15
Abnormalities
Effective in
some lines
Middleton, '15
Stylonychia
Fission-rate
Diverse strain
from one
THE PURE LINE AND SELECTION 137
THE RESULTS OF SELECTION WITHIN A PURE LINE Continued
Kind of
Pure line
Author
Organism
Character
selected
Result
Ackert, '16
Paramecium
Size
No effect
Root, '18
Centropyxis
Shell charac-
ters, fission-
rate
Effective
Hegner, '19
Arcella
Shell charac-
Diverse strains
ters
from one
Hanel, '08
Hydra
No.of tentacles
No effect
Lashley, '16
w
U It "
<
Woltereck, '09
Hyalodaph-
Length and
Temporary
nia
shape of
temperature
"head"
effect
Agar, '13
Simoceph-
No effect
alus
" '14
Aphids
t<
Partheno-
genetic
Ewing, '14
Length of
honeydew
prog-
tubes, an-
eny
tennae and
body
Kelly, '13
(
Length of an-
tennal j oints
a
Banta
Daphnids
Light reac-
Effective in
tions
one line
" '19
Simoceph-
Sex inter-
"Somewhat
alus
grades
effective"
Smith
Maize
Oil and pro-
tein content
Effective
Tower, '06
Potato
beetle
Pigmentation
No effect
May, '17
Zeleny,'20
Drosophila
Bar-eye
<
M
Ineffective
after 3 to 5
generations
McDowell, '15
M
Thoracic
bristles
Effective
Homo-
Reeves, '16
M
Thoracic
zygous
bristles
Crosses
Payne, '20
H
Thoracic
Effective for
bristles
several gen-
erations
Sturtevant,
M
Dichaet
'18
bristles
Effective
Pearl
Hen
Fecundity
No effect
Castle and
Phillips
Hooded rat
Coat pattern
Effective
138
GENETICS
In bacteria it is possible to isolate out variants
from a single strain but in none of the cases is the
origin of the supposed "clone" unquestionably from a
smgle bacterium as it would need to be in order to
form a pure line, so that what has occurred in all prob-
ability is the simple assortment of a pure line from a
population.
Among the protozoa, which reproduce asexually by
fission, much painstaking experiment and observation
95
FIG. 28. Eight pure races of Paramecium. The actual mean
length of each race is given in micra below the corresponding
outline. Magnified about 230 diameters. After Jennings.
has been made, notably by Jennings and various investi-
gators whom he has inspired.
For example, Jennings found that Paramecia differ
from each other in size, structure, physical character,
and rate of multiplication as well as in the environ-
mental conditions required for their existence and,
furthermore, that these differences, in an hereditary
sense, are "as rigid as iron."
With respect to the character of mean length he
THE PURE LINE AND SELECTION 139
was able to isolate eight races, or pure lines, whose
average size, drawn to scale, is shown in Figure 28.
Each of these pure lines produced a progeny which
exhibited a considerable range of fluctuating variation.
The offspring of pure line D, for example, varied from
256 to 80 micra x in length with an average of 176
micra, as shown in Figure 29, where samples of the
different classes of variants in pure line D are arranged
in a series.
256 <
Micra
- 80
FIG. 29. Diagram of a single race (D) showing the variation in
the size of the individuals. Magnified about 230 diameters.
After Jennings.
A single representative of each of the different classes
of variants out of all the eight pure lines bred by
Jennings is shown in Figure 30.
Each horizontal row represents a single race or
pure .line, the average size of which is indicated by the
sign +. The mean length of the entire lot, as shown
by the vertical line, is 155 micra. The total number
of individuals belonging to each size is not indicated, but
1 A micron is 1-1 000th of a millimeter.
140
GENETICS
in every horizontal line their number is more numerous
near the average for that line and less numerous at
155
210
310
FIG. 30. Diagram of the species Paramecium as made up of the
eight different races shown in Figure 28. Each horizontal
row represents a single race. The individual showing the
mean size in each race is indicated by a cross placed above it.
The mean for the entire lot is at the vertical line. The
magnification is about 24 diameters. After Jennings.
the extremes, thus forming the typical normal fre-
quency curves of fluctuating variability.
The significant fact about these series is that ex-
THE PURE LINE AND SELECTION 141
treme individuals selected from any pure line do not re-
produce extreme sizes like themselves, but instead, a
progeny varying according to the laws of chance
around the average standard of the particular line
from which it came. Thus quite independently of
Johannsen, Jennings arrived at the same general con-
clusion, namely, that selection within a pure line is with-
out effect.
But with Difflugia, another protozoan that secretes
for itself a jug-like shell, Jennings, after a characteris-
tically careful and prolonged study, has a different
story to tell. Difflugia proved to be a more favorable
form to study than Paramecium because it has numer-
ous distinctive shell characters which are all inheritable
to a high degree but are unchanged by growth and en-
vironment during the life of the individual, although
presenting variations from parent to offspring.
Jennings selected for (1) the number of spines on
the shell ; (2) the length of the spines ; (3) the diameter
of the shell; (4) the depth of the shell; (5) the number
of teeth surrounding the mouth; (6) the diameter of
the mouth. In two families, "one (#303) including
495 descendants of a single individual, and the other
(#314) including 1049 descendants of the original
parent, selection was effectire."
C. PARTHENOGENETIC PROGENY
Parthenogenetic animals furnish even better material
than unisexual clonal animals for testing the effective-
ness of selection in an unmixed line but here again the
142 GENETICS
conclusions of the investigators are not in entire har-
mony. There is no doubt that in most cases selection
within a parthenogenetic line is futile although Banta's
long continued observations upon daphnids seem to fur-
nish evidence of an opposite kind. Particular weight
should be given to this work because it presents one of
the longest pure lines that ever passed under the seeing
eye of a scientist. In some of his lines there have been
450 generations (1921) forming an unbroken line
extending over 10 years' time. If this pedigree were
translated into human generations of 30 years each it
would make a period of 13,500 years and would run
back over 100 centuries B. C. long before the very
beginnings of human history. There is no doubt that
many experiments in selection cannot be considered
decisive because they concern altogether too few gen-
erations, as compared with the time that has been at the
disposal of nature in accomplishing evolutionary
change.
D. HOMOZYGOUS CEOSSES
It is very difficult to find instances among animals
and plants where two individuals are homozygous in all
particulars. The nearest approach is "identical twins"
which arise from a single fertilized egg and consequently
are more nearly germinally alike, and can never cross since
they are always of the same sex.
It is useful, nevertheless, to consider pure lines result-
ing from homozygous crosses when limited to a single
character rather than to individuals, for of this con-
dition there are numberless instances.
THE PURE LINE AND SELECTION 148
a. Tower's Potato-Beetles
As an illustration of the effect of selection within
pure lines may be mentioned Tower's exhaustive experi-
ments on the Colorado potato-beetle Leptinotarsa
decemlineata. These beetles had been inbred for such an
extended period that they were presumably homozygous
for the character of color. Among the numerous cul-
tures which were under control, a considerable varia-
tion in color, nevertheless, made its appearance. For
convenience in classification these variations were
graded into arbitrary classes or graduated variants
ranging from dark to light.
When a male and a female from the extreme class at
the dark end of the series were allowed to breed together,
their progeny were not dark, but fluctuated in color
around the original average of the entire series. The
process of selecting each time an extreme pair of dark
parents was continued for twelve generations, as shown
in Figure 31, without in any way increasing the per-
centage of brunette potato-beetles in the progeny.
Thus in a pure line formed by the breeding of two
individuals, alike with respect to color, the selection of
an extreme variant was quite without effect in modifying
the color of the progeny.
b. Drosophila Bristles
Among the "hairs" on the scutellum of Drosophila
melanogaster there are four larger hairs or bristles,
as shown in Figure 32.
xn
XI
IX
vin
VII
VI
V
IV
II
FIG. 31. Diagram showing the ineffectiveness of selection through
twelve generations within a homozygous strain in the case of
the Colorado potato-beetle (Leptinotarsa). In each genera-
tion extremely dark specimens were selected as the parents of
the succeeding generation but the progeny always swung back
to the type. After Tower.
144
THE PURE LINE AND SELECTION 145
These four bristles are ordinarily strictly accounted
for in heredity but the occasional variation in their
number led MacDowell, and later others, to attempt to
establish by selection a new style in these bristly
decorations consisting of either extra
or fewer bristles. Apparent success
was the result in effective selection
among the offspring of parents
homozygous for the single character
of four bristles.
c. Pearl's Egg-laying Hens
In an experiment extending over
17 years and which involved nearly
5000 pedigreed birds, Pearl tried,
within a homozygous strain, to select biological Cin-
a hen that would produce 00 eggs Drawn from
annually instead of the ordinary Bridges Fi ^ C ' J *
number of 125, but without success.
d. Castle's Hooded Rats
Finally one of the most famous selection experiments
on record is that of extent of pigmentation, plus and
minus, in the hooded rat. This experiment involved
breeding an average of nearly twelve rats a day without
cessation for eight years and it has not only made the
Pied Piper of Hamelin roll over in his grave but has
kept biologists busy with explanations of the results,
for, like the four bristles on Drosophila's back, it ap-
146 GENETICS
parently furnishes evidence of modifications of an he-
reditary characteristic through selection following a
homozygous cross.
Castle succeeded in selecting two extreme races of
rats from his hooded stock, one possessing almost no
pigment and the other with the "hood" so extended
that it covered practically the entire body.
Is then the germ a variable thing that makes it possi-
ble to select effective differences out of a pure line, to
the discomfiture of Mendelians who build their house
on the rock of constancy of the germplasm, or can
these perplexing results be somehow explained?
7. CONCLUSION
At any rate it would be gratifying scientifically to
discover one fundamental law to which all these various
cases of pure line selection are accountable because in-
tellectual satisfaction always follows upon finding the
common denominator of things.
A unifying explanation that makes a single harmoni-
ous interpretation of these apparently diverse results,
based on the idea that all are reducible to Johannsen's
conception of the ineffectiveness of selection within a
pure line, has perhaps been reached in the theory of
modifying genes which will be considered in the next
chapter.
Certainly the pure line concept is a very useful tool
for the geneticist since with it the hereditary upset of
outside germplasm is eliminated. Consequently it is
THE PURE LINE AND SELECTION 147
of the utmost importance to know what can be done with
this tool. In any event the way of experimentation and
observation still lies open and what remains undiscov-
ered makes life worth living.
CHAPTER VII
THE FACTOR HYPOTHESIS
1. THE HEREDITARY UNIT
IN reducing any body of facts to a science, it is
first necessary to determine the underlying units out
of which the facts are made up.
Chemistry was alchemy until the chemical elements
were identified and isolated. Histology was terra
obscura until the cell theory brought forward "cells"
as the units of tissues. In the same way there could
be no science of genetics until the conception was de-
veloped that the individual is a bundle of unit char-
acters rather than a unit in itself. So it has come
about that geneticists speak of inheritance as applied
to unit characters rather than to individuals as a whole.
The apparent somatic unit characters, like the
color of the seed-coat or the length of the vine in
Mendel's peas, are conditioned by other intangible but
nevertheless real germinal units or determiners which
give rise to them. Mendel was apparently unaware of
the existence, in certain cases at least, of compound de-
terminers. His experiments led him to believe that
each character depends upon only a single determiner
for the reason that he worked on characters severally
belonging to different parts of the plant, but it has
148
j
:
THE FACTOR HYPOTHESIS 149
been ascertained within the last decade that some char-
acters require more than a single germinal determiner
to bring them to somatic expression. The converse
is also true, namely, that certain single determiners
may control more than one character. For instance,
the determiner for gray hair in rats also produces a
lighter color on the belly.
The idea of compound germinal determiners for a
single character has been termed the factor hypothesis
of heredity.
Hereditary germinal factors, that may sometimes
need to combine in order to produce a visible somatic
unit character, are known as genes ( Johannsen).
2. DIFFERENT KINDS OF GENES
There are various kinds of genes that bring about the
visible expression of unit characters in various ways.
An attempt to tabulate the kinds of genes is herewith
given.
SINGLE
Alternative
Allelomorphic
Presence or absence
PLURAL
Cumulative
Modifying
Complementary
Supplementary
Lethal.
When genes are derived from two parents, as in all
cases of sexual reproduction, they are always in pairs,
150 GENETICS
that is, one from each parent, and in the production of
a unit character they may act in single or in several
pairs.
If a single pair, the genes may be interpreted accord-
ing to either the allelomorphic or the presence-or-
absence hypothesis. In the first instance it is either
one thing or an alternative that produces the charac-
ter. For example, as in the case of the pea-vine, it is
either tallness or dwarfness. In the second instance,
the determiner of the character either is present or
it is not, and the resulting unit character is dependent
upon which of these two possibilities obtains. That is,
applied to the illustration just given, if the hereditary
factor or gene for tallness is present the pea-vine will
be tall but if there is no gene for tallness the plant
will be a dwarf. This condition is expressed by the term f
alternative genes and the operation of alternative
genes follows in the typical Mendelian fashion described
in Chapter V.
Under various kinds of plural determiners which in-
volve more than one pair of genes, cumulative genes
are those that are all alike in their separate effects
but which, acting together, alter the degree of expres-
sion that is given to the unit character. These will be
more fully described in Chapter VIII upon "Blending '
Inheritance."
Modifying genes are those germinal factors that are
without effect alone but which in conjunction with other
factors produce an alteration of those factors. They
may be (1) Complementary, when a factor is added
to a dissimilar factor in order that a particular charac-
THE FACTOR HYPOTHESIS 151
ter may appear; (2) Supplementary, when a factor is
added to a dissimilar factor already effective, with the
result that a character is modified or changed in some
way; (3) Lethal, so-called since they "cause the early
death of those gametes or zygotes in which such a fac-
tor is not balanced by a normal one" (Conklin).
It will be profitable to consider a few illustrations
of the factor hypothesis in some detail since it helps to
explain both the reappearance of old types and the for-
mation of new ones.
3. COMPLEMENTARY GENES
In the course of numerous breeding experiments
Bateson obtained two strains of white sweet peas,
Lathyrus, which, when normally self-fertilized, each
bred true to the white color. When these two strains
were artificially crossed, however, the progeny all had
purple flowers like the wild ancestral Sicilian type of
all cultivated varieties of sweet peas.
Here was apparently a typical instance of "rever-
sion" which would have delighted Darwin's heart, but
according to the factor hypothesis the true explanation
is this. The character of purple color is dependent
upon two independent genes which, though separately
heritable, are both required to produce it. Each of
these white strains of sweet peas possesses one of these
genes which can produce colored flowers only when
united with its complement, a proof of which appeared
upon interbreeding hybrid purples from such a cross.
In short, the color purple depends upon the action of
152
GENETICS
two complementary genes that follow the behavior of a
dihybrid. (See Chap. V, par. 10.)
The gametic formulae for the two strains of white
sweet peas used in this experiment are Cp and cP,
CP
cP
ep
11
12
16
A$
J
FIG. 33. Diagram to illustrate the possible progeny from two
heterozygous purple sweet peas according to data from
Bateson. C, color gene (large circles) ; c, absence of C
(small circles) ; P, pigment gene (large crosses) ; p, absence
of P (small crosses). In the zygotes within the checkerboard
squares the gametic symbols are superimposed.
respectively. C stands for a color gene without which
no color can appear, and c is the absence of this fac-
tor, while P represents a purple pigment gene which
finds expression in the somatoplasm only when taken
together with the color gene C. The small letter p
THE FACTOR HYPOTHESIS 153
stands for the absence of the purple pigment gene. It
will be seen that each of the white sweet peas the for-
mulse of which are given above lack one of the two
essential factors for purple color. When the two are
crossed, however, all the progeny are purple with the
formula CcPp.
These hybrid sweet peas upon gametic segregation
theoretically produce four kinds of gametes, CP, Cp,
cP, and cp which may combine as any other dihybrid
in sixteen different ways. In this case, however, these
combinations group themselves into only two pheno-
types, purple and white, as indicated in the accom-
panying diagram (Fig. 33) in which C and c are repre-
sented by large and small circles respectively, while
P and p are correspondingly indicated by large and
small crosses. The gametic symbols are superimposed
to form the zygotes.
The theoretical expectation here shown was closely
approximated in the actual results.
It may be noted in passing that the seven kinds of
white sweet peas resulting from the above cross, while
phenotypically alike, that is, in the zygotic symbols
of Figure 33, lacking either the large circle (color) or
the large cross (pigment), belong to three distinct groups
of genotypes as follows :
NUMBER OF
ZYOOTE IN
FIGURE 33
1
2
Without the pigment gene (large cross)
Without the color gene (large circle)
6 . 8 . 14
11 12 - 15
3
Without either pigment (large cross) or color
(large circle)
16
154
GENETICS
Among the purple flowers are the following four
genotypes :
1
2
3
4
NUMBER OP
ZYGOTE IN
FIGURE 33
Duplex for both color (large circle) and pig-
ment (large cross)
Duplex for color (large circle) but simplex for
pigment (large cross)
Simplex for color (large circle) but duplex for
pigment (large cross)
Simplex for both color (large circle) and pig-
ment (large cross)
1
2:3
3 *9
4 7 ; 10 13
4. SUPPLEMENTARY GENES
A. CASTLE'S AGOUTI GUINEA-PIGS
An illustration of a supplementary gene that acts
only in conjunction with some other to bring about a
modification, is the pattern gene demonstrated by
Castle in his guinea-pigs.
The wild gray, or "agouti," color of the hair of
certain guinea-pigs is due to the fact that pigment is
distributed along the length of each hair in a definite
pattern. The tip of a single hair is black followed
by a band of yellow, while most of the proximal part
which is more or less concealed by overlapping hairs
is a leaden color. The distribution of pigment in such
a pattern gives the characteristic gray, or agouti
color to the coat when taken as a whole.
Castle demonstrated the separate nature and be-
havior of such a pattern gene when he discovered that
THE FACTOR HYPOTHESIS 155
it is transmitted independently of pigment, which is
necessary to bring it to expression. He showed that
upon crossing a solid black guinea-pig, unquestionably
possessing pigment but no "pattern," with a white
albino guinea-pig having no pigment, some of the off-
spring "reverted" to the ancestral agouti, or "pat-
tern" type, thus proving that the pattern must be
carried in this case by the white or albino guinea-pig
as a factor independent of the color which is necessary
for its expression.
Another instance of the interaction of supplemen-
tary genes is seen in the spotting of piebald mice.
Cuenot discovered that such spotting is due to the
absence of a uniformity gene which if present causes
color to be uniformly distributed over the entire coat.
Both of these independent genes, spotting and uni-
formity, are real and not imaginary, since they may
be separately transmitted through albino animals in
the same way as the pattern gene mentioned above,
notwithstanding that in albinos both are hidden
through the absence of pigment, upon the presence of
which their visibility depends.
Whenever piebald or spotted animals appear in a
progeny derived originally from self-colored stock, it
is evidently due to the absence of such a "uniformity"
gene as has just been described.
Galton's theory of "particulate inheritance" (page
94) is now satisfactorily explained as true alterna-
156 GENETICS
tive inheritance in which the mosaic appearance is
caused by a Mendelian determiner, in this instance a
spotting gene or, in other words, the absence of a gene
for uniformity.
Miss Durham, in her work with mice, has demon-
strated an intensifying gene, the absence of which she
calls a diluting gene. The action of the former pro-
duces, as its name implies, intensity of color, while that
of the latter serves to lessen the degree of intensity in
which color appears.
These genes of intensity and diluteness, it should
be observed, do not in any way correspond to the
duplex and simplex condition of a dominant color
character, either of which would straightway appear
if crossed with an albino. The factors of intensity
and dilution of color are of an entirely different na-
ture, as they have been proven to be independently
transmissible through albinos where a color character
could not appear because of the absence of pigment.
The following illustration of this kind of supple-
mentary genes taken from Miss Durham's experiments
will serve to make the case clear. The symbols em-
ployed are:
B = black pigment which masks brown, or chocolate.
6 = the absence of B, consequently chocolate.
I = intensity gene.
i = dilution gene or absence of intensity.
C a complementary color gene acting with P.
p = a complementary pigment gene acting with (7.
BICP = black.
THE FACTOR HYPOTHESIS
157
BiCP blue or maltese (dilute black).
bICP = chocolate.
biCP = silver-fawn (dilute chocolate).
The results of crossing the hybrids formed from the
combinations indicated at the left in the table below are
shown at the right where the expectation is given in
parentheses after the actual results.
BLACK
(BICP)
BLUE
(BiCP)
CHOCO-
LATE
(bICP)
SILVER-
FAWN
(MCP)
Black (BICP) X Silver-fawn (MOP)
Blue (BiCP) X Chocolate (6/OP)..
Blue (BiCP) X Silver-fawn (biCP) .
9(9)
42(45)
0(0)
4(3)
16(15)
33(36)
3(3)
14(15)
0(0)
2(1)
8(5)
12(12)
It will be seen that the actual results, even when
such small totals are concerned, approximate very
closely the expectation and are entirely consistent.
D. CASTLE'S BROWN-EYED YELLOW GUINEA-PIGS
Furthermore Castle has shown that in guinea-pigs
there is an independent gene for extension of pigment
distinct from the uniformity gene already mentioned.
The absence of this extension gene ("restriction") is
manifested by a lack of black or brown pigment every-
where except in the eyes and to a slight extent in the
skin of the extremities, while the distribution of yellow
is wholly unaffected by it.
That such "extension" and "restriction" genes
really exist, is proven in the following way:
When a brown (chocolate) guinea-pig is crossed
with an ordinary black-eyed yellow one, the young are
all black pigmented, but by cross-breeding these hybrid
158
GENETICS
young four varieties are obtained in the next genera-
tion, viz., black, brown, black-eyed yellow, and brown-
10
bE \Be
Black
14
'be \ I Be
'. Yellow
IbE
BE \\bE
Black
BellbE
Black
11
bE \\bE
Chocolate
15
be I bE
Chocolate
|6e
Be \\be
Black-eyed Yellow
12
bE\\b*
Chocolate
16
be\ \be
3rown-eyed Yellow
FIG. 34. Diagram to illustrate the origin of a brown-eyed yellow
guinea-pig from two heterozygous black parents, based upon
Castle's experiments. The gene for yellow (Y) is present in
every gamete and is consequently duplex in every zygote but
is hidden whenever the gene B is present. B, black pigment
hiding brown or chocolate; b, chocolate (absence of B) ; E,
extension of B over the entire body hiding Y; e, restriction
of B to eyes alone thus exposing Y over the entire body.
eyed yellow, the latter a variety unknown before
Castle's experiment in breeding was made.
For the sake of clearness the formation of the
brown-eyed yellow is shown above in Figure 34.
THE FACTOR HYPOTHESIS 159
Symbols
B =* black pigment, hiding brown or chocolate.
b = absence of B, or chocolate.
Y = yellow pigment, hidden by B.
E = extension of B over entire body, hiding Y.
e = restriction of B to eyes alone, thus exposing Y over the
entire body.
O = complementary color gene acting with P to produce color.
P = complementary pigment gene acting with C to produce color.
(The genes C and P may be omitted for the sake of sim-
plicity, since they are present in each instance.)
First Cross
"Extended" chocolate (bEY) X black-eyed yellow
(BeY) = black (BbEeYY).
Second Cross
When these cross-breds are mated with each other,
they each form four kinds of gametes, BEY, BeY,
bEY, and beY, which unite into sixteen theoretical
genotypic possibilities, some of which are unlike (Fig. 34).
These fall into four phenotypes, nine black (BEY), three
black-eyed yellow (BeY), three chocolate (bEY), and
one brown-eyed yellow (beY). The actual results in
Castle's experiments gave all four kinds in close nu-
merical agreement with this expectation. The action
of extension and restriction genes is, therefore, plainly
a case of Mendelian dihybridism in which two inde-
pendent pairs of alternative characters are concerned.
E. BABBIT PHENOTYPES
Perhaps no better application of the factor hy-
pothesis, so far as supplementary genes are concerned,
may be found than in the case of the color of rabbits.
160
GENETICS
There are many varieties of rabbits with respect to
color, particularly among domesticated races. These
varieties are now quite explainable by the factor hy-
pothesis, as indicated in the table below. The sixteen
kinds of rabbits there catalogued have been obtained by
THE FACTOR HYPOTHESIS APPLIED TO COLORS or RABBITS
CONSTANT
GENES
ALTERNATIVE
GENES
GAMETIC
FORMULA
PHENOTYPIC CHARAC-
TER WHEN CROSSED
WITH THE SAME KIND
OP GAMETIC
COMBINATION
1
2
3
4
5
6
7
8
Br
B
Y
E
I
U
A
a
AUIEC[YBBr]
Gray
UIEC [YBBr]
Black
u
A
duIEC [YBBr]
Gray spotted
a
cwIEC [YBBr
Black spotted
i
I
U
A
AUiEC[YBBr
31ue-gray
a
aUiEC [YBBr
Blue (Maltese)
u
A
a
AuiEC [YBBr
Blue-gray spotted
auiEC [YBBr
Blue spotted
e
U
A
a
AUIeC [YBBr
JYellow (with white
[ belly and tail)
aUIeC [YBBr
f Sooty yellow (with
yellow belly and
tail)
u
A
AuleC [YBBr
Yellow spotted
a
auleC [YBBr
Sooty yellow spotted
i
U
A
AUieC [YBBr
Cream
a
aUieO [YBBr
Pale sooty yellow
u
A
a
AuieC [YBBr
Cream spotted
auieC [ YBBr
Pale sooty yellow
spotted
THE FACTOR HYPOTHESIS 161
Castle and other experimental breeders, as well as many
of the albino types that would double this list if c, or
the gene for absence of color, should be substituted for
C, the presence of color, in column 4 of the table on
page 160.
Explanation of Symbols in the Foregoing Table
Br = a gene acting on C to produce brown pigmentation.
B = a gene acting on C to produce black pigmentation.
Y = a gene acting on C to produce yellow pigmentation.
The three genes, Y, B, Br, are present in every rabbit
gamete and up to date have not been separable as inde-
pendent unit characters, although they ha ve been sepa-
rated out in guinea-pigs and mice. There are no brown
rabbits, because black always goes linked with brown
covering the brown factor. Yellow rabbits result, as
explained below, through the action of factor e.
C = a common color gene necessary for the production of any
pigment. It was discovered in 1903 by Cu6not.
c = the absence of C which results in albinos, regardless of
whaterer pigment gene may be present. By changing C
to c, sixteen kinds of albinos would be added to this
catalogue, an addition of one phenotype and sixteen
genotypes, all looking alike but breeding differently.
E = a gene governing the extension of black and brown pig-
ment, but not of yellow.
e = the absence of extension or restriction of black and brown
pigment to the eyes and the skin of the extremities only,
while yellow remains extended and visible. Demonstrated
by Castle in 1909.
7 = an intensity gene which determines the degree of pigmenta-
tion. It can be transmitted independently of C through
an albino. Discovered by Bateson and Durham in 1906.
i = the absence of intensity or dilution. Dilute black = blue.
Dilute yellow = cream. Dilute gray = blue-gray.
U = a gene for uniformity of pigmentation or "self-color" dis-
covered by Cuenot in 1904.
u = the absence of uniformity which results in spotting with
white.
A = a pattern gene for agouti, or wild gray color, which causes
the brown and black pigments to be excluded from cer-
tain portions of each hair, resulting in the gray coat.
When present in the rabbit, it is also associated with
white or lighter color on the under surfaces of the tail
and belly. It was demonstrated by Castle in 1907.
a = the absence of the agouti or pattern gene.
162 GENETICS
F. THE KINDS OF GRAY RABBITS
Each of the apparent kinds of gray rabbits indicated
in the foregoing table may be made up of various geno-
types. For instance, there are thirty-two different
genotypes, each of which is pheno typically a gray
rabbit. The zygotic formula for each of these thirty-
two possibilities is displayed in the next table, and it
will be seen that these range all the way from rabbits
homozygous in all their variable characters (No. 1)
to those homozygous in none (No. 32).
The progeny of these various types of gray rabbits
when inbred will consequently vary from the pure gray,
as in No. 1, to a gray from which sixteen possible
types of young may be expected as in No. 32.
Up to the time when Castle's paper upon the factor
hypothesis 1 was published in 1909, nine genotypic
kinds of gray rabbits had been obtained in his experi-
ments, whose genotypic formulae correspond to the
following numbers in the list: 1, 3, 6, 10, 13, 20, 22,
28, 29.
5. LETHAL GENES
Among mammals, as shown by Cuenot and confirmed
by Little, homozygous or pure yellow mice are un-
known although yellow individuals have long been ex-
ploited by fanciers. In other words, all kinds of yel-
low mice behave as if heterozygous or simplex with
respect to yellow color, for when any two yellow mice
1 "Studies of Inheritance in Rabbits." Carnegie Institution
Publications, No. 114, 1909. W. E. Castle in collaboration with
Walter, Mullenix and Cobb.
THE FACTOR HYPOTHESIS 163
THE KINDS OF GRAY RABBITS (Color only)
NUM-
BER OF
CjrENOTYPE
HET-
PHENOTYPES
ZYQOTIC FORMULA
EROZY-
GOTIC
When inbred, these kinds are produced
FAC-
TORS
1
AAUUIIEECC (YBBr] [YBBr]
None
X
2
AAUUIIEECc [YBBr][YBBr]
One
X
X
8
AAU UIIEeCC ( YBBr] [ YBBr]
One
X
x
4
AAUUHEECC [YBBr] (YBBr]
One
X
X
5
AAUuIiEECC [YBBr][YBBr]
One
x
x
6
AaUUIIEECC [YBBr] [YBBr]
One
X
X
7
AAUUIIEeCc [YBBr](YBBr]
Two
X
X
X
8
AAU UliEECc [ YBBr] ( YBBr]
Two
x
x
X
9
A A UuIIEECc ( YBBr] ( YBBr]
Two
x
X
X
10
AaU UIIEECc [ YBBr] [ YBBr]
Two
X
X
X
11
AAUUIiEeCC (YBBr] [YBBr]
Two
x
x
X
X
12
AA UuIIEeCC [YBBr] [YBBr]
Two
x
X
x
x
13
AaU UIIEeCC ( YBBr] ( YBBr]
Two
X
x
X
X
14
AAUuIiEECC (YBBr] [YBBr]
Two
X
X
x
X
15
AaUUIiEECC (YBBr] (YBBr]
Two
X
X
x
X
10
AaUuIIEECC [YBBr][YBBr]
Two
X
X
X
X
17
Aa UuIiEECC ( YBBr] ( YBBr]
Three
X
X
X
X
X
x
X
X
18
Aa UuIIEeCC [ YBBr] [ YBBr]
Three
X
X
x
X
X
X
X
X
19
Aa UuIIEECc [ YBBr] ( YBBr]
Three
X
X
X
X
X
20
AaU UliEeCC ( YBBr] [ YBBr]
Three
X
X
X
X
X
X
X
x
21
AaUUIiEECc [YBBr] [YBBr]
Three
x
X
X
X
x
22
AaUUIIEeCc [YBBr][YBBr]
Three
X
X
X
X
x
23
AAUuIiEeCC [YBBr][YBBr]
Three
X
x
X
x
X
X
X
x
24
A A UuIIEeCc [ YBBr] ( YBBr]
Three
x
X
X
X
x
25
AAUuIiEECc (YBBr] [YBBr]
Three
X
x
x
x
x
26
A A U UliEeCc (YBBr] [ YBBr]
Three
x
X
X
X
X
27
AAUuIiEeCc (YBBr] (YBBr]
Four
x
X
X
X
x
X
x
X
X
28
AaUUIiEeCc (YBBr] (YBBr]
Four
x
X
X
X
x
X
x
X
X
29
Aa UuIIEeCc [ YBBr] [ YBBr]
Four
X
X
X
X
x
X
X
X
X
30
AaUuIiEECc (YBBr] [YBBr]
Four
X
X
X
X
x
X
X
X
X
31
Aa UuIiEeCC [ YBBr] [ YBBr]
Four
X
X
X
X
x
x
X
X
X
X
X
X
X
x
X
x
32
AaUuIiEeCc [YBBr][YBBr]
Five
x
X
X
X
x
X
X
x
X
X
X
X
X
X
X
xx\
a
1
I
1
f
| Blue graj
w
g
| Blue gra?
w
I"
~
I
| Yellow si
i
1
P 3
5"
?
80
B
I
1
"S
1
I
I
I
1
i
164 GENETICS
are bred together they produce a certain percentage
of recessives lacking yellow which would not happen if
they were pure yellow. Hundreds of yellow individuals
have been tested but they always produce in addition
to yellow some non-yellow, that is, black, brown or gray
individuals. That the non-yellow individuals are re-
cessive is shown by the fact that when inbred, they
produce no yellow offspring, therefore, yellow is domi-
nant.
In a Mendelian monohybrid cross, as has been pre-
viously pointed out, the expectation is that in the sec-
ond generation one fourth of the offspring will be re-
cessives (DR X DR = DD + 2 DR + RR), but when
yellow mice are bred together, the percentage of re-
cessives approximates one-third instead of one-fourth.
