1 a

GENETIC VARIATIONS IN
RELATION TO EVOLUTION

M M M II

yiK ->Vc i''^ -i^.ii-

H 11^ tl ^

:i m

H. S, JENNINGS

-- 1

* i

%

6 "

M&RINE BIOLOGICAL LABORATORY.

Received d^g ,

.1.9.3.5.

Accession No. .

45286'

C^'^^"^y author Trndpul)

Place, .P.ri.n.ce.to.n.,.. N...J.*

*t* No book or pamphlet is to be removed from the Lab-
oratory without the permission of the Trustees.

' _I

" J

r-^

n

f— 1

^^= D

GENETIC VARIATIONS
IN RELATION TO EVOLUTION

LONDON: HUMPHREY MILFORD
OXFORD UNIVERSITY PRESS

J

GENETIC VARIATIONS

IN RELATION TO

EVOLUTION

A Critical Inquiry into the Observed
Types of Inherited Variation, in
Relation to Evolutionary Change

BY

H. S. JENNINGS

•^ . - w.-. -fjj^.

Henry IValten Professor of Z.oology ^> ,/• •

and Director of the Zoological Laboratory >*«« I I. I ^ ^ ^ y.

The Johns Hopkins University 1) "Sk^i r

PUBLISHED ON THE LOUIS CLARK VANUXEM FOUNDATION

PRINCETON 1935
PRINCETON UNIVERSITY PRESS

COPYRIGHT, 1935, PRINCETON UNIVERSITY PRESS

PRINTED AT THE PRINCETON UNIVERSITY PRESS
PRINCETON, NEW JERSEY, U. S. A,

FOREWORD

The chapters of this book are based on the mate-
rials presented in the Louis Clark Vanuxem Lectures,
given in December 1934, at Princeton University.

H. s. J.

PAGE

CONTENTS

I. Introductory. Structure of the Germ Plasm. Genetic
Variations Resulting from New Combinations of
Genes i

II. (i) Genetic Variations Resulting from Irregulari-
ties in the Structure and Operation of the Genetic
System. (2) Gene Mutations 27

III. Are the Gene Mutations Observed in Experimental
Genetics the Basic Material of Progressive Evo-
lution ? 58

IV. Genetic Variation in Unicellular Organisms, with
Special Relation to the Inheritance of Environmen-
tal Effects and Reactions 95

V. Genetic Variation in Unicellular Organisms, con-
tinued. What Is the Method of Operation of the
Genetic System in Bringing about Genetic Varia-
tions under the Long-continued Action of Environ-
mental Conditions ? 119

45286

CHAPTER I

Introductory. Structure of the Germ Plasm. Genetic
Variations Resulting from New Combinations of Genes

EXPERIMENTAL genetic science has said of late, with
some justification, that "the problem of heredity is
solved" ; solved in principle, though not in all its details. It is
solved to the extent that the nature of the processes which
underlie it are largely known; and so far as not known, they
are localized and open for study.

Can experimental genetic science say that the problem of
evolution is likewise in principle solved? Are the processes
which underlie evolutionary change known ? This question
forms the theme of the present study.

The processes which underlie organic evolution are bound
to be genetic variations of some sort; that is, changes in the
inherited characteristics. Experimental genetic science is now
able to detect genetic variations as they occur in living organ-
isms. Single genetic variations are isolated, the precise time
and place of their occurrence determined; they are subjected
to minute study in the laboratory. Within the last few years
men have learned how to produce these genetic variations;
this is perhaps the most important condition for determining
the nature of such changes. In a thousand laboratories men
are at work on these things ; genetic variation is now one of
the most active subjects in biological science. The observed
genetic variations are of a number of diverse types. Do these
known types furnish the material of evolution ? Are the single
variations steps in evolution ? Does genetic science therefore

2 GENETIC VARIATIONS

now see evolution occurring, as it sees inheritance occurring ?
Are the known types of genetic variation adequate to bring
about the salient features of organic evolution ? These are the
questions with which we wish to deal.

What are the salient features of organic evolution ? If we
could watch the procession of living things from the begin-
ning of life on the earth, what kinds of changes would we
see? We can I believe list the main types of change under
five heads:

1. First, evolution is transformism. Organisms do not re-
main constant, but slowly change with the passage of time, as
a developing child changes with the passage of time.

2. Second, evolution is diversification. Organisms trans-
form, not in one direction only, but in many directions. From
an early simple situation is produced that which we now see
about us: great numbers of animals and plants that are di-
verse in many ways.

3. Third, evolution brings about adaptation. Organisms
develop structures and functions that tend to keep them in
existence; to ward off dangers in a varied and changing
world, and to obtain conditions favorable for their internal
processes. Different organisms become adapted to different
conditions. Otherwise expressed, evolution produces struc-
tures and functions that bring about ends, of the sort that in
human beings become the object of conscious purpose. And
evolution produces, in man at least, some organisms whose
behavior is consciously purposive.

But, what is also important, there are produced, too, organ-
isms that are not adaptive; organisms that fail to develop or to
survive under the conditions of existence; a fact to which we
shall have to recur. Adaptation is produced in evolution, but
is not universal in the products of evolution.

INTRODUCTORY 3

4. Fourth, progressiveness is a salient feature of evolution.
Organisms were at first simple; in later ages some of these
have transformed into complex highly differentiated animals
and plants. From blobs of jelly-like material, responding
inadequately to a few of the forces of nature, have developed
organisms with elaborate sense organs and organs of
response; creatures whose behavior is minutely and ade-
quately correlated with processes in the surrounding world.

But progressiveness, like adaptation, is not a universal
feature of evolution. Highly developed organisms may de-
generate into parasites, and many organisms still exist that
have not developed beyond the lower grades of the organic
scale. Whatever brings about evolution must be able to pro-
duce progress; but it may produce also its opposite.

5. Fifth and finally, as to the method of progress when prog-
ress occurs, we must accept as a fact I believe the reports of
students of progressive evolutionary series among fossil organ-
isms such as are known in the development of the horse or
the elephants. They report that progress in such series is
continuous, gradual, by slight changes, not by sudden leaps.
Continuity, not discontinuity, is the characteristic of such
progressive series.^

Genetic variations, to be adequate to the production of
evolution, must then be of a sort to produce these five salient
features: transformism, diversification, adaptation, progres-
siveness and continuity.

And now, certain points as to the nature of genetic varia-
tions. In order that variations shall play a part in evolution,
they must be inherited ; they must affect, not single individ-
uals only but their descendants, becoming thus characteristics
of a race or species. Only such can be called genetic variations.

1 See the notes at the end of each chapter.

4 GENETIC VARIATIONS

There occur great numbers of variations that affect only
individuals, having no influence on their descendants. Color
becomes changed under action of the sun, but this does not
affect the offspring of the individual concerned; and there
are thousands of such variations.

Thus for a variation that plays a role in evolution — for a
genetic variation — two things are essential. There must be a
change. And there must be inheritance of the changed con-
dition. A non-genetic variation is like a change in the shape
of a crystal, resulting from filing off its angles and edges. Such
a crystal, when dissolved and recrystallized will produce
anew the original form, for its chemical constitution has not
been changed by the alteration in its shape. A genetic varia-
tion is like an alteration in the chemical constitution of the
crystal, causing it when dissolved and recrystallized, to appear
in a new form; a new form that is repeated as often as
crystallization occurs without further change in chemical
constitution.

Certain other relations play a great role in the results of
genetic variation. Variations occurring in certain parts of
organisms cannot be inherited. To influence later generations
they must affect certain particular parts of the body or of the
cell. In all higher organisms, reproduction does not occur by
mere division of the body of the parent into two equal parts,
as may occur in bacteria and protozoa. On the contrary, the
new individuals arise from single cells, "germ cells," which
are given off by the parents. These germ cells arise from
certain well defined parts of the body, the gonads or germ
glands. Thus any genetic variations that are to be carried on
to later generations must affect the germ cells. If a genetic
variation arises in some other part of the body — say in the
hand — this cannot be carried on into later generations, for
no part of the later generation arises from the hand. Whether

INTRODUCTORY 5

a genetic variation in the hand might in some way cause a
genetic variation in the germ cells is another question. But
unless the germ cells in some way are affected by the genetic
variation, it cannot be carried on to later generations.

Furthermore, even in any single cell, different parts of the
living material play very different roles in heredity and con-
sequently in genetic variation. It is know^n that the materials
that play the chief role in determining the characteristics of
organisms are those located in the central capsule of the cell
— the nucleus. By changing these materials in the nucleus of
the fertilized egg — leaving the rest of its materials unchanged
— the inherited characteristics of the entire individual pro-
duced from the egg are altered. To produce an effect on the
inherited characteristics, therefore, a genetic variation must
affect these materials. The materials within the nucleus con-
stitute what has been known in the past as the germ plasm;
genetic variations are changes in the constitution of the germ
plasm.

Until recently very little was known of the germ plasm; its
existence was assumed but speculative. But it has now been
identified, studied minutely, experimented on in many ways.
It is found to have an elaborate organization. On the nature
of this organization depend in large measure the kind of
genetic variations that occur, and their effect in evolution.
The first requirement therefore for understanding genetic
variation is to have before one a picture of this organization.
It is to such a picture that we first turn our attention.

An organism, as we know, begins its life as a single cell,
which we call the fertilized egg. The germ plasm, located
within the nucleus of this cell, is constituted of a large num-
ber of diverse materials, which we may call genie materials.
Each type of genie material is concentrated into a small
particle, commonly known as a gene. The entire germ plasm

6 GENETIC VARIATIONS

thus exists in the nucleus as a large number of separable
particles or genes, grouped into the chromosomes.^

These genie materials are extremely active in the life of the
organism. From within the nucleus the genes send out prod-
ucts of their action into the remainder of the cell — the cyto-
plasm. Here the genie products interact with each other and
with the cytoplasm itself, so becoming modified in many
ways, and producing many different things. In this way they
produce the various tissues of the body, and give rise to its
structural and physiological peculiarities; to what we call its
characteristics.

But of each of these genie materials there is retained an
unmodified part, a reserve portion, enclosed within the
nucleus. It is these unmodified reserves that are called the
genes. When cell division occurs, each of these unmodified
reserves, or genes, divides, so that each of the cells produced
receives a half of each gene; this grows again to form an
entire gene. Thus every cell of the body has the entire stock
of reserve materials or genes.

Genie materials therefore have two most important prop-
erties. First, they affect development, and thus help to deter-
mine the characteristics of the individual produced. And
second, each genie material assimilates, grows, and repro-
duces true to type; each gene, in other words, produces addi-
tional material of the same kind as itself.

Half of the genie materials within the nucleus come to the
fertilized egg from one of the two parents, half from the other
parent. They are built up in the nucleus into a complex
structural system, just as the nervous material is built up into
a complex nervous system. This system of genie materials we
may conveniently call the genetic system. Its peculiarities play
a great role in determining the nature of genetic variations, so
that we must examine them.

STRUCTURE OF THE GERM PLASM 7

The small particles into which the genie materials are con-
centrated — that is, the genes — are strung up in series, like
beads on a string. All those that come from the father form
one series or string; all those from the mother form another
series, another string of beads. Thus there are in the cell two
strings of genes. These constitute what I am calling the
Genetic System. This constitution of the genetic system is
shown as it appears under the microscope in Figure i, while
Figure 2 shows a diagram of its chief features.

The system is made a little more complex, but in a rela-
tively unimportant way, by the fact that each string, paternal
or maternal, is broken into a number of segments, which in
their condensed conditions are commonly called chromo-
somes (D, Figure i). Essentially, however, there is one string
of genes from the mother, one from the father, so that for the
present we may neglect the segments into which each is
broken. The different genes in either of the two strings are
diverse organic chemicals, with different properties, different
effects on the development.

Each gene, each different kind of material, has its regular
place in the string. Thus the different genes can be named or
numbered. Gene No. i has a certain definite effect on devel-
opment ; gene No. 40 another effect; gene No. 200 another, and
so on for all of them. Thus maps showing the relative posi-
tions of the different genes can be made; they have been made
for certain organisms that have been thoroughly studied.
Changing a single one of the genes — altering it, or substitut-
ing for it a different one — changes the course of development,
and causes the adult organism to have changed characteristics.
Altering one of the genes changes the color of the eye ; alter-
ing another changes the size or stature; altering another
changes the structure of a limb; altering another changes
certain physiological processes. Every feature of the organism,

8

GENETIC VARIATIONS

i

u

\

ir
tf oo

at

n

*»*

•i.

'VJS,

«•

*
«}

%, "

9

•9

gf

M$>»l*t S*'*

»•«

J)

Figure i. Characteristic features in the organization of the germ
plasm. The chromosomes and chromomeres, as seen under the micro-
scope, in different developmental stages. A and B. Chromosomes and
chromomeres in nuclei of the grasshopper, in the thread-like stage of
development, after Wenrich, 1916. The enlargements on the threads,
known as chromomeres, probably show the location of the different
genes: in any case their arrangement and position are the same as

STRUCTURE OF THE GERM PLASM 9

that known for the genes. C. Much enlarged chromosomal threads of
the lily, showing the chromomeres (genes?), after Belling, 1928.
D. The chromosomes in the condensed condition (chromomeres not
visible), in the female fruit fly, Drosophila melanogaster, after
Metz, 1916.

Structural or physiological, may be changed by altering one or
more of the genes ; by substituting one gene for another. The
entire practice of genetic science, of animal breeding, is an
elaborate technique for substituting one gene for another and
determining the consequences of such substitution.

Different individuals are found to have genes that differ
from each other in some of their properties; in consequence
of this, different individuals have different characteristics. To
this fact are due most of the diversities between individuals;
differences in features, in body form, in size and in struc-
ture; differences in physiology; differences in senses and
behavior; differences in character and personality. With in-
crease of knovi^ledge it has become more and more clear that
most differences of all kinds betw^een individuals are due to
differences in the materials of w^hich they are made; that is, in
the genes that they carry — rather than to differences in the
conditions under v^^hich they have lived.

When a new individual is formed— constituting at the
beginning a new fertilized tgg — this new individual gets half
his genes from his mother, half from his father. And now a
most important fact! The genes that he gets from one parent
— as from the mother — form a complete series, containing
genes for every function in development. This series contains
all the genes that are necessary for development and for pro-
ducing all the structures and functions of the body.

And the other chain of genes, that he gets from his other
parent, likewise contains all the genes that are necessary for

10

GENETIC VARIATIONS

Figure 2. Diagram of the organization of the germ plasm. Two
chains of genes, one (P) from the father, one (M) from the mother;
so that the genes are in pairs, one member of each pair from each
parent.

STRUCTURE OF THE GERM PLASM ii

the development of a complete individual w^ith all its struc-
tures and functions. That is, the new individual is double as
to his genes; he carries two complete sets/ each containing all
that is necessary for producing an individual. For every gene
in the maternal string there is a corresponding gene in the
paternal string (there are certain exceptions in some organ-
isms), so that the two strings are throughout parallel in
structure and action. Thus every kind of gene is represented
twice — by one example from the mother, the other from the
father. So the genes are in pairs — one gene maternal in origin,
the other paternal in origin. Each function, in development,
each effect, has two genes, either of which may perform it.
This doubleness of organisms in respect to their genes is one
of their most important features; it must never be forgotten.

But the two genes of a pair, though they have in general the
same function, may, and often do, perform that function in
a different manner. One of them, from the mother, tends to
produce eyes of a certain color — in man, for example, brown
eyes; while the one from the other parent may tend to pro-
duce eyes of another color — ^blue eyes, perhaps. One of a pair
may tend to produce a rapidly growing, vigorous body, the
other a slow-growing, weak body. One of a pair may tend to
produce acute hearing, the other defective hearing. One may
tend to produce blood of a certain type of the several possible
blood types, the other another blood type. And so of all their
effects.

In such cases, where the two corresponding genes differ,
usually one of the two prevails over the other. If one tends to
produce brown eyes, the other blue, the brown-eye gene pre-
vails and the individual has brown eyes; these have more
pigment than blue. If with respect to any function or struc-
ture one gene tends to produce a vigorous or normal individ-
ual, the other a weak or defective one, the one that tends to

12 GENETIC VARIATIONS

produce the vigorous or normal condition usually prevails
over the other; the individual is not weak or defective. In
general, it may perhaps be said that the gene that carries
development farthest usually prevails over the other, though
whether this is the invariable rule is doubtful. The gene that
prevails is said to be dominant; the one that does not prevail,
that produces no obvious effect, is said to be recessive. It is
only when two of the weak or recessive genes get together in
a pair — one such coming from each of the two parents — that
the weak or defective or recessive characteristic is produced
and appears in the individual.

Thus each individual is double with respect to his genes.
Each contains a number of genes that have no manifest
effect; a lot of recessive genes, whose effect is suppressed by
their dominant companion genes in the same pairs. Thus one
gene of a pair can become modified, become defective per-
haps, without producing any obvious effect on the character-
istics. But if two parents each contain such a defective gene,
in the same pair, then when they mate, some of their offspring
may get a pair of genes in which both members are recessive
— weak or defective — and these offspring will manifest the
weak or defective characteristics, though their parents did not.

Such then is the picture of the germ plasm, or of what we
may call the genetic system, a picture that we must keep in
mind in trying to understand genetic variation. In any indi-
vidual a double set of genes, frequently only one of each pair
of genes manifested, because dominant; the other recessive,
producing no effect (see Figure 2).

If we examine different individuals we find that the genes
of a certain pair — having a certain general function — may
exist in many different modifications in the different individ-
uals. So a certain gene pair, located at a particular point in
the chain, that affects the eye color in the fruit fly, is known

STRUCTURE OF THE GERM PLASM 13

to exist in at least a dozen different modifications, in different
individuals. One of the modifications produces red eyes,
another white eyes, another buff eyes, and others produce
various shades of red, or mixtures of red and yellow^, or
slightly tinged, or ivory eyes. Similar modifications are
known for many other genes, each producing a slightly dif-
ferent effect; they probably occur in all the genes.* In this
way, by diverse modifications of the same gene, a great num-
ber of different types of characteristics are produced ; different
eye colors, different forms, different sizes, different degrees
of strength and vigor, different physiological activities. Every
structure and function of the organism, without exception, is
thus capable of being produced in different grades and
qualities, through diversities among the genes.

If we look at any species of animal or plant as a unit, includ-
ing all the different individuals that compose it, we find then
the following situation as to genes: First, there is a great
number of different types of genes, strung up in a long
series. Each type has its special function, and each has its
definite place in the series; it is No. i or No. 40 or No. 200, or
the like. Any individual has two genes of each of the types.
But, second, each type has a number of different modifica-
tions, each modification performing the given function in a
somewhat different way. If the gene type is one that affects
the color of the eye, then the various modifications of that
type give in different individuals a whole series of different
eye colors. Since each individual has just two genes of a cer-
tain type, a single individual can contain only two of the
modifications of any type; other individuals may contain
other modifications.

Thus what characteristics are shown by any particular indi-
vidual depends on what combinations of the various types
and modifications of the genes are present in his cells. And

'm

14 GENETIC VARIATIONS

since the number of types and modifications is very great, the
number of diverse combinations of characteristics that may
occur is practically unlimited.

Observe now the relation of all this to genetic variations.
The genetic system presents three diverse aspects, each of
which may be altered; each of the three yields a distinctive
type of genetic variation. First, the genetic system is com-
posed of many diverse organic materials, the genes. These
materials might be altered, chemically or otherwise, giving
rise to new characteristics; this would be the fundamental
type of genetic variation. The genes are arranged in a definite
structure, the genetic system. This structure may be altered;
this gives a second type of genetic variation. Third, within
the framework of this structure, there are diverse combina-
tions of genes, in different individuals. Changing the combi-
nations of genes that are present gives a third type of genetic
variation.

Genetic variations of all these types are known, or at least
are commonly held to be known; they have been described in
detail. About one of the three types, however, serious doubt
has recently arisen; this is a matter with which we shall deal
later (see Chapter III).