Little, in a total of over 1200 young produced by yel-
low parents, obtained almost exactly two-thirds yel-
low. This apparent exception to the Mendelian ratio
finds an explanation, however, when it is assumed that
D (yellow) is a lethal gene when present m duplex
(DD) form. The DDs drop out entirely which leaves
the proportion approximately two DRs and one RR.
This supposition is further supported by the fact that
the litters of young from yellow mice are, on an aver-
age, only three-fourths as large as normal litters of
mice, which is exactly what would be expected if one-
fourth of the possible gametic combinations (DD) fail
to produce offspring. Moreover, evidence of the death
m utero of the pure yellow mice has been produced by
Ibsen and Stiegleder, '17.
THE FACTOR HYPOTHESIS 165
Morgan and his associates have demonstrated the
existence of over twenty different lethal factors in
Drosophila which when inherited from both parents
not only prevent the development of any unit charac-
ters but also doom the individual to death. Only
heterozygotes for such lethals, who receive the death
warrant from one parent alone, may escape and hand
on this fatal determiner.
In plants lethal genes have been demonstrated by
Baur in snapdragons and by Lindstrom in maize. In
these instances the lethal factor is a lack of chlorophyll
which is not fatal if inherited from a single parent
because the deficiency is supplied by a gene for
chlorophyll from the other parent, but when the lack
comes from both parents it produces a seedling unable
to survive.
Recently'G. H. Shull has demonstrated the existence
of two balanced recessive lethal factors in one pair of
the fourteen chromosomes in (Enothera, one pair pro-
ducing a lethal effect in the zygote, the other pair
destroying the gametes. This fact explains many of
the hitherto confusing ratios obtained in breeding this
classical plant.
"Such lethal factors modify the expected Mendelian
ratios and greatly complicate the study of genetics,
but they do not destroy its fundamental principles,
indeed when properly understood they furnish one of
the strongest proofs of the truth of the factorial
theory of heredity" (Conklin).
166 GENETICS
6. MODIFYING GENES AND SELECTION
The recognition of modifying genes has furnished
an explanation for the apparent effectiveness of selec-
tion within a pure line without assuming germinal
inconstancy.
The gene itself, like that producing the hooded pat-
tern of Castle's rats, is constant but it is accompanied
by a halo of modifying genes likewise constant which
have no somatic expression except when the original
factor for hooded pattern is present. These modifying
genes are simply potential increasers or diminishers of
the hooded-pattern gene. In the absence of the pattern
gene there is nothing to increase or diminish and con-
sequently there is no way to demonstrate the modifying
factors. They are not imaginary things, however, for
their separate existence and transmissibility have been
demonstrated from many sides. What selection within
the progeny of the homozygous cross of hooded rats
or bristly flies accomplishes is simply the elimination
or addition of either plus or minus modifying genes,
according as the attempt is being made to increase
or decrease the hooded pattern of pigmentation or the
number of bristles.
If this explanation stands the test of further investi-
gation then we are still dealing in heredity with con-
stant dependable units, such as the chemist finds in his
elements, and it may be said that all genetic roads
lead to the Rome of gene-constancy.
However, it is well to remember that Darwin did not
revolutionize the concept of evolution until he broke
THE FACTOR HYPOTHESIS 167
down the idea of constancy of species and that geology
did not come into its own until Lyell substituted for
constancy the molding hand of incessant change.
No doubt there is a substratum of unity underlying
all of these processes and Mendelians, therefore, may
still retain their constant characters undismayed and
have, at least, the three following ways left by means
of which it is possible to get results in selection:
(1) By the isolation of pure lines if the stock is
hybrid (heterozygous) ;
(2) By the elimination or addition of modifying
genes if the stock is pure (homozygous) ;
(3) By mutation of the genes.
It should be repeated that change by mutation does
not beg the question of constancy of the genes. A
mutation is not a changed gene. It is the substitution
of an entirely different one t
CHAPTER VIII
BLENDING INHERITANCE
1. RELATIVE SIGNIFICANCE OF DOMINANCE AND
SEGREGATION
OF the three fundamental principles which underlie
"Mendel's law," namely, segregation, independence of
unit characters, and dominance, the principle of domi-
nance has been found to hold true in a surprising num-
ber of cases and in relation to very diverse organisms,
notwithstanding the fact that its universal application
is by no means assured.
Mendel himself noted certain exceptions to the law
of dominance, and his followers have pointed out with
increasing emphasis that it is subject to many modifi-
cations. It is now understood, indeed, that segrega-
tion, not dominance, is the most essential factor in the
Mendelian scheme.
2. IMPERFECT DOMINANCE
It frequently occurs that dominance is so imperfect
that a heterozygous, or simplex, dominant may be dis-
tinguished at once by simple inspection from a homo-
zygous, or duplex, dominant, whereas the test of cross-
ing with a recessive is necessary whenever dominance
is complete, as has been previously explained. The
168
BLENDING INHERITANCE 169
single dose of the determiner in such a case has plainly,
then, less phenotypic effect than a double dose.
There are many instances of imperfect dominance
among flowering plants. Correns* red and white four-
o'clocks with pink offspring (p. 106) is a case in hand.
A classic illustration of imperfect dominance among
animals is the "blue Andalusian fowl," the hereditary
behavior of which is illustrated below (Fig. 35). It
will be seen that when two blue Andalusian fowls,
Andalusian Andalusian
j i r i
Black Andalusian Andalusian Splashed White
\
I: . I II
Black Andalusian Andalusian Spl. White Spl.White
Andalusian
FIG. 35. The heredity of the blue Andalusian fowl, an illustra-
tion of "imperfect dominance."
I
characterized by a mottled plumage, are bred together,
they produce three kinds of offspring in the ratio of
1 : 2 : 1. Twenty-five per cent are clear black, 50 per
cent are blue Andalusian, and 25 per cent are white
"splashed" with black. Both the black and the
splashed white fowls from this cross prove, upon fur-
ther breeding, to be homozygous, while the blue Anda-
lusian itself is heterozygous and can, therefore, never
be made to breed true. In order to produce 100 per
cent of blue Andalusian chicks, it is necessary simply
to cross a splashed white with a black Andalusian.
There is nothing in this case to indicate whether the
170 GENETICS
black or the splashed white should be regarded as the
homozygous dominant, since dominance is imperfect.
In either case the heterozygous blue Andalusian is at
once evident in the first filial generation without further
crossing.
A similar case of imperfect dominance is furnished
by the roan color of cattle which results when red and
white are crossed. If two roans are mated, they pro-
duce red, roan, and white offspring in the proportion
of 1 : 2 : 1, thus showing that roan is a heterozygous
character in which the dominance of red is imperfect.
Even in cases of apparently perfect dominance it is
sometimes possible by close inspection to detect differ-
ences between a pure dominant (DD), Figure 17, and
a heterozygous dominant (DR) when a superficial ex-
amination is not sufficient to distinguish them.
Morgan cites a Drosophila cross between "ebony"
and "sooty" wings wherein the F 2 generation ranges
from ebony to sooty in an inseparable transition but
it proves, nevertheless, to be of three classes in the
proportion of 1 : 2 : 1, as further breeding tests show.
3. DELAYED DOMINANCE
A character which is really dominant is sometimes
so late in manifesting itself in the individual growth
of the offspring that it may properly be termed a
delayed dominant.
Dark-haired individuals often do not acquire their
definitive hair color until adult life, and it is common
knowledge that the eyes of an infant for a consider-
BLENDING INHERITANCE 171
able period provoke no little speculation among ador-
ing relatives as to "whose eyes" they are.
According to Davenport, when a white Leghorn
fowl is crossed with a black Leghorn, white being
dominant in this case, chicks are produced that are
white with black flecks in their plumage. These black
flecks, however, disappear at the time of the first molt.
The complete dominance of white is, therefore, simply
delayed.
4. "REVERSED" DOMINANCE
In certain instances there seems to be a reversal of
dominance, as may be illustrated by Lang's results with
snails (Helix). *He has proven in his experiments that
red snails are generally dominant over yellow snails,
although in certain cases there is apparently an
exception to the rule, for snails with yellow shells
dominate those with red shells.
Davenport has shown too that although extra toes
are usually dominant over the normal number in poul-
try, yet, in something like 20 per cent of the cases, the
normal number is dominant.
It sometimes occurs that a character which is domi-
nant in one species may be recessive in another. Horns
are dominant in sheep, but recessive in cattle. White
color is recessive in rodents and sheep, but dominant
in most poultry and in pigs.
Again Morgan describes a Drosophila that possesses
a gene for abnormally banded abdomen which does not
come to somatic expression unless the flies are supplied
with fresh food and a proper amount of moisture.
172 GENETICS
When the food becomes dried up and there is a mini-
mum of moisture the banding on the abdomen disap-
pears. Here is a reversal of dominance but the gene
itself is not affected since the same flies which are
hybrid with respect to the character of banding, show
a difference according to the environment of food and
moisture, in the amount of banding given expression.
Notched margin in leaves is dominant in nettles and
recessive in the celandine. Again a negative character
may be the dominant one in a pair of allelomorphs.
For example, the bob-tail of the Manx cat is dominant
over the ordinary long tail of the cat; the reduced
number of three digits in guinea-pigs is dominant over
four digits ; the polled condition is dominant over horns
in cattle; the rumpless fowl is dominant over the fowl
with a rump, and brachydactyly in man, that is, fingers
or toes with only two joints each, is dominant over
the three- jointed arrangement.
5. POTENCY
Davenport seeks to explain modifications in typical
dominance as variations in the potency of determiners.
He defines potency as follows: "The potency of a
character may be defined as the capacity of its germi-
nal determiner to complete its entire ontogeny."
That is, if the potency of a determiner, for some
reason, is insufficient, there may be either an incom-
plete or delayed manifestation of the character in
question, or it may fail entirely to develop.
The variations of potency may be grouped into three
BLENDING INHERITANCE 173
general categories according to the degree of their
manifestation ; namely, total potency, partial potency,
and failure of potency.
A further word of explanation for each of these
three kinds of potency seems desirable at this point.
A. TOTAL POTENCY
This is complete Mendelian dominance in which
even the heterozygotes produced by a simplex dose of
a character are indistinguishable phenotypically, that
is, by inspection, from the homozygotes produced by a
duplex dose of the same character. It is as if a single
bottle of black ink poured into a jar of water was
just as effective as two bottles of ink, in forming an
opaque fluid.
Even in the cases of apparently complete dominance,
however, refined methods of examination or analysis
may make it possible to distinguish the duplex from
the simplex condition without recourse to breeding.
Darbishire has shown, in the case of Mendel's smooth
and wrinkled peas, that the two kinds of smooth prog-
eny from the Fj hybrid upon microscopic examination
show a difference in their starch grains that indicates
at once which is homozygous and which is heterozy-
gous. Moreover, in the power of absorption, hybrid
smooth peas (DR) are intermediate between their
pure dominant smooth (DD) and pure recessive
wrinkled (RR) parents.
Blakeslee has demonstrated a chemical method of
distinguishing unseen genetic differences in the appar-
174 GENETICS
ently similar flowers of the black-eyed susan (Rud-
beckia hirta). When placed in a solution of KOH,
the yellow cones of one kind turn a purplish-black,
while the other kind turns red.
B. PARTIAL POTENCY
Partial potency covers all cases of incomplete
dominance, such as those of the four-o'clock (Mira-
bilis) and blue Andalusian fowls, where a simplex dose
of a determiner does not produce the same visible effect
as a double dose.
The dominant prickly jimson weed (Datura), when
crossed with a recessive glabrous variety of the same
plant, produces cross-breds in the first generation
which show only a few prickles (Bateson) (Baur),
following the law of partial potency.
Banded and uniformly colored snails also, when
crossed together, produce snails with shells showing
only a pale banding (Lang).
Numerous further instances of incomplete domi-
nance could be cited.
C. FAILURE OF POTENCY
If for any reason a determiner fails to accomplish
its possibilities in whole or in part, then the character
in question may never become evident, and the result,
so far as appearances go, is the same as if it was a
recessive lacking the determiner entirely.
That the failure of potency is not identical with the
absence of a determiner can usually be demonstrated
BLENDING INHERITANCE 175
by further breeding, because dominants failing in po-
tency, which are either of the formula DD or DR,
may, if bred inter se, give a various progeny among
which the dominant character D is likely to become
manifest again, while recessives, of the formula RR,
on the contrary, will invariably give offspring which
all agree in the entire absence of the character in
question.
Davenport cites an extreme case of failure of po-
tency in one of two rumpless cocks from the same blood.
The character of rumplessness is due to an inhibitor
of tail development. That these two cocks both pos-
sessed this character was demonstrated by the entire
absence of any tail in either case. The inhibiting de-
terminer for tail growth was so weak in cock No. 117,
however, that, to quote Davenport's exact words: "In
the heterozygote the development of the tail is not
interfered with at all, and even in extracted dominants
it interfered little with tail development, so that it
makes itself felt only in the reduced size of the uro-
pygium and in-bent or shortened back. But in No. 116
the inhibiting determiner is strong. It develops fully
in about 47 per cent of all the heterozygotes and in
extracted dominants may produce a family in all of
which the tail's development is inhibited."
Here were two birds of the same blood, phenotyp-
ically alike and presumably genotypically alike, which
because of an individual difference in the potency of
the determiner for rumplessness produced quite differ-
ent results in their offspring although bred to precisely
the same array of hens.
176 GENETICS
6. BLENDING INHERITANCE
In the instances of imperfect dominance given above,
where the progeny of unlike parents present an inter-
mediate condition, it is found that, upon cross-breed-
ing these offspring, segregation into the grandparental
types occurs just as truly as in instances of complete
dominance.
In poultry, for example, when Cochins, which are
"booted," and Leghorns, which are clean-shanked, are
crossed, booting of an intermediate grade of four re-
sults, on a scale in which ten represents complete boot-
ing, and zero, no booting or clean shank (Davenport).
The character of booting and its alternative absence,
however, segregate out in true Mendelian fashion when
these hybrids are subsequently crossed together. It is
evident that dominance plays only a secondary role in
such cases, and that the all-important factor is segre-
gation.
Are there, then, any cases where true fusion of
hereditary parental traits occurs, in other words,
where segregation in the second filial generation does
not appear? Does the "melting-pot of cross-breed-
ing" ever "melt" the characters thrown into it?
It was formerly believed that diverse parents gener-
ally produce intermediate offspring, and that this
intermediate condition continues without any segrega-
tion at all in the form of "blending inheritance," but
within the last decade apparent cases of blending in-
heritance have been thrown out of court one after the
other by the Mendelians. Bateson, in an inaugural
BLENDING INHERITANCE 177
address at Cambridge University in 1908, stated that
what was once believed to be the rule has now become
the exception. He goes on to say: "One clear excep-
tion I may mention. Castle finds that in a cross be-
tween the long-eared lop rabbit and a short-eared
breed, ears of intermediate length are produced; and
that these intermediates breed approximately true."
Let us examine this "one clear exception" a little
more closely.
7. THE CASE OF RABBIT EAES
As a typical example of blending inheritance in
rabbit ears the following case may be cited :
A female Belgian hare with an ear-length of 118 mm.
was crossed with a male lop-eared rabbit with an ear-
length of 210 mm. The average of these ear-lengths
is 164 mm. Five offspring from this pair had ear-
lengths, when adult, approximating this average as
follows: 170, 170, 166, 156, 170, of which two were
females and three were males. When from this litter
one of the females measuring 170 mm. in ear-length
was subsequently crossed with her brother having an
ear-length of 166 mm., two litters were produced in
which the individuals when adult attained ear-lengths
of 170, 166, 168, 160, 172, and 168 mm. These
results are represented diagrammatically in Figure 36.
This illustration is typical of many other breeding
experiments made by the same investigators 1 upon the
1 Castle, in collaboration with Walter, Mullenix and Cobb.
"Studies of Inheritance in Rabbits." Carnegie Institution Publi-
cations, Washington, No. 114^ 1909.
178
GENETICS
Offspring of 1 and 7
Offspring of 3 and 5
$ 1 3 I 5 5
FIG. 36. A case of three generations of ear-length in rabbits.
0-6, average ear-length of the first filial generation (FJ.
a'-b', average ear-length of the F 2 generation derived from
1 and 7. Data from Castle, in collaboration with Walter,
Mullenix and Cobb.
ear-length of rabbits which included 70 different litters
of rabbits containing 341 individuals. In none of these
experiments could the blend in the second filial genera-
BLENDING INHERITANCE 179
tion be called perfect, but it may at least be said that
evidence of segregation, that is, a return to one or the
other of the parental types, was much less apparent
than evidence of blending.
Furthermore, crosses were made in which lop ears
of various fractional lengths were obtained as desired,
including |, i, f, i, f , f , and lengths. Not one of
these fractional lengths apparently segregated in sub-
sequent generations after the Mendelian fashion, but
all bred approximately true.
Moreover, ears of one half lop length, for instance,
were obtained in three ways: first, by crossing full-
length lops with short-eared rabbits as indicated in the
first cross of the case cited above; second, by crossing
one half lop lengths together, demonstrated by the
second cross in the illustrative case given, and third,
by mating J and f lop lengths. Theoretically, -J and J-
as well as f and f lop lengths would also produce
\ lop lengths, for in all of the crosses that were made
the length of ear behaved in a blending fashion.
These results were based, not upon a single measure-
ment of each specimen, which might be open to con-
siderable error, but upon daily measurements from the
time the rabbits were two weeks old until their ears
ceased to grow at about twenty weeks. The growth
curves drawn from these daily measurements showed
continually an intermediate or blending condition in
progeny derived from diverse parents.
A Mendelian explanation of this apparently excep-
tional case of blending inheritance has been suggested
by Lang based upon the result of Nilsson-Ehle's dis-
180 GENETICS
coveries while breeding wheats at the Agricultural
Experiment Station of Svalof in Sweden.
8. THE NILSSON-EHLE DISCOVERY
Nilsson-Ehle, 1907, found in breeding together
different strains of wheat that a certain wheat with
brown chaff crossed with a white-chaffed strain yielded
only brown-chaffed wheat in the first generation.
These heterozygous or hybrid brown-chaffed wheats
when crossed with each other produced, not the ex-
pected proportion of three brown to one white, but
fifteen brown to one white. This was not explainable
as the chance result of a single cross, but was the con-
clusion drawn from fifteen different crosses, all of the
same strains, that yielded a total progeny of 1410
brown-chaffed to 94 white-chaffed plants, which hap-
pens to be exactly the proportion of fifteen to one.
In other experiments it was discovered that although
dominant red-kerneled strains of wheat crossed with
white-kerneled varieties usually gave the three-to-one
proportion upon segregation in the second filial genera-
tion, yet one particular strain of red-kerneled Swedish
wheat in the second generation gave approximately
sixty-three red to one white-kerneled strain.
The explanation of these two unexpected results is
this. In the case of brown-chaffed wheat there are two
independent determiners for the character of brown
color, and these simply follow the Mendelian laws for
a dihybrid, while in the case of the red-kerneled -wheat
there are three independent determiners for the charac-
BLENDING INHERITANCE
181
ter of red color, each of which is able to give red color
to the wheat. Taken together, these three red color
determiners behave cumulatively, following the law of a
trihybrid.
For example, if a brown-chaffed wheat with the
formula BB', in which B and JB' each represent a
brown - chaffed
factor, is
crossed with a
white - chaffed
wheat of the
formula bb' 9 in
which b and b'
each represent
the absence of
B and B' re-
spectively, then
all the progeny
of this cross
will be brown-
chaffed, having
the zygotic for-
mula BB'W.
When upon ma-
turation the gametes form out of the germ-cells from
such hybrids, the following four combinations are pos-
sible, and no others : BB', Bb', bB', bb'. These repre-
sent, therefore, the possible gametes present in each sex
of the first filial generation, and upon intercrossing
they may combine into sixteen possible zygotes to form
the second filial generation, as shown in Figure 37.
BB 1
Bb'
bB'
bb'
BB 1
Bb 1
bB'
bb 1
BB 1
BB'
BB 1
BB'
BB'
BB'
Bb'
bB'
bb'
Bb'
Bb 1
Bb'
Bb'
Bb'
BB 1
Bb 1
bB'
66'
bB'
bB'
bB'
bB'
bB'
BB'
Bb 1
bB'
66'
bb'
bb'
bb'
bb'
66'
SL
(o)
FIG. 37. Diagram of the possible combina-
tions in the F 2 generation of brown-chaffed
wheat according to experiments of Nilsson-
Ehle. B and B' are cumulative factors for
the brown-chaff character; b and b' denote
the absence of B and B' respectively.
182 GENETICS
The numbers in the squares in Figure 37 indicate how
many times a brown determiner is present in each
zygote. It will be seen that only one out of the sixteen
possibilities lacks a brown-chaff factor, and this one will
43210
Number of doses of the brown determiner
FIG. 38. The distribution of the sixteen possibilities resulting
when two similar determiners (brown-chaff) act together as
a dihybrid.
consequently produce only white chaff, while the re-
maining fifteen possibilities, each of which has at least
a single determiner for brown, will all yield brown
chaff.
The brown-chaff factor, moreover, is present in
BLENDING INHERITANCE
183
varying doses among these fifteen possibilities, as indi-
cated by the numbers in the squares. It is evident,
therefore, that several shades of brown will be repre-
n
0e0
000
0e
060
o
000
o e e
000
000
6
CD e
000
5
060
000
5
00
5
060
000
4
00
4
060
00
4
060
000
3
0o
000
o
5
0e
00
4
e
000
4
o
00
4
060
00
3
0*
3
060
3
000
2
(D o0
000
0e0
5
00
e0
4
0e0
060
4
<D
0e0
4
060
060
3
06-0
3
000
00
3
000
000
2
000
o0
5
e
o
<i
e
o0
4
o>
*0
4
060
3
06
o
3
060
3
000
60
2
d)e e>
000
0eo
4
c
0e0
3
e
0e
3
060
3
06 C
060
2
060
2
060
0..
006
o r
0o
000
000
4
00
3
0e0
o e
3
3
060
2
o
e
2
o0
00
2
000
00
e
0)00
060
4
e
000
3
060
6
3
060
3
060
060
2
060
2
000
000
2
000
00
1
000
e
060
o 00
060
000
060
060
3
o
2
2
1
1
1
FIG. 39. Diagram to illustrate Nilsson-Ehle's case of trihybrid
red wheat. The large screw-heads each represent a single
determiner for the red character. The small screw-heads
symbolize the absence of the red character, or white. The
number in each square indicates how many doses of the "red"
determiner is present. For further explanation see text.
sented depending upon the number of doses of the
brown determiner in each instance.
Figure 38 shows how these different shades of brown
arrange themselves in the manner of a frequency curve
of fluctuating variation with the greatest number in
GENETICS
the halfway class and the least numbers at the two
extremes. In this instance six out of sixteen individu-
als of the second generation theoretically present a
perfect "blend" between the original brown- and white-
chaffed grandparents, although complete segregation
has actually occurred.
The same explanation holds true as displayed in
Figure 39 for the trihybrid case of red- and white-
kerneled wheats in which only one white-kerneled to
sixty-three red-kerneled individuals appear in the sec-
000 + e o=0o0e0o
Pure red -f- white =3 Hybrid red
FIG. 40. The result of crossing white wheat with trihybrid red
wheat.
ond filial generation. The number of red determiners
in each zygote is indicated by the figure at the bottom
of each square. The large screw-head symbols with
vertical, horizontal and diagonal slots each represent
an independent determiner for red kernel, while the
small screw-heads symbolize the absence of each of
these determiners, or white kernel. When the pure
strain of red-kerneled wheat is crossed with a pure
strain of white-kerneled wheat, the first generation is
all a heterozygous red of a somewhat lighter shade
than the original pure red strain as shown diagram-
matically in Figure 40.
When plants of this heterozygous sort are crossed
together, they yield plants producing red-kerneled and
white-kerneled wheats in the proportion of sixty-three
to one. The sixty-three kinds of red wheats are of
BLENDING INHERITANCE
185
#
varying degrees of redness and may be classified after
the manner of fluctuating variations with the greatest
number of kinds at the
intermediate degree be-
tween pure red and pure
white. (See Figure 41.)
In order to test whether
the sixty-four kinds of
wheats produced in the
second filial generation, as
theoretically displayed in
Figure 39, really contain
separable, though indis-
tinguishable, determiners
for red-kernel, Nilsson-
Ehle produced families of
the third filial generation
by self-crossing plants of
the second generation. It
was to be expected that, if
these hybrid wheats of the
second generation carried
one, two, three, or more
determiners for a red ker-
nel as the theoretical
tables in Figures 39 and
41 demand, their progeny
would be distributed with
reference to the number of
red- and white-kerneled
individuals, in the follow-
ing ratios :
5 4
3
#
FIG. 41. The distribution of
the sixty-four possibilities in
the F 2 generation when three
similar determiners act to-
gether as a trihybrid.
186 GENETICS
3 red to 1 white when 1 heterozygous determiner for red
is present.
15 red to 1 white when 2 "1 heterozygous deter-
63 red to 1 white when 3 > miners for red are
All red to no white when 4 or more J present.
Among seventy-eight sample families of the third
generation inbred to test this theoretical conclusion,
the actual results were:
8 families giving the ratio of 3 red to 1 white.
15 families giving the ratio of 15 red to 1 white.
5 families giving the ratio of 63 red to 1 white.
50 families giving the ratio of all red to no white.
It has been actually demonstrated therefore, in the
case of this particular strain of wheat: (1) that the
factors producing red kernels are several in number;
(2) that they act independently of each other in
heredity; (3) that these several independent factors
segregate; and (4) that any one red factor acting
alone produces a "red" result.
The Nilsson-Ehle principle of cumulative determiners
has been confirmed in America by East in a masterly
series of breeding experiments upon maize.
In connection with the Nilsson-Ehle principle, it
will be seen that the possible number of intergrades
between the two extremes increases rapidly as the num-
ber of duplicate determiners increases. Thus with six
duplicate determiners for the same character present,
the ratio of possible dominants to recessives in the
second filial generation would be 4095 to 1. The re-
appearance of this single recessive among 4095 domi-
BLENDING INHERITANCE 187
nants would be extremely unlikely, and it might easily
be mistaken for a mutation or a freak. Apparent
blends of all intermediate degrees, however, would be
sure to appear. Yet these are not blends in the
"melting-pot" sense at all, but strictly cases of
Mendelian dominance and segregation.
9. THE APPLICATION OF THE NILSSON-EHLE EXPLA-
NATION TO THE CASE OF RABBIT EAR-LENGTH
The so-called blending rabbit ears, along with other
similar cases, can now be made to fall into line, as
pointed out by East and by Lang, with the Mendelian
law of segregation.
If we assume that the long ear of the lop rabbit has
only three independent but equal determiners for excess
length, the case becomes one of Mendelian trihybridism
with cumulative factors, which works out like Nilsson-
Ehle's red-kerneled wheat in the following manner:
In general the average for full lop ear-length may
be placed at 220 mm. and for the ordinary short-
eared rabbit 1 at 100 mm. The difference, or the excess
length of the lop ear, is 120 mm., which, according to
the trihybrid formula, corresponds to the six doses of
the character symbolized in the upper left-hand square
in Figure 39 by six large screw-heads, three coming
from each parent respectively. If all of these inde-
pendent determiners are* equal as regards excess ear-
length, each factor would represent an excess of 20 mm.
J Not the Belgian hare, as cited in the illustration given in
Figure 36. The Belgian hare has typically a somewhat longer
ear than the ordinary short-eared rabbit.
188
GENETICS
above the normal ear-length found in short-eared rab-
bits, that is,
220 mm. 100 mm.
= 20 mm.
6
When according to this computation a lop (20
mm. X 6 + 100 mm. = 220 mm.) and a pure short-
eared rabbit (20 mm. X + 100 mm. = 100 mm.)
are crossed, if imperfect dominance occurs, which is a
very common phenomenon, it is true that the offspring
might present a "blended" appearance. If now these
cross-breds of the first generation prove to be tri-
hybrids with respect to excess ear-length, there would
be sixty-four possibilities in their progeny segregating
out just as in the red-kerneled wheat.
These possibilities would be arranged in the fol-
lowing frequencies:
NUMBER OF EXCESS
EAR-LENGTH
DETERMINERS
NUMBER OF CASES
OCCURRING OUT OF 64
TOTAL LENGTH IN
MILLIMETERS OF
EARS RESULTING
6
1
220
5
6
200
4
15
180
3
20
160
2
15
140
1
6
120
1
100
Since the average litter among rabbits is about five,
the chances that these five rabbits will breed true to
their hybrid parents and form a perfect blend between
their grandparents is 20 out of 64, while the chance
of their being like either grandparent is only one out
of 64.
BLENDING INHERITANCE 189
It should be noted further that 50 out of 64, or 77
per cent, of these hybrids of the second filial genera-
tion would have an ear-length between 140 and 180,
thus approximating a "blend" closely enough to be so
classified upon a casual inspection.
If it should be found, moreover, that excessive ear-
length in rabbits is due to more than three duplicate
determiners, the possibilities of getting anything but
an apparent blend would be much decreased.
Furthermore, the fact that the fractional ear-
lengths of the hybrid rabbits in Castle's experiments
bred approximately true in the second and subsequent
filial generations, may also be explained by the Nilsson-
Ehle hypothesis.
For example, half lop lengths, according to this
explanation, are those with three doses of the deter-
miner for excess ear-length. It follows that the
progeny of two rabbits each carrying three doses of
a determiner will likewise, after the reduction during
the maturation of the germ-cells, have three doses of
the determiner
It would be interesting to breed rabbits having ears
of one-eighth lop length in which, according to the
foregoing hypothesis, there presumably would be
present only a single determiner for excess ear-length,
with ordinary short-eared rabbits having no excess
ear-length, in order to see if the expected Mendelian
three-to-one proportion for a monohybrid would ap-
pear in the progeny.
190 GENETICS
10. HUMAN SKIN COLOR
Finally, in man the skin-color of mulattoes, in hy-
brids between blacks and whites, has often been men-
tioned as a case of blending inheritance since mulattoes
are commonly supposed to produce mulattoes when
they mate together or a blending degree of color when
they mate with some one whose shade of color is unlike
their own.
This matter has been carefully and extensively
studied by Davenport and Danielson l who came to the
conclusion that the pure-blooded negro of the West
Coast of Africa possesses two pairs of duplicate genes
for black pigmentation (A ABB) which, though sepa-
rately heritable, are cumulative in effect. The corre-
sponding formula for black pigmentation in a normal
white is adbb. When black (AABB) and white (aabb)
are crossed, the formula for the mulatto will be AaBb
in which half the total amount of black pigment of both
parents is present.
The result of crossing two mulattoes is shown by
checkerboard diagram in Fig. 42.
The figures in the corners of the squares indicate
the total amount of black pigment in each case upon
the supposition that A = 19, B = 1 6, # = 2 and
b = 1, these values being determined by the color-top
method described by Davenport and Danielson.
In the table on page 191 is a classification of these
possibilities according to the amount of black pigment
1 Heredity of Skin Color in Negro and White Crosses. Pub.
No. 188 of the Carnegie Inst. of Washington,
BLENDING INHERITANCE
191
present, and in Figure 43 is a graphic representation
of the numerical chances for skin-color in spite of
segregation when two mulattoes mate.
AB
Ab
aB
ab
AB
Ab
aB
ab
AB
a b
70
55
53
Ab
AB
55
Ab
40
-
aB
AB
53
/Ab
21
ab
a b
B
23
21
6
FIG. 42. Checkerboard to show the different expected shades of
black color in the possible offspring of two mulattoes. A =
18; B = 17; a = 2; b = l percent of black pigment. Data
from Davenport and Danielson.
The case thus is similar to that of Nilsson-Ehle's
brown-chaffed wheat already described, so that the
possible range of offspring of a mulatto pair is all the
way from black to white. Theoretically, any one of
five degrees of pigmentation, including the extremes
of black and white, may be expected. The chances
192
GENETICS
A CLASSIFICATION OF HYBRID SKIN COLORS ON THE BASIS OF THE
FACTOR HYPOTHESIS
Genes
All absent
Gametic
Formula
Color
Relative
Fre-
quency
Range of
Percent of
Pigment in
Offspring
Popular Names
(Jamaica)
aabb
White
1:16
0-11
Pass-for-white
Mustifino
Mustifee
Octaroon
One present
Aabb
aaBb
Light
4:16
12-25
Quadroon
Two present
AAbb
AaBb
aaBB
Medium
6:16
26-40
Mulatto
Three present
AABb
AaBB
Dark
4:16
41-55
Mangro
Sambo
All four present
AABB
Black
1:16
56-78
Negro
which any one of these five degrees of color has of
reappearing in a child of mulatto parents is indicated
in Fig. 43. It is evident that there is more likelihood,
in point of fact six chances out of sixteen, that a
child from mulatto parents will be mulatto than any-
thing else, and this expectation ordinarily agrees with
the realization, but there are four chances out of six-
teen that it will be either darker or lighter than its
parents and one chance out of sixteen that it will be
as dark or as light as its black or white grandparents.
Davenport and Danielson show several illuminating
photographs of large families of children from mulatto
parents in which a manifest inequality of color shade
among the different children is apparent as would be
expected according to this explanation.