We shall take up these three types of genetic variation in
the reverse order from that in which I have named them. We
shall take up first, changes in the combinations of genes
present; then changes in the structure of the genetic system,
reserving the supposed changes in the material of the genes
for our third topic. Beyond this I plan to present certain other
genetic variations not precisely fitting into this classification.
In the limits of these pages I shall have to confine myself to
types of variation observed to occur in living organisms. We
shall look at each type in relation to evolution.

NEW COMBINATIONS OF GENES 15

GENETIC VARIATIONS RESULTING FROM NEW COMBINATIONS

OF GENES

There is in nature a great mechanism, always operating, for
the production of new combinations of genes, and thereby
giving rise to genetic variation on a grand scale. The opera-
tion of this mechanism is what we call biparental reproduc-
tion, with its consequences that we call Mendelian inheri-
tance. This is perhaps the most potent means that is known
to us for yielding genetic variation and promoting evolution-
ary advance. Its operation therefore we must sketch.

Heredity is commonly thought of as a matter of resem-
blance, rather than of diversity ; resemblance between related
individuals. But Mendelian heredity produces diversity as
well as resemblance. Production of diversity is indeed perhaps
its most conspicuous feature; it might equally well be called
Mendelian variability. From a single pair of parents there
arise by Mendelian inheritance offspring of many diverse
types. Any two unrelated human parents doubtless have the
capability of producing offspring of a thousand diverse types.
Certain features of this process are of great interest for the
problem of genetic variation and evolution.

Each individual, as we have seen, has a certain combina-
tion of genes, some dominant, some recessive ; half of them
have come from his mother, half from his father. This defi-
nite combination of genes results in producing, under the
given life conditions, a certain set of characteristics, structural
and physiological. They give a body of a certain size, form
and structure, with functions that are performed in a certain
way; with a certain type of mental characteristics and be-
havior. They give an individual with a definite personality.

Now, when individuals reproduce, this combination of
genes is taken apart, and a new one is made by putting to-

i6

GENETIC VARIATIONS

P M P M

Father

Hother

<

SOME

of the

GERM
CELLS

SOME

ofihe

OFFSPRING

• Q

1 A

o • » • c

Figure j. Diagram to illustrate the recombination of genes when
offspring are produced. Above are shown, in the two rectangles, a
small portion of the double string of genes in each parent; in each
string but eight of the many genes are shown. In each parent the
genes of paternal origin (P) are represented as solid, those of maternal
origin (M) in outline. In the middle row are shown a few of the
many possible different combinations of genes in the germ cells, pro-

NEW COMBINATIONS OF GENES 17

duced by taking one gene of each pair for each germ cell. In the lower
row the three rectangles show three of the great number of diverse
combinations producible by union of one germ cell from the father,
one from the mother.

gether half of the genes from each of the two parents. This
gives a combination of genes that is different from that of
either of the two parents; so that the child will have a differ-
ent set of characteristics from either parent. There may thus
be produced a combination of genes never before produced,
giving an individual with characteristics different from any
that have before existed. To this method of operation is due a
great deal of the diversity that exists in nature. To it is due
at least a certain amount of what we may call evolutionary
progress; just how much it is difficult to say in the present
state of our knowledge. Examination of some of the rules of
this method of producing new combinations will help in
grasping it.

The great rule for the production of a new individual by
two parents is this : Each parent gives to the child one member
of each of his own pairs of genes. This is indicated in the
diagram of Figure 3. If the parent has a thousand pairs of
genes, he gives to each child one member of each; a thousand
single genes in all. The other parent gives also one member
of each of his thousand pairs, so that the child gets again from
both together a thousand pairs.

Each parent may give to the child either member of any of
his pairs of genes. We may designate the two members of any
of the parent's pairs as maternal (M) or paternal (P), accord-
ing as they have come to him from his mother or his father.
The child may get then either the maternal gene or the
paternal gene. Frequently, perhaps usually, these two mem-
bers of the pair are different in their effects. Commonly one

1 8 GENETIC VARIATIONS

is dominant, one recessive. The two members may differ (in
diverse cases) in many gradations of effect, or in many quali-
ties. One may tend to produce blue eyes, one brown eyes. One
may tend to produce vigor, the other weakness; and similarly
for all structures and functions. The child gets one of the two,
not the other; he may get either. Some of the offspring get
one of the two, some get the other.

Furthermore, the different pairs of genes of the same parent
are in a measure independent in the way they are distributed
to the offspring. If the child gets the paternal gene of one
pair, he may get either the maternal or the paternal gene of
any other pair.^ Thus there is a great number of possible
different combinations of genes from one parent that may
pass to diverse children. If there were but two pairs of genes
there would be four diverse combinations, any one of which
might be received by the offspring. For four pairs the number
of different combinations is i6; for ten pairs it is over a
thousand ; for one hundred pairs the number is unimaginably
great (it is two to the hundredth power). Since for higher
organisms the number of pairs of genes is much above one
hundred, the number of different combinations of genes that
may be received by the different children from one parent is
very great.

From the other parent the diverse children may receive an
equally great number of different combinations of genes. Any
combination from one parent may meet in the child any
combination from the other. The result is that the number
of different types of offspring that might be produced from
a single pair of parents is almost inexpressibly great. Each of
these types of offspring would have a different set of genes
and a different set of characteristics.

Thus reproduction from two parents is a great process of
producing individuals with new combinations of genes and

NEW COMBINATIONS OF GENES 19

new combinations of characteristics. Every combination of
genes that is mathematically conceivable may thus be pro-
duced. Every possible combination of the characteristics that
depend on the genes may thus be brought into existence. To
take a simple example, if one parent has blue eyes and light
complexion, the other dark eyes and dark complexion, then
among their descendants vi^ill be some like the fair parent,
some like the dark parent, some w^ith blue eyes and dark
complexion, some with dark eyes and light complexion. If
the parents differ in many characteristics instead of but two,
the number of different combinations of characteristics
among their descendants is very great.

In this production of new combinations through reproduc-
tion from two parents, there are three types of consequences
that are of interest for evolution.

First, in this way there are produced combinations of genes,
and therefore combinations of characteristics, that have never
before existed.

Second, in this way single qualities, powers or dimensions,
may be increased beyond the degree in which they have
before existed.

Third, as a result of the recombinations of genes, character-
istics may appear in the descendants that have never occurred
in their ancestors, or in any other organism.

In these three ways therefore, evolutionary change is
brought about by reproduction from two parents. We will
consider each of the three types of change separately.

I. By biparental reproduction, any quality or peculiarity of
any individual or any stock may become combined with any
features of any other individual or stock with which it can
unite for parenthood. Thus are produced combinations of
characteristics that have never before been brought together.
This plays a great role in relation to the fitness or adaptiveness

20 GENETIC VARIATIONS

of organisms. The result may be individuals superior in some
respects to any that have before existed, so that a distinct evo-
lutionary advance has been made. One stock may have
strength and agility, but also poor sight and hearing. Another
may have acute senses, but be weak and slow. By the mating
of the two, some offspring may be produced that have all the
superior qualities combined: strength, agility and acute
senses. Such combinations have a great advantage over either
of the parent stocks; a superior type has been produced.
Superior combinations are thus produced in thousands of
different ways, by the mating of diverse stocks.

Of course inferior combinations are produced too in the
same manner. By mating the stock that is strong and agile
but has poor sight and hearing, with the one that is weak and
slow but has acute senses, some combinations are produced
with all the inferior qualities; they are weak, slow, and have
poor senses. Such individuals are not successful in the strug-
gle of life; they soon disappear, leaving only the better com-
binations. Along with the production of new combinations
by biparental reproduction, there is selective elimination on
a large scale, getting rid of the poor combinations, retaining
those that are efficient. Thus through the formation of new
combinations, along with selective elimination, progress is
brought about.

2. The second effect of biparental reproduction is the
strengthening or increase of single powers and qualities,
beyond the previously existing degree. It is known that size,
strength, vigor, acuteness of senses, and other faculties or
characteristics that play an important role in life, do not
depend on single genes only; each is known to be influenced
by many diverse genes. Some of these genes are of superior
type in some individuals, others are superior in other individ-
uals. Thus one individual or race may carry a certain number

NEW COMBINATIONS OF GENES 21

of genes that tend to produce vigor and mental power. An-
other may carry another and different set of genes that produce
a degree of vigor and powder. When these two are mated, some
of the descendants receive both sets of genes that give vigor
and mental power; they show therefore a vigor and power
far beyond that of their parents. By successive matings of this
sort, any particular capability or quality may be carried far
beyond the degree in which it formerly existed. What may be
the limits of progressive evolution occurring in this way no
one can say.

3. Third and perhaps still more important, as a result of
these recombinations of genes, characteristics may appear in
the descendants that have never occurred in their ancestors.
Examples of such results are abundant in animals and plants.
They are perhaps most obvious in relation to colors. To take
a well known example, we may begin with two races of
primroses, one red, the other white." Crossing the two, we
find in their descendants individuals having many other
colors: magenta, pink, many different shades of red, and
flowers with colors in blotches. In place of two colors we
have in the descendants a dozen or more different colors.
These are obviously due to the chemical interaction of the
materials derived from the genes of the two parent races. The
different descendants get different combinations of the genes
of the two parents. In development these genes give off
materials into the cytoplasm. There they interact, producing
many different results in different cases. There are known
great numbers of examples of the production of new colors,
and also of new forms, in this way. The same method of
action must apply to other characteristics that are less conspic-
uous and less readily defined; they must apply to vigor,
strength, special capabilities, and mental power. How far

22 GENETIC VARIATIONS

such interaction of gene products may go in producing new
characteristics, not before seen, no one can say.