Blending inheritance, then, is probably nothing more
BLENDING INHERITANCE
193
than Mendelian alternative inheritance in which two or
more similar genes, duplicate or cumulative, are con-
cerned. One explanation instead of two or three,
White
Light
Dark
Black
14641
FIG. 43. Diagram to show the expectation of color and its fre-
quency in the cross between two mulattoes.
therefore, is sufficient to dispose of an array of appar-
ently diverse phenomena.
"We may therefore conclude," says Conklin, "that
the Mendelian law of heredity, especially as regards
segregation of inheritance factors, is of universal oc-
currence that there is no other type of inheritance."
CHAPTER IX
OLD TYPES AND NEW
1. THE DISTINCTION BETWEEN REVERSION AND
ATAVISM
THERE are two ways in which types of animals or
plants that are different from the present ones may
be conceived to arise, namely, by the reappearance of
old types and by the formation of new ones. In the
reappearance of old types a distinction may be drawn
between reversion and what has been termed atavism.
Atavism, or "grandparentism," may be defined as
skipping a generation with the result that a particular
character in the offspring is unlike the corresponding
character in either parent, but instead, resembles the
character in one of the grandparents.
In reversion, on the contrary, a character reappears
which has not been manifest perhaps for many gen-
erations, although it was actually present in some
remote ancestor. J. Arthur Thomson's definition of
reversion is : "All cases where through inheritance there
reappears in an individual some character which was
not expressed in his immediate lineage, but which had
occurred in a remoter, but not hypothetical, ancestor."
This distinction between atavism and reversion be-
comes clearer by illustration.
194
OLD TYPES AND NEW
195
If heterozygous brown-eyed individuals mate, there
is one possibility in four that their offspring will have
blue eyes unlike their own, but like the two blue-eyed
Grandfather Grandmother Grandfather
Grandmother
Homozygote Hobiozygote
. Duplex^ NuUiplex
Heterozygote
Simplex
Homozygote
Duplex
Heterozygote
Simplex
Homozygote
Nulliplex
FIG. 44. Three generations of a Mendelian monohybrid. The
outlines represent the somatoplasms with the phenotypic
character on the outside. The black symbols inclosed within
the somatoplasm stand for the germplasm in the form of
gametes. The short dotted arrows indicate the relation be-
tween germplasm and somatoplasm. The long dotted arrows
indicate possible recombinations of germplasms.
grandparents. Such a blue-eyed child would be an
instance of atavism. The explanation of this appar-
ently inconsistent hereditary behavior is perfectly
simple in the light of the Mendelian ratios, as shown
196 GENETICS
diagrammatically in Figure 44, in which the circles
represent the blue-eyed and the squares the brown-
eyed character.
This figure also illustrates what typically occurs in
the formation of Mendelian monohybrids of the first
and second filial generations. The squares are symbols
for the dominant characters, while the circles are sym-
bols for the recessive characters. When the two are
superimposed, the circle recedes from view. The large
outside figures indicate the somatoplasm, therefore the
phenotype. The small inclosed figures indicate the
germplasm, therefore the genotype. The short dotted
arrows indicate what it is that determines the somato-
plasm in each case, while the long dotted arrows show
what possible recombinations of germplasms can be
made. Child No. 4 is an "extracted recessive" derived
from dominant parents, but with one recessive grand-
parent on each side. It is a case of "atavism," or
taking after the grandparent. Notice that atavism
can occur only by alternative inheritance.
To quote Davenport: "In the majority of cases
atavism is a simple reappearance in one fourth of the
offspring of the absence of a character due to the
simplex nature of the character in both parents."
An illustration of reversion would be the reappear-
ance of the ancestral jungle-fowl pattern in domestic
poultry or of the slaty blue color of the ancestral
rock-pigeon among buff and white domestic pigeons,
for the ancestral character or characters in this type
of hereditary behavior, as said before, reappear only
after a lapse of many generations.
OLD TYPES AND NEW 197
2. FALSE REVERSION
"Around the term 'reversion/ " Bateson observes,
"a singular set of false ideas have gathered them-
selves." In proof of this statement there may be cited
at least five categories of apparent reversion which
properly ought not to be classed as true reversion.
A. ARRESTED DEVELOPMENT
Feeble-mindedness is not reversion to ancestral
forms of less intelligence, but an instance of arrested
development when, for some reason, the individual fails
to accomplish his normal cycle of development.
Likewise harelip in man is not a case of reversion to
rabbit-like ancestors in which harelip is the normal
condition, but it is ordinarily due to an arrest or fail-
ure of certain embryonic steps essential to the develop-
ment of the usual form of human lip.
B. VESTIGIAL STRUCTURES
These are the vanishing remains of characters that
were formerly of significance. They do not represent
something latent that is now reappearing, for they
have never yet disappeared phylogenetically, and con-
sequently they cannot be regarded as true reversions.
The muscles under the scalp which enable those
persons possessing them to wiggle the ears; the pala-
tine ridges in the roof of the mouth of many babies
and some adults which resemble the ridges in the roof
of a cat's mouth ; the vermiform appendix, a necessary
198 GENETICS
part of the digestive apparatus of many animals but
fraught so often with evil consequences to man; these
and scores of similar characters, which, taken together,
make man in the eyes of the comparative anatomist a
veritable old curiosity shop of ancestral relics, are the
last traces of characters which formerly had a sig-
nificance in some of man's forbears. Having lost their
usefulness, these structures still hang on to the ana-
tomical household as pensioners. They have not been
recalled from the past, but have always been with us,
although of diminishing importance. In no sense,
therefore, can they be called reversions.
C. ACQUIRED CHARACTERS RESEMBLING ANCESTRAL ONES
Sometimes the drunken descendant of a drunken
great-grandparent has acquired this characteristic
through his own initiative quite aside from any ances-
tral contribution to his germplasm. This is not rever-
sion, but reacquisition resembling the ancestral con-
dition.
Again, tame animals that run wild acquire habits
resembling those of their wild ancestors, but this is
not necessarily reversion. It is the natural response
of feral animals to the conditions of wild life.
D. CONVERGENT VARIATION
The European hedgehog, Erinaceus, an insectivore,
the American porcupine, Erithizon, a rodent, and the
Australian spiny anteater, Echidna, a monotreme, are
all mammals which have developed in a similar manner
OLD TYPES AND NEW 199
the very peculiar device of dermal spines. There is no
reason, however, for regarding this character as due
to descent from a common spiny ancestor. It is not
reversion to an ancestral type, but rather a case of ^
convergent variation. Similarity does not always indi-
cate genetic continuity.
In the case of birds albinism, melanism and flavism
are modifications of ordinary pigmentation which
appear irregularly among many different species as
pathological "sports," but no one of these conditions
can be regarded as reversions to ancestral white, black,
or yellow types.
E. REGRESSION
Galton's "law of regression" refers to the widespread
phenomenon already explained of a constant swinging
back to mediocrity which the breeder must oppose with
continual selection in order to maintain the standard
of any particular strain. We have seen that within
a "pure line," regression is complete and that in popu-
lations made up of a mixture of pure lines it is a fac-
tor that must invariably be considered. Regression,
however, has to do with fluctuating variations and does
not bring about a permanent change of type. It
should, therefore, not be confused with reversion.
3. EXPLANATION OF REVERSION
Darwin, who did not always differentiate between
reversion and atavism, suggested that reversion was
due sometimes to the action of a more natural en-
200 GENETICS
vironment, as in the case of animals set free after hav-
ing been in captivity, and sometimes to hybridization,
since there seems to be a general tendency of hybridized
organisms to "revert" to ancestral types.
It is now known that reversion, like atavism, is/
simply a case of latent characters becoming apparent'^
according to the Mendelian principle of segregation, i
To quote Davenport: "There is nothing more mys-
terious about reversion, from the modern standpoint,
than about forming a word from the proper combina-
tion of letters."
4. THE ART AND SCIENCE or BREEDING
The art of breeding animals and plants has been
practiced from very early times while the science of
maintaining old types and initiating new ones is of
comparatively recent origin.
Some of the methods that have been employed with
varying degrees of success are:
A. Mass selection;
B. Pedigree breeding ;
C. Inbreeding;
D. Hybridization;
E. Genotypic selection.
A. MASS SELECTION
The natural thing to do in the maintenance or im-
provement of cultivated plants and domestic animals is
to select seeds from the best looking plants and to
OLD TYPES AND NEW 201
breed together from the flock or herd those animals
which appear most desirable. This has been the
method from the beginning and there is a reason for
the considerable degree of success that has followed
this obvious mode of procedure. The method, however ,i
has its limitations because it is entirely phenotypic andS
the breeder is sure to find with Dryden that "all as I
they say that glitters is not gold."
Two methods of mass selection, as applied to plants,
may be mentioned that differ in the extent to which
the environment is recognized as a contributing factor.
a. The Method of Hallet
The English wheat-grower Hallet formulated this
method in 1869 and it has been in common use for a
long time. It consists in placing the organisms to be
bred in the very best possible environment and then
choosing those individuals making the best showing
as the stock from which to breed further, a procedure
based upon the deep-seated belief that acquired char-
acters are inherited.
For example, in a field of wheat, plants near the
edge of the field which, from lack of crowding or by
reason of proximity to an extra local supply of fer-
tilizer or any other favorable environmental factor,
make a more vigorous growth than their neighbors,
are selected in the hope that the gains made by them
will be maintained in their offspring.
We have seen that it is very questionable whether
acquired characters due to environmental conditions
202 GENETICS
play any role whatever in heredity. The phenotypic
character does not always indicate what the germ-
plasm will subsequently do, and when the true geno-
typic constitution of the germplasm is still further
masked by the temporary fluctuations caused by a
modified environment it is increasingly difficult to select
wisely from the display of variants those which will
produce the best ancestors for the future stock.
That this common procedure of selecting the best-
appearing animal in the flock and the biggest ear of
corn in the bin has met with a large degree of success
in the past is due entirely to the fact that in many
instances the phenotypic character is an actual express
sion of the genotypic constitution. This is not always
the case, however, and we cannot now fail to see thali
the method is blind and full of error. Its successes\
are due to the indirect results of chance rather than j
to a direct control of the factors of heredity.
b. The Method of Rimpau
Contrasted with the method of Hallet of augment-
ing acquired characters and then selecting from them
the best display, is the method of Rimpau, who ex-
perimented for two decades with various grains and,
finally, among other results, produced the famous
Schlandstedt barley.
Rimpau's method is to sow grain under ordinary
conditions with a minimum rather than a maximum
amount of fertilizer and then to select individuals,
neither from the rich spots nor from the edges of the
OLD TYPES AND NEW 203
field where there is little crowding, but from situations
where the environmental conditions are ordinary or
even unfavorable. Individuals making a good showing
under such usual, or even adverse, conditions are wor-
thy by nature rather than by nurture and are conse-
quently most desirable as progenitors of future stock.
By this method the attempt is not to keep the progeny
of single individuals separate, but to mass together
the best as they appear under ordinary normal environ-
ment.
This again is an indirect method of procedure,
although the character of the germplasm is more
nearly hit upon in this way than by Hallet's method,
since the mask of temporary accessory modifications/
is stripped so far as possible from the somatoplasm, \
and the phenotype made to approximate the geno-/
typical constitution.
B. PEDIGREE BREEDING
Mass selection, or the choosing of a number of indi-
viduals out of a population to be the progenitors of
the next generation, is subject to repeated backsliding
to mediocrity and consequently the selection must be
made over and over again in each generation. A
greater degree of success than is possible by this
method has followed attempts to isolate single self-
fertilizing individuals that manifest the desired quali-
ties and to establish pedigrees from this isolated stock.
This is Johannsen's method of the pure line and is
particularly applicable to self-fertilizing plants, al-
204 GENETICS
though it may be extended to clones, parthenogenetic
lines and to homozygous crosses.
A quotation from the memoirs of the Manchu em-
peror K'ang-Hsi, 1662-1723, translated from VEm-
pire Chinois, E. R. Hue, will illustrate an early
application of the pedigree method. "On the first day
of the sixth moon I was walking in some fields where
rice had been sown to be ready for the harvest in the
ninth moon. I observed by chance a stalk of rice
already in ear. It was higher than all the rest and
ripe enough to be gathered. I ordered it brought to
me. The grain was very fine and well grown, which
gave me the idea to keep it for a trial and see if the
following year it would preserve its precocity. It
did so. All the stalks which came from it showed ear
before the usual time and were ripe in the sixth moon.
Each year has multiplied the produce of the preceding,
and for thirty years it is the rice which has been
served at my table. It is the only sort which can ripen
north of the great wall, where the winter ends late and
begins very early ; but in the southern provinces, where
the climate is milder and the land more fertile, two
harvests a year may be easily obtained, and it is for
me a sweet rejection to have procured this advantage
for my people"
In the last century the isolation of pure lines was
practiced notably by the Englishman LeCoutour, who
isolated 150 varieties of wheat and by the Scotchman
Shirreff, who worked with various cereals.
In recent years the principle has been extensively
applied with remarkable results, particularly by
OLD TYPES AND NEW 205
Nilsson of Svalof in Sweden, upon peas, potatoes,
clovers, grasses and grains.
Among others in America, Hays has isolated pedi-
grees of wheat at the Minnesota Agricultural Experi-
ment Station, which within a decade have been grown
on thousands of acres and have "made possible the
increased production of wheat throughout the northern
States and Canada."
An isolation method that has been successfully ap-
plied to the sugar-beet industry is that of Vilmorin.
The seeds from each plant to be tested are sown in
separate beds from which upon maturity samples are
taken and tested for sugar content. The plants from
the bed furnishing the sample containing the highest
percentage of sugar are then used as the seed pro-
ducers for the next generation. In this way by con-
tinual selection an improved strain is maintained.
C. INBREEDING
When breeding is kept up between individuals of
the same stock it tends to perpetuate or preserve the
distinctive characteristics of that stock, a practice
that was advocated in the Mosaic law, "Thou shalt
not let thy cattle gender with a diverse kind; thou
shalt not sow thy field with mingled seed." (Levit.
XIX :19.)
Numerous experiments to test the effect of inbreed-
ing have been carried out upon various organisms.
Darwin, for instance, planted morning-glories,
derived from the same stock of seeds, in two
206 GENETICS
beds which were laid out side by side, that is, in an
environment as nearly the same as possible, but with
half of the beds screened from insects which usually
transfer pollen from flower to flower. In the screened
half where all insects were excluded the flowers were
of necessity self -fertilized, while in the exposed half
they were presumably cross-pollinated by insects which
had free access to them. The seeds produced in the
two beds were kept separate and the experiment was
continued for ten years, so that at the end of that time
two lots of morning-glories, one self-fertilized for ten
generations and the other presumably cross-pollinated
for the same length of time, were obtained for com-
parison. The criterion Darwin used was the vigor of
the plants as shown by the length of the vine. He
found that the cross-pollinated plants were to the self-
pollinated ones as 100 to 53, and his conclusion was,
consequently, that cross-pollination is beneficial and
self-pollination is detrimental.
Ritzema-Bos inbred rats for twenty generations.
For the first ten generations the average number of
young per litter was 7.5, while for the last ten genera-
tions it fell to 3.2.
Weismann inbred mice for twenty-nine generations
and obtained a parallel result. For the first ten gen-
erations the average number per litter was 6.1, for the
second ten generations 5.6, and for the last nine gen-
erations 4.2.
Dr. Helen King, on the other hand, practiced close
inbreeding with white rats for 40 generations com-
prising over 20,000 individuals obtained by mating
OLD TYPES AND NEW 207
brothers and sisters from the same litter at the end of
which time the animals were larger and more vigorous
than those not inbred.
Shull found in growing Indian corn that loss of
vigor results from continual self-fertilization, and
many breeders have had similar experiences with ani-
mals as well as, other plants.
In the case of the pomace fly, Drosophila, Castle
inbred brother and sister for fifty-nine generations
without diminishing the fertility of the line.
Hornaday cites the case of the deer in the royal
herd at Windsor which arose from one male and two
females introduced from New Zealand in 1862. The
herd now numbers 20,000 and shows no signs of de-
terioration.
No arbitrary law with respect to the effects of in-
breeding upon vigor and fertility can be laid down,
therefore, which will apply equally to all cases.
Nature has secured, often by elaborate devices, a
separation of the sexes, especially among the higher
organisms, and in consequence there has arisen an
unavoidable necessity of outcrossing. The intricate
adaptations existing between insects and flowers, for
example, seem to be directed entirely toward insuring
outcrossing among plants.
There are, on the other hand, various well known
provisions in nature to insure inbreeding. The ma-
jority of plants are probably self-fertilized while her-
maphroditic animals, which sometimes at least are self-
fertilized particularly among the lower forms, are very
common.
208 GENETICS
The whole matter of inbreeding and the part it plays
in emphasizing defects has received a fresh interpre-
tation in the light of Mendelism.
There is a widespread popular belief that inbreeding
is injurious and that it is necessary to outcross in
order to maintain the vigor and avoid the defects of
any line, but inbreeding in itself may not necessarily
be injurious. The consequence of inbreeding as shown
by the working of Mendelian laws is that latent or
recessive characters tend to become homozygous and
so brought to the surface, while outcrossing brings
about the formation of heterozygous traits which mask
recessive characters and render them ineffective.
In the case of mankind, consanguineous marriage
of various degrees has long been forbidden by law or
custom in many races, particularly among the Jews,
Mohammedans, Indians and Romans. On the other
hand, the Persians, Greeks, Phoenicians and Arabs have
freely practised inbreeding, while one of the longest
of known human pedigrees, a royal line of Egypt, is
notorious for close inbreeding, even to the mating of
brother and sister.
There has been a greater degree of inbreeding in
the Puritan stock of New England than is commonly
realized. David Starr Jordan points out that a child
of to-day, supposing no inbreeding of relatives had
occurred, would have had in the time of William the
Conqueror, thirty generations ago, 8,598,094,592
living ancestors. If this theoretical supposition were
really so, it would seem quite possible for every New
Englander to-day to have at least one ancestral
representative who won glory under William.
OLD TYPES AND NEW 209
The difference between the unthinkable number
given above and the actual number of probable an-
cestors alive thirty generations/, : ago emphasizes the
fact that inbreeding must have occurred freely.
Cousin-marriages, although producing a high per-
centage of defects, do not necessarily produce unde-
sirable traits. They simply bring , out latent or re-
cessive characters for the reason that under these con-
ditions defect meets defect instead of the opposite
normal condition which would dominate the defect and
cause it not to appear.
Since a recessive trait may be properly regarded as
the absence of a positive dominant character, it more
frequently stands for an undesirable feature than
otherwise. Thus it comes about that inbreeding, by
combining negative features, may "produce" a defec-
tive strain.
Outcrossing always increases heterozygous combi-
nations in the germplasm and covers up undesirable^
recessive traits through the introduction of additional \
dominant traits. Inbreeding, on the contrary, tends ;
to simplify the germplasm, that is, to make it more ;
homozygous, and so to bring recessive defects to the ;
surface.
D. HYBRIDIZATION y/
Among the first to use the powerful tool of hybridi-
zation were Koelreuter, 1733-1806, in Germany, and
Knight, 1758-1838, in England. These pioneer trans-
gressors of the Mosaic law cited in the foregoing para-
graph, opened up a broad road to the army of the
Mendelians who were to follow them. Not only have
210 GENETICS
individuals of two varieties showing hereditary differ-
ences been hybridized but successful crosses have been
artificially brought about between individuals belong-
ing to different species, to different genera and even
to different groups still more distantly related to each
other.
It may be possible to point out at least two general
methods of utilizing hybridization.
a. The Method of Burbank
This is a method of greatly increasing the number
of variants by promiscuous hybridization and then of
eliminating all except those of a desired phenotypic
combination. Indirectly it depends upon the principle
of the segregation of unit characters which makes
possible rearrangements of these characters according
to the laws of chance. The characters themselves re-
main unchanged, since nothing new is produced by
hybridization except new arrangements of existing
characters.
The spectacular success of Luther Burbank in
"creating" new plant forms is due largely to his very
extensive hybridizations, his skill in detecting among
the varying progeny the winning phenotype and his
ruthless elimination of the great majority of variations
that do not quite fill his requirement.
The successful combinations mustjDe propagated in
most instances asexually by grafting, cuttings, bulbs,
etc., rather than sexually through the medium of
seeds, because new genotypes which will breed true are
OLD TYPES AND NEW
not necessarilyj^oLated by this^ procedure. The conse-
quence is that Burbank's method cannot be utilized
in animal breeding to any great extent where the main-
tenance of a desirable strain by asexual propagation
is out of the question.
It will be seen that this method is fortuitous and to
a certain extent unscientific in that no one can repeat
the exact conditions of the experiment and arrive at
the same results. It depends upon the chance mixing
up of a large number of possibilities and then in not
being distracted or blinded by the good while selecting
the best. In the hands of a skilful plant breeder
with unlimited resources at his command it may
result in much practical achievement, but it does
not particularly illuminate the path of other breeders
who wish to repeat the experiment. It is after all
selection of phenotypes and, therefore, forever open to
error, since phenotypes do not always indicate what
the behavior of their constituent genotypes will be in
heredity.
b. The Method of Mendel
The method of Mendel, like the foregoing, depends
upon hybridization with the difference that the desired
combination is sought directly by definite predetermined
crosses, according to the expectations of the Mendelian
ratios, rather than through the random result of for-
tuitous combinations. It is a method which has been
rendered possible by the determination of Mendel's laws
of dominance, and of the independence and segregation
of unit characters which give to the experimental
GENETICS
breeder definite expectations and a method of procedure.
If, upon hybridization, the desired character be- -
haves like a recessive, then all that is necessary to\
establish a pure stock exhibiting the character in ques- (
tion, is to breed two recessives together, because reces- J
sives are always homozygous and, regardless of their /
ancestry, breed true.
On the other hand, if the desired character proves ,
to be a dominant, then it is necessary to determine
whether it is present in a duplex or a simplex condition ; j
in other words, whether it is homozygous or hetero-
zygous, for only homozygous organisms breed true.
Establishing a strain consists, consequently, in making
an organism homozygous.
The test to determine whether a dominant character
is homozygous or heterozygous, that is, whether it
will breed true or not, can be made by a single cross
according to the procedure outlined in paragraph 8
of Chapter V. If, upon crossing the individual to be
tested with a recessive, it produces an entirely dominant
progeny, then its germplasm is duplex for this charac-
ter, and it will always reproduce the character in
either duplex or simplex condition according to what-
ever cross may be made with it. When crossed, for
instance, with another duplex dominant like itself, a
pure homozygous strain of the character in question
will be perpetuated.
If, on the contrary, the dominant character to be
tested proves to be simplex or heterozygous, as de-
termined by the fact that, when crossed with a re-
cessive, 50 per cent of the progeny are recessive, then
OLD TYPES AND NEW
it requires more than a single generation to establish
a homozygous dominant strain.
In random inbreeding of diverse strains if the re-
A
(cessives are constantly eliminated as they appear, a
/population is gradually obtained which is composed
\ of an increasing number of dominants so that after
only a few generations the chances are much reduced
that recessives will again appear, which means the prac-
tical purity of the strain.
E. GENOTYPIC SELECTION
/" The success, however, of any method of originating
\ new types of organisms or of improving old ones must
/ depend in the long run upon the selection of germinal
differences.
j The difficulty here of course lies in the fact, that we
may only know the potential germplasm from its per-
formance in producing somatoplasm, but Mendelism
with its analysis of the genes through breeding^ cer-
tainly has gone a long way toward making genotypic
selection possible and definite. Moreover, the preserva-
tion and exploitation of mutations when they are known
is certainly along the line of genotypic selection, since
mutations when isolated may become the progenitors of
desirable new lines. Accordingly until the secret of
the origin of mutations is solved the work of the suc-
cessful breeder consists to a very large extent in simply
taking what mutations nature spontaneously furnishes
to him rather than in attempting to force nature into
producing something new.
214 GENETICS
5. HETEROSIS
When hybrids are formed the first hybrid generation /
not only shows more variability but also more vigor \
than the parental strains and this vigor is in proportion k
to the number of factors in which the parents differ, be-
cause in hybridization there is a pooling of hereditary
resources. Such hybrid vigor is termed hetergsis.
East and Hayes describe, for example, a cross be-
tween two different wild varieties of tobacco in which
the average height of over fifty plants of each of the
two wild parents was 31 and 54* inches respectively.
Of an equal number of hybrids of the first generation
the average height was over 67 inches under the same
environmental conditions. Shull and East, working
separately upon maize, came to the same conclusion,
namely, that the first hybrid generation following an
artificial cross is decidedly more vigorous than the
parental stocks from which it is derived. This is shown
in Figures 45 and 46.
The mule is a notorious hybrid that possesses more
"kick" than its parents.
FIGS. 45 and 46. Results of crossing two inbred strains of corn.
At the left in Fig. 45 are two inbred varieties. The tall corn
at the right is the result of crossing them. In Fig. 46, the
basket at the right represents the average production of two
inbred strains after three generations of inbreeding 61
bushels per acre. The basket at the left shows the first gen-
eration results from crossing them 101 bushels per acre.
After East and Hayes.
CHAPTER X
THE CARRIERS OF THE HERITAGE
1. INTRODUCTION
HEREDITY, as has been shown in the introductory
chapter, is essentially a matter of continuity between
\ succeeding generations of living organisms. This con-
tinuity may be direct, as when a mother protozoan
divides into two daughters, or it may be indirect, as
illustrated by the relationship of a father and son,
an uncle and nephew, or any other relatives of varying
degrees of kinship which, taken singly or collectively,
are somatoplasms derived from a common stream of
gern.plasm.
It is the purpose of the present chapter to consider
this material continuity between succeeding genera-
tions and to discover, if possible, just what are the
carriers of the heritage from one generation to another.
To this end it will be necessary in the first place to
take up what is meant by the "cell theory."
2. THE CELL THEORY
In 1838^839 the "cell theory" of Schleiden and
Schwann, which affirms that all organisms, both plant
and animal, are made up of cellular units, had its birth.
215
216 GENETICS
Robert Hooke, as early as 1665, had described
"little boxes or cells distinguished from one another"
which he saw in thin slices of cork, and to him is due
the rather unfortunate use of the term "cell" which
has survived in biological writings to this day. The
reason this term is unfortunate is because walls, which
are ordinarily the characteristic feature of any cell,
such as a prison cell, are usually the least important
part of the structure of a living cell, often indeed
being entirely absent.
3. A TYPICAL CELI,
A typical undifferentiated cell is represented dia-
grammatically in Figure 47. Near the center of the
Cell wall
Cytoplasm
Centrosome
Nuclear membrane
Nucleus
Chromatin network
FIG. 47. Diagram of a typical cell.
cell the nucleus is shown surrounded by a nuclear mem-
brane. The nucleus, in common with the enveloping
cytoplasm, is made up of living substance called proto-
plasm (Hugo von Mohl, 1846), and around the whole
there is usually formed a wall or membrane which
THE CARRIERS OF THE HERITAGE
serves to separate one cell from another. Within the
protoplasm there may be a considerable amount of non-
living substance in the form of salts L pigments, oil-
drops, water, and other inclusions of various kinds.
The nucleus is to be regarded as the headquarters
of the whole cell, since changes which the cell under-
goes seem to be initiated in it, while cells deprived of
their nuclei cannot long survive. A single instance will
serve to show the vital part which the nucleus plays
in the life-history of the cell. In 1883, Gruber found
that after rocking a thin cover-glass back and forth in
a drop of water containing a collection of the proto-
zoan Stentor, which has a long chain-like nucleus,
these tiny animals could thus be cut into fragments,
which would in some instances recover from the opera-
tion and regenerate into complete individuals. Only
those pieces, however, which contained a fragment of
the nucleus regenerated into new Stentors, while pieces
of relatively large size which lacked a fragment of
nuclear substance very soon disintegrated.
The nucleus, it should be said, is made up of more
than one substance, a fact that is easily demonstrated
by processes of staining, in which certain dyes, through
chemical union, stain a part but not the whole of the
nuclear substance. The part most easily stained is
called chromatin, that is "colored material," and during
certain phases of cell life the chromatin masses to-
gether within the nucleus into visibly definite structures
or bodies termed chromosomes.
Throughout all the various cells that make up the
individuals of any one species these chromosomes ap-
218 GENETICS
pear to be practically constant in number with some
exceptions to be mentioned later in connection with sex.
This law of the constant chomosome number for any
species was first stated by Boveri in 1900.
The chromosomes of different organisms vary in
number from two in the worm Ascaris up to perhaps
1600, according to Haecker ('09), in certain radiolaria.
A recent list records the number of chromosomes typical
for 960 different animals. 1 Species which apparently
are closely related may differ widely with respect to the
number of their chromosomes, while species of unques-
tionably remote relationship may have an identical
number of chromosomes in each of their cells. The
number of chromosomes characteristic for a species,
therefore, is in no way an index to the complexity or
degree of differentiation of the species.
Besides the nucleus there may often be identified in
the cytoplasm of the animal cell a tiny body known as
the centrosome. At certain times in the life-cycle of
a cell the centrosome becomes the focal point of pecul-
iar radiating lines, which play an important part in
the behavior of the cell, particularly during the period
of division.
Every cell passes through a cycle of life which may
be compared with that common to individuals. It is
born from another cell; passes through a vigorous
youth characterized by growth and transformation;
attains maturity when the metamorphoses of its earlier
life give place to a considerable degree of stability ; and
finally, after a more or less extended period of normal
Vowr. of Morphology, vol. 34, pp. 1-67, 1920.
THE CARRIERS OF THE HERITAGE
activity reaches old age, and death completes the
cycle. In most instances, however, before this final
phase is reached, the cell gives place to daughter-
cells through fission, after the manner of most proto-
zoans, and a new cell cycle is begun.
Sometimes the road of differentiation has been trav-
eled so far that it is apparently impossible, as in the
case of the complicated brain-cells, to retrace these
steps of differentiation and begin again. In such in-
stances the outfit of cells provided in the embryo
determines the numerical limit of the cells available
throughout life. When this supply is exhausted no
more cells appear to replace those which have been
worn out.
4. MITOSIS
The ordinary process by which two cells are made
out of one is termed mitosis. It occurs constantly,
and particularly during growth, in all cellular organ-
isms. A series of diagrams, modified from Boveri, illus-
trating the typical phases of mitosis is given in Figures
48 to 57.
The restmg cell (Fig. 48) is characterized by the
presence of a nuclear membrane, a single centrosome,
and by a chromatin network within the nucleus. In the
beginning of the pro phase (Fig. 49) the centrosome
has divided into two parts, while in the early prophase
(Fig. 50) the two centrosomes have moved farther
apart and definite separate chromosomes have formed
out of the chromatin network. The prophase^ proper
(Fig. 51) is marked by the vanishing of the nuclear
220
GENETICS
membrane and the more compact form of the chromo-
somes. At the end of the prophase (Fig 52) the chro-
mosomes have come to lie at the equator of the cell,
fig. 48. The Resting Cell Kg. 49. Beginning Propncae Flg.60. Early Prophets
F1g 61. Prophase Fig. 52. End of Prophase Fig. 63. Mataphate
Fig. 64. Beginning Anaphase
Fig. 66. Anaphase
Fig. 66. Beginning Telophase Kg. 67. End of Telophaw
Pioi. 48-57. Diagrams illustrating mitosis. After Boveri.
being connected by the mantle fibers with the centro-
somes, each of which now occupies a polar position.
In the metaphase (Fig. 53) the chromosomes split
lengthwise, and at the beginning of the anaphase (Fig,
THE CARRIERS OF THE HERITAGE
54) these half-chromosomes commence to separate from
each other and to move toward the poles, while the
mantle fibers shorten. During the anaphase (Fig. 55)
the cell body lengthens and begins to divide, while the
migration of the half-chromosomes toward the poles is
completed. In the begirmmg of the telophase (Fig.
56) the half-chromosomes grow until they attain full
size and the division of the cell body into two parts
becomes complete. The mantle fibers have disappeared
and the nuclear membrane begins to reform around
the chromosomes. Finally, at the end of the telophase
(Fig. 57) the nuclear membrane becomes complete,
the chromosomes break up into a chromatin network,
and two resting cells take the place of the single one
with which the process began (Fig. 48).
5. SEXUAL REPRODUCTION
The mechanism by means of which two cells unite
to make one in sexual reproduction is quite as com-
plicated as that of mitosis by which one cell is trans-
formed into two.
In sexual reproduction there are two kinds of germ-
cells, the egg_and the ^ermatozoan respectively, which
take part in producing a new organism. These cells
are structurally unlike each other in nearly every par-
ticular, but each is a true cell, which von Kolliker made
clear as early as 1841, and each has typically the same
number of chromosomes in its nucleus, a fact more re-
cently determined by van Beneden in 1883.
The egg-cell is often supplied with one or more
222 GENETICS
envelopes of protective or nutritive function, and it is
usually distended with stored up yolk, in consequence
of which it is comparatively large and stationary.
The result is that whatever locomotion is necessary to
bring the two cells together for union devolves upon
the sperm-cell. Consequently the sperm-cells are prac-
tically modified into nuclei with locomotor tails of cyto-
plasm, and frequently, in addition, with some structural
modification for boring a way into the egg-cell. They
are, moreover, much more numerous than the egg-cells,
so that although many go astray, never fulfilling their
mission, the chances are nevertheless good that some
one of them will reach the egg and effect fertilization.