All these three methods of operation occur in man as well
as in other organisms. They occur in relation to mental and
temperamental qualities as well as to others. One parent may
be industrious and ambitious, but fail in life because he is
mentally dull and slow. The other may be quick and acute
mentally, but fail because he lacks industry and ambition.
Among their descendants would be some that combine the
industry and ambition of one of the parents with the acute-
ness and quickness of the other. Such descendants would be
superior. By fusing thus all the best qualities of numbers of
different ancestors, persons far beyond the usual level would
be produced. Again, as we have seen, existing powers and
capabilities may in this manner be increased in degree,
and even qualities produced that did not before exist. Such
methods of operation are doubtless the origin of the individ-
uals that are superior to most of humanity, the geniuses, the
Goethes and Leonardos.

Thus this method of operation by the formation of new
combinations through the mating of two individuals is
powerful in producing variation. A very large proportion of
the variety found among organisms is due to it. If reproduc-
tion occurred only by the division of single individuals, as
may occur in plants and many lower animals, the world
of living things would be relatively simple; there would be
comparatively few living types. The variety among organ-
isms is multiplied a thousandfold — perhaps a millionfold —
by reproduction from two parents instead of from one.

And reproduction from two parents has, as we have seen,
great potentialities for the production of new types, and so
for bringing about progressive evolution. It is certainly one
of the greatest factors in evolutionary progress.

NEW COMBINATIONS OF GENES 23

But for the production of new types that are permanent,
that bequeath their peculiarities to their descendants, this
method of operation suffers under a very great limitation.
This is because the interaction that produces the new charac-
teristics occurs between the products of the genes in the
cytoplasm, not between the genes themselves. When repro-
duction occurs, all these cytoplasmic gene products are lost
with the death of the body ; only the unaltered reserves, only
the unmixed genes themselves, go into the germ cells that
produce the new generation. And further, in reproduction
from two parents, the combination of genes that gave origin
to the new characteristics is taken apart and the genes are
recombined in new ways, in the next generation. Thus the
new types, with their perhaps superior characteristics, are
destroyed as types, through the same process that produced
them. The new or superior types so produced are taken apart
at each reproduction; their components are scattered and
combined again to form other types. No type produced by
recombination of genes is permanent, so long as reproduction
from two parents occurs in every generation.

But in many organisms, including the higher plants and
many lower animals, reproduction from two parents is not the
invariable method. Reproduction may take place by the
division of single individuals. This gives a method of perpet-
uating the new characteristics and new combinations that
have been produced in the way we have described. In such
reproduction from a single parent, the existing combination
of genes is not taken apart. Each gene divides, half of it
passing into the cells of each of the new individuals. All the
individuals produced by this method of uniparental repro-
duction have therefore the same combination of genes; a
combination identical with that in the original parent. A
great number, an entire stock or race, is produced, all having

24 GENETIC VARIATIONS

the same combination of genes, all having the same hereditary
characteristics.

It is through this method that is produced much of the
astonishing variety seen in cultivated plants. This is the
reason v^^hy the horticulturist is able to progress so much more
rapidly than the animal breeder in developing new types. By
mating together different strains he produces a great num-
ber of diverse combinations, some of them differing very
greatly from the original stocks. These combinations he
multiplies and makes permanent through reproduction from
the single parents — by cuttings, bulbs, and similar methods.

But this method cannot be used w^ith the higher animals
and man. Here reproduction from two parents is the only
method. Every combination produced in one generation is
taken apart in the next. A slow and inefficient substitute for
uniparental reproduction in making the new combinations
permanent is found in inbreeding — the mating of close rela-
tives. Individuals having the same, or nearly the same combi-
nations of genes, are mated. Their offspring necessarily differ
much less from their parents and from one another in their
combinations of genes than in the usual case of mating be-
tween parents with very diverse sets of genes. By continuing
for many generations the mating of close relatives, stocks may
be produced that are nearly uniform and permanent — until
they are again crossed with other stocks. In this way even in
the higher animals stocks are producible that are lasting, and
that differ greatly from the stocks from which they took
origin. By selecting and isolating the best of these, great im-
provement can be brought about. Some time, some ruthless
race, like the Japanese, will apply this procedure to man, with
possibly startling results.

Biparental reproduction and inheritance thus plays an im-
portant role in evolution. It brings about transformism, it

NEW COMBINATIONS OF GENES 25

brings about diversification. There being given a great variety
of different types of genes, it brings about adaptation and
progress — through the method of selective survival and elim-
ination. But it requires the previous existence of many diverse
types of genes, and it does not produce these diverse types.
What it can do is limited by the potentialities of the different
genes already existing. These as yet unrealized potentialities
of the genes now existing are doubtless very great, and these
may in the future be brought to realization by the making of
new combinations among them. But the actual production
of diverse types of genie materials is required; it is the essen-
tial feature of evolution. To understand evolutionary progress,
the method of producing new types of genie materials must
be sought.

To this search for the fundamental factor in evolution
much effort has been devoted. The sought-for changes in the
nature of genes have been called mutation; and many discov-
eries of mutation have been announced. But often the
announcements have been later discovered to be errors; the
changes observed were something much less fundamental.
This is a history which at the present moment appears to be
repeating itself on a grand scale.

But in this search, other types of genetic variation, of only
less importance, have been discovered. Great numbers of
genetic variations have been found to be the result of irreg-
ularities in the structure and operation of the genetic system.
These things we examine in the next chapter.

NOTES ON CHAPTER I

^ Page 3. See the discussion of this matter in H. F. Osborn, "Aristo-
genesis, the Creative Principle in the Origin of Species," American
Naturalist, Vol. 68, 1934, pp. 193-235.

2 Page 6. For detailed accounts of the constitution of the germ plasm —
the chromosomes and genes — see some modern text-book of cytology;
for example, L. W. Sharp's Introduction to Cytology (New York,
1934), or E. B. Wilson's The Cell in Development and Inheritance
(New York, 1928). For details as to their relation to genetics, some
detailed text-book on that subject may be consulted, such as the author's
Genetics (New York, 1935), or Sinnott and Dunn's Principles of
Genetics (New York, 1932). A simplified account of both of these mat-
ters will be found in the present author's The Biological Basis of
Human Nature (New York, 1930).

^ Page II. Practically all statements in biology are subject to exceptions.
A certain limited number of needed genes may be lacking in the set
supplied by one of the parents.

* Page 13. As will be shown in Chapter III, some or all of these "mod-
ifications" of a particular gene may possibly be simply the same gene
in different relations to other genes, and yielding different effects on
account of these different relations. For the present the text follows the
account commonly given.

^ Page 18. There is a tendency for two genes that are close together in
the maternal series or in the paternal series to remain together in the
offspring. But this is not fixed or absolute; genes that are side by side
may separate into different offspring.

® Page 21. Description and plate on this in Wm. Bateson, Mendel's
Principles of Heredity, 1909, p. 294 and Plate VI.

CHAPTER II

(i) Genetic Variations Resulting from Irregularities in
THE Structure and Operation of the Genetic System

(2) Gene Mutations

IN SEARCHING for the method by which new kinds of
genes are produced, investigators have come upon another
type of genetic variation. The genetic system is a mechanism,
operating in a complicated fashion. As the cells divide and
other operations occur, the genetic system appears under the
microscope to be of an astonishingly machine-like character.
And like other mechanisms it is subject to getting out of
order, to accidents, to breaks and irregularities of operation.
When it fails thus to act in the expected manner, it causes
changes in the characteristics of the individuals that are
produced; it yields genetic variations.^

This production of genetic variations by breaks and irregu-
larities in the genetic system plays at present a great role in
genetic science, and it seems destined to play a still greater
role in the future; one till recently unsuspected.

As you recall, the different materials of the genetic system —
the genes — are strung up in long strings (Figures i and 2).
These strings are divided into segments, which in their con-
densed condition we call the chromosomes. Since a set of
genes comes to the individual from each of his two parents,
there are two sets of genes, and therefore two sets of chromo-
somes in his cells. The chromosomes, like the genes of which
they are composed, are therefore in pairs; one member of
each pair derived from the individual's father, one from his
mother.

h-^-ifLlS^sst^

28

GENETIC VARIATIONS

When the individual himself forms germ cells, the two
chromosomes of any pair separate into different germ cells;
the maternal member of the pair going to one germ cell, the
paternal one to another germ cell (Figure 4). Thus each germ
cell receives but one chromosome of each pair (and conse-
quently but one gene of each pair). Later tvi^o germ cells
(sperm and ovum) unite, and as each brings a member of
each pair of chromosomes, the new individual or fertilized
egg again has its chromosomes (and genes) in pairs, as shown
in Figure 4.

aren

is

Germ Cells Fertilized Egg

Figure 4. Diagram of formation of germ cells and their union to
form the fertilized egg, to show the separation of the chromosomes of
each pair into different germ cells, and the re-formation of the pairs in
the fertilized egg.

We come now to the accidents and aberrations that give
rise to genetic variations. The complex and delicate operations
that we have just described do not always take place accu-

RESULTS OF IRREGULARITIES

29

lately or according to specifications. In rare cases, in forming
germ cells, the two members of a pair of chromosomes fail to
separate (Figure 5, 5). Then one germ cell is formed that has
two members of this pair, while another has no member at
all of this pair. This gives rise to many further irregularities.

A

B

C

D

Parents

Germ Cells FertLlized Eggs

Figure 5, Diagram showing how the number of chromosomes some-
times becomes changed through irregularities of division. The cell A
divides in the usual way, producing germ cells containing one chromo-
some from each of the seven pairs. In the division of the cell B, both
members of the second pair go to one of the germ cells, which thus has
eight chromosomes, while the other germ cell has no chromosome of
the second pair. By union of the normal germ cells from A with the
changed germ cells from B, as indicated by the arrows, a germ cell C
is produced that has three chromosomes of the second pair, and
another germ cell D with only one chromosome of the second pair.

When the germ cell with two chromosomes of a pair unites
at fertilization with one of the usual germ cells, an individual
is formed that has three chromosomes instead of the usual
two (Figure 5, C), for this pair; the genes of this pair are

30 GENETIC VARIATIONS

then in threes instead of in twos. And when the germ cell
with no chromosome of this pair unites with a typical germ
cell, an individual is formed with only one chromosome of
that pair; its genes are single instead of in pairs (Figure 5, Z)).

Many cases of this sort have been discovered, in many
different animals and plants. It is found that such changes in
the numbers of chromosomes present alter the development,
and so yield individuals having different characteristics from
the usual ones. An individual with three chromosomes in
place of the two of one of its pairs differs in certain respects
from the typical individuals; an individual with but one
chromosome in place of one pair differs in another way.
These things have been most thoroughly studied in certain
plants, particularly the Jimson weed, Datura,^ and the famous
evening primrose, Oenothera.

Just what changes in the inherited characteristics are made
depends on which pair of chromosomes it is that has an
additional chromosome, or that lacks one of its chromosomes.
The genes of the different chromosomes are diverse mate-
rials, with different functions, different effects on characteris-
tics. Hence adding a chromosome to one pair has a very
different effect from adding it to another pair; and the same
is true of subtracting a chromosome from a pair. Since any
organism has, as a rule, several or many different pairs of
chromosomes, a large number of different heritable types can
be produced in such ways.

An example will give an idea of the variety introduced into
organisms in this way. In Datura, the Jimson weed, there are
twelve pairs of chromosomes. Adding a third chromosome to
any one of the pairs gives in each case different characteristics.
Hence there are twelve different varieties producible through
adding a chromosome in different cases to the twelve different
pairs. Also there are twelve other varieties formed by taking

RESULTS OF IRREGULARITIES

31

away one chromosome in different cases from each of the
twelve different pairs. Many of these twenty-four different
varieties have been produced and studied by Blakeslee and
his associates.

In a similar way, two of the pairs may have a third member
added to them. For twelve chromosomes, there are sixty-six
different ways in which this can be done. This gives sixty-six
other differing varieties, many of which have been produced
and identified. And there are sixty-six other varieties pro-
ducible by subtracting one chromosome from two pairs, in
the sixty-six possible combinations of two from twelve.

When individuals with three chromosomes in place of
one of the pairs produces germ cells, some of these germ
cells have two chromosomes of that pair, others but one
(see Figure 6). When two germ cells unite, each con-

Parents

Germ Cells Fertilized Egg

Figure 6. Diagram of the method by which there are produced indi-
viduals with four chromosomes in place of one of the pairs (in this
case the second pair, counting from the left).

32 GENETIC VARIATIONS

taining two chromosomes of that pair, we get an indi-
vidual that has four chromosomes belonging to one of its
pairs ; the genes of these chromosomes are thus in fours. Such
individuals have different characteristics from any of the
others of which we have spoken. There are in the Jimson
weed twelve different varieties obtainable by giving four
chromosomes in place of two, in different cases, to each of
the twelve chromosome pairs. And sixty-six more types can
be formed by giving four chromosomes to each of two of the
pairs, in each of the sixty-six different combinations of two
from twelve.

Thus the number of diverse types of organisms producible
in these ways is becoming large. We have thus far enumerated
two hundred and thirty-four possible types. It might appear
that we could go on multiplying varieties indefinitely in this
way, by adding or subtracting one or two, to each of three
pairs, then to each of four pairs and so on. But nature fails us
when this is attempted. If more than two pairs are thus
changed by addition or subtraction, yet not all the twelve are
thus changed, the individual cannot live; it seems then to be
unbalanced and so incapable of development.

Yet if all the chromosome pairs lack one chromosome — or
if all have one or two additional chromosomes — then all be-
ing changed equally, there is no lack of balance, so the indi-
viduals do not die, but live and develop. Our diagram of
Figure 7 shows various different types that are producible in
these ways. In the diagram there are represented but four
pairs of chromosomes instead of the twelve of Datura. The
normal individuals, known as diploids (B), have four pairs.
Others are known in many organisms in which the chromo-
somes are single instead of paired; these are known as hap-
loids. Others have three chromosomes of each kind (in place
of but two or one); others have four of each kind (tetra-

E
F

H

RESULTS OF IRREGULARITIES 33

n=4

Haploid

6 SSSS Dif)io'id 2a=8

Trifjlold

3n=12

D SSSS Tetraploid 4n=16

2(1-1=7

Trisomic 2n+l»9

Doubte Trisotnic Z n*2'l0

Zn*l-1=8

Figure 7. Diagrams showing some of the different groupings of
chromosomes, that may produce living organisms having different
hereditary characteristics. In this case the normal grouping is shown
at B; four pairs of chromosomes. At A the chromosomes are single
instead of in pairs, at C they are in threes, at D in fours. In E to / are
various other groupings such as have been observed in some cases.

34 GENETIC VARIATIONS

ploids). All these different types live and develop; they are
known in many organisms. Figure 7 represents also other
types that live: those in which one or two pairs have an addi-
tional chromosome, and the like.

By adding or subtracting one or two chromosomes to the
different groups ("pairs") in these tetraploids or other types,
and in other ways, a great number of varieties are producible
from a single type. Blakeslee and his associates have produced
and studied in this way eighty-nine different varieties in
Datura, the Jimson weed. And these are but a fraction of
those that are possible. Computations show that in Datura
there must be three thousand six hundred and twenty differ-
ent possible types producible from the single typical variety,
by thus altering the grouping of the chromosomes and genes.
There is reason to believe that all of these can be produced
and that they will live.

The conditions thus carefully studied in Datura are not
exceptional; they are found in many other organisms, par-
ticularly in plants. They have been very extensively studied
also in Oenothera, the evening primrose. Here again there are
many different varieties or types, resulting from changes in
the grouping of the chromosomes and genes. You will re-
member that it was on the evening primrose that Hugo De
Vries made the famous studies that he published about the
year 1900 in the great two-volume work, Die Mutationsthe-
orie: a work which may be said to constitute the beginning
of modern genetics. Most of the changes in inherited charac-
teristics that De Vries described — what he called mutations —
are now known to be the result of changes in the grouping of
the chromosomes and genes, of the sort that we have sketched.

Most of the varieties produced thus by altering the numbers
and grouping of the chromosomes are not entirely permanent
when reproduction occurs from two parents. The irregulari-

RESULTS OF IRREGULARITIES 35

ties in the chromosome numbers cause irregularities in the
processes of germ cell formation and fertilization, with the
result that the offspring produced are not all like their par-
ents; many of them indeed cannot live. It is only the cases in
which all the chromosome pairs are doubled — so that there
are four in place of each pair — that are inclined to reproduce
themselves quite normally, so forming a lasting variety.

But in many plants reproduction may occur from a single
parent, by cuttings, bulbs, runners, offsets and the like; these
do not disturb the chromosome arrangements. Hence many
of these irregularly formed varieties may be multiplied and
perpetuated in this way. In cultivated plants that multiply
thus vegetatively, beyond doubt such alterations in the num-
ber and grouping of the chromosomes play a large role in
producing the great variety that is to be observed. If one goes
to a Dahlia show or to an exhibition of Chrysanthemums or
other cultivated flowers, he is quite overwhelmed by the
extraordinary variety of forms, sizes and colors. Such an
exhibition is an astonishing display of genetic variations. The
differences between the differing varieties is often much
greater than between many wild species found in nature.
Part of these great differences are the result of crossing differ-
ent stocks, thus producing new combinations of genes and of
characteristics, and producing actual new characteristics, in
the way described earlier. But many of the different types are
the result of increasing or decreasing the numbers of certain
of the chromosomes, or of all the chromosomes; of altering
the grouping of the chromosomes, in the way we have just
considered. A very large share of the variety seen in cultivated
plants is due to such changes in grouping and combination of
genes and chromosomes; resulting on the one hand from
crossing different stocks ; on the other from aberrations in the
operation of the chromosomal mechanism.

36 GENETIC VARIATIONS

There is strong evidence too that changes of this type play
a certain role in producing different stocks, different varieties
and perhaps different species, in plants that live in a state of
nature. Plants ordinarily have one set of chromosomes from
the female parent, another set, with an equal number, from
the male parent. Such usual individuals may be said to have
tw^o sets of chromosomes; technically they are called diploids.
As we have seen, at times individuals are produced having
but one set instead of two; such are known as haploids. Other
results of irregularities in distribution produce individuals
with three sets, or with four sets. If we call the number of
chromosomes in one full set by the letter n, then in ways we
have indicated, there may be produced individuals carrying
either n, 2n, yt, or 4« chromosomes. The production of all
these different types has been observed in closely studied
plants, like Datura; the different types produced having dif-
ferent characteristics, in flowers, fruits, stems or leaves.

In wild plants it is found that many closely related varieties
or species have chromosome differences of just these kinds.
Thus, in the roses the number of chromosomes in a set was
seemingly originally 7. Varieties of roses are found having 14
chromosomes (two sets); 21 chromosomes (three sets);
28 chromosomes (four sets); even 35, 42 and 56 chromo-
somes; that is, five sets, six sets, eight sets. Again in different
varieties or species of wheat the numbers go by multiples of
7; some have 14 chromosomes, others 28, others 42. The
species and varieties of Chrysanthemums go by sets of 9;
different ones have 18, 36, 54, 72, and 90 chromosomes. In
different Oenotheras again the chromosomes go by sets of 7;
in Potentillas by 8's and so on.

Similar conditions are found too in animals. In some differ-
ent species of starfish the chromosomes are based on sets of
9; there are species having 18, others with 36. Some varieties

RESULTS OF IRREGULARITIES 37

of Cyclops have 6 chromosomes, others 12. A very large num-
ber of cases of this type are known. It appears that changes
in the number of sets of chromosomes present plays an impor-
tant role in bringing about differences of type and variety, in
both w^ild and domesticated animals and plants.

There is another method by which changes in the structure
of the genetic system, changes in the grouping of the genes,
produce genetic variations and add to the variety in the
organic world. In any given species or variety, the genes are
grouped into a certain definite number of chromosomes; in
man 48, in the Jimson weed 24, in the fruit fly 8, in the horse
worm 2, and so on. This grouping of the genes has an effect
on inheritance, for all the genes that are grouped in a single
chromosome tend to go together from parents to descendants;
and similarly all the characteristics that depend on these genes
that are grouped together tend to go together in inheritance.
It is this tendency of characters dependent on genes in the
same chromosome to go together that is called linkage.