Ordinarily only one sperm enters the egg, but when
several succeed in penetrating into the egg-cytoplasm
only one proceeds to combine with the egg nucleus, that
is, only one sperm nucleus is normally concerned in the
essential process of fertilization.
It was formerly thought by the school of "ovists"
that in fertilization the essential process is a stimula-
tion of the all important egg by the sperm. The
opposing school of "spermists," on the other hand, re-
garded the egg simply as a nutritive cell the function
of which is to harbor the all important sperm. It is
now known that both the egg- and the sperm-cell are
equally concerned in fertilization, which consists in the
union of their respective nuclei within the cytoplasm of
the egg.
THE CARRIERS OF THE HERITAGE 223
6. MATURATION
Certain preliminary changes of a preparatory na-
ture, termed maturation, regularly precede the union
of the nuclei of the two sex-cells in fertilization.
These maturing changes result in reducing the outfit,
of chromosomes in each sex-cell to one-half the originalx
number, a process which is necessary in order to main-/
tain the chromosomal count which is characteristic/
for any particular species and which is known to exist
unbroken from generation to generation. If there were ;
no such reduction, then the fertilized egg, formed by
the union of egg and sperm nuclei, would contain double
the characteristic number of chromosomes, and during
the formation of a new individual, the number in all
the cells arising by mitosis from such a fertilized egg
would likewise be double. When the germ-cells of
such individuals unite in fertilization, the original num-
ber of chromosomes would be quadrupled, and so on in
geometric progression throughout subsequent genera-
tions. In 1883, too late for Darwin to learn of it,
van Beneden discovered the important fact that the
mature germ-cells, as expected, actually contain only
half the normal number of chromosomes.
The mature egg- or sperm-cell, with half its normal
number of chromosomes, is termed a gamete (marry-
ing cell), while the fertilized egg which is formed by
the union of two gametes (mature egg- and sperm-
cell), and which consequently has the characteristic
number of chromosomes, is called a zygote (yoked cell).
A diagrammatic representation of the process of
GENETICS
Fio. 58. Diagram to show typical maturation and fertilization.
maturation is shown in Figure 58. The number of
chromosomes (not shown in the diagram) remains con-
THE CARRIERS OF THE HERITAGE 225
stant in each germ-cell respectively until the division
of second spermatocytes into spermatids which are
subsequently transformed into spermatozoa, and of the
second oocytes into mature eggs and second polar cells,
when it is reduced to one half the normal number. As
spermatozoan and mature egg unite in fertilization,
the original number of chromosomes is restored in the
fertilized egg (zygote).
7. FERTILIZATION
The stages concerned in a typical case of fertiliza-
tion, according to Boveri, are illustrated in Figures
59 to 67.
In Figure 59 the "head" and the "middle piece"
of the sperm-cell have penetrated into the egg cyto-
plasm, while in Figure 60 the tail of the sperm-cell
has become lost and the middle piece, which furnished
the centrosome, has rotated 180 so that it lies between
the nucleus, or head, of the sperm-cell and that of the
egg-cell. Figure 61 shows an increase in the size of
the sperm nucleus and a division of the centrosome into
two parts which begin to migrate towards the poles.
This process of polar migration of the centrosomes
is carried further in Figure 62 as well as the increase in
the size of the sperm nucleus, until in Figure 63 the
process is complete so that the centrosomes have as-
sumed a polar position and the sperm nucleus is equal
in size to the egg nucleus and lies in contact with it.
In Figure 64 the chromatin network of the two nuclei
has formed into an equal number of chromosomes which
236
GENETICS
Fig. 59. Entry of Sptrm Fig. 60. Loss of.'SpenmTail
Fig. 61. Division of
Centrosome
KB. 62. Approach of $p,erm Hg, 63. Star ease of Sjperm, Fig. S^Formation of
Nucleus itPacleus Chromosomes
Big. 66. Splitting of Chromosomes
Fig, 66. Anaphase
Fag. 67. Two-celled Stage
FIGS. 59-67. Diagrams illustrating fertilization. After Boveri.
THE CARRIERS OF THE HERITAGE 227
in each case is half the number characteristic for the
species. Figure 65 shows the complete disappearance
of the nuclear membrane, a process that had already
begun in the preceding figure, and also the arrangement
of the chromosomes, connected with mantle fibers, in the
equatorial plane where the former split longitudinally.
In Figure 66, when the half chromosomes thus formed
pull apart and migrate toward the poles, the segmenta-
tion of the fertilized egg has begun, and there finally
occurs, as shown in Figure 67, the two-celled stage
following fertilization in which each cell contains the
normal number of chromosomes, half of which came
from the egg and half from the sperm.
8. PARTHENOGENESIS
Fertilization is by no means an essential process in
the formation of a new individual, even in those ani-
mals which produce both eggs and sperm. Many
animals and plants reproduce parthenogenetically,
that is, the egg-cell may develop without first uniting J
with a sperm-cell. In these instances the chromo-
somes of the egg are not halved during maturation, and \
the offspring, therefore, have the same number of
chromosomes as the parent, since they are simply
fragments of the parent.
Professor Loeb, by the use of certain chemicals,
has succeeded in doing artificially what apparently
is not ordinarily accomplished in nature, namely, mak-
ing an egg that normally requires fertilization develop
parthenogenetically.
228 GENETICS
V
9. THE HEREDITARY BRIDGE
Whatever may ultimately prove to be determiners
of the hereditary characters which appear in successive
generations, it is obvious that, in any event, such
determiners must be located in the zygote, that is, in
the fertilized egg. This single cell is the actual bridge
of continuity between any parental and filial genera-
tion. Moreover, it is the only bridge.
In the majority of animals the egg develops en-
tirely outside of and independent of the mother, thus
limiting to the egg-cell itself all possible maternal
contributions to the offspring. Although there is
abundant evidence that half of the filial characteristics
come from the male parent, the only actual fragment
of the paternal organism given over to the new indi-
vidual is the single sperm-cell, which unites with the egg
in fertilization, and the whole of this even is not usually
concerned in the process of fertilization. The entire
factor of heritage is packed into the two germ-cells
derived from the respective parents and, in all prob-
ability, into the nuclei of these germ-cells, since the
nuclei are apparently the only portions of these cells
that invariably take part in fertilization. To the new
individual developing by mitosis from the fertilized egg
into an independent organism, the factors of environ-
ment and response referred in to Figure 1 are subse-
quently added.
When it is remembered that the human egg-cell
is only about Vizsth of an inch in diameter, a gigantic
size as compared with that of the human sperm-cell,
THE CARRIERS OF THE HERITAGE 229
and, furthermore, when one passes in rapid review
the marvelous array of characteristics which make up
the sum total of what is obviously inherited in man,
the wonder grows that so small a bridge can stand
such an enormous traffic. A sharp-eyed patrol of
this bridge as the strategic focus of heredity is proving
to be one of the most effective points of attack in the
entire campaign of genetics.
10. THE CHROMOSOME THEORY
Certain investigators, who seek a morphological basis
for heredity, regard the chromosomes as the carriers
of the heritage ; in other words, as the source of the de-
terminers of ontogeny or the effective factors in the
process of differentiation.
A few of the grounds for this theory are briefly
indicated below.
First: In spite of the great relative difference in
size between the egg-cell and the sperm-cell, in heredity
the two are practically equivalent, as has been re-
peatedly shown by making reciprocal crosses between
the two sexes. The only features that are apparently
alike in both the germ-cells are the chromosomes. The
inference is, therefore, that they contain the determiners
which are the causal factors for the equivalence of
adult characters in heredity. The existence of an extra
chromosome in probable connection with the matter
of sex is, as will be pointed out later, an exception to
the exact chromosome equivalence of the two sexes,
which only goes to strengthen the supposition that the
230 GENETICS
chromosomes are the carriers of hereditary qualities
since extra chromosomes are always associated with
the character of sex.
Second: The process of maturation, which always
results in halving the chromosome material of the
germ-cells as a preliminary step to fertilization, is a
series of complicated manoeuvers not practised by other
cells. During this process no other part of the cells
appears to play so consistent and important a role
as the chromosomes. Provided they act as hereditary
carriers, their peculiar behavior during maturation is
just what is needed to bring together an entire comple-
ment of hereditary determiners out of partial contri-
butions from two parental sources.
Third: Sometimes abnormal fertilization occurs, as
in the case when two or more sperm-cells, instead of
one, enter the egg cytoplasm and unite with the egg
nucleus. This unusual performance has been artifi-
cially induced by chemical means in the case of sea-
urchins' eggs. The fertilized egg, or zygote, thus
formed with an excess of male chromosomes, results in
the development of abnormal larvae. It is thought that
a causal connection may exist, therefore, between the
additional male chromosomes in the fertilized ovum and
the abnormalities of the progeny.
Fourth: The fact that chromosomes may retain
their individuality throughout the complicated phases
of mitosis, as has been proven in some instances, agrees
with the corresponding fact that certain characteristics
of the somatoplasm maintain their individuality from
generation to generation.
Moreover, certain chromosomes in the fertilized egg
THE CARRIERS OF THE HERITAGE
have been identified with particular features in the
adult developing from that egg. Tennent summarizes
his work on Echinoderms (1912) by the statement
that from a knowledge of the chromosomes in the
parental germ-cells, particular characters in the adult
hybrids may be predicted, and, conversely, that
from the appearance of sexually mature hybrids the
character of certain chromosomes in their germ-cells
may be predicted.
Again, the correlation of a particular chromosome in
the germ-cells with a definite adult character, namely
sex, has been repeatedly demonstrated in connection
with the so-called "extra chromosome" to which refer-
ence has already been made.
Fifth: Finally, excellent evidence of a definite causal
connection between certain chromosomes of the germ-
cells and particular somatic characters has been fur-
nished by certain critical experiments upon the eggs
of sea-urchins. Boveri found that he was able in some
instances to shake out the nuclei bodily, chromosomes
and all, from the mature eggs of the sea-urchin,
Splicer echinus, and when there was added in sea water to
such enucleated eggs the sperm-cells of an entirely
different genus of sea-urchin, namely, Echwws, the
Echinus sperm-cells entered the Sphcerechmus eggs,
which had been robbed of their nuclei, and from this
peculiar combination larvae developed which exhibited
only Echinus characters!
Such cumulative circumstantial evidence as the fore-
going has convinced many that in the chromosomes we
have visibly before us the carriers of heredity.
In any event the supposition that the chromosomes,
GENETICS
with certain chemical reservations, are the morpho-
logical carriers of the heritage, forms an excellent
working hypothesis, and this chapter may suitably be
closed with a quotation from Professor Wilson, whose
brilliant work in the entire field of cytology makes it
possible for him to speak with authority. "In my
view studies in this field are at the present time most
likely to be advanced by adopting the comparatively
simple hypothesis that the nuclear substances are actual
factors of reaction by virtue of their specific chemical
properties ; and I think that it has already helped us
to gain a clearer view of some of the most puzzling
problems of genetics.'*
CHAPTER XI
THE ARCHITECTURE OF THE GERMPLASM
1. DROSOPHILA, THE BIOLOGICAL CINDERELLA
JUST as the bacteriologist firmly believes that guinea-
pigs were specially created for serological experimenta-
tion, so the geneticist has come to realize that the
banana-fly, Drosophila melanogaster, to which repeated
reference has already been made, was designed for
disclosing the secrets of the "architecture of the germ-
plasm" (Weismann).
This tiny ubiquitous fly (Fig. 32), which hovers
around bruised fruit without regard to place, is so
small and harmless that it does not even qualify as
a pest. It has proved, nevertheless, to be a veritable
bonanza to the geneticist. It has many well-defined
characters that can be observed under the microscope
and it lives successfully upon a bit of banana in a milk
bottle plugged with cotton. Every ten or eleven days
a pair produces two to three hundred descendants that
in turn are ready to produce similar families of their
own so that the investigator who begins with them needs
to be an expert bookkeeper in order to be able to record
his results. Although, like Cinderella, Drosophila
comes from the humble environment of the garbage can,
yet this fly has easily outstripped all its sister competi-
233
GENETICS
v
tors for genetical honors, until to-day it stands prob-
ably as the most famous experimental organism in the
whole world.
Prof. T. H. Morgan of Columbia University is the
most conspicuous leader in the investigation of Droso-
phila. In his laboratory -over ten millions of these
animals, which literally "breed like flies," have passed in
review under the microscope while pedigrees of over
three hundred generations have been obtained and
recorded. In no other plant or animal has the remark-
able parallelism between the segregation of Mendelian
characters in experimental breeding and the behavior
of the chromosomes been so completely demonstrated.
2. LINKAGE
Drosophlla has only four pairs of chromosomes al-
though more than three hundred different characters
have been found in the flies themselves, a fact which
makes it at once evident that many genes, or character-
determiners, must be located together in each chromo-
some.
Experimental breeding of Drosophila shows that
there is not always complete independent assortment of
the different characters that enter into a cross, as
Mendel found to be true for the different characters of
peas with which he experimented.
Genes located together in any one chromosome are
likely to stay together during the conjugation of the
chromosomes and the subsequent separation of the
members of homologous pairs in the process of matura-
ARCHITECTURE OF THE GERMPLASM 235
tion. This hanging together of neighboring genes of
the same chromosome throughout the complicated pro-
cess of meiosis is termed linkage.
It is extremely fortunate for the evolution of our
knowledge of the mechanism of heredity that Mendel
happened to work upon characters located in separate
chromosomes and so was able to establish the law of
the independent segregation of unit characters before the
apparent contradiction, that is, linkage, became known.
If he had come upon the confusing phenomenon of linkage
first, the discovery of the laws of Mendelism, in all prob-
ability, would have been long delayed.
Bateson and Punnett called attention to linkage as
early as 1906 under the name of "coupling" in the case
of certain characters of sweet peas. A vague general
knowledge of many groups of correlations, such as
deafness and defective teeth going along with blue
eyes and albinism in cats, had for a long time existed.
In Drosophila, the brilliant and extensive investiga-
tions of Morgan and his co-workers have resulted in
definitely placing something like two hundred characters
in four linkage groups corresponding to four pairs of
chromosomes. The limitation of linkage groups to the
number of chromosome pairs found in the organism
is proving to be one of the fundamental principles of
heredity.
Moreover, it has been shown by reciprocal crosses
that linkage when it occurs is not due to some relation
per se between the genes but simply to the fact that
the linked genes chance to lie together in the same chro-
mosome. In other words, if two characters enter a
236
GENETICS
cross together from one parent they will stay together
in the offspring, and if they enter from separate parents
they remain separate in the offspring.
t
ft*
ft*
ft*
GL
bv
GL
Gv
bL
bv
bv
bv
bv
bv
GL
Gv
bL
bv
bv
bv
bv
bv
GL
Gv
bL
bv
bv
bv
bv
bv
GL
Gv
bL
bv
bv
bv
bv
bv
FIG. 68. Checkerboard to show the result of crossing a gray-long,
black-vestigial hybrid fly back to a black-vestigial recessive.
G = gray-body ; L = long-wings ; b = black-body ; v = ves-
tigial wings.
3. THE MODUS OPERANDI OF LINKAGE
The way linkage works out may best be made clear
by illustrations from Morgan. When an ordinary
wild-type fly with gray body and long wings is crossed
with a fly showing the two mutations of black body and
ARCHITECTURE OF THE GERMPLASM 287
vestigial wings, the hybrids of the first generation are
all like the wild-type parent because gray body and
long wings are dominant over black body and vestigial
wings.
Parents
Gametes
Gamete*
FIG. 69. Typical linkage in Drosophila. Symbols as in Fig. 68.
Data from Morgan.
When a male of one of these hybrid flies is crossed
back with a recessive black-vestigial female, if Mendelian
segregation took place there would be four possible
kinds of offspring, as shown in Figure 68, viz., gray-
238 GENETICS
long; gray-vestigial; black-long and black-vestigial.
The actual experiment, however, shows but two classes
of offspring, viz., gray-long and black- vestigial, like the
two grandparents (Fig. 69). In other words, gray-
body and long-wings entering the cross from one parent
stay linked together as do also black-body and vestigial
wings. The method of crossing the hybrid back to
the recessive is the common procedure in order to bring
out what is latent in the hybrid, for the recessive, since
it does not dominate or conceal anything, allows what-
ever is present in the hybrid being tested to appear.
The Mendelian practice of crossing the FI hybrids
together tends to conceal linkage and perhaps has
prevented its earlier recognition.
The reciprocal cross is shown in Figure 70. In
this case likewise, whatever goes in together comes
out together and no new combinations appear.
4. CROSS-OVER
When a gray-bodied long-winged female hybrid, such
as is produced by crossing gray-long and black-vesti-
gial together in the preceding experiment, is crossed
back to a recessive black-vestigial male, there are pro-
duced four kinds of offspring, gray-long and black-
vestigial like the grandparents and two new combina-
tions, gray-vestigial and black-long. These four types
of F 2 are what would be expected upon free assort-
ment of all the gametes and they should all occur in
equal numbers or in the proportion of 1 : 1 : 1 : 1. See
Figure 68. Instead, as an actual result of extensive
ARCHITECTURE OF THE GERMPLASM 239
crosses of this kind, Morgan obtained 41.5% each of
gray-long and black-vestigial and 8.5% each of
'Parents
Gametes
Gametes
FIG. 70. Typical linkage in Drosophila. Reciprocal to the case
shown in Fig. 69. Data from Morgan.
the new combinations of black-long and gray-vestigial.
(See Figure 71.) The new combinations represent
cross-overs or breaks in the linkage of the genes within
the chromosomes.
Although this superficially resembles the free assort-
ment or segregation of typical Mendelian crosses, it
240
GENETICS
Parents
Gametes
Gametes
Expectation ]
tnMendelian }
Assortment J
.Artwtl Results- 41. 5 #
Linkage
Cross-overs
FIG. 71. Typical cross-over in Drosophila. Symbols as in Fig. 68.
Data from Morgan.
is quite a different thing since Mendelian segregation
involves whole chromosomes while cross-overs involve
only parts of chromosomes. The percentage too of the
ARCHITECTURE OF THE GERMPLASM
different classes resulting in the F 2 generation from
hybrids is different in typical Mendelian segregation
and in cross-overs.
Furthermore, the percentage of cross-overs varies in
different crosses. For example, when white-eyed yel-
low-bodied flies are crossed with normal wild-type red-
. eyed gray-bodied individuals, the resulting hybrids re-
semble wild red-eyed, gray-bodied flies. When such a
female hybrid is crossed back to a recessive white-eyed
yellow-bodied male, the offspring show only one per-
cent of cross-overs, that is, white-eyed, gray-bodied and
red-eyed, yellow-bodied individuals, and 99% of linkage,
that is, white-eyed, yellow-bodied and red-eyed, gray-
bodied (Fig. 72).
Another percentage of cross-over, that between
white-eye and miniature-wing was found to be 33. It is
obvious that in any case the cross-over will never ex-
ceed 50%.
Jennings has said: "The studies of 'crossing-over'
promise to bring us into closer touch with the actual
, details of the hereditary mechanism than any other
phenomena now under examination."
5. How DO CROSS-OVERS OCCUR?
In germ-cells before maturation, homologous mater-
nal and paternal chromosomes pair off and usually come
to lie side by side. This is the phenomenon of syndesls
or conjugation. During this temporary contact there
seems to be an opportunity for such an exchange of
parts as cross-over breeding demonstrates does actually
99$ \%
FIG. 72. A case of one percent cross-over in Drosophila. Gray-
body and red-eyes are represented by stippling and solid black
respectively. Yellow-body and white-eyes are unshaded. After
Sharp, from Morgan's data.
242
ARCHITECTURE OF THE GERMPLASM 243
occur. Syndesis has been repeatedly observed and
sometimes two chromosomes are seen even to twist
about each other. When separation comes after this
embrace the two original chromosomes may simply
unwind and so regain their former condition unchanged,
or they may break and fuse in such a way that one (A)
has a part of the other (B), and the remaining parts
show a corresponding fusion, as indicated in Figure 73.
This is the chromosomal explanation (Chiasmatype
FIG. 73. Diagram to show cross-over between two homologous
chromosomes. After Muller.
theory of Janssens) of the cross-over phenomena
known to the experimental breeder.
6. INTERFERENCE
The varying percentages of cross-overs between dif-
ferent pairs of genes led Morgan and his associates to
attempt the localization of genes within the chromo-
somes. The idea, as suggested by Bridges in 1914, is
simply this, that the farther apart two genes are in
the chromosome the more likely they are to cross
over and to exchange places with their homologous
genes during syndesis.
244
GENETICS
Of course if they lie very close together in the chro-
mosome they are apt to be found finally on the same side
regardless of the twisting of the paternal and maternal
chromosomes about each other. This is evident in
_ Figure 74 where the invisible
genes are represented hypo-
thetically by letters placed
within the chromosomes. Cross-
over is more likely to occur be-
tween A and E which lie at the
extremes of chromosome I than
between A and B which are
closer together.
Again, when genes lie close
together they theoretically in-
terfere with the crossing over
of neighboring genes as pointed
out by Muller and confirmed by
subsequent breeding experi-
ments. In Figure 74, for ex-
ample, if crossing-over took
place between the pairs Cc and
Dd, breaking the linkage be-
tween C and D and between c
and d, it would prevent another
break of linkage between BC and be. This is the phe-
nomenon of interference. It follows that the nearer
together two pairs of genes involved in cross-over are
located, the greater will be the interference.
FIG. 74. Interference.
Two homologous chro-
mosomes during syn-
desis. When there is
a cross-over between
Cc and Dd, it inter-
feres with another
cross-over near by be-
tween Cc and Bb.
ARCHITECTURE OF THE GERMPLASM 245
7. THE ARRANGEMENT OF THE GENES
Morgan assumes that if one per cent of cross-overs
occurs this may be made to represent one arbitrary
unit of distance between the two genes in question.
Haldane proposes to call this unit of cross-over a
morgan. In the illustration of black-body and vestigial-
wing where there was 17% of cross-over it is assumed
that the genes for these two characters are 17 units, or
morgans, apart in the chromosome.
Following up this fertile idea it becomes possible
even to map the location of the genes in the chro-
mosomes. Sturtevant was the first to make such a
map for the genes in the "sex chromosome" of Droso-
phila.
This has been followed by maps of the other chro-
mosomes, after breeding a total of several million flies
and analyzing the data which include . altogether the
behavior of over a hundred different genes.
The relative location of the genes has been determined
by the following method. If for example two genes,
A and B,, upon breeding back to the recessive show 5%
of cross-overs with a and b, while B and C show 20%
with their allelomorphs, b and c, then when A and C
are bred together with a and c, they should give
either the sum (5 + 20 = 25%) or the difference
(20 5 = 15%) of cross-overs.
For example, in an actual experiment, yellow-body
and white-eye gave 1.2% cross-overs while white-eye
and bifid-wing gave 3.5% cross-overs. When yellow-
body and bifid-wing were tested they met the expecta-
246 GENETICS
tion and gave 4.7%, or the sum of the other two per-
centages, as shown in Figure 75.
If upon breeding yellow and bifid a percentage of
2.3% had been obtained instead of 4.7% as was ac-
tually found, then the order of the genes would have
been yellow-bifid-white instead of yellow-white-bifid.
In the eloquent frontispiece of The Mechanism of
Mendelian Heredity, by Morgan, Sturtevant, Bridges
and Muller, there are drawn four straight parallel lines
representing the "chromosome maps" of Drosophila
yellow
white 4.7%
9
bifid'
FIG. 75. An illustration of the proof of gene-localization from
cross-over percentages obtained by breeding.
as known in 1915. It is doubtful if in any book there
may be found four straight lines that mean so much.
The work of gene-localization is quite comparable to
that done by mathematicians and astronomers in deter-
mining the distances that separate the stars in the
heavens from each other and is perhaps equally in-
comprehensible to the layman. In gene-localization it
is the infinitely small instead of the infinitely great
that one must observe. When it is remembered that
Drosophila is a very tiny fly; that occupying only a
small part within its abdomen are paired reproductive
organs; that each of these reproductive organs in the
13.7
16.7-
20.0.
43.0-
65.0-
^ruby
cross-vcinless
club
inged
tan
vermilion
tt0r:brt*tZM
sable
44.4- garnet
63.6-
3!6- - small-eye
9.5- s/twed
-cleft
-2.0-r telegraph
star
aristaless
4.0-
9.0-
13.0-
14.0-
220J-
28.0-
ae.o-
83.0-
35 Jr
48.5-
B8.0
expanded
0K
pink-wing
streak
eream-b
dacha
ski
apterous
purple
eafranin
61.0 - trefoil
65.0- vestigial
66.6. - telescope
67.0" Xdoaft
76.0" -^dachsous
77.0 roe^"
0.0-1- roughoid
26.3.
70.
89.
95.<
beaded
88.0 - ftwB^V
6.o4- pwpteoid
are
plexus ,
lethal lla
brown
blister t
FIG. 76. Chromosome maps of Drotophila. After Sharp.
247
1
eyeless
.0.0
divergent
Oitnoraxota
69.0- - alass
e n BJO.O: -kidney
m *mS ^jz% ad
63.5 - delta
65.6- ->.JiatrZess
o 5 -f
an
72.0- white oeeUi
86.6 - rough
f\
248 GENETICS
male is made up of several tubules; that within these
tubules may eventually be found the sperm cells with
plenty of room to move about; that within a single
sperm cell is the nucleus ; that after half of the contents
of the nucleus has been disposed of there remain four
chromosomes ; that within each chromosome beyond the
range of vision there are hundreds of genes and that it
has been possible in a single chromosome to determine
not only the relative arrangement of over thirty genes
but also to find out the relative distance between these
Fewle Male S enes > * wil1 be realized
that the analysis of the
germplasm has gone a
long way.
In Figure 76, taken from
FIG. 77. The chromosomes of Sharp's "Introduction to
Drosophila melanogaster.
After Bridges. Cytology," are repre-
sented the four chromo-
some maps of Drosophila corrected to November, 1920.
The four visible chromosomes of Drosophila corre-
spond to the four linkage groups of characters obtained
by experimental breeding and it is a striking fact that
no character has yet appeared that cannot be assigned
to one of these four linkage groups. The relative
length of the four "maps," which has been determined
from the carefully worked-over data acquired by years
of riotous breeding for cross-overs, agrees remarkably
with the relative differences in the actual size of the
chromosomes as measured under the microscope. The
four pairs of chromosomes in a male Drosophila mel-
anogaster are represented in Figure 77.
ARCHITECTURE OF THE GERMPLASM 249
8. LINKAGE IN OTHER ORGANISMS
The phenomenon of linkage has already been observed
in various other organisms besides Drosophila. Even
in Mendel's classic peas White demonstrated four link-
age groups of characters and seven pairs of chromo-
somes. It is doubtful if Mendel himself ever heard of
chromosomes for he died in 1886 and Boveri's pioneer
work on chromosomes had only then recently appeared.
A list of a few of the organisms in which linkage has
been reported is given below.
Organisms that show linkage Author
Sweet pea Bateson and Punnett
Snapdragon, Wheat Baur
Primula Altenberg. Gregory
Maize Emerson. Breggar.
Lindstrom. Jones
Tomato Jones
Beans. Oats Surface
Evening primrose Shull
Drosophila virilis 1
busckii I Metz
repletaj
Silkworms Tanaka
Grouse locust Nabours
Pigeon Cole and Kelly
Rat. Mouse .Castle and Dunn
Rat Ibsen
Rabbit ...Castle
CHAPTER XII
SOMATOGENESIS
1. THE HEREDITARY TUNNED
The earlier studies in heredity were concerned with
the comparison of successive individuals, or somato-
plasms. This phenotypic method has attained a con-
siderable degree of success through the analysis afforded
by Mendelism.
A different and still more recent method of attack
upon the problem of heredity deals not with individuals
but with chromosomes which are generally acknowledged
to be the living springs from which flow the streams
of inheritance. Such an intensive cytological study
of the germplasm has revealed a mechanism that ex-
plains to a marvelous extent the results of the experi-
mental breeder.
The demonstration of the parallel between the be-
havior of the germplasm as seen in the chromosomes
and the performance of the somatoplasm as exhibited
in the end results of experimental breeding, is one of
the most impressive scientific achievements of our times.
There is an undoubted causal connection between
the genotype and the phenotype at the extremes of the
hereditary pageant but between these extremes, that is,
between the fertilized egg and the adult, investigators
are as yet by no means as confident or well-informed.
250
SOMATOGENESIS 251
It is as if heredity was represented by a long under-
ground tunnel. We are in the light at either end
and have made out to a considerable degree the details
at the entrance and exit, but we are still largely in
darkness throughout the passage-way itself.
The science of embryology has given us a series of
flash-light pictures of what goes on in the tunnel of
development but of necessity its contribution has been
largely morphological. Consequently the geneticist
still awaits some torch-bearer who will reveal how an
invisible gene within a chromosome can give form and
substance to a definite visible unit character in an or-
ganism. Probably genetics has contributed more to
embryology than embryology to genetics in the past but
it is quite likely that the account will be more than
balanced in the future.
The way in which germ-cells come by their potent
hereditary components, rather than how they make use
of them, has been the first and most natural problem
to engage the attention. The solution which satisfies
most biologists, who have considered the evidence, has
been found in the idea of the contmmty of the germ-
plasm, that is, that hereditary genes are not the product
and result of the body carrying them but are lineal
descendants of ancestral genes which have been housed
temporarily in other bodily domiciles in the past.
The familiar miracle of how hereditary genes work
together to produce a new plant or animal is farther
from a satisfactory solution, yet there is no doubt that
some of the impending great discoveries in genetics are
sure to be exactly in this field.
25* GENETICS
2. PREFORMATION AND EPIGENESIS
How does germplasm transmute into somatoplasm?
Historically there have been two conspicuous at-
tempts to solve the riddle of differentiation, neither of
which gives intellectual satisfaction any longer in the
light of what is known to-day.
The first held sway in the 17th and 18th centuries
under the guise of the preformation theory which as-
sumes that development is simply the unfolding and en-
larging of what was already present in the germ in
miniature. This has been called the theory of "emboite-
ment" or "infinite encasement,' 5 because, not only is
the miniature plant or animal supposed to be packed
within the germ-cell like the embryo plant between the
cotyledons of the bean seed, but within each miniature
also it is supposed that the next generation is encased,
and the next, ad mfinitum. Aided by a poor microscope
and a good imagination the theory of preformation
was carried to such an extreme that a mannikin or
"homunculus" was actually figured by Hartsoeker
seated within the head of a human spermatozoan !
The second attempt to solve the riddle of develop-
ment resulted in the theory of eplgenesis which goes
to the other extreme, maintaining that organization
gradually appears out of an absolutely simple undiifer-
entiated germ. This theory had its most influential ex-
position in "Theoria Generationis" by C. F. Wolff in
1759. "The mistake in the doctrine of preformation
was in supposing that germinal parts were of the same
kind as adult parts; the mistake of epigenesis was in
maintaining a lack of specific parts in the germ."
SOMATOGENESIS 253
(Conklin). Neither of these two conceptions is in ac-
cordance with the facts as known to-day.
3. WHAT is SOMATOGENESIS?
Development is not simply the unfolding or assort-
ment of what is already present in the germ nor is it
the miraculous writing of something new upon a clean
slate. Rather it is the orderly initiation and sequence
of new structures and functions conditioned by the
interaction of the germinal elements present in the egg
or ovule.
Thus somatogenesis is the study of the emergence of
bodily structure out of hereditary sources. Like the
evolution of species, which has so enthralled the minds
of thinking men, somatogenesis in a parallel way is the
evolution of the individual. No doubt each of these
epic histories will eventually furnish the key and vo-
cabulary to the other.
Both somatogenesis and garnet ogenesis, which con-
cerns the origin of the germ-cells themselves, are cyto-
logical in their terminology, and are referable to the
germplasm (see Figure 78), as contrasted with the
Mendelian and biometric aspects of genetics which are
not primarily cytological but are, on the contrary,
statistical in method, dealing directly with somato-
plasms.
4. THE FACTORS IN SOMATOGENESIS
Somatogenesis deals with the interaction of at least
two sets of factors, viz., (1) hereditary, and (2) en-
vironmental.
54
GENETICS
Hereditary factors have been described and have re-
ceived the major share of attention in the preceding
pages. Environmental factors may upon occasion,
CYTOLOGICAL
________________
Deals with the origin and development of the individual
The new method of attacking heredity
ijnpv /o sdnoMJ snojouinu itfim, 8JDQ
IVDIASI.LVJ.S
FIG. 78. Any-side-up Diagram of Genetical Sciences.
however, cause enormous modifications in somatogenesis
although the limits of variation are set by hereditary
genes. For example, genes under any environmental
circumstances whatsoever never allow an egg with the
heritage of a worm to develop into a bird, nor do human
SOMATOGENESIS 255
genes freighted with the handicap of idiocy ever pro-
duce an intellectual leader.
5. THE ROLE OF GENES IN SOMATIC DIFFERENTIATION
An essential feature of cellular differentiation is the
unequal division of material, both quantitatively and
qualitatively. When we trace the complicated adult
organism backward step by step to the fertilized egg
from which it started we see that its complexity has
arisen largely through this process of unequal division.
Moreover, each stage in the "process of becoming" is
conditioned upon what has already happened in pre-
ceding stages, since differentiation is a forward-moving
sequence of events. Just as the roof of a house must
follow and not precede the erection of walls which are
placed on a foundation previously prepared, so the he-
reditary matter in the gene must pass through a long
series of preliminary steps of differentiation before
finally coming to manifest fruition in the soma.
Weismann, who by the process of logic rather than
experimentally located the germinal substance in the
nucleus of the germ-cell, assumed an elaborate theoreti-
cal system of "biophores," "ids," "idants," etc., where-
by a differential distribution of the nuclear substance
of the germ-cells to the various somatic cells is supposed
to occur. This is diagrammatically shown in Fig-
ure 79.
Subsequent discovery and confirmation of the facts
of mitosis, however, have shown that the germplasm
does not influence the development in this way, for
56
GENETICS
everything indicates that the entire machinery of
mitosis is directed toward securing an equal division
of the heredity-determining chromatin to the two
daughter-cells at each division. Ordinarily the entire
SOMATOGENESIS 257
chromatic complex is handed down from cell-generation
to cell-generation in the development of the soma re-
gardless of the type of tissue to be formed. The ques-
tion now logically follows : How can identical germinal
substance give rise to different products in different
cells? How can a nerve cell, for example, so depart
from its embryonic spherical form that its cytoplasm
becomes drawn out into enormously attenuated neu-
rones tingling with neuro-fibrils, while a cartilage cell,
with the same outfit of germinal determiners in its
nucleus, commits cytological suicide by the excessive
secretion of its cell wall ?
DeVries in his theory of "intra-cellular pangenesis"
(1889) proposes, as a way out of this dilemma, enzy-
matic "pangenes," of which each nucleus contains a
complete set, that escape into the cytoplasm and so
control its differentiation, an explanation "which
nearly meets the present requirements and fits pres-
ent knowledge." It is the cytoplasm and not the
nucleus that differentiates, although the directing
stimulus for differentiation comes from the nucleus.
This conception is diagrammatically shown in Fig-
ure 80, which figure, furthermore, explains how the
stamp of the germplasm upon the somatoplasm can
influence not only immediate cell-division but all subse-
quent ontogenetic divisions until the adult structure
results.
6. "CYTOPLASMIC INHERITANCE"
While the germinal determiners in the chromosomes
are being apportioned to the daughter cells in mito-
sis with strict impartiality, the cytoplasm surrounding
258
GENETICS
the nucleus does not meet the same fate. The unequal
distribution of the cytoplasm, even in the early cleavage
stages of somatogenesis, is quite apparent. Moreover,
SOMATOGENESIS 259
in the cytoplasm of the fertilized egg of many forms
qualitative differences may already be detected that
prophesy clearly the course which differentiation is
to take. Conklin cites the illuminating case of the
ascidian Styela, in whose egg the cytoplasm in differ-
ent regions varies distinctively in color so that these
parts may be unquestionably followed in subsequent
cleavage and their fate definitely discovered.
For example, the peripheral area of the cytoplasm
of this egg containing yellow coloring matter finds its
way into the cleavage cells which become muscles and
mesoderm; a gray area is differentially assorted into
cells that become nervous system and notochord; a
slate-blue part proves to be the source of epidermal
cells and a region of colorless substance gives rise to
ectoderm cells.
Most egg-cells are more reticent than Styela in
revealing the part that their cytoplasm is to play in
ontogenesis, but it has been possible in many instances
to trace cell-lineage through the cleavage stages until
the results of differentiation are unmistakable in the
tissues.
The fact that so many eggs clearly show polarity
and indicate the future symmetry of the organism be-
fore development has begun at all is further evidence of
the important part that the cytoplasm plays in soma-
togenesis. For example, when the fertilized frog's egg
divides for the first time into the two-cell stage these
two cells are the ancestors of the right and left sides
of the animal and the cleavage plane between them
marks the future long axis of the body.
260 GENETICS
Thus while the chromosomes with their invisible
genes are the ultimate determiners of heredity, the
enveloping cytoplasm that surrounds the nucleus, par-
ticularly of the egg-cell, may be the immediate arbiter
of the differentiation processes that characterize soma-
togenesis. "In short," as Conklin says, "the egg cyto-
plasm determines the early development and the sperm
and egg nuclei control only the later differentiations.
. . . The chromosomes are chiefly concerned in
heredity, the cytoplasm in development."
There is nothing in what has been said of "cyto-
plasmic inheritance," however, to conflict with the
generalization that the real determiners of heredity
are germinal, for it is the genes in the nucleus of the
parent germ-cell that gives the character to the egg
of the daughter-cell, both to its nucleus and to its cyto-
plasm, although the latter in turn influences particu-
larly the early stages of somatogenesis. In an ex-
cellent criticism of the role of nucleus and cytoplasm
as vehicles of heredity, Dunn 1 concludes : "For de-
velopment, its mechanism is but grossly known, but we
have learned enough of the determinative effect of the
nucleus and of the possibilities of interaction betweei)
cytoplasm and nucleus to foster a suspicion that one
day the governance of the chromosomes over develop-
ment will be explained in physical terms."
7. THE PHYSICAL STAGE-SETTING
During development the organism is beset on all sides
by various external physical factors, which are more
*Amer. Nat. vol. LI, 1917, p. 286.
SOMATOGENESIS 261
or less necessary to its life, and the modification of these
factors brings about a corresponding variation in the
normal progress of somatogenesis.
These external factors, such as temperature, mois-
ture, light, chemical solutions, pressure, etc., may ac-
celerate, retard or even inhibit the normal course of
events, but invariably such external environmental fac-
tors contribute largely to the end result of somatogene-
sis.
It is quite likely that many kinds of monsters and
defective organisms are the result not of defective
heredity but of alterations in the normal constellation
of physical factors which constitute the environment of
the developing organism.
8. THE INTERNAL ENVIRONMENT
Not only is somatogenesis hedged about by external
modifying factors but there is also an internal environ-
ment that controls to a large degree the behavior of
hereditary factors and determines how they shall come
to expression in the somatoplasm.
The obvious way in which growth is dependent upon
the intake and use of food, and the abnormal outcome
following an unnatural chemical situation within the
body, such as the presence of poisons, are illustrations
of what is meant by internal environment. Perhaps the
best illustration of this is furnished by the endocrine
glands in mammals. Twenty years ago very little was
known with certainty about the part that these ductless
glands play in the organism, but they have become so
262 GENETICS
important in modern medical research that endocrin-
ology is now recognized as a very lusty infant in the
family of biological sciences.
The chief endocrine structures in man are the thy-
roids, parathyroids, the two functionally and anatomi-
cally distinct lobes of the pituitary gland, pineal gland,
thymus, adrenal glands, portions of the pancreas
and the various sex glands (testes, prostate, ovaries,
etc.). These structures are physiological regulators
and have to do with the growth and development not
only of the body but also of the mind. Human in-
stincts, emotions, mental and psychic states are stimu-
lated, inhibited, altered and complicated by endocrine
action. The endocrines, therefore, constitute a large
part of the machinery through which heredity must act
to bring about its results and consequently it is possible
to control, to a considerable extent, the development
and behavior of man through the internal secretions
produced by these glands. "Some people are born with
so stable an endocrine relation," says Bandler, "that
nothing will alter the normal interaction of the endo-
crine glands ; others inherit or acquire endocrines so
unstable or deficient that nothing else can elevate them
to the threshold of the normal."
9. THE RATE or DEVELOPMENT
No doubt one essential feature in the development
of an organism is differentiation or the unequal assort-
ment of material as already mentioned in a preceding
paragraph, but another factor in somatogenesis is
SOMATOGENESIS 263
surely the time element as it appears in the acceleration
or retardation of the processes concerned. Not all
tissues or organs develop at the same rate. Some out-
run others necessarily in order to prepare the way for
what follows. Under normal conditions in ontogene-
sis things swing into place in the nick of time to make
the next step possible. When these rhythms are upset,
just as when Field Marshal Grouchy at Waterloo
failed to swing his troops into line at the critical mo-
ment, then there results a Waterloo in the organism.
To any one who has followed in detail the intricate
stages of ontogenesis in some organism, conditioned as
it is by its indispensable and modifiable environmental
complex, the wonder grows that the successes are so
many and the disasters so few.
10. CONCLUSION
It is not enough for the geneticist to know the chro-
mosomal machinery at the beginning of his story and
the Mendelian moral at the end of it. Between these
two fields of investigation lies the no-man's land of
somatogenesis which forms an important part of the
hereditary tale.
The processes of somatic differentiation are so amen-
able to experimental interference that no doubt future
investigators will continue to be attracted to the cul-
tivation of this promising field of genetics.
CHAPTER XIII
THE DETERMINATION OF SEX
1. PREVALENT IDEAS
THE mechanism of sex determination has been a
matter of speculation since time immemorial and many
erroneous as well as impossible ideas remain even to-day
in the mind of the layman. These speculations fall
into three categories, according to whether the belief
is held (1) that the sex of the offspring is predeter-
mined in the egg; (2) that it is determined at the time
of fertilization; or (3) that it is not determined until
after the zygote has been formed.
All the older experiments on sex were based on the
last of these suppositions. It was believed that by
varying the nutrition of the developing embryo either
sex, as desired, could be obtained. This belief was ap-
plied even to human beings. Experiments on tadpoles
seemed to give definite positive results, but we now
know that the death rate in these experiments was so
large that the results may be more truly explained as
due to differential mortality.
Others held that the age or vigor of the parent de-
termines the sex, the older or more vigorous of the two
parents tending to impress its sex upon the offspring.
Yet another belief, and one still held by many, re-
264
THE DETERMINATION OF SEX
gards the freshness or staleness of the egg as the im-
portant factor in predetermining sex. According to
this idea it is thought that an egg shortly after ovula-
tion tends to produce a female, while one that remains
some time in the oviduct tends to produce a male.
The idea that two distinct types of eggs are formed
is not altogether new. Thus, entirely without biological
foundation, the theory has been propounded that one
ovary gives rise to male-producing eggs and the other
forms female-producing eggs. Equally without founda-
tion is the theory that in one testis male-determining
spermatozoa are produced and in the other, female-
determining spermatozoa.
Modern theories of sex determination hold to the
first and second of the three possibilities mentioned
above. If there are two kinds of eggs, male-producing
and female-producing, then the sex of the individual is
already fixed at the time of the extrusion of the first
polar cell, before the sperm-nucleus has united with the
egg-nucleus in fertilization. If there are two kinds of
sperm, male-determining and female-determining, then
sex depends upon the type of sperm uniting with the
ovum, and it may, therefore, be said that sex is deter-
mined at the time of fertilization.
2. SEX CHROMOSOMES
Our present day stand on sex determination is based
entirely on direct observation, both cytological and ex-
perimental. In 1902, an unpaired chromosome was ob-
served by McClung in the testes of certain Orthoptera.
266 GENETICS
This he called a sex-determiner. The association of this
chromatic body with sex determination proved to be
a discovery of primary importance. In fact it opened
a new era in cytology and heralded the beginning of
a large number of experiments and much profitable dis-
cussion dealing with the mechanism of sex determina-
tion.
In many groups of animals there is an unpaired
chromosome in the male, called the ^-chromosome,
which may be seen in the somatic cells, in the sperma-
togonia and in the spermatocytes (Fig. 81). In the
female cells, both somatic and germinal, the #-chromo-
some is paired. During the process of spermatogenesis
the autosomes, that is, the remaining chromosomes
which have nothing to do with the determination of sex
pair to form tetrads, in the first spermatocyte division,
but are later reduced to dyads, in the second sperma-
tocyte division, when the ^-chromosome passes un-
divided to one of the second series of spermatocytes
(Fig. 81). The other spermatocyte of the second
division accordingly receives no part of the sex-deter-
mining material. During the division into spermatids
the former second spermatocyte, now freighted with
the .r-chromosome, gives rise to a?-bearing cells which
form the female-determining sperm, while the other
second spermatocyte, which did not receive an iT-chro-
mosome, gives rise to two male-determining spermato-
zoa.
Hence, any zygote receiving two sets of autosomes
and two ^-chromosomes becomes a female, while a
zygote receiving two sets of autosomes and only one
THE DETERMINATION OF SEX 267
^-chromosome becomes a male. Since both types of
sperm are ordinarily formed in equal numbers, the
FEMALE
Secondary
Oocyte
Matured
Ovum
FIG. 81. Sex determination in the case of heterogametic males.
chances that a male- or a female-determining sperm
will reach the egg in the process of fertilization, are
268
GENETICS
equal and the resulting zygotes, therefore, are approxi-
mately 50 per cent male and 50 per cent female
(Fig. 82).
A. THE Y-CHROMOSOME
The foregoing is the simplest case of sex-determina-
tion known and, while this is the fundamental type,
still there are many variations of the mechanism. For
example, the ^-chromosome may have a "y" partner
in the male cells, in which case, if n = the haploid, or
FIG. 82. Diagram to show how numerical equality of the sexes
results when one parent is homozygous (the female in this
instance) and the other is heterozygous for the sex character.
halved, set of autosomes in a given animal, then the
following formula holds true:
2n -f- xy = male and 2n + xx = female.
In the spermatogonia of animals maturating in this
manner, half the spermatids receive an ^-chromosome
and half a ^-chromosome, the latter being the male-
determining spermatozoa.
In certain other cases the ^-chromosome may be
represented by several discrete, i.e., separate, compo-
nents, and it may or may not have a ^-chromosome
THE DETERMINATION OF SEX 269
associated with it in the male cells. Thus, in Gelas-
tocoris, a hemipteron, the male is represented by the
formula 2n + kx + y and the female by 2n -f- &r.
Here "n" equals fifteen, so that the male diploid num-
ber of chromosomes is thirty-five and the female,
thirty-eight.
Until recently the ^-chromosome has not been known
to carry specific genes for bodily characters. Indeed,
this chromosome has been generally regarded as merely
a degenerate ^-chromosome that has lost its sex genes
and most of its other genes as well. That it is
essential to the typical development of those species
where it is normally present has been proven in the
non-disjunction experiments of Bridges. A male
Drosophila without the ^-chromosome, for instance, is
sterile.
In many forms it is not unlikely that there is no sex-
determining mechanism visible even with the aid of the
best microscopes, but, nevertheless, it is probable that
x- and ^/-chromosomes exist, and that the ^-chromo-
somes are practically equal to the ^-chromosomes in
size, differing from them merely in the absence of
specific genes.
B. SEX GENES
In the female, except in those cases where difference
in chromosomal size is present, the ^-chromosomes can-
not always be distinguished from ordinary autosomes
and it is furthermore known from breeding experiments
that they bear many genes for characters having noth-
ing to do with sex.
270 GENETICS
That there are specific genes in the z/-chromosome
which, working in conjunction with autosomal genes
are capable of producing males, females or interme-
diates, in cases where the normal relationship is upset,
has been indicated very clearly, especially by Bridges in
recent experiments on DrosophUa.
Sex, in other words, is now put upon a basis of
specific genes. We are, therefore, entirely rid of the
older ideas that the tf-chromosome is composed of a
different kind of chromatin from that found in the
autosomes and that the sex of the zygote depends upon
the amownt of a?-chromatin it receives.
C. HETEROGAMETIC FEMALES
The reverse of the foregoing mechanism, in which
two kinds of sex-determining sperm are present, is
found in the Lepidoptera and birds. In these groups
the presence of 2n -f- xx constitutes a male and 2n -\-x,
a female. The formulae in these cases are usually writ-
ten 2n -f- zz and 2n + 2, in order to distinguish them
from those of heterogametic males.
The cytological proof for the s-chromosomes is not
as strong as for the o?-chromosomes, since both avian
and lepidopteran chromosomes are peculiarly difficult
to study. Nevertheless, the facts are well borne out by
breeding experiments in both groups.
Definite results have been reached by Seiler and also
by Doncaster in experiments with moths, showing that
two types of ova are produced, namely, one which,
after extruding the ^-chromosome into the polar cell
THE DETERMINATION OF SEX 271
MALE FEMALE
%
Mature Ova
FIG. 83. Sex determination in the case of heterogametic females.
and becoming fertilized, produces females, and another
which, retaining the ^-chromosome, produces males.
It is obvious that in this case (see Fig. 83), the sex
272 GENETICS
of the zygote depends entirely upon the method of
maturation of the ovum, the retention or expulsion of
the ^-chromosome being the deciding factor in the de-
termination of sex. If in any way maturation can be
controlled by factors exerting an influence either from
within the egg itself or external to it, then sex ratios
may be altered from the normal 50 : 50. This has
been done by Seller in the case of moths by controlling
the temperature of the developing ova at the critical
time in the process of maturation.
The control of maturation offers a possible explana-
tion of such sex ratios as have been obtained by Riddle
in his forced breeding experiments on doves, where
females are produced in the latter part of the breeding
season from large eggs and males in the early part
of the season from small eggs.
3. SEXUAL CYCLES
A. APHIDS AND PHYLLOXEBANS
Most enlightening observations on the determina-
tion of sex by means of influencing maturation, have
been made upon aphids and phylloxerans by Morgan
and by Von Baehr. It is well known that in the case
of Aphis fertilized eggs always produce females. Under
favorable conditions both males and females are pro-
duced parthenogenetically, the males, however, always
arising from smaller eggs than the females.
It has been observed too that in these smaller eggs
(Fig. 84) an entire ^-chromosome is extruded in the
giving off of the one polar cell, leaving in the egg
THE DETERMINATION OF SEX 273
APHID-PHYLLOXERAN CHROMOSOME CYCLE
Soma of stem-mother and
migrantfemales (her daughters)
Parthenogenetic eggs
of migran t females
giving rise ^ v *
to JT ^^
Anaphase of the single
maturation division
Fio. 84. The chromosome cycle in parthenogenesis of aphids and
phylloxerans.
2n + # chromosomes (five in number) and that such
an egg forms a male. On the other hand, in the larger
274 GENETICS
parthenogenic eggs no whole ^-chromosome is extruded
into the single polar cell given off and consequently the
egg, retaining 2n -\- xx chromosomes ( six in number) ,
develops into a female.
In the spermatogenesis of these forms it has been
found that only one secondary spermatocyte develops
from each primary spermatocyte, namely, the one
which receives the J7-chromosome. Thus, only two in-
stead of four spermatids result from a primary sper-
matocyte and these two form female-determining sper-
matozoa. The "winter eggs" of these insects have two
maturation divisions reducing the chromosomes to the
haploid condition. The female diploid number is re-
stored upon fertilization.
It would seem, therefore, that in the phylloxerans
and aphids at least, maturation is actually controlled
by the size and composition of the egg.
B. ROTIFERS AND DAPHNIDS
It is unfortunate that the rotifers and daphnids,
which lend themselves so favorably to breeding experi-
ments, are not as favorable cytological material as the
homopterons, for it is not at all unlikely that their
sex-determination rests upon a similar basis to that
above described.
In rotifers and daphnids, as in homopterons, fer-
tilized eggs give rise to females, whereas during par-
thenogenesis both females and males may arise, the
latter coming from smaller eggs than the former.
These facts are all the more interesting for the reason
THE DETERMINATION OF SEX 275
that Whitney and A. F. Shull, each working separately
on rotifers, have been able, through modification of ex-
ternal conditions, to alter the normal cycle of repro-
duction, by causing the continuance of the partheno-
genetic process beyond the normal limit.
It seems evident that, through the modification of
external conditions, they have succeeded in influencing
the type of egg produced. If this case is really paral-
lel to that of Aphis and Phylloxera, then the type of
egg artificially produced ought thereafter to control
its own maturation.
In daphnids, where parthenogenesis alternates with
the sexual cycle, at least three kinds of eggs are pro-
duced; (1) thick-shelled, fat-laden, ephippial eggs
which must be fertilized in order to develop; (2) thin-
shelled, glycogen-laden, parthenogenetic eggs, which
develop into females without fertilization; and (3)
thin-shelled, smaller, parthenogenetic eggs which de-
velop into males. The type of egg produced, as
shown by Smith, may be influenced by temperature and
also by food. It is not improbable that we may yet
discover in the maturation of these ova differences in
chromosomal behavior correlated with each type of
ovum and the sex of the resulting offspring.
C. THE HONEY BEE
Closely allied to the problem of the sex cycle, as
described in experiments with the homopterons, is the
question of sex-determination as observed in the
hym^noptera.
276 GENETICS
Even before chromosomes were known, Dzierzon
postulated that males of this group (drones) are
formed from unfertilized eggs, and females (worker
and queens) from fertilized eggs, a view which has
been substantiated by both cytological and genetical
observations. Newell has shown that in the cross be-
tween Italian (gray) queens and German (dark)
drones, as well as in reciprocal crosses, the male off-
spring are purely maternal, while the females are
hybrid in character. Cytological observations by
Petrunkevitch and by Nachtsheim have also estab-
lished the validity of the Dzierzon theory.
Coupled with this, studies on the spermatogenesis
of hymenoptera have revealed the fact that the sper-
matogonia possess solely the haploid number of chro-
mosomes, and in order, therefore, that this number be
not further reduced in the process of maturation, only
one division of chromatin takes place. In the first
spermatocyte division of the honey bee all the chro-
matin passes to a single chromosome, only a minute
degenerate non-chromatic globule being formed at the
other pole of the spindle. In the second spermatocyte
division the chromatin divides but one of the sper-
matids is very small and degenerates. Thus, instead
of four spermatids, only one is formed and this one
contains the haploid number of chromosomes.
Variations of this process are found in other
hymenoptera which frequently result, in the formation
from the larger second spermatocyte, of two separate
spermatids each possessing the haploid number of
chromosomes.
THE DETERMINATION OF SEX 277
4. P<XLYEMBRYONY
Closely allied to the chromosomal basis of sex are
the facts of polyembryony, for when more individuals
than one are formed from a single ovum they are inva-
riably of the same sex. Classical examples are para-
sitic hymenoptera, principally of the families Procto-
trypidoe and Chalcididce, in which thousands of indi-
viduals often result from a single egg. Other examples
are the quadruplets formed in the nine-banded arma-
dillo, Tatusia, and identical or monochorial twins in
man and other mammals.
In the case of mammals the type of sperm, either
with or without the ^-chromosome, is undoubtedly the
deciding factor in sex determination, for the reason
that when all of the chromosomes of the zygote divide
normally the sex of the resulting individuals must be
the same. In other ways also they will be genetically
identical.
In hymenoptera sex depends entirely upon whether
fertilization or parthenogenesis takes place. A fer-
tilized egg will result in females and an unfertilized one
in males, a supposition based upon direct cytological
observation. The facts of polyembryony thus offer
strong substantiation to the idea of chromosomal de-
termination of sex.
5. SEX-LINKED INHERITANCE
The association of Mendelian characters with par-
ticular chromosomes is nowhere better shown than in
278 GENETICS
the case of sex-linked characters, the genes for which
are undoubtedly located in the sex-chromosomes, and
whose inheritance follows exactly the distribution of
these chromosomes. About thirty genes of this kind
have been discovered in Drosophila alone. (See Fig.
76, the left hand line.)
Sex-linked inheritance, which means that genes for
characters other than sex are associated with a par-
ticular sex, i.e., are carried in the same chromosome
that bears the sex-determining genes, should not be
confused with sex-limited characters, i.e., with secon-
dary sexual characters that are found in one sex only
but the genes for which may be located in any chro-
mosome.
An example of a dominant sex-linked character is
the red eye of Drosophila. The manner of its inherit-
ance is as follows.
If a red-eyed female is mated with a white-eyed
male (Fig. 85), the Fj generation are all red-eyed,
and when members of the F 1 generation are inbred the
F 2 generation shows the expected proportion of three
red-eyed individuals to one white-eyed. However, a
peculiar result appears inasmuch as all of the white-
eyed individuals are males. Thus, one half of the F 2
males are white-eyed like their grandfathers while all
of the Fj females are red-eyed because the character
of white-eyes is covered up when the gene for red is
present. The eggs of F A females, however, which
eliminate the genes for red eyes in the polar body
THE DETERMINATION OF SEX 279
during maturation and are then fertilized by a sperm
bearing a ^-chromosome, mature into white-eyed off-
spring.
The reciprocal cross of white-eyed females with red-
Red-eyed Female White-eyed Male
Parental
XX
XT
, .y
Gametes
'*
XX
T fp>
XY ^
jr
Gametes
Tjt "V "V V V "V V V XT
x"o _^-, _". ** ** **_ -^ -t- .*
FIG. 85. Criss-cross inheritance. The underscored X means the
presence of the genes for red eyes in the sex-chromosome.
The male is heterozygous.
eyed males, gives an entirely different result (Fig.
86.) It will be seen that in this case the F t females
are red-eyed like their fathers, while the males are
white-eyed like their mothers. In the F 2 generation
half of the males and half of the females are white-
eyed and the others are red-eyed, due to the fact that
the male mechanism which has only one ^-chromosome,
280
GENETICS
is capable of bearing the gene for red in only half of
its germ-cells. The F females, which normally carry
two ^-chromosomes, all receive an tf-chromosome from
their father and are consequently red-eyed, while the
White-eyed Female Red-eyed
XX
XY
Gametes
Gametes
XX
XX
XY
XY
FIG. 86. Criss-cross inheritance. The reciprocal cross to that
shown in Fig. 85. All individuals with underscored X have
red eyes. The male is heterozygous.
Fj males all receive a single ^-chromosome from their
white-eyed mother and are, therefore, themselves white-
eyed.
B. COLOR-BLINDNESS IN MAN
This criss-cross type of inheritance has long been
known in man, color-blindness being perhaps the best
THE DETERMINATION OF SEX 281
known example of a sex-linked character, behaving in
its inheritance exactly as that of red-eye in Drosophila.
That color-blind females are so rare is due to the fact
that it requires a duplex, or homozygous, dose of the
determiner for color-blindness to produce a color-blind
female, while only a simplex, or heterozygous, dose is
Gametes
Gametes
FIG. 87. General diagram for sex-linked inheritance. The under-
scored symbol (X) represents a sex determiner with some
other character (as color-blindness) linked with it.
needed to produce a color-blind male. These facts
agree perfectly with the idea that the female is homo-
zygous and the male heterozygous with respect to sex,
and that the factor for color-blindness is linked with
the determiner for sex. Sex-linked inheritance, as
shown in this case, may be illustrated by the diagram
above (Fig. 87) in which, for the sake of simplicity,
only sex chromosomes and the determiners for color-
282
GENETICS
blindness are represented. Underscored X represents
a color-blind determiner linked to a sex chromosome.
From this diagram, which agrees substantially with
the facts, it is apparent that a color-blind male mated
to a normal female will produce no color-blind off-
spring, although the females will be "carriers" of
color-blindness, that is, will possess the factor in sim-
plex form and will, therefore, carry it for the female
in a latent condition.
PARENTS
EXPECTED OFFSPRING
$
Normal
9
Color-blind
8
Color-blind
$
Carrier
Normal T
Carrier
color-blind
% normal
\ carrier
normal
Color-blind
Normal
Normal
Carrier .
Color-blind
Color-blind
Color-blind
Color-blind
Color-blind
Carrier
$ color-blind
\ normal
\ color-blind
\ carrier
The sons of such a mating having a normal mother
and a color-blind father will be absolutely free from
the defect and cannot produce color-blindness in any
of their offspring when mated with a normal strain.
If, however, the "carrier" daughters from such a
parentage, who are genotypically heterozygous for
color-blindness but phenotypically normal, mate with
normal individuals, the expectation is that one half
of the sons, and none of the daughters will be color-
blind, but that one half of these daughters will carry
THE DETERMINATION OF SEX 283
the color-blind determiner in simplex form, that is, in
a condition ineffective for producing color-blindness
in female individuals.
All of the various possibilities in the inheritance of
Barred Male
Female
Z
Gametes
Gametes
Z Z
Z O
FIG. 88. Sex-linked inheritance, with the female heterozygous.
The "barred" character is indicated by underscored letters.
color-blindness according to the sex-linked interpreta-
tion are indicated in the table on page 282.
C. THE BARBED PLYMOUTH BOCK
In animals in which the female is heterogametic
(Lepidoptera and birds) sex-linked characters are like-
284
GENETICS
wise known to exist and, in fact, were first discovered
in moths by Doncaster. In these cases it is the fe-
male instead of the male that possesses the mechanism
whereby the character in question can be present only
once.
Black Male Barred Female
Gametes
Z Z
z
zo
Gametes
Z O
z z
z o
FIG. 89. Sex-linked inheritance, with the female heterozygous.
Reciprocal cross to that shown in Fig. 88. The "barred" char-
acter is indicated by the underscored gametes.
For example, "barring" is a dominant sex-linked
trait in poultry, as shown in Figures 88 and 89.
In the cross shown in Figure 88 all the males and half
the females in the F 2 generation are barred while in
the reciprocal cross shown in Figure 89 the Fj males are
barred because they have a ^-chromosome from their
THE DETERMINATION OF SEX 285
maternal side, while the F x females are black because
their single s-chromosomes came from their black
father.
6. NON-DISJUNCTION
A striking confirmation of the chromosomal inter-
pretation of sex is furnished by the phenomenon of
non-disjunction discovered in 1913 by Bridges. In
attempting to explain certain unexpected ratios which
he obtained in a long series of breeding experiments
upon white-eyed Drosophilas, Bridges found that his
results would be more intelligible if what he termed
"non-disjunction" was assumed to occur.
By non-disjunction is meant that both the #-chro-
mosomes instead of disjoining and going normally to
the two poles during the last maturation division,
remain attached to each other and pass together to
one pole leaving the other pole without any #-chromo-
some. In consequence, half the mature eggs should be
provided with two d?-chromosomes and half with none
at all. Cytological examination of these unusual flies
showed that this was what actually did sometimes
happen.
The progeny of non-disjunctional white-eyed females,
as shown in Figure 90 taken from Sharp's "Introduc-
tion to Cytology", show a theoretical diversity of
characters which is borne out in the results of actual
breeding. Morgan sums the matter up when he says :
"An abnormal distribution of sex-chromosomes goes
hand in hand with an abnormal distribution of all sex-
linked factors."
286
GENETICS
THE DETERMINATION OF SEX 287
Explanation to Figure 90
"Non-disjunction and its results in Drosophila. The two large
circles in the first row represent male and female flies producing
sperms and eggs respectively. Non-disjunction in the female gives
2 kinds of eggs, with XX- and no sex-chromosomes, instead of
the normal single kind with one X. At fertilization there are
possible 4 combinations rather than 2, as shown in the large circles
of the second row. Owing to the several ways in which her 3 sex-
chromosomes may be distributed at maturation, the female repre-
sented by the third circle produces 4 kinds of eggs. When mated
to a normal male (below the horizontal line) with two kinds of
sperms, 8 combinations are possible (last row). Numbers 1, 4
and 5 are normal flies and give the usual type of progeny. Num-
bers 2, 6 and 7, owing to the presence of 3 sex-chromosomes, give
exceptional results when bred. Types Numbers 3 and 8 do not
appear in the cultures, probably because they die very early.
The original male has red eyes and the original female white
eyes. Red eyes (represented by the dots) appear in every fly
bearing the X-chromosome of the original male."
(Diagram by Sharp based on data from Bridges and Morgan.)
7. SECONDARY SEXUAL CHARACTERS AND HORMONES
It will be seen from the preceding illustrations that
the primary differences between the sexes is in the kind
of gametes which they form. The female is an egg-
producer, the male a sperm-producer. In many ani-
mals especially invertebrates, it is very difficult to dis-
tinguish males from females without first examining
the gonads, although there is no lack of forms in which
one can with ease distinguish the sexes solely by ex-
ternal appearances.
Very often this sexual dimorphism is confined, first,
to the genitalia or to accessory apparatus used in
copulation, oviposition, or rearing of the young; and
second, to extra genital characteristics not associated
directly with reproduction, such as color, ornamenta-
tion, and the like. Both of these types of sexual
288 GENETICS
dimorphism are, however, secondary to gamete pro-
duction.
In mammals and birds these so-called secondary
sexual characters are found to be largely dependent
for their proper development upon the normal presence
and activity of the gonads. For example, castration
of young male mammals results in individuals lacking
in many ways the attributes of normal males.*^Among
cattle and horses, which have undergone this opera-
tion, the fiery males become docile and lack the thick
neck common to their kind. They also put on fat more
readily. In man the voice fails to change, the beard
is weak, the epiphyses of the bones do not fuse and the
spirit is dulled. Females deprived of ovaries early in
life fail to develop normal mammary glands, while
certain of their skeletal characters are likewise much
altered. Extensive experiments have proved that in
birds and mammals secretions of the gonads, known
as hormones, are essential to normal development. The
castration of young male rats followed by ingraft-
ing of ovaries causes these individuals to become femin-
ized in character.