Now, it sometimes happens that a single chromosome be-
comes broken into two, thus adding one to the number of
groups of genes. And sometimes two chromosomes belonging
to the same pair, or two belonging to different pairs, become
united, so decreasing the number of groups, or chromosomes,
into which the genes are united. Such changes have been
observed in organisms under cultivation in the laboratories.
They are now producible on a large scale by subjecting organ-
isms to radiations. These changes produce alterations in the
method of inheritance in the organisms in which they occur,
for they alter the linkage, or tendency of characteristics to go
together to the same individual.

And when we compare related species or varieties, we often
find that they have differing groups of chromosomes, such
that one could have been derived from another in one of these

38 GENETIC VARIATIONS

ways; that is, by breaking or union of certain chromosomes.
This is for example the case in different species of Drosophila
and other closely related genera. Metz has given excellent
figures of the chromosome groups in different species of
Drosophila, some of which we have copied in Figure 8. Type
A, in the common fruit fly, has four pairs of chromosomes,

)> . II. M. il.
II , llT II . ~

H

Figure 8. Diagrams showing the chromosome groups of eight differ-
ent related species of flies, after Metz, 1916.

one pair straight, two pairs F-shaped, one pair small. Type B,
in another species, is like Type A except that the small chromo-
somes have united with some of the larger ones, leaving but
three pairs. Type C, found in a third species, is like Type A ex-
cept that the chromosomes of one of the F-shaped pairs have
broken at the point of the V into two straight chromosomes;
so there are in this five pairs. In Type F, both the F-pairs of
Type A have broken into two pairs of straight chromosomes,
giving six pairs instead of four. The various other types result
from similar rather simple changes. There are many other
cases of related species having chromosome groups that thus
differ by slight but definite modifications of a common type.

RESULTS OF IRREGULARITIES 39

In other cases chromosomes of the different pairs are found
to have exchanged parts ; they have exchanged blocks of genes
(Figure 9, A to D). Furthermore, in ways that w^e shall
examine later, the order of the genes in a part of a chromo-
some sometimes becomes reversed, while in the rest of the
chromosome the order remains unchanged (Figure 9, E and

B

E

abcjclefglh tjk Im

e d li L J k I m
-« — • — • — • — • — • — • — •

t

Figure 9. Diagrams of the exchange of genes between chromosomes
of different pairs (^ to D), and of changes in the order of the genes in
a part of the chromosome (E and F). In E and F the order of the genes
is given by the letters; in F the order of the genes between c and h is
reversed as compared with E.

F). In this way genes that are at a distance from each other in
some individuals are brought close together in other indi-
viduals. These changes in the relation of the genes, and the
accompanying disturbances, result in striking changes in the
method of inheritance, and also, as appears probable from
recent discoveries, in the characteristics that are inherited.
The genetic variations so caused appear to play a very great
role; some of the so-called gene mutations are apparently of

40 GENETIC VARIATIONS

this type. Further discussion of these will be reserved till we
take up explicitly the subject of mutations.

But to conclude the matter with which we are now deal-
ing, it is found that a single species may thus consist of a great
jiumber of diverse stocks, differing in the way their genes are
arranged in the chromosomes, but hardly at all in other ways.
It turns out on the whole that the way the genes are arranged
in the chromosomes is less stable than had been supposed. The
genetic system not infrequently undergoes changes in its
structure, in this way giving rise to many slightly differing
races.

To recapitulate then, it is clear that much genetic variety in
organisms is due to changes in the groupings and numbers
of the chromosomes and genes. Such changes occur in plants
and animals that are under study in gardens and laboratories.
They cause in many cases marked changes in the characteris-
tics of the organisms; in their forms, structures, physiology,
and methods of reproduction and inheritance. And different
types found in nature or in domestication show differences in
their chromosome groups that are exactly such as would be
produced in these ways. We here catch nature at work differ-
entiating genetic types, producing strains and varieties that
differ in their inherent constitutions. We thus know certainly
what underlies great numbers of genetic differences among
organisms. We are able here to see evolution occurring; at
least evolution in the sense of diversification.

But it is clear that in all these alterations in the combination
and grouping of the genes, there is nothing that tends directly
to bring about adaptation. In the formation of new combina-
tions that accompanies biparental reproduction, and in the
new grouping of chromosomes and genes that results from
irregularities in the operation of the genetic system, unadap-
tive conditions are as readily produced as adaptive ones;

GENE MUTATIONS 41

indeed, doubtless in the large majority of cases the results are
the reverse of adaptive. Individuals produced with unadaptive
groupings of genes and characters simply fail to develop, or to
reproduce, leaving only such nev^ groupings as can live and
develop under the conditions. If by chance any of these have a
particularly favorable constitution, these may multiply and
become numerous ; in this vi^ay an adaptive result may some-
times come to light. Selective elimination occurs on a grand
scale in these changes; this is clearly observable in the
laboratory and garden studies. Such adaptation as may result
is the consequence of selective elimination and selective
propagation.

Furthermore, the tw^o types of genetic variation that wc
have thus far considered — recombinations of genes, and
changes in the structure of the genetic system — do not involve
any change in the constitution of the genes themselves. They
merely rearrange the genes that already exist. In order that
they shall be effective, they require the existence of many
different kinds of genes. In some way diversity must arise
among the genes themselves, giving rise to new types of genie
materials; otherwise there could be no result from recombina-
tions and regrouping of the genes. We are therefore put again
upon a search for a third type of genetic variation, which shall
change the constitution of the genes themselves.

2. GENE MUTATIONS

What have been believed to be the required actual changes
in the constitution of genes, giving rise to new types of genes
that make possible progressive evolution, have been observed
on an extensive scale; they constitute what are known as
gene mutations.^

In theory or concept, a gene mutation is a change in the
constitution of a single gene; a rearrangement of its constit-

42 GENETIC VARIATIONS

uent atoms so as to yield a chemically and physiologically
altered material. Thus an essentially new gene is held
to be produced, yielding new characteristics, structural or
physiological.

Such change in the constitution of the genes would appear
to constitute the primary and fundamental type of genetic
variation. Without it, recombination of genes appears power-
less to produce effects, for without it there are no diverse
kinds of genes to recombine. If the changes in gene material
are of appropriate character, they supply the remaining factor
that is logically necessary for understanding evolutionary
change. The transmutation, or "mutation," of the genie mate-
rials, will, if it is of appropriate nature, supply new unit genes.
Biparental reproduction shuffles and recombines the different
genes so produced, giving organisms with diverse combina-
tions of characteristics, on which selective survival and elimi-
nation may act. New lasting groupings of the genes may be
given by irregularities in the operation of the genetic system
that we have before described. Thus with mutations of appro-
priate character, there is logically a complete scheme for
evolution that will yield both adaptation and progress.

But the fundamental question remains: Are the observed
changes of genes indeed of such a nature as will yield progres-
sive evolution? With that question in mind, we proceed to
an examination of the changes in genie materials that are
known as gene mutations. (We shall follow common usage
in employing the term gene mutation, although on later
pages its appropriateness as a designation will be called in
question.)

In practice, for a change in the inherited characteristics to
be identified as a gene mutation, certain conditions must be
fulfilled. It must be possible to show that the change is not the
result of ordinary recombinations, such as occur in Mendelian

GENE MUTATIONS 43

inheritance. It must be shown farther that the change in the
genetic system that gives rise to the change in characteristics
is limited (at least mainly) to a single one of the genes; and
that it is not simply the loss of a gene. These conditions are
not by any means easy to fulfil. They can be fulfilled only in
organisms in which there has been acquired a detailed knowl-
edge of the structure and operation of the genetic system, so
precise that the locations of the single genes can be deter-
mined; knowledge that would permit maps to be made,
showing the relative positions of the genes. The effects of
particular genes on the characteristics must be known, so that
the effect of changes in these genes can be detected. Above
all, the results of typical recombinations among the genes,
such as occur in biparental reproduction, must be known.
And finally, the results of chromosomal irregularities must
have been adequately investigated. Until these things are
known, it is impossible to distinguish the results of recombi-
nations and irregularities, on the one hand, from the actual
changes in single genes on the other. There are very few
organisms in which these conditions have been even approxi-
mately fulfilled. The best known of these is Drosophila
melanogaster, the fruit fly. As will appear in the sequel, it
seems doubtful whether there has been hitherto any organism
whatever of which knowledge was adequate for distinguish-
ing gene mutations from other kinds of changes.

However, many changes have been observed that affected
mainly individual genes, and that are commonly called gene
mutations. Whatever their character and whatever their rela-
tion to evolution, they are in themselves of great interest and
importance. A great body of detailed knowledge on these has
accumulated. We shall first deal with the main features of
this established knowledge. Next we shall make an explicit
examination of the question that is of underlying interest for

44 GENETIC VARIATIONS

evolution: whether the observed changes in the genes do
indeed provide the fundamental materials for progressive
evolution.

The changes knovv^n as gene mutations have been found to
occur in genes that affect all parts and functions of the organ-
ism. Indeed, in the organism in which thecourseof heredity is
best known — the fruit fly — practically all knowledge of
heredity is based on the mutations that have occurred. The
ordinary "wild" individuals are extremely uniform; mating
two of these, the progeny are uniform and there is little
opportunity for the study of the rules of inheritance. But
when a gene becomes mutated, and the individual carrying
jt is mated with another in which that gene is not mutated,
the descendants display all the rules and proportions of
Mendelian heredity. In this way the course of heredity for
hundreds of structural and physiological characteristics has
been worked out; in each case there has been a mutation in
a gene affecting the characteristic.

The same single gene becomes mutated in different indi-
viduals in different ways, so as to give different characteris-
tics in each case. So, in Drosophila there is a gene located
near one end of the X-chromosome, which cooperates with
pther genes in producing the color of the eye. If this particu-
lar gene and the others that work with it are in their usual or
"normal" condition, the color of the eye is red. If this gene
is mutated in a certain way (the other genes remaining un-
changed), the eye color changes to white. Other mutations in
this same gene give other eye colors, and in this way, by differ-
ent mutations of this single gene a whole series of eye colors
have been produced, some eleven or twelve in all. These eye
colors, resulting from different mutations of this one gene,
have been given the following names, beginning with those

GENE MUTATIONS 45

showing the deepest colors: red, coral, blood, cherry, eosin,
apricot, ivory, tinged, buff, ecru, white, and ultra-white.