Perhaps no better case of the influence of hormones
is known than that of the "free martin," adequately
explained by the observations of Lillie. He found in
cattle that when the chorionic coverings of twin em-
bryos of opposite sex fuse so that the blood vessels
anastomose, the more rapidly developing male embryo
sends out hormones into the circulation which inhibit
the normal development of the female embryo. The
much modified female embryo may then be born as a
THE DETERMINATION OF SEX 289
free martin in which the ovaries tend to form tubules
quite like those of a testis.
In birds the activities of the gonads likewise control
to a large extent the development of the secondary
sexual characters, as has well been shown by Goodale
and by Morgan in castration and transplantation ex-
periments on ducks and fowls. Most striking is the
case of female birds which, when castrated while still
young, develop male plumage and posture.
It has been clearly demonstrated that the genes for
secondary sexual characters lie in the autosomes and
thus both male and female have determiners for the
secondary sexual characters of both sexes. For ex-
ample, normally in cases where the male is heteroga-
metic, the presence of a single ^-chromosome in all of
its cells, together with the endocrine secretion of its
gonads, causes the male genes for secondary sexual
characters to develop and those of the female to be
suppressed. By castration and transplantation the
normal condition may be upset and the female secon-
dary sex genes brought into action.
The whole problem of sex-hormones is very compli-
cated since it has been shown that the secretion from
the gonads is merely one link in the chain of endocrine
factors which tend to set into action the genes for
determining secondary sexual characters. In the de-
velopment of sex in the vertebrates the genes for the
production of these sex-hormones are second in impor-
tance only to those genes which determine whether ova
or sperm shall be formed in an individual.
290 GENETICS
8. THE EFFECT OF PARASITISM ON SEX
It has been well demonstrated in insects that castra-
tion, even of very young individuals, produces no effect
upon the secondary sexual characters when the animal
reaches its adult form. Even the implantation of
gonads of the opposite sex results in no change. The
growth and development of the soma seems to be fixed
by the chromosomal complex and does not appear to
be influenced by the action of any sex-hormone. Altera-
tions of secondary sexual characters may occur, how-
ever, by means of parasitism, as shown by experiments
on Crustacea and insects.
Among Crustacea the best case of this kind per-
haps is that of the crab Inachus, the male of which
when parasitized by the cirripede Sacculina, as de-
scribed by Smith, becomes similar to the normal female
in the form of its claw, abdomen and abdominal ap-
pendages.
Among insects Thelia bimaculata, described by Korn-
hauser, is a good example. Parasitized males resemble
females even to the minute structure of their chitinous
integument. Such alterations are due, very likely, to
an entire upset in the metabolism of the host, changing
the internal environment so fundamentally that the
genes for the male secondary sexual characters fail to
find the conditions necessary for their expression in
the developing soma.
9. GYNANDROMORPHS AND SEX INTERGRADES
In insects and Crustacea abnormal individuals occa-
sionally appear, presenting both male and female
FIG. 91. "A gynandromorph mutillid wasp, Pseudomethoca cana-
densis, male on right side, female on left." From Morgan's
"Heredity and Sex," by permission of the Columbia University
Press.
THE DETERMINATION OF SEX
characters. Sometimes the demarcation is exactly
median, one-half being male and the other female.
Such forms are true gynandromorphs. (Fig. 91.)
There are cases, however, where the division may be
either dorso-ventral or antero-posterior, and still
others which show a patchwork of male and female
parts, these latter being mosaic or inter-sex individuals.
Examples of such sex-intergrades have been found
among moths as described by Goldschmidt and by
Banta among daphnids.
Insect gynandromorphs do not necessarily have the
gonad of the corresponding sex in their respective
halves, showing that the soma is not moulded by sex-
hormones.
The cause of gynandromorphism has been studied by
Boveri and by Morgan. Boveri claims to have found
in gynandromorph bees of crossed races that the male
half was maternal, and the female half hybrid. Obvi-
ously, if after the division of the egg-nucleus, a sperm
unites with one of the daughter nuclei that half will be
female, whereas the sister nucleus, developing partheno-
genetically, will form a male half purely maternal in
origin.
This explanation certainly holds good for some cases
but Morgan finds in Drosophila that male portions of
gynandromorphs often bear paternal characters, genes
of which are in chromosomes other than the #-chromo-
some. He concludes, therefore, that at times an ^-chro-
mosome is lost during the meiosis of a female zygote,
leaving a nucleus that fails to get two ^-chromosomes,
which, consequently, develops into the male portion of
the gynandromorph.
292 GENETICS
Similarly a misplaced ^-chromosome in a primary
germ-cell may cause testes to form in a female. Such
a case of gynandromorphism in Thelia (Kornhauser)
proved upon actual chromosome count to have one
^-chromosome missing.
It is rather difficult to offer any simple mechanical
explanation for the mosaics or sex-intergrades of
moths and daphnids. Goldschmidt has attempted to
explain his results upon a quantitative basis, assign-
ing values for the determiners for maleness and female-
ness, and adding the assumption that the strength of
these determiners varies in different races. Thus, the
crossing of a strong male race with a weak male race
brings about an upset of normal conditions, establish-
ing a new balance of factors so that neither one sex nor
the other predominates. An expression of two sets of
genes, therefore, is brought about in various parts of
the organism.
Bridges* recent work on triploid races of Drosophila
seems to indicate that when the normal relation of the
autosomal genes to the sex-genes of the ^-chromosome
is upset, either by the preponderance of one or the
other, then sex abnormalities of many sorts may be
expected.
10. HEEMAPHEODITISM
One of the most obscure problems of the entire sex
question is that of hermaphroditism, or the production
of ova and sperm by a single individual. Instances of
this condition are found normally occurring in many
groups of invertebrates, such as coelenterates, cteno-
THE DETERMINATION OF SEX 293
phores, flat-worms, round-worms, annelids, molluscs
and crustaceans. It is, however, the exception rather
than the rule and must be viewed as a modification of
the bisexual condition necessitated to insure insemina-
tion in animals poorly adapted to bring about typical
fertilization of the eggs.
Sometimes hermaphrodites are female in appearance
and again they resemble more closely the males of the
group to which they belong. In certain Nematodes,
for example, Rhabdites aberrant an occasional male is
found among thousands of hermaphrodites of female
appearance. In this worm Miss Krueger has shown
that occasionally there is a failure of one chromosome
to become incorporated in one of the second sper-
matocytes. Spermatozoa resulting from such deficient
spermatocytes may be the cause of these occasional
male zygotes. Since our knowledge of the chromosomes
in hermaphroditism is deficient, it is hardly worth
while at present to speculate on the mechanism which
produces such individuals.
That the sexual tendencies of hermaphroditic forms
are often in a sensitive balance, influenced by external
conditions, is shown by the experiments of Baltzer on
Bonellia and by Gould on Crepidula.
In the marine worm Bonellia there are produced
minute motile larvae with hermaphroditic possibilities.
If these free-swimming larvae find the proboscis of a
female Bonellia they attach themselves thereto and
develop into minute males after a parasitic existence of
about four days. If, however, no proboscis is encoun-
tered, the motile larva sinks to the bottom and develops
294. GENETICS
into a female. In this case we may say that some
secretion from the female stimulates the development
of the male potentialities and suppresses those of the
female. In fact, intermediates were produced by
Baltzer by allowing larvae to become attached to a
proboscis temporarily and then removing them at in-
tervals of less than four days.
In Crepidula plana, a hermaphroditic gasteropod
which is normally protandric, that is, producing first
sperm and afterward ova, Gould has shown that the
presence of older individuals during the female phase
of development causes the production of sperm in such
young individuals as, when isolated omit sperm produc-
tion, developing instead the female phase and pro-
ducing ova. Here there is an animal in sensitive bal-
ance influenced by a secretion which probably comes to
it through the sea-water from individuals in the female
phase of reproduction.
The problem of hermaphroditism, its mechanism and
relationship to bisexual reproduction, is well worthy
of intensive study. From such exceptions to the gen-
eral rule we may hope to learn much about the normal
mechanism of sex-determination.
11. CONCLUSION
Finally, one may ask, can sex ever be controlled?
There seem to be two avenues of approach to this
problem.
The maturation, and thereby sex, in forms in which
the female is heterogametic, may be controlled by ex-
THE DETERMINATION OF SEX 295
ternal conditions, as in the case of Seller's moths and
Riddle's doves. When, however, the male is hetero-
gametic, it would be possible to control sex only by
some agency which would differentially aid or inhibit
the progress of one of the two kinds of sperm peculiar
to this kind of an organism in its approach to, or
penetration of, the ovum.
CHAPTER XIV
THE APPLICATION TO MAN
1. THE APPLICATION OF GENETICS TO MAN
HUMAN civilization goes hand in hand with the de-
gree of successful interference which man exerts upon
the natural forces surrounding him.
Primitive man was overwhelmed and outmastered
by his environment, but civilized man harnesses nature
to do his will. Savages are not proficient in the arts
of cultivating plants and domesticating animals, while
these are the very things upon which human progress
fundamentally depends. The degree of civilization of
any people is closely correlated with the degree of their
success in exercising a conquering control over plants
and animals. Any knowledge of the laws of heredity,
therefore, as applied by man, either directly to himself
or indirectly to animals and plants, is a distinct con-
tribution to human progress.
In 1900 the National Association of British and
Irish Millers, as Kellicott points out, being dissatisfied
with the quality and quantity of the annual wheat
yield, engaged Professor Biffen to apply his knowledge
of heredity to the practical problem of improving their
wheat crop. The characters desired were a short full
head, beardlessness, high gluten content, immunity to
296
THE APPLICATION TO MAN 297
rust, strong supporting straw, and a large yield per
acre. In the short time that has elapsed, Professor
Biffen has succeeded in producing strains of wheat that
combine all these desirable characters to a remarkable
degree. Such an immediate result would not have been
possible before 1900, when the rediscovery of Mendel's
law revolutionized man's knowledge of the action of
heredity in nature.
This same knowledge which has made possible the
improvement of wheat may be applied with certain
reservations to the breeding of man, for there is no
reasonable doubt that man belongs in the same evolu-
tionary series with all other animals, as Darwin showed,
and is consequently subject to the same natural laws
to a considerable degree.
It must be admitted that thus far in the progress of
civilization more attention has been directed to the
scientific breeding of animals and plants, little as that
has been, than to the scientific breeding of man. Let
us hope that the future will have a different story to
tell!
2. MODIFYING FACTORS IN THE CASE OF MAN
There are certain qualifying factors that make the
problems of genetics somewhat different in the case of
man than in other organisms.
For example, mankind has come to be partially ex-
empt from some of the natural laws which affect other
organisms. Thus with respect to the workings of
natural selection man is partially under "grace" rather
than "law." Nature no longer "selects" good eyes in
298 GENETICS
man by long, patient, and devious processes when poor
eyes are made good almost instantly by a visit to the
oculist. She has long since given up providing natu-
ral weapons of defense for those who have the wits to
supply themselves more efficiently with artificial means
of self-preservation, and she no longer attempts to
improve the natural powers of locomotion of those
who are able to tame a horse to ride upon, or who
build steamships, railroads, automobiles and aero-
planes, thus accomplishing at once what would require
ages at least to evolve.
Neither does the law of the survival of the fittest in
its original sense apply equally to man and to other
organisms. Human society to-day protects its unfit
in hospitals, asylums, and through various philan-
thropies, while physicians devote themselves to the art
of prolonging life beyond the period of usefulness.
We do not desire these results of our modern
civilization to be otherwise, but the fact remains that
some of the most inflexible and universal "natural
laws" are ineffective in the case of man, and it is profit-
able to bear this in mind when applying the laws of
genetics to man.
The laboratory for human heredity is the wide
world, and it is obvious that the experimental method
which has proven so effective in studying the heredity
of animals and plants is impracticable in the case of
man. The consideration of human heredity, therefore,
must always be largely from the statistical side, con-
sisting in an analysis of experiments already performed
rather than in arbitrarily initiating new experiments.
Such institutions as insane asylums, prisons, sani-
THE APPLICATION TO MAN 299
tariums, and homes for the unfortunate are excellent
foci for studying certain phases of human heredity,
because they are simply convenient places where the
results of similar dysgenic experiences have been
brought together.
3. EXPERIMENTS IN HUMAN HEREDITY
A. THE JUKES
A classic example of an experiment in human hered-
ity which has been partially analyzed by the statistical
method is that furnished by Dugdale in 1877 in the
case of "Max Jukes" and his descendants. At that
time it included over one thousand individuals, the
origin of all of whom has been traced back to a shift-
less, illiterate, and intemperate backwoodsman who
started his experiment in heredity in western New York
when it was yet an unsettled wilderness.
In 1877 the histories of 540 of this man's progeny
were known, and that of most of the others was partly
known. About one third of this degenerate strain died
in infancy, 310 individuals were paupers who all to-
gether spent a total of 2300 years in almshouses, while
440 were physical wrecks. In addition to this, over
one half of the female descendants were prostitutes,
and 130 individuals were convicted criminals, including
7 murderers. Not one of the entire family had a com-
mon school education, although the children of other
families in the same region found a way to educational
advantages. Only 20 individuals learned a trade and
10 of these did so in state's prison.
It is estimated that up to 1877 this experiment in
300 GENETICS
human breeding had cost the state of New York over
a million and a quarter dollars, not including the drink
bill, and the end is by no means yet in sight.
The discovery in 1911 of Dugdale's original manu-
script giving the real names and localities of the mem-
bers of the Jukes clan made it possible to follow up
the later history of this famous strain of undesirable
human germplasm. This was done by Dr. A. H.
Esterbrook, who published the results of his investiga-
tions under the title of "The Jukes in 1915," * after
personally visiting every individual whom he was able
to trace.
Since Dugdale's time the Jukes, now in the eighth
generation, have been forced to disperse from their
original habitat because the cement mining industry
upon which most of them formerly depended for a
livelihood was abandoned with the introduction of
Portland cement. Esterbrook has recorded 2094 indi-
viduals bearing Jukes' blood who were scattered
through fourteen states. Of 748 living descendants of
Max Jukes over 15 years of age, he found 76 who were
socially adequate; 255 doing fairly well; 323 "typical
degenerates," and 94 whom he left unclassified due to
lack of sufficient information. He says : "The re-
moval of Jukes from their original habitat to new re-
gions is beneficial to the stock itself, as better social
pressure is brought to bear on them and there is a
chance for mating into better families," and Daven-
port, commenting on the entire matter, adds, "The
most important conclusion that may be drawn from
Carnegie Institution of Washington, Pub. 240, 1916, pp. 85.
THE APPLICATION TO MAN 301
Dr. Esterbrook's prolonged study of the Jukes forty
years later is that not merely institutional care nor
better environment will cause good social reactions in
persons who are feeble-minded or feebly-inhibited, al-
though on the other hand, better stimuli will secure
better reactions from weak stock than will poor stimuli.
. . . The chief value of a detailed study of this sort
lies in this : that it demonstrates again the importance
of the factor of heredity."
B. THE DESCENDANTS OF JONATHAN EDWAEDS
In striking contrast to the case of Max Jukes is
that of Jonathan Edwards, the eminent divine, whose
famous progeny Winship describes as follows: "1394
of his descendants were identified in 1900, of whom
295 were college graduates; 13 presidents of our
greatest colleges, besides many principals of other im-
portant educational institutions; 60 physicians, many
of whom were eminent; 100 and more clergymen,
missionaries, or theological professors ; 75 were officers
in the army and navy ; 60 were prominent authors and
writers, by whom 135 books of merit were written and
published and 18 important periodicals edited; 33
American states and several foreign countries and 92
American cities and many foreign cities have profited
by the beneficent influence of their eminent activity;
100 and more were lawyers, of whom one was our most
eminent professor of law; 30 were judges; 80 held
public office, of whom one was vice-president of the
United States ; 3 were United States senators ; several
302 GENETICS
were governors, Members of Congress, framers of state
constitutions, mayors of cities, and ministers to foreign
courts ; one was president of the Pacific Mail Steamship
Company; 15 railroads, many banks, insurance com-
panies, and large industrial enterprises have been in-
debted to their management. Almost if not every de-
partment of social progress and of public weal has
felt the impulse of this healthy, long-lived family. It
is not known that any one of them was ever convicted
of crime."
Similarly Galton, in "Hereditary Genius," points out
in his analysis of one hundred celebrated persons that
they had 3 great-grandfathers; 17 grandfathers; 31
fathers; 48 sons; 14 grandsons and 3 cousins who also
were celebrated.
C. THE KAI/LIKAK FAMILY
A more convincing experiment in human heredity
than the foregoing, since it concerns the descendants
of two mothers and the same father, is furnished by the
recently published history of the "Kallikak" family. 1
During Revolutionary days, the first Martin Kalli-
kak, the name is fictitious, who was descended from
a long line of good English ancestry, took advantage
of a feeble-minded girl. The result of their indulgence
was a feeble-minded son who became the progenitor of
480 known descendants of whom 143 were distinctly
feeble-minded, while most of the others fell below
mediocrity without a single instance of exceptional
ability.
'"The Kallikak Family." H. H. Goddard. The Macmillan Co.
THE APPLICATION TO MAN 303
"After the Revolutionary war, Martin married a
Quaker girl of good ancestry and settled down to live
a respectable life after the traditions of his fore-
fathers. From this legal union with a normal woman
there have been 496 descendants. All of these except
two have been of normal mentality and these two were
not feeble-minded. . . . The fact that the descendants
of both the normal and the feeble-minded mother have
been traced and studied in every conceivable environ-
ment, and that the respective strains have always been
true to type, tends to confirm the belief that heredity
has been the determining factor in the formation of
their respective characters."
Other recent extensive studies of dysgenic lines in-
clude the "Nams," the "Hill Folk," the "Pineys" of
New Jersey, the "Ishmaels" of Indiana and the
"Zeros" of Denmark.
4. MORAL AND MENTAL CHARACTERS BEHAVE
LIKE PHYSICAL ONES
These instances of human breeding show unmistak-
ably that "blood counts" in human inheritance, even
though the hereditary unit characters that lead to
these general results have not yet been analyzed with
the clearness that is possible in dealing with the char-
acters of some animals and plants.
There is of course no question of moral and mental
traits in plants, and the role that these play in animals
is not easy to determine; but in man the case is un-
doubtedly much more important and complex, since
304 GENETICS
mental and moral characteristics have a large share
in making man what he is. The brute acts according
to his inherited organization ; man is urged by his but
may act according to a higher, moral law. There is,
however, no fundamental scientific distinction which
can be drawn between moral, mental, and physical
traits, and they are undoubtedly all equally subject to
the laws of heredity.
For instance, as an illustration of the heritability
of non-physical traits, in the Jukes pedigree three of
the daughters of Max impressed their peculiar moral
and mental characteristics in a distinctive way upon
their offspring. To quote Davenport: "Thus in the
same environment, the descendants of the illegitimate
son of Ada are prevailingly criminal; the progeny of
Belle are sexually immoral; and the offspring of Effie
are paupers. The difference in the germplasm deter-
mines the difference in the prevailing trait." As
Woods observes : "The most interesting and even
startling thing has been the ease with which heredity
has been able to bear the brunt of explaining the gen-
eral make-up of character."
5. THE CHARACTER OF HUMAN TRAITS
Of the mental, moral, and physical traits which are
heritable in man, some must be regarded as generally
desirable, some as indifferent, and others as defects to
be avoided if possible. In general the majority of
human traits, those which together make up man as
distinguished from other animals, do not particularly
THE APPLICATION TO MAN 305
claim the attention because they are so universal.
Some which stand out from the mass, such as the
physical traits of eye-color and the color and charac-
ter of hair, may be regarded as indifferent so far as
the welfare of the individual is concerned, while others
like skin color and certain racial features that charac-
terize particular strains of "blood" may, under certain
circumstances, work a social handicap upon their pos-
sessors according to the traditions of the community in
which they appear.
A long list of desirable mental traits might be enu-
merated that seem in a general way to be subject to
the laws of inheritance, although they have not yet
undergone the careful analysis demanded by modern
genetics which deals in unit characters rather than in
lump inheritance.
Musical, literary, or artistic ability, for example,
mathematical aptitude and inventive genius, as well as
a cheerful disposition or a strong moral sense are
probably all gifts that come in the germplasm. They
may each be developed by exercise or repressed by
want of opportunity, nevertheless they are fundamen-
tally germinal gifts.
A genius must be born of potential germplasm.
There are no "self-made men." Each has within from
his ancestry, the potentiality of whatever he becomes.
No amount of faithful plodding application can com-
pensate for a lack of the divine hereditary spark at
the start.
306 GENETICS
6. HEREDITARY DEFECTS
Undesirable hereditary traits are frequently defects
due to the absence of some character. For instance,
albinism, which occurs in several kinds of animals and
also in man in one out of every 20,000 individuals
(according to Elderton), is due to the absence of pig-
ment in the skin, hair and eyes. Albinic individuals
have poor eyesight because they are unable to stand
strong light, being without protective pigment in the
eyes. This peculiarity of albinism behaves as a reces-
sive character both in man and in other animals. An
albinic individual may, therefore, marry a normal indi-
vidual without fear of producing albino children, al-
though the children of such a mating would carry
heterozygous germplasm with respect to albinism, and
in cousin marriages might subsequently produce some
albino children.
Davenport, in his work on "Heredity in Relation to
Eugenics," brings together a long catalogue of human
hereditary defects, although in most instances they
are extremely difficult of accurate analysis. This is
true, first, because these defects so often probably de-
pend upon a combination of determiners rather than
upon a single one, and, second, because the available
data are usually scattered and incomplete.
Deafness, for example, is a defect which is heredi-
tary though exactly to what degree, it is at present
impossible to state. The following table taken from
the extensive work of Fay (1898) upon "Marriage of
the Deaf in America" gives some idea of the results of
different matings lumped together statistically.
THE APPLICATION TO MAN 307
CONDITION OF PARENTS
PERCENTAGE OF
DEAF OFFSPRING
Both born deaf . . . ...
25.9
One born deaf, one with acquired deafness .
6.5
One born deaf one normal .
11.9
Both with acquired deafness
2.3
One with acquired deafness, one normal . .
2.2
That two parents born deaf do not produce more
than 26 per cent of deaf children is probably due to
the fact, first, that each parent is in all likelihood het-
erozygous for deafness and that, second, the same com-
bination of factors which is the cause of the parental
defect on either side of the pedigree does not happen to
recombine after segregation to form the new individual.
Deafness will be produced in the offspring only when
matings occur in which the proper factors are com-
bined. Such an undesirable result is much more likely
to happen if both parents come from the same, or
related, hereditary strains than if they are derived
from families in no way connected by blood.
Herein lies the biological objection to cousin mar-
riage which tends to bring together, and thus to per-
petuate, like defects. Outcrossing, on the contrary,
through the law of dominance, tends to conceal defects
and to prevent their expression.
If the patent parental characters were all that re-
appeared in the offspring, the marriage of near kin
would present fewer difficulties. It is the "skeleton in
the closet" that makes trouble. Elderton gives a case
of haemophilia where the direct line was free from taint
308 GENETICS
but collaterals showed the disease latent for six gen-
erations.
Inbreeding is often the result of proximity. Insular
or isolated communities, slums in cities, where those of
one language herd together, or hovels in the back-
woods, where degenerates of a kind are kept in intimate
association, as well as asylums of various sorts in
which similar defectives are promiscuously housed
under the same roof, are all potent agencies to insure
human inbreeding.
Similarly, localities which have been devastated by
migrations of the most effective blood, as, for example,
parts of Ireland or many rural villages in New Eng-
land, are frequently characterized by a population
showing a large percentage of defectiveness. The able-
bodied and ambitious go forth into the world to seek
their fortunes, while the deficient in body or spirit are
left behind where, under the spell of proximity, they
perpetuate their deficiencies.
The part that improved transportation has played
in mixing up populations and in counteracting the
effects of stagnation on human heredity, through in-
breeding under the inertia of proximity, is very great.
There were, obviously, geographic reasons for the
well-known love story of Adam and Eve. Before the
days of railroads, cousin-marriages were much more
frequent than they are now.
Many cases of human defects, such as imbecility or
insanity, are extremely difficult of analysis from the
standpoint of heredity because, in the first place, the
defective conditions descriptively included under these
THE APPLICATION TO MAN 309
vague terms are made up of a multitude of diverse con-
ditions each of which must have a different array of
determiners and, in the second place, because any one
definite sort of insanity or imbecility may be condi-
tioned by a variety of factors.
However, the difficulty of the problem is no reason
for abandoning the attempt to reach its solution and
to learn, if possible, "whence come our 300,000 insane
and feeble-minded, our 160,000 blind or deaf, the
2,000,000 that are annually cared for by our hospitals
and homes, our 80,000 prisoners and the thousands of
criminals that are not in prison, and our 100,000 pau-
pers in almshouses and- out" (Davenport).
7. THE CONTROL OF DEFECTS
The method of possible control of human defects
depends upon whether they are positive or negative,
that is, dominant or recessive. In those cases where
a given defect is due to a single determiner the
Mendelian expectation for the possible offspring
arising from various matings is indicated in the
table on page 310 in which D stands for the defect
and d for its absence.
If the defect is positive and in a duplex or homo-
zygous condition in one parent, as in 1, 2, and 4
all the offspring will possess it regardless of the ger-
minal constitution of the other parent. In two cases
only, namely, in 3 and 5, where the defective parent is
heterozygous, is there any chance of unaffected off-
spring, and even in these cases the defect is quite as
310 GENETICS
THE MEXDELIAX EXPECTATION- FOR DEFECTS
IF THE DEFECT is POSITIVE
(dominant)
IF THE DEFECT is NEGATIVE
(recessive)
When both
parents show
the defect
1
DDXDD=allDD
dd X dd=all dd
2
DDXDd=$DD + lDd
3
Dd X Dd=\DD + %Dd + \dd
When one
parent only
shows the
defect
When neither
parent shows
the defect
4
DD X dd=a\l Dd
dd X DD=all Dd
5
DdXdd=$Dd + ldd
ddXDd=$Dd + $dd
6
ddXdd=alldd
DDXDD=sdlDD
7
DdXDD = $DD + $Dd
8
DdXDd=DD + lDd + \dd
likely to appear as not. It is obvious that the only
way to rid germplasm of a dominant defect is by con-
tinued mating with recessive individuals. By this
method it is possible in time to shake off the defect.
When it once disappears in any individual, it will never
return unless crossed back to a similar defective domi-
nant strain.
In other words, such a recessive extracted from a
heterozygous ancestry will breed just as true as a
recessive which was pure from the start. In both in-
stances there is an entire absence of the character in
question, and it is clear that this character can there-
after never again reappear, since something cannot be
derived from nothing.
On the other hand, if a defect is negative, depending
upon the absence of a normal dominant determiner, as
is usually the case with defects, it behaves as a
Mendelian recessive, that is, it is always apparent in
individuals developing from the homozygously defective
germplasm.
THE APPLICATION TO MAN
311
It is certain, for example, that an imbecile which has
arisen from homozygous defective germplasm carries
only the determiner for imbecility in his own germ-
plasm, and when two such recessives mate, nothing but
imbecile offspring can result, for recessives breed true.
Nothing plus nothing equals nothing.
For practical purposes it is unimportant to know
FIG. 92. Pedigree chart illustrating the law that two defective
parents have only defective offspring. A, alcoholic; (7,
criminalistic ; d, died; F f feeble-minded; T, tubercular.
After Goddard.
whether or not feeble-mindedness, or any similar defect,
is Mendelian in behavior. The fact that it is heredi-
tary is enough.
An illustration of this principle is given in the
above pedigree (Fig. 92) furnished by Goddard, 1910.
The result is quite different, however, when one parent
only shows the defect. If the other parent is a normal
homozygote, as in Case 4 of the accompanying table,
all the offspring will be normal in appearance, but with
the bar sinister of defectiveness in their germplasm,
while if it is heterozygous (Case 5), one half of the
progeny will be defective.
312 GENETICS
Finally, when neither parent shows defectiveness but
one carries the defect as a heterozygote (Case 7), then
there will be no defective children, while if both parents
are heterozygous there is one chance in four that the
offspring will be defective.
As a matter of fact, defectives usually mate with
defectives for the simple reason that normals ordi-
narily avoid them, so it comes about that streams of
poor germplasm naturally flowing together tend to the
inbreeding of like defects.
Davenport * lays down the following general eugenic
rules for the guidance of those who would produce
offspring wisely: "If the negative character is, as in
polydactylism and night-blindness, the normal char-
acter, then normals should marry normals, and they
may be even cousins. If the negative character is
abnormal, as imbecility and liability to respiratory
diseases, then the marriage of two abnormals means
probably all children abnormal; the marriage of two
normals from defective strains means about one quar-
ter of the children abnormal; but the marriage of a
normal of the defective strain with one of a normal
strain will probably lead to strong children. The
worst possible marriage in this class of cases is that
of cousins from the defective strain, especially if one
or both have the defect. In a word, the consanguineous
marriage of persons one or both of whom have the same
undesirable defect, is highly unfit, and the marriage
of even unrelated persons who both belong to strains
1 Davenport. Rep. of Amer. Breeder*' A*oc., Vol. VI, p. 431,
1910.
THE APPLICATION TO MAN 313
containing the same undesirable defect is unfit. Weak-
ness in any characteristic must be mated with strength
in that characteristic; and strength may be mated
with weakness."
In short, the eugenical Cupid does not tell one so
often whom to select for a partner as whom to avoid.
CHAPTER XV
HUMAN CONSERVATION
1. How MANKIND MAY BE IMPROVED
THERE are two fundamental ways to bring about
human betterment, namely, by improving the indi-
vidual and by improving the race. The first method
consists in making the best of whatever heritage has
been received by placing the individual in the most
favorable environment and developing his capacities
to the utmost through education. Such enterprises
may be included under this head as improving sanita-
tion, controlling disease, insuring health, safe-guarding
human life, banishing child-labor, lessening drudgery
of all kinds, substituting something better for the
slums, championing the weak, reforming penal institu-
tions, maintaining charitable organizations, cultivat-
ing true temperance, dispelling ignorance and length-
ening life. The second method consists in seeking a
better heritage with which to begin the life of the
individual.
The first method is immediate and urgent for the
present generation. The second method is concerned
with ideals for the future, and consequently does not
usually present so strong an appeal to the individual,
314
HUMAN CONSERVATION 315
The first is the method of euthenics, or the science
of learning to live well. The second is eugenics, which
Galton defines as "the science of being well born."
Every gain in eugenics, it need hardly be said, will
make euthenics more effective but the reverse cannot
be affirmed.
These two aspects of human betterment, however,
are inseparable. Any hereditary characteristic must
be regarded, not as an independent entity, but as a
reaction between the germplasm and its environment.
The biologist who disregards the fields of educational
endeavor and environmental influence, is equally at
fault with the sociologist who fails sufficiently to real-
ize the fundamental importance of the germplasm.
Without euthenic opportunity the best of heritages
would never fully come to its own. Without the
eugenic foundation the best opportunity fails of ac-
complishment. The euthenic point of view, however,
must not distract the attention now, for the present
chapter is particularly concerned with the program of
eugenics.
2. HUMAN ASSETS AND LIABILITIES
In an attempt to take account of human stock
Dr. H. H. Laughlin, of the Eugenics Record Office,
has made the following eugenical classification based
on the manner in which families assemble in their off-
spring heritable traits which determine for their pos-
sessors (a) social adjustment and (b) special talent
or defect.
316 GENETICS
I. Persons of genius;
II. Persons of special skill, intelligence, courage,
unselfishness, enterprise or strength;
III. Persons constituting the great normal middle
class, the "people";
IV. Socially inadequate persons.
The first three groups constitute those eugenically
fit from sterling inheritance, who produce the socially
valuable nine-tenths of humanity among civilized peo-
ple, and in the last group are the eugenically unfit from
defective inheritance who produce the socially inade-
quate or the "submerged tenth" of humanity.
Among persons of genius Dr. Laughlin would in-
clude the 5000 persons most splendidly equipped by
nature throughout historic times, as, for example,
Aristotle in philosophy, Newton in science, Pasteur in
medicine, Dante in poetry, Shakespeare in drama, and
Cecil Rhodes in business. Reckoning that since civili-
zation began there have been born and reared in civil-
ized countries approximately thirty billion persons, the
expectation of a genius is about 1 : 6,000,000.
In the second group are included the "natural and
acknowledged leaders in all lines of human endeavor,
the "Who's Who people." The incidence of these in
the total population is possibly 1 : 6,000.
The third group, the "people," constitute nine-
tenths of all, since the first two classes, although their
influence is very great, are numerically negligible,
while the fourth group is made up of the residue or the
socially inadequate, namely, (1) feeble-minded; (2)
HUMAN CONSERVATION 317
pauper; (3) inebriate; (4) criminalistic ; (5) epileptic;
(6) insane; (7) asthenic or weak; (8) diathetic, or
predisposed to disease; (9) deformed; (10) cacses-
thenic, that is, with defective sense organs.
Laughlin concludes ; "The task of eugenics is (1) to
encourage fit and fertile matings among those persons
most richly endowed by nature and (2) to devise prac-
ticable means for cutting off the inheritance lines of
persons of naturally meagre or defective inheritance."