Other genes in Drosophila and in other organisms, are
known thus to have been changed by mutation in many dif-
ferent ways; many series of modifications of a single gene,
like that just listed, are known.

The change induced by the mutation of a gene may be
great, or it may be very small. The eye of the fruit fly is
changed by a certain mutation from red to white; a great
change. By another mutation it is changed in a barely detect-
able way, from red to "coral" or "blood." Based largely on the
so-called mutations of De Vries — which were in fact mainly
the result of gross chromosomal irregularities — there has been
a prevalent tradition that mutations are sudden extensive
changes, "saltations," like the change from red eye to white
eye. It was natural that the first mutations (Observed should be
these very conspicuous changes: sudden loss of wings or of
eyes ; or other marked changes in structure. Such conspicuous
mutations are indeed not infrequent.

But as studies became more exact and detailed, it was
found that many mutations produce extremely minute
changes, not to be detected without great care. For example,
there are known in the fruit fly a number of gene mutations
whose only discoverable effect is to cause a very faint lighten-
ing of the color of the eye in case the individual containing
them has eosin-colored eyes. If the eyes are not eosin-colored,
these mutations have no discoverable effect. The number of
gene mutations having such very slight effects is much
greater than that of those having marked effects ; most muta-
tions produce almost imperceptible changes.

The majority of mutations are recessive in heredity. That
is, if in one of the pairs of genes, one of the two has been
mutated, the other not, the mutated gene produces no mani-

46 GENETIC VARIATIONS

fest effect on the individual; the latter remains quite un-
changed. Thus one or many gene mutations may have
occurred in the chain of genes of a given individual, but so
long as only one gene of any pair is affected, there may
be no manifest effect. Only vv^hen both members of a pair of
,genes are mutated does the recessive mutation produce its
bodily effect. A small proportion of the mutations that occur
are dominant; in this case even when but one member of a
pair of genes has mutated, the individual is changed by the
mutation.

As to their absolute frequency, in ordinary observation the
changes knov^^n as gene mutations seem to occur very rarely.
Yet when large numbers of individuals are examined for
long periods, and the eye is carefully trained for the detection
of mutations, the number occurring in even a short period of
years is very considerable. In the fruit fly, during twenty-five
years of observation, such changes have been seen to occur in
several hundred different genes. In relation to the course of
evolution, twenty-five years is an infinitesimally small period,
so that relative to the course of evolution, gene mutations are
numerous and of frequent occurrence. If they fulfill other
requirements, their numbers are sufficient to account for
evolution.

As to the occurrence of gene mutations, a mutation occur-
ring at a given time appears to affect only a single one of the
genes in the chain of genes. Further, a mutation occurs in
only one of the two like genes that are present in any cell.
The genes are in pairs, the two members of a pair having
similar or identical constitutions and effects. But the occur-
rence of a mutation affects but one of these, leaving the other
unchanged. It can even be shown that when a single gene has
split in two, in preparation for cell division, one of the two
halves may become mutated while the other does not. All

GENE MUTATIONS 47

these relations show that whatever is causing the mutation is
precisely localized and extremely restricted in its action.

Since most mutations are recessive in their effects, and since
only one gene of the pair of genes present is mutated, as a
rule the mutation produces no manifest effect on the indi-
vidual in which it occurs. The fact of its occurrence cannot be
detected in that individual. It is only when by the processes
of mating in ordinary reproduction, two of the recessive
mutations get together in the same pair, in some of the de-
scendants of this individual, that the mutation appears to view.

Thus, when the recessive mutation becomes manifest, it
has actually occurred, as a rule, at least two generations earlier,
in one of the ancestors of this individual. This relation intro-
duces much difficulty into the study of the time and place at
which mutations occur. Many mutations do not come to light
for several or many generations; indeed there must exist in
any organism many mutations that have never come to light.
In the course of generations a large number of recessive muta-
tions may have collected in the chromosomes, none of them
producing any efTect, because each is accompanied by a
normal or unmutated gene of the same pair. If now inbreed-
ing — the mating of close relatives — occurs, a number of these
mutated genes may be brought together into the same pairs,
in one individual ; thereupon the effect of all these mutations
suddenly become manifested in that individual. Since most
mutations are harmful in their effects, this individual will be
weak and defective, perhaps in several different ways. This
is the reason why inbreeding is harmful; if there were no
gene mutations, apparently there would be no disadvantage
in the mating of close relatives. One of the chief results of
biparental reproduction — what might be called one of its
functions — is that it largely nullifies the effect of harmful
mutations.

48 GENETIC VARIATIONS

In view of the fact that mutations may remain long hidden,
it is fortunate for purposes of study of the occurrence of muta-
tions that in many organisms a certain number of genes in
one of the chromosomes, known as the X-chromosome, are in
one of the two sexes not paired, but single. In most organisms
that have been extensively studied from this point of view, it
is the male that has a certain number of genes that are single,
in the X-chromosome. If any of these unpaired genes become
mutated, in the fertilized egg or in the cells from which the
ovum is derived, the effect of the mutation appears at once, in
case the individual developed from that egg is a male. Thus
the time and place of the occurrence of gene mutations has
been studied most extensively in the genes of this X-chromo-
some. In recent years, however, an elaborate and adequate
technique has been devised for bringing to light almost at
once the effect of mutations in the genes of any of the
chromosomes. In this way much has been learned as to the
time and place of the occurrence of mutations, and as to the
agents causing them.

It has been discovered that a gene mutation may occur in
any cell of the body, at any period in the life of the organism.
The mutation occurs in but one single cell. If this is the single
cell of the fertilized egg, then all the cells of the body devel-
oped from that egg will carry the mutated gene. If the cell in
which the mutation occurs is one of the two cells into which
the fertilized egg divides, then only half the body will carry
that mutation. If the mutation occurs in a cell at a still later
stage of development, only a small part of the body will show
the mutation. Many cases of such kinds have been thoroughly
studied, cases in which one part of the body shows the muta-
tion, the rest not. The general upshot of all these studies is to
show, as remarked before, that gene mutations may occur in
any cell of the body, at any period in life.

GENE MUTATIONS 49

From this it follows that many gene mutations occur that
have no manifested effect whatever; they do not change the
characteristics of the individual in which they occur, nor those
of his descendants. For example, consider a mutation in a
gene that aflfects the color of the eye. Eyes derived from the
cell in which this mutation occurs will be changed in color.
But many of the cells of the body do not produce eyes; in-
stead they produce legs, wings, parts of the body, or the like.
Mutation of an eye-color gene in such cells will thus have no
effect on eye color, no visible effect on any characteristic. It
can probably be said with truth that most eye-color gene
mutations have no effect on eye color, for most of them
occur in cells that do not produce eyes. Doubdess there occur
great numbers of such mutations that yield no visible effect;
any individual may carry many such mutated genes. All such
mutations, present in body cells only, are totally lost at the
death of the individual carrying them.

When a mutation has occurred, is the change a permanent
one? Or may the genetic system later return to its original
condition ?

The great majority of mutations appear to be permanent
changes. They are inherited by the progeny of the mutated
individuals, and such inheritance continues for an indefinite
number of generations.

But in recent years a considerable number of cases have
been discovered in which the change is not permanent. Hav-
ing mutated in a certain way, producing its usual effects on
the characteristics, the genetic system later changes back to
normal, so that the original characteristics are restored. Some-
times such a reversion to normal is produced by subjection of
the mutated genes to radiation. In other cases the mutated
genes show a marked tendency to revert to the original condi-
tion; reversion occurs in a considerable proportion of them,

50 GENETIC VARIATIONS

without the action of radiation. A considerable number of
mutations that behave in this way have been discovered. This
is a matter of much interest in relation to the nature of muta-
tions; some examples will therefore be mentioned.

Demerec* found in a species of Drosophila that a mutation
affecting a certain gene located at a particular point in the
X-chromosome caused the body to take on a reddish tinge,
instead of the normal gray. Another mutation of this same
gene caused the body to become yellowish in color. When
these two differently mutated genes were brought together in
the same cells, it was found that in about one-fifth of all the
cases in which this was done, the genes that had mutated to
reddish were transmuted back to the original condition, so
that they produced again the original gray color. Thus the
mutation that produced the reddish color was not a perma-
nent change. In some way the presence with it, in the same
cell, of the gene that had mutated to yellow caused in some
cases the reversal of the change to reddish.

The proportion of the cases in which the reddish gene
changes back to normal was found to differ in different fam-
ilies. By selection from the different families and further
breeding, Demerec obtained families in which the reddish
mutation never returns to normal ; others in which the rever-
sion occurs very frequently.

Other reversible mutations were found in the same animal.
A mutation of a certain gene ("miniature") in the X-chromo-
some caused the wings to be small.^ In later generations this
mutated gene frequently reverts to normal. By selection from
different families it was possible to obtain some stocks in
which the return of the mutated gene to normal takes place
very rarely, the mutation remaining permanent in most cases.
In such a group there were but five reversions in 11,609 cases.
In other stocks the reversion is very frequent, occurring in up

GENE MUTATIONS 51

to 50 per cent of the cases. The reversion may occur in either
the germ cells, or in certain of the body cells. When the
reversion occurs in the germ cells, the offspring of the mutant
(small-winged) individuals are of the normal large- w^inged
type. When the reversion occurs in the body cells, those parts
of the wings derived from these reverted cells are of the nor-
mal large type, while the remaining parts of the wings are of
the miniature type. Thus mosaic individuals are produced,
portions of the wings large, other parts small, in the same
individual. In certain families reversion from mutated to
normal occurred frequently in the body cells, yielding mosaic
individuals; but never in the germ cells.