3. MORE FACTS NEEDED
Since the point of attack in human heredity must
be largely statistical, it is of the first importance to
collect more facts. Our actual knowledge is confused
with a mass of tradition and opinion, much of which
rests upon questionable foundations. The great pres-
ent need is to learn more facts ; to sift the truth from
error in what is already known; and to reduce all
these data to workable scientific form. Much progress
is being made in this direction, owing to the impetus
given by the revival of Mendel's illuminating work, but
as yet the science of eugenics is in its infancy.
Eugenics, being a biological science, its truths can-
not be arrived at by arbitration and discussion, and no
doubt the entire eugenic movement has suffered much
at the hands of its over-enthusiastic friends. There is
a wide difference between eugenic zeal and eugenic
knowledge and wisdom.
"If there is one thing to be deprecated little less than
ignorance or indifference," says Sir John MacDonald,
318 GENETICS
"it is science in a hurry, eagerness to go to market
with one's crops before they are fully ripe."
The most systematic and effective attempt in this
country to collect reliable data concerning heredity
in man has been initiated under the leadership of
Dr. C. B. Davenport in connection with what is now
the Department of Genetics of the Carnegie Institu-
tion of Washington. This began in 1910 as the Eu-
genics Record Office, with a staff of expert field and
office workers and an adequate equipment of fire-proof
vaults, etc., for the preservation of records, at Cold
Spring Harbor, Long Island, New York, under Dr.
H. H. Laughlin as superintendent. "The main work of
this office is investigation into the laws of inheritance
of traits in human beings and their application to eu-
genics. It proffers its services free of charge to per-
sons seeking advice as to the consequences of pro-
posed marriage matings. In a word, it is devoted to
the advancement of the science and practice of eu-
genics." Already a considerable number of publica-
tions have been issued from the Eugenics Record
Office.
The Volta Bureau, founded about thirty-five years
ago in Washington by Dr. Alexander Graham Bell, is
collecting data with reference to deafness and has now
systematically arranged particulars concerning the
history of over 20,000 individuals. In England, also,
the Galton Laboratory for Eugenics, founded in 1905,
is systematically collecting facts about human pedi-
grees and publishing the results in a compendious
"Treasury of Human Inheritance."
HUMAN CONSERVATION 319
Besides these special bureaus of investigation, innu-
merable facts about the inheritance of particular traits
are being incidentally brought together and made avail-
able in various institutions and asylums throughout
the world immediately concerned with the care of de-
fectives of different types. It is in connection with
such institutions for defectives that much of the most
successful "field work" is being accomplished in the
United States.
4. FURTHER APPLICATION OF WHAT WE KNOW
NECESSARY
Human performance always lags behind human
knowledge. Many persons who are fully aware of the
right procedure do not put their knowledge into prac-
tice. It follows, therefore, that any program of eu-
genics which does not grip the imagination of the
common people in such a way as to become an effective
part of their very lives is bound to remain largely an
academic affair for Utopians to quarrel and theorize
over.
It is not enough to collect facts and work out an
analysis and interpretation of them, for, important
as this preliminary step is, it must be followed by a
convincing campaign of education.
The lives of the unborn do not force themselves
upon the average man or woman with the same insis-
tency as lives already begun. In the midst of the
overwhelming demands of the present, the appeal of
posterity for better blood is vague and remote. If
320 GENETICS
every individual regarded the germplasm he carries
as a sacred trust, then it would be the part of an
awakened eugenic conscience to restrain that germ-
plasm when it is known to be defective or, when it is
not defective, to hand it on to posterity with at least
as much foresight as is exercised in breeding domestic
animals and cultivated plants.
The eugenic conscience is in need of development,
and it is only when it becomes thoroughly aroused in
the rank and file of society as well as among the lead-
ers, that a permanent and increasing betterment of
mankind can be expected.
5. RESTRICTION OF UNDESIRABLE GERMPLASM
A negative way to bring about better blood in the
world is to follow the clarion call of Davenport and
"dry up the streams that feed the torrent of defective
and degenerate protoplasm."
The education of the feeble-minded, the cure of the
insane and the reform of the criminal are all euthenic
not eugenic means of relief. Some idea of the extent
of the drag of the "submerged tenth'* upon human so-
ciety may be gained from the following table, the data
for which are derived from the U. S. census. 1
The burden of the three undesirable D's, "defectives,
dependents and delinquents," upon human society is
by no means entirely represented in the dollar-column
of this table. Each individual recorded is a human
being, the member of some family and community,
1 Statistical Directory of State Institutions for the Defective,
Dependent and Delinquent Classes. Washington, 1919.
HUMAN CONSERVATION
321
STATE INSTITUTIONS FOR DEFECTIVE, DEPENDENT AND
DELINQUENT CLASSES
Institutions for
No. of
Inmates
Jan. 1,
No. of
Institu-
tions
Expenditures
for maintenance
and operation
1916
in 1915
1. Insane
199,340
147
$36,312,662.20
2. Criminalistic
95,985
170
21,244,892.00
3. Dependent
45,373
84
9,675,932.37
4. Tuberculous
7,187
45
3,539,454.95
5. Feeble-minded
19,298
27
3,341,442.85
6. Deaf
6,826
33
1,893,490.09
7. Epileptic
6,097
9
1,345,821.57
8. Feeble-minded and epi-
leptic
6,984
9
1,285,500.05
9. Blind
3,118
28
1,066,973.14
10. Blind and deaf
2,233
12
615,468.41
11. Inebriate
615
3
232,080.62
12. Deformed
601
4
206,747.23
13. Criminalistic and depen-
dent
814
1
105,705.86
14. Feeble-minded, blind and
deaf
191
1
67,051.73
15. Blind, deaf and dependent
215
1
59,649.67
16. Leprous
114
2
56,118.19
TOTAL
394,991
576
$81,048,990.93
which must be more or less directly borne down by the
unfortunate one. Moreover, the unfortunates who are
in institutions are but a small percentage of the total
number in the population who are not in institutions.
It should be remembered that although heredity plays
an important part in such life-tragedies it is not en-
tirely to blame for these depressing data.
The restriction of undesirable additions to our
human stock may be partially accomplished, at least
in America, by employing the following agencies:
control of immigration; more discriminating marriage
GENETICS
laws ; a quickened eugenic sentiment ; sexual segrega-
tion of defectives; and finally, drastic measures of
asexualization when necessary. Providing for the eu-
genic elimination of defectives is as truly a civic duty
as administering charity to them after they are born.
A. CONTROL OF IMMIGRATION
The enforcement of immigration laws tends to debar
from the United States not only many undesirable indi-
viduals, but also incidentally to keep out much poten-
tially bad germplasm that, if admitted, might play
havoc with future generations.
For example, during the year of 1908, 65 idiots,
121 feeble-minded, 184 insane, 3741 paupers, 2900
individuals having contagious diseases, 53 tuberculous
individuals, 136 criminals, and 124 prostitutes were
caught in the sieve at Ellis Island alone and turned
back from this country by the immigration officials
in spite of the fact that an average of only 8 cents a
head was expended upon inspection.
These 7000 and more individuals probably were the
bearers of very little germplasm that we are nationally
not better off without.
Eugenically, the weak point in the present applica-
tion of immigration laws is that criteria for exclusion
are phenotypic in nature rather than genotypic, and
consequently much bad germplasm comes through our
gates hidden from the view of inspectors because the
bearers are heterozygous, wearing a cloak of desirabil-
ity over undesirable traits.
HUMAN CONSERVATION
It is not enough to lift the eyelid of a prospective
parent of American citizens to discover whether he has
some kind of an eye-disease or to count the contents of
his purse to see if he can pay his own way. The offi-
cial ought to know if eye-disease runs in the immi-
grant's family and whether he comes from a race of
people which, through chronic shiftlessness or lack of
initiative, have always carried light purses.
In selecting horses for a stock-farm an expert horse-
man might rely to a considerable extent upon his
judgment of horseflesh based upon inspection alone, but
the wise breeder does more than take the chances of
an ordinary horse-trader. He wants to be assured of
the pedigree of his prospective stock. It is to be
hoped that the time will come when we, as a nation,
will rise above the hazardous methods of the horse
trader in selecting from the foreign applicants who
knock at our portals, and that we will exercise a more
fundamental discrimination than such a haphazard
method affords, by demanding a knowledge of the germ-
plasm of these candidates for citizenship, as displayed
in their pedigrees.
This may possibly be accomplished by having trained
inspectors located abroad in the communities from
which our immigrants come, whose duty it shall be to
look up the ancestry of prospective applicants and to
stamp desirable ones with approval.
This should be done by our own government and not
by labor contractors or steamship companies 'who are
not actuated by any eugenic considerations. More-
over, any immigration law requiring certificates from
324 GENETICS
foreign governments would seriously interfere with our
getting many desirable foreigners to come to this
country.
The national expense of such a program of genea-
logical inspection would be far less than the mainte-
nance of introduced defectives, in fact it would greatly
decrease the number of defectives in the country. At
the present time this country is spending over one hun-
dred million dollars a year on defectives alone, and
each year sees this amount increased.
The United States Department of Agriculture
already has field agents scouring every land for desir-
able animals and plants to introduce into this country,
as well as stringent laws to prevent the importation
of dangerous weeds, parasites, and organisms of vari-
ous kinds. Is the inspection and supervision of human
blood less important?
B. MORE DISCRIMINATING MARRIAGE LAWS
Every people, including even the more primitive
races, make customs or laws that tend to regulate
marriage. Of these, the laws which relate to the
eugenic aspect of marriage are the only ones that
concern us in this connection. "Marriage," says
Davenport, "can be looked at from many points of
view. In novels as the climax of human courtship;
in law largely as two lines of property descent; in
society, as fixing a certain status ; but in eugenics,
which considers its biological aspect, marriage is an
experiment in breeding."
HUMAN CONSERVATION 325
Certain of the United States have laws forbidding
the marriage of epileptics, habitual drunkards, paupers,
idiots, the insane, feeble-minded, and those afflicted
with venereal diseases. It would be well if such laws
were not only more uniform and widespread, but also
more rigidly enforced.
The fact that much marriage taboo already exists
regardless of laws which effectually hinder or prevent
certain kinds of undesirable matings, forms a basis of
hope for future control.
It is quite true that marriage laws in themselves do
not necessarily control human reproduction, for ille-
gitimacy is a factor that must always be reckoned with ;
nevertheless such laws do have an important influence
in regulating marriage and consequent reproduction.
Marriage laws may, however, sometimes bring about
a deplorable result eugenically, as in the case of forced
marriage of sexual offenders in order to legalize the
offense and "save the woman's honor." To compel,
under the guise of legality, two defective streams of
germplasm to combine repeatedly and thereby result in
defective offspring just because the unfortunate event
happened once illegitimately, is fundamentally a mis-
take. Darwin says : "Except in the case of man him-
self hardly any one is so ignorant as to allow his worst
animals to breed."
C. AN EDUCATED SENTIMENT
A far more effective means of restricting bad germ-
plasm than placing elaborate marriage laws upon our
326 GENETICS
statute-books is to educate public sentiment and to
foster a popular eugenic conscience, in the absence
of which the safeguards of the law must forever be
largely without avail since our best hope lies not in
compulsion but in voluntary effort.
Such a sentiment already generally exists to a large
extent with respect to incest, and the marriage of
persons as noticeably defective as idiots or those
afflicted with insanity, and also in America with respect
to miscegenation, but a cautious and intelligent ex-
amination of the more obscure defective traits, exhib-
ited in the somatoplasms of the various members of
^families in question, is largely an ideal of the future.
Under existing conditions non-eugenic considerations
such as wealth, social position, etc., often enter into
the preliminary negotiations of a marriage alliance,
but an equally unromantic caution with reference to
the physical, moral, and mental characters that make
up the biological heritage of contracting parties is
less usual.
The scientific attitude is not necessarily opposed to
the romantic way of looking at things. If the bandage
across the eyes of blind Cupid is allowed to slip a little
in so important and far-reaching an operation as
"falling in love" it is perhaps just as well. The dia-
logue in "Two Gentlemen of Verona" between Julia
and Lucetta is quite to the point where the eager and
curious Julia says to her maid,
"But say, Lucetta, now we are alone,
Woulds't thou counsel me to fall in love ? "
HUMAN CONSERVATION
and the canny Lucetta makes reply,
"Aye, Madam, so you stumble not unheedfully."
This advice is simply "organized common sense," and
romance, which dispenses with this balance-wheel, al-
though it may be entertaining and always exciting at
first, is sure to be disappointing in the end. Marriages
may be "made in heaven," but, as a matter of fact,
children are born and have to be brought up on earth,
and there is nothing particularly romantic in defective
children who might better never have been born. It fol-
lows without saying that it will be much easier to stamp
out bad germplasm when an educated sentiment be-
comes common among all people everywhere.
D. SEGREGATION OF DEFECTIVES
Persons with hereditary defects, such as epileptics,
idiots, and certain criminals, who become wards of
the state, should be segregated or confined in comfort
so that their germplasm may not escape to furnish
additional burdens upon society. "We have become
so used to crime, disease and degeneracy that we take
them for necessary evils. That they were, in the
world's ignorance, is granted. That they must remain
so, is denied" (Davenport).
"The great horde of defectives once in the world
have the right to live and enjoy as best they may
whatever freedom is compatible with the lives and free-
dom of other members of society," says Kellicott, "but
society has a right to protect itself against repetitions
of hereditary blunders."
328 GENETICS
There is one grave danger connected with the ad-
ministration of our humane and commendable philan-
thropies for the unfortunate, since it frequently hap-
pens that defectives are kept in institutions until they
are sexually mature or are partly self-supporting,
when they are liberated only to add to the burden of
society by reproducing their like.
Furthermore, if defectives of the same sort are col-
lected together in the same institutions, unless sexual
segregation is strictly maintained, they may by the
very circumstance of proximity tend to reproduce
their kind just as defectives in any isolated community
tend to multiply. There is much misplaced philan-
thropy that is euthenic but not eugenic. The tempo-
rary troubles of the individual may be alleviated only
to make possible a future addition to the burden of
society.
David Starr Jordan cites the interesting case of
cretinism which occurs in the valley of Aosta in
northern Italy, to prove the wisdom of the sexual
segregation of defectives. Cretinism is an hereditary
defect connected with an abnormal development of
the thyroid gland which results in a peculiar form of
idiocy usually associated with goitre.
"In the city of Aosta the goitrous cretin has been
for centuries an object of charity. The idiot has re-
ceived generous support, while the poor farmer or
laborer with brains and no goitre has had the severest
of struggles. In the competition of life a premium
has thus been placed on imbecility and disease. The
cretm has mated with cretm, the goitre with goitre,
HUMAN CONSERVATION 329
and charity and religion have presided over the union.
The result is that idiocy is multiplied and intensified.
The cretin of Aosta has been developed as a new species
of man. In fair weather the roads about the city are
lined with these awful paupers human beings with
less intelligence than a goose, with less decency than
the pig."
Whymper, writing in 1880, further observes: "It is
strange that self-interest does not lead the natives of
Aosta to place their cretins under such restrictions as
would prevent their illicit intercourse; and it is still
more surprising to find the Catholic Church actually
legalizing their marriage. There is something horribly
grotesque in the idea of solemnizing the union of a
brace of idiots, and, since it is well known that the dis-
ease is hereditary and develops in successive genera-
tions the fact that such marriages are sanctioned is
scandalous and infamous."
Since 1890 the cretins have been sexually segregated,
and in 1910 Jordan reported that they were nearly
all gone.
E. DRASTIC MEASURES
A fifth method of restricting undesirable germplasm
in the case of confirmed criminals, idiots, imbeciles, and
rapists may be mentioned, namely, the extreme treat-
ment of either asexualization or vasectomy. The lat-
ter is a minor operation confined to the male which
occupies only a few moments and requires at most only
the application of a local anaesthetic, such as cocaine.
There are probably no disturbing or even inconvenient
330 GENETICS
after effects from this operation. It consists in remov-
ing a small section of each sperm duct and is entirely
effectual in preventing subsequent parenthood.
In the female the corresponding operation, which
consists in removing a portion of each Fallopian tube,
is much more severe, but not impracticable or dan-
gerous.
According to Laughlin who has carefully collected
data on the subject, in ten of the fifteen states which
have enacted eugenical sterilization statutes the law is
still (1921) on the statute books, unattacked b}
courts and so still available for use. From the be-
ginning of legal sterilization in the United States in
1907 until January 1st, 1921, a total of 3233 caco-
genic persons have been made sexually sterile under
the several statutes.
Laughlin goes on to point out that "the nature of
administrative machinery, which will work and which
will fail, is, from the experiments already made, fairly
well known, so that if the principle of eugenical sterili-
zation has public support, practically any state legis-
lature can, if it chooses, enact a well-functioning law."
The possibility of the abuse of sterilization if legal-
ized is, however, so great that this extreme method of
last resort will be for a long time doubtless of very
questionable application.
6. THE CONSERVATION OF DESIRABLE GERMPLASM
The eugenic ideal may be approached not only nega-
tively by the restriction of undesirable germplasm, but
HUMAN CONSERVATION 331
also positively by the conservation of desirable germ-
plasm.
The various ways in which this improvement of
society may be brought about are:
A. BY ENLARGING INDIVIDUAL OPPORTUNITY
Much good human germplasm goes to waste through
ineffectiveness on account of unfavorable environment
or lack of a suitable opportunity to develop.
Every agency which contributes toward increasing
the opportunity of the individual to attain to a better
development of his latent possibilities is in harmony
with a thoroughly positive eugenic practice. Thus
better schools, better homes, better living conditions,
in short, all euthenic endeavor, directly serves the
eugenic ideal by making the best out of whatever ger-
minal equipment is present in man.
B. BY PREVENTING GERMINAL WASTE
Much good protoplasm fails to find expression in
the form of offspring because one or the other of pos-
sible parents is cut off either by preventable death or
by social hindrances. To avoid such calamities is a
part of the positive program of eugenics.
a. Preventable Death
War, from the eugenic point of view, is the height
of folly, since presumably the brave and the physically
fit march away to fight, while in general the unqualified
333 GENETICS
stay at home to reproduce the next generation. When
a soldier dies on the battlefield or in the hospital, it
is not alone a brave man who is cut off, but it is the
termination of a probably desirable strain of germ-
plasm.
David Starr Jordan has presented this matter very
clearly. He points out that the "man with a hoe"
among the European peasantry is not the result of
centuries of oppression, as he has been pictured, but
rather the dull progeny resulting from generations of
the unfit who were left behind when the fit went off to
war never to return.
Benjamin Franklin, with characteristic wisdom, sums
up the situation in the following epigram: "Wars are
not paid for in war time; the bill comes later."
b. Social Hindrances
There are many conditions of modern society which
act non-eugenically.
For instance, the increasing demands of profes-
sional life prolong the period necessary for prepara-
tion, which, with the "cost of high living," tends
toward late marriage. In this way much of the best
gennplasm is very often withheld from circulation
until it is too late to be effective in providing for the
succeeding generation.
Certain occupations such as school-teaching and
nursing by women are filled by the best blood obtainable,
yet this blood is denied a direct part in molding pos-
terity, since marriage is frequently either forbidden or
HUMAN CONSERVATION 333
regarded as a serious handicap in such lines of work.
Advertisements concerning "unincumbered help" and
"childless apartments" tell their own deplorable tale.
One of the darkest features of the dark ages from
a eugenic standpoint was the enforced celibacy of the
priesthood, since this resulted, as a rule, in withdrawing
into monasteries and nunneries much of the best blood
of the times, and this uneugenic custom still obtains in
many quarters to-day.
C. BY SUBSIDIZING THE FIT
It is possible that if some of the philanthropic en-
deavor now directed toward alleviating the condition
of the unfit should be directed to enlarging the oppor-
tunity of the fit, greater good would result in the
end. In breeding animals and plants the most notable
advances have been made by isolating and developing
the best, rather than by attempting to raise the stand-
ard of mediocrity through the elimination of the worst.
One leader is worth a score of followers in any com-
munity, and the science of genetics surely gives to edu-
cators the hint that it is wiser to cultivate the excep-
tional pupil who is often left to take care of himself
than to expend all the energies of the instructor in
forcing the indifferent or ordinary one up to a passing
standard. The campaign for human betterment in the
long run must do more than avoid mistakes. It must
become aggressive and take advantage of those human
mutations or combinations of traits which appear in the
exceptionally endowed.
334 GENETICS
7. WHO SHALL SIT IN JUDGMENT?
In the practical application of a program of eu-
genics there are many difficulties, for who is qualified
to sit in judgment and separate the fit from the unfit?
There are certain strongly marked characteristics
in mankind which are plainly good or bad, but the
principle of the independence of unit characters dem-
onstrates that no person is wholly good or wholly bad.
Shall we then throw away the whole bundle of sticks
because it contains a few poor or crooked ones? Is it
wise to burn the barn in order to kill the rats?
The list of weakling babies, for instance, who were
apparently physically unfit and hardly worth raising
upon first judgment, but who afterwards became power-
ful factors in the world's progress, is a notable one and
includes the names of Calvin, Newton, Heine, Voltaire,
Herbert Spencer and Robert Louis Stevenson.
Dr. C. V. Chapin recently said with reference to
the eugenic regulation of marriage by physician's
certificate: "The causes of heredity are many and
very conflicting. The subject is a difficult one, and
I for one would hesitate to say, in a great many cases
where I have a pretty good knowledge of the family,
where marriage would, or would not, be desirable."
Desirability and undesirability must always be re-
garded as relative terms more or less undefinable. In
attempting to define them, it makes a great difference
whether the interested party holds to a puritan or
a cavalier standard. To show how far human judg-
ment may err as well as how radically human opinion
HUMAN CONSERVATION 335
changes, there were in England, as recently as 1819,
233 crimes punishable by death according to law.
One needs only to recall the days of the Spanish
Inquisition or of the Salem witchcraft persecution to
realize what fearful blunders human judgment is
capable of, but it is unlikely that the world will ever
see another great religious inquisition, or that in ap-
plying to man the newly found laws of heredity there
will ever be undertaken an equally deplorable eugenic
inquisition.
It is quite apparent, finally, that although great
caution and broadness of vision must be exercised in
bringing about the fulfillment of the highest eugenic
ideals, nevertheless in this direction lies the future
path of human achievement.
8. EUGENICS, NOT "BLUEGENICS"
Eugenics has been called the "dismal science" by ro-
mantic people who chafe under the restrictions of
common sense, and by conscientious individuals who
are depressed by the appalling hereditary blunders
made by mankind, but, as a matter of fact, eugenics
presents the brightest hope for the future of humanity.
Some of the unattractiveness of the eugenical program
lies in the fact that it calls for results in the distant
future in which there can be little or no personal par-
ticipation, and often at the expense of present day
comforts. It is a lofty ideal of altruism and patriot-
ism, and in the words of Major Leonard Darwin, "an
ideal to be followed like a flag in battle without thought
of personal gain"
336 GENETICS
9. THE MORAL
Race-preservation, not self-preservation is the first
law of nature. Because the laws of heredity work re-
lentlessly within predetermined limits is iio reason for
branding eugenics with the mark of a fatalistic philo-
sophy that would avoid personal responsibility. The
Florida orange-grower who uses his intelligence and
plants frost-resisting varieties to replace those over-
taken by frost does not blame fate for his losses. It
is never fatalistic to seek to find out the true determin-
ing causes of a disaster and to apply the obvious
remedy. As Osborn has said: "To know the worst
as well as the best in heredity; to preserve and select
the best, these are the most ^essential forces in the
future evolution of human society."
Our hereditary endowment may be something given
us without our consent and connivance and the accident
of our birth may determine very largely the environment
in which we must work out our salvation but there
lies a sleeping giant of possibility in everyone, and,
whether we have one talent or five or ten, the individual
response we make is our own and we alone are respon-
sible for it.
Finally, to quote the wise^words of Huxley, "To
learn what is true in order to do what is right" is the
summing up of the whole duty of man, for all VhoTtlre
not able to satisfy their mental hunger with the east
wind of authority."
BIBLIOGRAPHY
A few recent works of a general nature are listed below.
Several of these books, particularly those that are starred, con-
tain bibliographies of technical papers and other original sources
of information.
*Babcock, E. B., and R. E. Clausen, 1918. "Genetics in Relation
to Agriculture." New York.
Bateson, W. 1894. "Materials for the Study of Variation."
London.
1913. "Problems of Genetics." Yale Univ. Press.
1913. Mendel's "Principles of Heredity." New York.
Baur, E. 1914. "Einfiihrung in die Experimentelle Vererbungs-
lehre." 2te Auf. Berlin,
Castle, W. E. 1911. "Heredity in Relation to Evolution and
Animal Breeding." New York.
* Castle, W. E. 1920. "Genetics and Eugenics." 2nd Ed. Har-
vard Univ. Press.
Castle, W. E.; J. M. Coulter; C. B. Davenport; E. M. East;
W. L. Tower. 1912. "Heredity and Eugenics." Chicago.
Conklin, E. G. 1915. "Heredity and Environment." Princeton
Univ. Press.
Correns, C. 1912. "Die neuen Vererbungsgesetze." Berlin.
Coulter, J. M. 1914. "Fundamentals of Plant-Breeding." New
York and Chicago.
Darbishire, A. D. 1912. "Breeding and the Mendelian Discov-
ery." London.
Darwin,, C. 1859. "The Origin of Species by Means of Natural
Selection, or the Preservation of Favored Races in the Struggle
for Life." New York.
1868. "The Variation of Animals and Plants under Domesti-
cation." 2nd Ed. New York.
Davenport, C. B. 1911. "Heredity in Relation to Eugenics."
New York.
337
338 GENETICS
Doncaster, L. 1911. "Heredity in the Light of Recent Research."
Cambridge Univ. Press.
1914. "The Determination of Sex." Cambridge Univ. Press.
East, E. M. 1907. "The Relation of Certain Biological Princi-
ples to Plant Breeding." Bull. 158 Conn. Agric. Sta.
East, E. M. and D. F. Jones. 1919. "Inbreeding and Outbreed-
ing." Philadelphia.
Galton, F. 1883. "Inquiries into Human Faculty." New York.
1889. "Natural Inheritance." London.
1892. "Hereditary Genius." London.
Gates, R. R. 1915. "The Mutation Factor in Evolution." Lon-
don.
Goddard, H. H. 1912. "The Kallikak Family." New York.
1914. "Feeble-mindedness ; Its Causes and Consequences."
New York.
Godlewski, E. 1909. "Das Vererbungsproblem im Lichte der
Entwicklungsmechanik." Leipzig.
Goldschmidt, R. 1911. "Einfiihrung in die Vererbungswissen-
schaft." Leipzig.
Guyer, M. F. 1916. "Being Well Born." Indianapolis.
Haecker, V. 1912. "Allgemeine Vererbungslehre." Braun-
schweig.
*Hall, Gertrude E. 1913. "Eugenics and Social Welfare." Bull.
No. 3, State Board of Charities, Albany, N. Y.
Hays, W. M. 1902-4. "Breeding Plants and Animals." Minne-
apolis.
Johannsen, W. 1913. "Elemente der exakten Erblichkeitslehre."
2te Auf. Jena.
Kellicott, W. E. 1911. "The Social Direction of Human Evolu-
tion." New York.
Kronacher, C. 1912. "Grundziige der Zuchtungsbiologie."
Berlin.
Lang, A. 1914. "Die experimentelle Vererbungslehre in der
Zoologie seit 1900." Jena.
Lock, R. H. 1911. "Variation, Heredity and Evolution." Lon-
don.
Lotsy, J. P. 1906-1908. "Vorlesungen iiber Descendenztheorien."
Jena.
1916. "Evolution by Means of Hybridization." The Hague.
BIBLIOGRAPHY 339
Mendel, G. 1865. "Versuche iiber Pflanzen-hybriden." Verb. d.
Naturf. Vereins in Briinn. Translated in Castle, 1920.
Montgomery, T. H. 1906. "The Analysis of Racial Descent in
Animals." New York.
Morgan, T. H. 1913. "Heredity and Sex." New York.
1916. "A Critique of the Theory of Evolution." Princeton
Univ. Press.
*Morgan, T. H. 1919. "The Physical Basis of Heredity." Phila-
delphia.
Morgan, T. H., A. H. Sturtevant, H. J. Muller, and C. B. Bridges.
1915. "The Mechanism of Mendelian Heredity." New York.
Newman, H. E. 1921. "Readings in Evolution, Genetics and
Eugenics." Univ. of Chicago Press. Chicago.
Pearl, R. 1915. "Modes of Research in Genetics." New York.
Plate, L. 1913. "Vererbungslehre." Leipzig.
Popenoe, P. and R. H. Johnson. 1918. "Applied Eugenics."
New York.
Punnett, R. C. 1919. "Mendelism." Fifth Ed. New York.
Reid, Archdall 1905. "The Principles of Heredity." London.
Reid, G. A. 1910. "The Laws of Heredity." London.
Rignano, E. 1911. "Upon the Inheritance of Acquired Charac-
ters." Translated by B. C. H. Harvey. Chicago.
Saleeby, C. W. 1909. "Parenthood and Race Culture; An Out-
line of Eugenics." New York.
1914. "The Progress of Eugenics." New York and London.
Schallmayer, W. 1910. "Vererbung und Auslese im Lebenslauf
der Volker." Jena.
Schneider, K. G. 1911. "Emfiihrung in die Descendenztheorie."
Jena.
Schuster, E. 1912. "Eugenics." London.
Semon, R. 1912. "Das Problem der Vererbungslehre erworbener
Eigenschaften." Leipzig.
Sharp, L. W. 1921. "An Introduction to Cytology." New York.
Thomson, J. A. 1908. "Heredity." London.
Watson, J. A. S. 1912. "Heredity." New York.
Weismann, A. 1904. "The Evolution Theory." London.
1893. "The Germplasm. A Theory of Heredity." English
translation by W. N. Parker and Harriet Ronnfeldt. New
York.
340 GENETICS
Woods, F A. 1906. "Mental and Moral Heredity in Royalty."
New York.
de Vries, H. 1905. "Species and Varieties, their Origin by
Mutation." Chicago.
1901-3. "Die Mutationstheorie." Leipzig.
Ziegler, H. E. 1918. "Die Vererbungslehre in der Biologic und
in der Soziologie." Jena.
INDEX
Aberrations, chromosomal, 58
Ability, artistic, 305
literary, 305
musical, 305
Abnormal fertilization, 230
ACKERT, 137
Acquired characters, 62, 91
Adam and Eve, 308
AGAR, 137
Agouti, 154, 161
Aggregate mutation, 51
Albinism, 52, 199, 235, 306
Albino animals, 49, 155, 161
Alcoholism, 80, 198, 311
Allelomorphs, 99, 107, 149
Alpine flora, 77
Alternative genes, 149, 150, 160
inheritance, 94
ALTENBERG, 249
Amblystoma, 78, 102
Ammonites, 56
Amphimixis, 36, 69
Anaphase, 220, 221, 226, 267,
271, 273
Ancon sheep, 49
Andalusian fowl, 169, 174
Angora, 102
Annelids, 293
Ant-eater, spiny, 198
Anti-body, 89
Aphids, 137, 272, 273, 275
Appendix, vermiform, 21, 197
Apples, greening, 11
Arabs, 208
Arcella, 137
Architecture of germplasm, 233
ARISTOTLE, 316
Arithmetical mean, 23, 26, 42
Armadillo, 277
Arrested development, 197
Artistic ability, 305
Ascaris, 14, 218
Ascidian, 259
Asexualization, 329
Asexual spores, 12
Asthenic, 317
Atavism, 194
Autosomes, 266
Average deviation, 26, 27
Axolotl, 78
Azaleas, double, 47
BABCOCK, 51, 57, 85
Babies, weakling, 334
Bacteria, 80, 136, 138
BALLS, 102
BALTZER, 293, 294
Banana fly (see Drosophila)
Banded shell, 102
BANTA, 137, 142, 291
BARBER, 136
Bare neck in poultry, 50
Bar-eye, 53, 137
Barley, 102, 136, 202
Barring, 283, 284
BATESON, 3, 22, 37, 39, 46, 50,
95, 96, 100, 102, 106, 151,
152, 161, 174, 176, 197, 235,
249
Battle scars, 75
BAUR, 38, 102, 165, 174, 249
Beans, 123, 124, 125, 128, 136,
249
Beardlessness, 102
BEECHER, 56
Beech leaves, 25, 31, 33
purple, 47
Beetles, potato, 33, 36, 137, 143,
144
Begonia, 10, 11
BEINHART, 55
Belemnites, 56
341
342
INDEX
Belgian hare, 177, 187
BELL, 318
BELLING, 59
Bertillon system, 19
BIFFEN, 102, 296, 297
Bimodal polygons, 29
Biometry, 23, 254
Birds, 199, 270, 283, 288
Birthmarks, 83
Black-eyed susan, 174
BLAKESLEE, 59, 173
BLARINGHEM, 76
Blending inheritance, 93, 168
Blind, 309, 321
Bluegenics, 335
Bob-tail, 172
Bonellia, 293
Booted poultry, 176
BOVERI, 14, 218, 219, 220, 226,
231, 249, 291
Brachydactyly, 102, 172
Branched habit, 102
Breeding, experimental, 254
pedigree, 200, 203
BHEGGAR, 249
BRIDGES, 53, 59, 145, 243, 246,
248, 269, 270, 285, 287, 292
Bristles, dichaet, 137
thoracic, 137, 143, 145
BROOKS, 63
Brown-eyed yellow, 157
Bryozoans, 11
Bud variations, 55
BUBBANK, 210, 211
Cabbage butterflies, 51
Cacaesthenic, 317
Calf, two-headed, 22
CALKINS, 136
Callosities, 78
CALVIN, 334
Canaries, 102
Capaella, 77
Carnation, 47
Carnegie Institution, 318
Carriers, 282
CASTLE, 7, 85, 86, 96, 102, 115,
117, 118, 137, 146, 154, 157,
158, 159, 161, 162, 166, 177,
178, 189, 207, 249
Castration, 289
Cats, 235
tailless, 49, 172
Cattle, 102, 205, 288
color, 170
hairless, 50
hornless, 49, 172
Causes of mutation, 57, 60
variation, 34
Celandine, 47, 172
Celibacy, 333
Cell differentiation, 72
germ, 221, 55
polar, 54, 224, 267, 271, 273
sex, 224
theory, 215
typical, 216
wall, 216
Centropyxis, 137
Centrosome, 216, 218
Cereals, 204
Cerebral hernia, 50
CHAPIN, 334
Characters, acquired, 62
congenital, 68
dominant, 195
germinal, 67
individual unit, 117
mental, 303
moral, 303
prenatal, 68
recessive, 119, 196
secondary sexual, 287
somatic, 67
Chelidonium, 47
Chiasmatype theory, 243
Chimaera, 55
Chinese women, feet of, 15
Chromatin, 216, 217
Chromosomal aberrations, 58
Chromosomes, 217, 226, 243, 248
cycle, 273
Chromosome, extra, 229
maps, 247, 248
number, 218
sex, 245, 265
theory, 229, 254
x, 266, 268, 27?, 277, 280, 285
y, 268, 279
z, 283, 284
INDEX
343
Chrysanthemum, 31, 32
Cinderella, 145, 233
Circumcision, 75
CLAUSEN, 57, 85
Clones, 130, 134, 135, 136,
138
Clovers, 205
Cirripede, 290
Coat pattern, 137
COBB, 162, 177, 178
COCKEREL, 48
Cocoon, 102
Coefficient of variability, 26
Coelenterates, 292
CCLE, 249
Coleus, 136
C 'olios, 51
Color blindness, 280, 282
eye, 17, 20, 53, 102, 170, 195,
235, 305
gene, 152, 161
roan, 170
skin, 190
Comb, 50
Combinations, 38, 39, 51
Complementary genes, 149, 150,
151
Congenital characters, 68
CONKLIN, 63, 76, 151, 165, 193,
253, 256, 258, 259, 260
Connecting links, 2
Consanguineous marriage, 208,
312
Conscience, eugenic, 320
Conservation of germplasm, 330
human, 314
Continuity, 13, 251
Constants, 25, 26
Convergent variation, 198
CORRENS, 96, 100, 102, 169
Cotton, 52, 102
Coupling, 235
Cousin marriage, 209, 307
Crab, 290
Crepidula, 293, 294
Crested head, 102
Cretinism, 328
Criminals, 299, 304, 309, 311,
317, 320, 321, 322, 329
Crippled toes, 75
Criss-cross inheritance, 279,
280, 281'
Crosses, homozygous, 132, 133,
134, 136, 142
negro-white, 190
Cross-over, 238, 240, 241, 242,
243
Crustacea, 35, 290, 293
Ctenophores, 292
CUENOT, 96, 155, 161
Cumulative genes, 149, 150
Cupid, 313, 326
Curly hair, 111
Curves, skew, 31
Cycle, chromosome, 273
life, 56
sexual, 272
Cytoplasm, 216, 257
Cytoplasmic inheritance, 257
Daisy, 31, 32
DANIELSON, 190, 191, 192
DANTE, 316
Daphnids, 137, 142, 274, 291,
292
DARBISHIRE, 100, 102, 173
DARWIN, 2, 3, 20, 23, 24, 34, 37,
40, 41, 51, 61, 65, 70, 73, 95,
96,, 105, 151, 166, 199, 205,
206, 223, 297, 325
DARWIN, Maj. L., 335.
Datura, 59, 174.
DAVENPORT, 28, 50, 96, 102, 111,
171, 172, 175, 176, 190, 191,
192, 196, 200, 300, 304, 306,
309, 312, 318, 324, 327
DAVIS, 46 .
Deaf, 309, 321
Deafness, 235, 306, 318
Death, 10, 56
Deaths, preventable, 331
Deer, 207
Deer-mouse, 52
Defectives, segregation of, 327
Defective teeth, 235
Defects, control of, 309
hereditary, 306
Deformed, 317, 321
trees, 76
Degressive species, 46
344
INDEX
Department of Agriculture, 324
Dependent, 321
Desirable germplasm, 330
Determination of sex, 264, 267
DETTO, 70
Development, arrested, 197
rate of, 262
Deviation, standard, 26, 27
average, 26, 27
DEVRIES, 39, 41, 42, 43, 44, 45,
46, 52, 56, 96, 102, 124, 257
Diathetic, 317
Differentiation, cell, 219
somatic, 255, 256, 258
Difilugia, 136, 141
Dihybrid, 107
Diluting gene, 156
Dimorphism in sex, 287
Dinosaurs, 56
Disease, 80
Dissolution of hybrids, 46
Disuse, effects of, 78
Dogs, hairless, 50
tailless, 49
Dominance, 119, 168
delayed, 170
incomplete, 174
imperfect, 168
reversed, 171
Dominant, 97, 99, 196, 212
DONCASTER, 270, 284
Double flowers, 47
Doves, 272, 295
Drastic measures, 329
DRINKARD, 102
Drosophila, 51, 52, 53, 101, 137,
143, 145, 165, 171, 179, 207,
233, 234, 235, 237, 239, 240,
242, 245, 246, 247, 248, 249,
269, 270, 278, 281, 285, 287,
291
DROWNE, 11
Drugs, effect of, 82
DRYDEN, 201
Ducks, 289
DUERDEN, 52
DUGDALE, 299, 300
DUNK, 249, 260
Duplex, 106, 195
Duplicate genes, 190
DURHAM, 156, 161
Dwarf peas, 98, 99, 100, 105, 107
Dyads, 266
DZIERZON, 276
EAST, 136, 186, 187, 214
Ears of rabbits, 177, 187
Echidna, 198
Echinoderms, 231
Echinus, 231
EDWARDS, JONATHAN, 301
Effects of drugs, 82
use and disuse, 78
Egg, 221
abortive, 224
ephippial, 275
fertilized, 15, 55, 58, 80, 223
frog's, 259
human, 228
laying, 145
mature, 224
parthenogenetic, 273, 275
winter, 273, 274
EHRLICH, 82
EIMER, 21
EI.DERTON, 306, 307
Elementary species, 41
Ellis Island, 322
Embryology, 251, 254
EMERSON, 48, 249
Emperor K'ang Hsi, 204
Endocrine factors, 261
Endocrinology, 262
Environment, 46, 60, 76, 296,
304, 315
Environmental factors, 261
Enzymatic pangenes, 257
Ephippial eggs, 275
Epigenesis, 252
Epileptic, 317, 321, 325
Erinaceus, 198
Erithyzon, 198
ESTERBROOK, 300, 301
Eugenics, 315, 317, 335
Eugenic conscience, 320
Record Office, 315
Euthenics, 315
Evening primrose, 42, 52, 53, 56
EWINO, 137
Experimental breeding, 254
INDEX
345
Extension gene, 157
External factors, 261, 275
Extracted recessives, 310
Extra chromosomes, 229
toes in poultry, 50, 171
Eye color, 17, 20, 53, 102, 170,
195, 235, 305
Factor hypothesis, 148
endocrine, 261
environmental, 254
external, 261
modifying, 297
mutation, 58, 60
Falling in love, 326
False reversion, 197
FARRABEE, 102
FAY, 306
Feathers, 50
Fecundity, 137
Feeble-mindedness, 197, 302,
303, 309, 311, 316, 320, 321,
322, 325
Feet of Chinese women, 75
Feral animals, 198
Fertilization, 224, 225, 226, 268
abnormal, 230
self, 97, 134
Fertilized eggs, 55, 58, 223
Fever, Texas, 82
Fibers, mantle, 220
Filial generation, 103
FISH, 118, 145
Fission, 10, 11, 87
Fission rate, 136, 137
Flatworms, 293
Flavism, 199
Fluctuation, 42, 129
Four-leaved clover, 20
Four-o'-clock, 106, 169, 174
Fowls, 89, 171, 289
blue Andalusian, 169, 174
jungle, 60, 196
rumpless, 172
FRANKLIN, 322
Freaks, 51
Free martin, 288
Frequency polygon, 29
Frog's eggs, 259
FEUWIRTH, 136
GAGE, 71
GALTON, 23, 24, 65, 93, 94, 121,
122, 130, 155, 199, 302, 318
Laboratory of Eugenics 318
Gametes, 223
Gametic mutation, 53
Gametogenesis, 254
Garlic, 136
GATES, 46, 59
Gelastocoris, 269
Gemmules, 11, 73
Generation, filial, 103
parental, 103
spontaneous, 9
Gene, 149
allelomorphic, 149
alternative, 149, 150, 10
arrangement of, 245
color, 152, 161
complementary, 149, 150, 151
constant, 160
cumulative, 149, 150
diluting, 156
duplicate, 190
extension, 157
kinds of, 149
lethal, 149, 151, 162
localization of, 246
intensifying, 156
modifying, 146, 149, 150
pattern, 154, 161
pigment, 152
plural, 149
restriction, 156
sex, 269
single, 149
supplementary, 149, 151, 154
uniformity, 155
Genetics, 3
Genius, 305, 316
Genotypes, 105, 110, 111, 113,
116, 196, 250
Genotypic selection, 200, 213
Gemmules, 11, 73
Germ-cells, 221
Germplasm, 12, 14, 16, 72, 113,
196, 254, 315
architecture of, 233
continuity of, 13, 14, 251
conservation of, 330
346
INDEX
Gtermplasm, desirable, 330
undesirable, 320
theory, 84
GEROTJLD, 51
GODDARD, 302, 311
GOLDSCHMIDT, 25, 26, 201, 292
GOODALE, 289
Goose, 60
GOULD, 293, 294
Graduated variants, 143
Grains, 205
Grasses, 205
Greeks, 208
Greening apples, 11
Green peas, 100, 107
GREGORY, 59, 136, 249
GRIFFON, 76
GROUCHY, 263
Grouse locust, 249
Growth, 10
GRUBER, 217
Guinea-pigs, 71, 85, 86, 102,
115, 118, 154, 155, 161, 172,
233
GUYER, 89
Gynandromorphs, 51, 290, 291
HAECKER, 102, 218
Haemophilia, 307
Hair, 305
angora, 102
color, 20, 111, 170
curly, 111
form, 111
Hairlessness, 50
HALDANE, 245
HALLET, 201
HANEL, 137
Hare, Belgian, 177, 187
Harelip, 197
HARTSOEKER, 252
HATAI, 49
HAYES, 48, 55, 214
HAYS, 205
Hedgehog, 198
HEGNER, 137
Helix, 171
HEIKE, 334
Helianthus, 48
Hemiptera, 269
Hen, 137, 145
Hereditary bridge, 228
character, 8
defects, 306
factors, 254
tunnel, 250
unit, 148
Heredity, definition of, 6
Hereford cattle, 49
Heritage, 4, 6
Hermaphroditism, 292
Hernia, cerebral, 50
Heterogametic female, 270, 271
male, 267
Heterosis, 214
Heterozygote, 105
Hill Folk, 303
Hindrances, social, 332
HODGE, 79
Homozygote, 105
Homozygous crosses, 132, 133,
134, 136, 142
Honey-bee, 275
Hooded rat, 137, 145
HOOKE, 216
Hormones, 89, 287, 288
HORNADAY, 207
Hornless cattle, 49, 172
Hornlessness, 102
Horns, 102, 171, 172
Horses, 75, 102, 323
hairless, 50
pacing, 102
trotting, 102
Hue, 204
Human assets, 315
betterment, 314
conservation, 314
egg, 228
liabilities, 315
skin color, 190
srferm, 228, 252
stature, 121
traits, 302
HURST, 100, 106
HUXLEY, 336
Hyalodaphnia, 137
Hybridization, 200, 209
Hydra, 137
Hymenoptera, 275, 276, 277
INDEX
347
IBSEN, 164, 249
Identical twins, 18, 142
Identification, personal, 19
Idiots, 322, 325, 329
Illegitimacy, 325
Imbecility, 308, 329
Immigration, 322
Immortality, 13
Immunity, 82
to rust, 102, 296
Imperfect dominance, 168
Impressions, maternal, 83
Inachus, 290
Inbreeding, 54, 200, 205, 308
Incomplete dominance, 174
Inconstant species, 46
Independent assortment, 103
unit characters, 117
Indians, 208
Induction, parallel, 70, 80, 91
somatic, 70, 91
Inebriate, 317, 321, 325
Inheritance, 6
alternative, 94
blending, 93, 94
biological, 7, 63
criss-cross, 279, 280, 281
cytoplasmic, 257
particulate, 94, 155
sex-limited, 278
sex-linked, 277, 283, 284
Inhibitor, 175
Insane, 305, 317, 320, 321, 322,
325
Insanity, 308
Instinct, 79
Intensifying genes, 156
Interference, 243, 244
Intergrades, sex, 290, 291, 292
Intracellular pangenesis, 257
Inventive genius, 305
I^otnoea, 205
Ishmaels, 303
JANSSENS, 243
Japanese art, 19
JENNINGS, 29, 30, 87, 88, 93,
123, 136, 138, 139, 140, 141,
241
Jews, 208
Jimson weed, 59, 174
JOHANNSEN, 24, 105, 122, 123,
124, 125, 127, 128, 129, 130,
132, 135, 141, 146, 149, 203
JOHNSON, 83
JOLLOS, 136
JONES, 249
JORDAN, 208, 328, 329, 332,
Jukes, 299, 304
Juglans, 51
Jungle fowl, 60, 196
Kallikak, 302
KAMMEEER, 77
KEEBLE, 48
KELLICOTT, 296, 327
KELLY, 137, 249
KELVIN, 8
Kernel, mealiness of, 136
KING, 206
KLEBS, 35, 36
KNIGHT, 209
KOELREUTER, 209
KGRNHATTSER, 290, 292
KREUGER, 293
LAMARCK, 34, 42, 64, 65
Lamb, legless, 20
LANG, 102, 171, 174, 179, 189
LASHLEY, 137
Lathyrus, 151
LAUGHLIN, 315, 316, 317, 318,
330
Law, Mendel's, 97, 100, 101, 102,
117, 168, 211
of regression, 121, 122, 130
of segregation, 103, 187
Leaves, beech, 25, 31, 33
serrated, 102, 172
smooth-margined, 102
LE COUTOUR, 204
LEHMANN, 55
Lemna, 136
Lentils, 136
Lepidoptera, 270, 283
Leprous, 321
Leptinotarsa, 143, 144
Lethal genes, 149, 151, 162, 165
Liability to disease, 312, 317
Light reactions, 137
LILLIE, 288
348
INDEX
LlNDSTROM, 165, 249
Linkage, 234, 236, 239, 249
Linnaean species, 1
Lint, 102
Literary ability, 305
Live-for-ever, 35, 36
LITTLE, 162
Localization of genes, 246
LOCK, 100
LOEB, 227
LOTSY, 46
Lupines, 136
LYELL, 167
MACDONALD, 317
MACDOUGAL, 46
MACDOWELL, 137, 145
Maize, 102, 137, 165, 186, 207,
214, 249
Mammals, 288
Man, 102
primitive, 296
Mangro, 142
Mantle fibers, 220
Marriage, consanguineous, 208,
312
cousin, 209, 307
laws, 324
taboo, 325
Mass selection, 200
Maternal impressions, 83
Mathematical aptitude, 305
Maturation, 53, 113, 223, 224,
230, 268
MAY, 53, 137
McCLUNG, 265
MEADER, 136
Mealiness of kernel, 136
Mean, arithmetical, 23, 26, 42
Meiosis, 54
Melanism, 199
Melting-pot, 93, 176, 186
Membrane, nuclear, 216
MENDEL, 3, 23, 95, 97, 98, 102,
106, 107, 110, 114, 119, 148,
168, 173, 211, 234, 235, 249,
317
Mendel's law, 97, 100, 101, 102,
117, 168, 211, 297
Mendelism, 93, 250, 254
MENDIOLA, 136
Mental characteristics, 303
Merino sheep, 49
Metaclonosis, 55
Metaphase, 220, 267, 271, 273
METZ, 52, 249
Mice, 75, 101, 102, 161, 206, 249
hairless, 50
in abnormal temperature, 76
intensified, 156
piebald, 95, 155
ricin-immune, 82
spotted, 52, 155
waltzing, 102
yellow, 162
Micron, 139
MIDDLE-TOST, 136
MILTON, 9
Mirabilis, 106, 174
Mitosis, 219
Mode, 26, 27
Modifications, 38, 61
Modifying genes. ^46, 149, 150,
166
Mohammedans, 208
Mold, 83
Mollusks, 293
Monohybrid, 99, 103, 19tfL
MONTGQMERY, 67, 69
Moral characters, 303
MORGAN, LLOYD, 73
MORGAN, T. H., 53, 97, 102, 103,
132, 165, 170, 171, 234, 235,
236, 237, 239, 240, 242, 243,
245, 246, 272, 285, 287, 289,
291
Morning-glories, 205
Mosaics, 292
Mcths, 70, 270, 272, 284, 292
Mudpuppy, 79
Mulatto, 190, 191, 192
Mule, 214
MULLENIX, 162, 177, 178
MULLER, 54, 244, 246
Multiple mutation, 51
variation, 20
Museum of heredity, 132
Musical ability, 305
Mustifee, 192
Mustifino, 192
INDEX
349
Mutation, 23, 38, 39, 40, 51, 55,
61, 62, 91, 254
aggregate, 51
causes of, 57, 60
factor, 58, 60
gametic, 53
kinds of, 51
multiple, 51
origin of, 53
parallel, 52
recurrent, 53
reverse, 53
single gene, 51
somatic, 53, 55
theory, 41
zygotic, 53, 55
Mutilation, 75
NABOURS, 279
NACHTSHEIM, 276
NAEGELI, 20, 21, 95
Nams, 303
Narcissus, 59
Natural selection, 41, 42, 51,
297
Nature, 63, 203
Necturus, 79
Negro, 192
Negro-white cross, 190
Nematodes, 293
Neo-Darwinian, 65
Neo-Lamarckian, 65
Neotony, 78
Nettles, 102, 172
NEWELL, 276
NEWTON, 316, 334
Nicodemus, 72
Night-blindness, 312
NILSSON, 136, 205
NILSSON-EHLE, 179, 180, 181,
183, 186, 187, 191
NlTOBE, 19
Non-disjunction, 59, 269, 285,
286, 287
Nuclear membrane, 216
Nucleus, 916
Nulliplex, 107, 195
Nurture 63, 203
Oats, 136, 249
Octaroon, 192
(Enothera, 43, 44, 45, 59, 165
Oil-gland, 50
Oocyte, 224, 267, 271, 273
Oogonia, 224, 267, 271, 273
Oranges, navel, 11
Origin of individual, 2
life, 8
mutation, 53
species, 3, 41
Orthoptera, 265
OSBORN, 336
Ostrich, 52
Outcrossing, 207, 209, 307
Ovarian transplantation, 85, 86
OVID, 9
Ovists, 222
Ovum, mature, 267, 271, 273
Oyster-borer, 28, 29
Pacing horses, 102
Pangenesis, 70, 73, 90, 91, 105
intracellular, 257
Palatine ridges, 197
Papaver, 47
Parallel induction, 70, 81, 91
mutation, 52
Paramerium, 29, 30, 87, 88, 136,
137, 138, 140, 141
Parasitism, 290
Parental generation, 103
Parrot, 20
Participate inheritance, 94, 155
Parthenogenesis, 132, 134, 227
Parthenogenetic eggs, 273, 275
progeny, 132, 134, 136, 141
Partial potency, 174
PASTETJR, 9, 82, 316
Pattern genes, 154, 161
Paupers, 299, 305, 309, 317, 322
PAYNE, 137
PEARL, 136, 137, 145
PEARSON* 24, 25, 31
Peas, 98, 110, 136, 148, 205, 234,
249
dwarf, 98, 99, 100, 105, 107
garden, 100
green, 100, 107
smooth, 100, 107, 173
sweet, 151, 152, 235, 249
tall, 98, 99, 100, 105, 107
350
INDEX
Peas, wrinkled, 100, 107, 173
yellow, 100, 107, 110
Pebrine, 82
Pedigree breeding, 200, 203
Pendulum, 40
PenicilUum, 83
Peromyscus, 52
Persians, 208
PETRUNKEVITSCH, 276
Petunia, 47
Phacechaerus, 78
Phaseolus, 123
Phenotype, 110, 111, 113, 116,
159, 196, 250
Phenotypic selection, 201
PHILLIPS, 85, 86, 137
Phoenicians, 208
Phylloxera, 272, 273, 275
Physiological regulators, 262
Piebald mice, 95, 155
Pied Piper, 145
Pigeon, 196, 249
Pigment gene, 152
production, 136
Pigs, 171
Pineys, 303
Plesiosaurs, 56
Plural genes, 149
Pomace fly (see Drosophila)
Polarity, 259
Polydactylism, 312
Polyembryony, 277
Polygons, bimodal, 29
frequency, 29
Poppy, Shirley, 47
Polar cells, 54, 224, 267, 271,
273
POPEXOE, 83
Population, 127
Porcupine, 198
Poultry, 50, 71, 89, 102, 124,
171, 176
rumpless, 172, 175
tailless, 49
Potato, 136, 205
beetles, 33, 36, 137, 143, -144
Potency, 172
failure of, 174
partial, 174
total, 173
Preformation, 252
Prenatal characters, 68
influences, 83
Presence or absence theory,
106, 149
Preventable death, 321
PRICE, 102
Primitive man, 296
Primrose, double, 47
evening, 249
giant, 48
Primula, 59, 249
Prisoners, 309
Progeny, parthenogenetic, 132,
134, 136, 141
Progressive species, 45
Prophase, 219
Prostitutes, 299, 322
Protophyta, 10
Protoplasm, 216
Protozoa, 10, 13, 86, 138
Proximity, 308, 328
Pseudomethoca, 291
Pug jaws, 50
PUKNETT, 99, 235, 249
Pure lines, 121, 122, 123
Puritan stock, 208
Purple beech, 47
Quadroon, 192
Quail, 79
Rabbit, 16, 89, 101, 162, 163.
249, 268
color of, 160
ears, 177, 187
gray, 162
lenses, 89
lop-eared, 177
phenotypes, 159
Race preservation, 336
Radiolaria, 218
Rate of development, 262
Rats, 149, 206, 249
hooded, 139, 145, 166
Recessive, 99, 119, 196, 212
extracted, 310
Recurrent mutation, 53
Red-eyed yellow mice, 52
REDFIELD, 67
INDEX
351
Reduction division, 58, 59
REEVES, 137
Regression, 121, 122, 125, 130,
199
Regulators, physiological, 262
REID, 66
Reinfection, 82
Relative variability, 28
Repair, 9
Reproduction, sexual, 12, 221
Reproductive period, 56
Resistance to poisoning, 82
Response, 4, 6
Restriction gene, 157
Retrogressive varieties, 45
Reversed dominance, 171
Reverse mutation, 53
Reversion, 131, 151, 194
explanation of, 199
false, 197
Rhabdites, 293
RHODES, 316
Rice, 204
Ricin immunity, 82
RIDDLE, 71, 272, 295
RIMPAU, 202
Ringer's solution, 89
RITZEMA-BOS, 206
Roan color, 170
Rodents, 171
Romans, 208
ROOT, 137
Roses, 47
Rotifers, 274, 275
Round worms, 293
Rudbeckia, 174
Rudimentary wing, 53
Rumplessness, 102, 172, 175
Rust, immunity to, 102, 296
Sacculina, 290
Salamander, 77, 102
Salamandra, 77
Sambo, 192
SA TINDERS, 102
Scale of success, 6
Scalp muscles, 197
SCHAFFNER, 48
SCHLEIDEN, 215
SCHWANN, 215
Sea-urchin, 230, 231
Secondary sexual characters,
287
Sedum, 35, 36
Segregation, 103, 113, 119, 168
of defectives, 327
SEILER, 270, 272, 275
Selection, 121, 166
genotypic, 200, 213
mass, 200
natural, 41, 42, 51
phenotypic, 201
pure line, 133, 136, 137
Self fertilization, 97, 134
Sex cells, 224
chromosome, 245, 265
cycle, 272
determination of, 264, 265
genes, 269
intergrades, 290, 291, 292
linked, 277, 283, 284
Sexual reproduction, 12, 221
SHAKESPEARE, 63, 316
SHARP, 242, 247, 248, 285, 287
Sheep, 18, 48, 75, 171
Shirley poppy, 47
SHIRREFF, 204
SHULL, A. F., 275,
SHULL, G. S., 46, 76, 102, 105,
106, 165, 207, 214, 249
Siamese twins, 50
Silkworms, 82, 102, 249
Simocephalus, 137
Simplex, 107, 195
Single genes, 149
SITKOWSKI, 70
Skew curves, 31
Skin color, 190
Simplex, 107, 195
SMITH, 89, 137, 275, 290
Smooth-margined leaves, 102
Smooth peas, 100, 107, 173
Snails, 28, 33, 102, 171, 174
Snapdragon, 102, 165, 249
Social hindrances, 332
Somatic characters, 67
differentiation, 255
Somatic induction, 70, 91
mutation, 53, 55
variation, 21
352
INDEX
Scmatogenesis, 250, 253, 254
Somatoplasm, 12, 14, 72, 113,
196, 254
Soy beans, 136
Species, cycle, 56
definition of, 1
degressive, 46
elementary, 41
Linnaean* 1
inconstant, 46
origin of, 3, 41
progressive, 41, 45
SPENCER, 79, 334
Sperm, human, 228, 252
Spermatid, 224, 267, 271
Spermatocyte, 224, 267, 271,
273
Spermatogonia, 224, 267, 271,
273
Spermatozoa, 221, 224, 226, 267,
271, 273
Spermists, 222
Sphcerechinus, 231
SPILLMAN, 102
Spines, dermal, 199
Spiny ant-eater, 198
Sponges* 10, 11
Spontaneous generation, 9
Spores, 12
Sports, 20, 40, 42, 199
Spotted mice, 52, 155
SPRENGER, 47
Spurs, 50
Standard deviation, 26, 27
STANDFUSS, 55
Starchy kernel, 102
Starfish, 22, 25, 26, 33
Statoblasts, 11
Stature, human, 121
Stentor, 217
Sterilization, 330
STEVENSON, 334
STIEGLEDER, 164
Stock, 47
STOCKARD, 20
STOCKING, 136
STOMPS, 59
STOUT, 136
Struthio, 52
STURTEVANT, 59, 137, 245, 246
Styela, 259
Stylonychia, 136
Submerged tenth, 320
Subsidizing the fit, 333
Sudan red, 71
Sugar beet, 205
Sugary kernel, 102
SUMNER, 52, 76
Sunburn, 76
Sunflower, 48, 102
Supplementary genes, 149, 151,
154
SURFACE, 136, 249
Survival of the fittest, 298
Susceptibility to disease, 312,
317
rust, 102
Sweet peas, 151, 152, 235, 249
Synapsis, 58
Syndesis, 241
Taboo, 325
Tadpoles, 78
Taillessness, 49
Tails, docked, 75
Tall peas, 98, 99, 100, 105, 107
TANAKA, 249
Tattooing, 75
Tatusia, 277
Telophase, 220, 221
TENNENT, 231
Tetraploidy, 59
Tetrads, 266
Texas fever, 82
Thelia, 290, 292
Theromorphs, 56
Theory, cell, 215
chiasmatype, 243
chromosome, 229, 254
mutation, 41
pre formation, 253
Thigh-bones, absent, 50
THOMSON, 194
Thoracic bristles, 137, 143,
145
Thumb prints, 19
Tineola, 70
Tobacco, 48, 52, 214
Toenails, absent, 50
extra, 50
INDEX
353
Toes, crippled, 75
extra, 50, 171
reduced, 172
webbed, 50
Tomato, 52, 59, 102, 249
Total potency, 173
TOWER, 33, 36, 137, 143, 144
TOYAMA, 102
Training, 5
Traits, human, 304
Transmission of disease, 80
Transplantation of ovaries, 85,
86, 289
Treasury of human inheritance,
318
Trees deformed by wind, 76
Triangle of life, 3, 4
Trihybrid, 114, 183, 185
Trilobites, 56
Trotting horses, 102
TSCHERMAK, 96, 100, 102
Tuberculosis, 20, 68, 80, 311,
321, 322
Twins, identical, 18, 142
Siamese, 50
TYNDALL, 9
Unhanded shell, 102
Unbranched habit, 102
Undesirable germplasm, 320
Uniformity gene, 155
Unit characters, 2
hereditary, 148
Urosalpinx, 28, 29
Use, effects of, 78
VAN BENEDEN, 221, 223
Variability, coefficient of, 26
relative, 28
Variants, graduated, 143
Variation, 17, 254
abnormal, 22
bud, 55
causes of, 34
continuous, 22
convergent, 198
definite, 21
discontinuous, 22
fluctuating, 23, 24, 42, 51, 62,
199
Variation, fortuitous, 21
germinal, 22
graduated, 34
harmful, 21
hereditary, 23, 61
indefinite, 21
integral, 33
indifferent, 20
kinds of, 19
morphological, 19
multiple, 20
non-hereditary, 23
normal, 22
orthogenetic, 21
physiological, 19
psychological, 20
qualitative, 23
quantitative, 22
single, 20
somatic, 21
useful, 20
universality of, 18
Varieties, regressive, 48
retrogressive, 45
Variety, 38
Vasectomy, 329
Venereal diseases, 325
Verbena, 48
Vermiform appendix, 21, 197
Vestigial structures, 197
VILMORIN, 135, 136, 205
Viola, 18
VOGLER, 136
Volta Bureau, 318
VOLTAIRE, 334
VON BAEHR, 272
VON KOELLIKER, 221
VON MOHL, 216
Walnut, 51
WALKER, 136
WALTER, 29, 162, 177, 178
Waltzing mice, 102
War, 331, 332
Wart-hog, 78
Wasp, 291
Waterloo, 263
Weakling babies, 334
Webbed toes, 50
WEBBER, 132
354
INDEX
WEISMANN, 10, 35, 65, 67, 69,
71, 73, 74, 75, 84, 88, 206,
233, 255, 256
Wheat, 102, 135, 136, 180, 191,
201, 204, 249, 296
WHITE, 100, 249
WHITNEY, 275
WHYMPER, 329
WlEDERSHEIM, 75
William the Conqueror, 208
Willows, 76
WILSON, 232
Wings, absent, 50
cut, 53
rudimentary, 53
WINKLEB, 59
WINSHIP, 301
WINSLOW, 136
Winter eggs, 273, 274
WOLF, 136
WOLFF, 252
WOLTZBZCK, 34, 35, 137
WOODS, 304
Worms, flat, 293
nematode, 14, 218
round, 293
silk, 82, 102, 249
WRIGHT, 48
Wrinkled peas, 100, 107, 173
X-chromosome, 266, 268, 272,
277, 280, 285
Y-chromosome, 268-279
Yellow mice, 162
Yield per acre, 136
Youth, 56
Z-chromosome, 283, 284
ZEDERBAUR, 77
ZELENY, 137
Zeros, 303
ZlEGI-ER, 47
Zygote, 223
Zygotic mutation, 53, 55
354
INDEX
WEISMANN, 10, 35, 65, 67, 69,
71, 73, 74, 75, 84, 88, 206,
233, 255, 256
Wheat, 102, 135, 136, 180, 191,
201, 204, 249, 296
WHITE, 100, 249
WHITNEY, 275
WHYMPER, 329
WlEDERSHEIM, 75
WlLKS, 47
William the Conqueror, 208
Willows, 76
WILSON, 232
Wings, absent, 50
cut, 53
rudimentary, 53
WINKLER, 59
WINSHIP, 301
WINSLOW, 136
Winter eggs, 273, 274
WOLF, 136
WOLFF, 252
WOLTEHECK, 34, 35, 137
WOODS, 304
Worms, flat, 293
nematode, 14, 218
round, 293
silk, 82, 102, 249
WRIGHT, 48
Wrinkled peas, 100, 107, 173
X-chromosome, 266, 268, 272,
277, 280, 285
Y-chromosome, 268-279
Yellow mice, 162
Yield per acre, 136
Youth, 56
Z-chromosome, 283, 284
ZEDERBAUR, 77
ZELENY, 137
Zeros, 303
ZlEGLER, 47
Zygote, 223
Zygotic mutation, 53, 55
^^^
BIOLOGY
LIBRARY
UNIVERSITY OF CALIFORNIA LIBRARY