Extreme cases of frequent, reversible, and inconstant muta-
tions are known in plants. Typical examples of these are seen
in variegated maize or Indian corn." In the variegated varie-
ties, parts of the same ear of corn show seeds of different
colors: red, yellow, white. Even single seeds on the ear may
be variegated; splashed or streaked with two or three colors
in small areas. The variegated condition is known to be due
to a single gene that differs from one of the genes in the
uniform varieties.

When the variegated plants are bred together they give a
number of different sets of offspring, with different types of
coloration. Some of the offspring have the seeds mainly of one
color; some mainly of another color; some much variegated,
some all of one color. The diversities of color tend to be inher-
ited, showing that the differences in color are due to differ-
ences in the genes. If from a variegated ear there be selected
two seeds, one lighter in color, the other darker, the former
gives lighter offspring, the latter darker offspring — each still
with much variation. From a single seed of a certain single
color a plant may be produced having diverse colors in its

52 GENETIC VARIATIONS

diflferent seeds, and these colors are (partly) heritable in later
generations.

Thus in these plants the genes frequently mutate so as to
yield at one time one color, at others another. The mutations
occur in many different ways, yielding many different colors
and patterns of variegation.

These facts as to frequent, reversible and inconstant muta-
tions raise questions as to the essential nature of mutations.
What is it that is really occurring here ? To this question we.
return.

What are the agents that produce mutation? For many
years after mutations w^ere found to occur, nothing was
known as to the agents that cause them. A few genes mutated,
as a few of the atoms of a radiant metal disintegrate; in both
cases no outside agent appeared to be at work. There seemed
to be no relation between the environment and the produc-
tion of genetic variations.

But of late this situation has changed. The connection be-
tween environmental action and genetic variation has been
made. Certain outer agents are found to play a role in produc-
ing gene mutations. In 1927 H. J. MuUer' discovered that by
subjecting developing organisms to a powerful dose of X-rays,
the genetic system was altered in many ways. The X-rays
injure the animals; they kill a large proportion of them; they
leave many of the rest sterile. Still other survivors are found
to have suffered changes that leave their descendants with
hereditary abnormalities, weaknesses and other peculiarities;
and these are found to be the result of changes in the genetic
system caused by the radiations, changes that are known as
gene mutations. The radiations have caused the chromosomes
to break, caused pieces from different chromosomes to stick
together; destroyed some of the genes; produced irregularities
of many kinds. And amid all this wreckage, they have pro-

GENE MUTATIONS 53

duced a lot of gene mutations, to which are due the inherited
abnormalities, weaknesses and other aberrations. The number
of visible mutations produced under the action of radiation is
about seventy-five times as great as occurs without radiation.

This effect of radiations in producing mutations was con-
firmed by others and extended to other organisms. Radiations
from radium have the same sort of effect as X-rays; they
greatly increase the number of gene mutations. Gene muta-
tions have been produced by radiations in maize, barley,
tobacco, cotton, various animals, and many other organisms.

By radiation the same types of mutations are produced as
occur spontaneously. In many cases the mutations due to the
radiations are identical with others that had occurred without
radiation. Others induced by radiation differ from any before
observed. The same gene may be caused by radiation to pro-
duce repeatedly in different individuals the same mutation.
Since all kinds of mutations are induced by radiation, it is
clear that there is here no relation between the kind of muta-
tion produced and the kind of environmental agent that
induces them. The mutations induced by radiation are inher-
ited in later generations, as are those that occur spontaneously.

The number of mutations produced in a given number of
individuals is found to increase in proportion to the energy of
the radiation; that is, to its intensity and duration, up to an
energy that causes injury so great that the radiated germ cells
do not develop.

The question arises as to whether the "spontaneous" muta-
tions are not partly or wholly the result of radiations occur-
ring under natural conditions. It was found that fruit flies
kept in a mining shaft in which natural radiation from the
minerals present was abundant gave a larger proportion of
mutations than those kept under usual conditions, indicating

54 GENETIC VARIATIONS

that the natural radiations are indeed effective in producing
mutations. But a very careful study by MuUer and Mott-
Smith, based on the fact that the proportional number of
mutations varies as the intensity and duration of the radiation,
showed that the number of mutations that occur under nat-
ural conditions is much greater than can be accounted for by
the frequency of natural radiations. It is clear therefore that
some other agent is at w^ork.

And another agent has been identified. Muller found that
the number of mutations is slightly increased when organ-
isms are kept at a high temperature — ^29 degrees in place of 20
degrees. Then Goldschmidt, and after him Jollos/ employed
heat that was destructive; up to 37 degrees. Again, as in the
case of radiation, most of the organisms are killed. Again the
survivors show injuries and weaknesses of various kinds. And
again the progeny of survivors show inherited abnormalities
and peculiarities which are found to be due to gene mutations.
The same types of mutations are produced by heat as occur
spontaneously and under radiation; they are inherited in the
same manner.

Certain other matters of great interest have been reported
by Goldschmidt and Jollos in connection with the mutations
induced by heat. In some of their experimental cultures very
great numbers of mutations were produced. It was further
reported by these authors that the heat produces changes in
the bodies of the individuals subjected to the heat, and that
in later generations similar inherited changes appear as a
result of gene mutations. However, these later mutations did
not as a rule appear in the descendants of the particular indi-
viduals that had been directly changed by the heat.

In the continuation of this work by Jollos, a further rela-
tion, of the very greatest interest, is reported. Subjection of
one generation to a high temperature causes a slight inherited

GENE MUTATIONS 55

effect, due to a slight mutation of a particular gene. Subjec-
tion of those changed descendants to the high temperature
for another generation is reported to increase the effect and
alter the gene still more. By continued subjection to high
temperatures for many generations a series of gradations is
thus produced, each one hereditary at ordinary temperatures,
but passing from slight changes in earlier generations to very
great ones in later generations. Thus, in one series of experi-
ments, a certain gene was so altered, in the first generation
subjected to heat, as to cause a slight darkening of the body
color. Subjection to heat in later generations increased the
effect on this gene until finally the body was black. In another
series, the red color of the eye was, by subjection of successive
generations to heat, caused to become lighter and lighter
until it was yellowish, and until finally entirely white eyes
were produced. The mutations to which these changes in
eye color were due were all in a certain gene, that is perhaps
the best known gene in the fruit fly. Thus what could be
called directed mutations were induced; series of changes in
a certain direction for successive generations; what might be
called experimental orthogenesis.

Furthermore, according to Jollos, the nature of the muta-
tions was dependent to some extent on the nature of the con-
ditions in which the organisms were subjected to the heat.
Those in moist conditions, for example, gave a different set
of mutations from those in dry conditions.

These rather revolutionary conclusions are, however, not
confirmed by the extremely extensive and long-continued
work of Plough and Ives.^ These authors confirm the fact
that high temperatures increase the frequency of mutations.
But they do not verify the other conclusions of Jollos; and
particularly they find no indications of the directed muta-

56 GENETIC VARIATIONS

tions, or orthogenetic series reported by JoUos. As matters
stand, the more radical conclusions of Jollos cannot be con-
sidered established.

The discovery of agents producing mutation brings us
squarely to the question in which these lectures are primarily
interested. Both the known agents are powerfully injurious.
Both kill a large proportion of the individuals exposed to
them ; both leave many of the others abnormal or imperfect.
Both damage the genetic system, producing changes that
make it impossible for the individuals to develop. And both
produce inherited changes that are manifested mainly as
abnormalities, weaknesses and imperfections.

Are these the type of changes on which evolution is based ?
Amid the carnage engendered by blighting radiation and
unbearable heat, do we find produced the changes that mean
progressive evolution? The gene mutations produced by
radiation and by heat are of the same types as occur sponta-
neously. It is commonly assumed and urged that the muta-
tions observed in experimental genetics are the raw material
of evolutionary change; are visible steps in evolution. Is this
to be considered established ? This question we take up in our
next chapter.

NOTES ON CHAPTER II

1 Page 27. Changes and irregularities in the genetic system in relation
to the differentiation of varieties, species, etc., have been very fully
described and figured in C. C. Hurst's The Mechanism of Creative
Evolution (New York and Cambridge, 1932). See also T. H. Morgan,
The Theory of the Gene (New Haven, 1928) and The Scientific
Basis of Evolution (New York, 1932).

^ Page 30. The work on Datura is by Blakeslee and his associates. See
A. F. Blakeslee and J. Belling, "Chromosomal Mutations in the Jimson
Weed, Datura stramonium," Journal of Heredity, Vol. 15, 1928, pp.
195-206; and the references there given.

^ Page 41. For general accounts of gene mutations, see the references
given in Note i, above.

* Page 50. See M. Demerec, "Reddish — A Frequently 'Mutating' Char-
acter in Drosophila virilis," Proceedings of the National Academy of
Sciences, 1926, Vol. 12, pp. 11-16.

^ Page 50. See M. Demerec, "Miniature-Alpha — A Second Frequently
Mutating Character in Drosophila virilis," Proceedings of the National
Academy of Sciences, 1^26, Vol. 12, pp. 687-90.

^ Page 51. For an account of variegation in Maize, see W, H. Eyster,
"A Genetic Analysis of Variegation," Genetics, 1924, Vol. 9, pp.
372-404.

^ Page 52, See H. J. Muller, "Artificial Transmutation of the Gene,"
Science, 1927, Vol. 66, pp. 84-7, and "Types of Visible Variations In-
duced by X-Rays in Drosophila," Journal of Genetics, 1930, Vol. 22,

PP- 299-334-

® Page 54. See V. Jollos, "Studien zum Evolutionsproblem. I. Ueber die

experimentelle Hervorrufung und Steigerung von Mutationen bei

Drosophila melanogaster," Biologisches Zentralblatt, 1930, Bd, 50, pp.

541-54-

^ Page 55. See H. H. Plough and P. T. Ives, "Heat-Induced Mutations

in Drosophila," Proceedings of the National Academy of Sciences, 1934,

Vol. 20, pp. 268-73.

CHAPTER III

Are the Gene Mutations Observed in Experimental
Genetics the Basic Materials of Progressive Evolution?

WE TAKE up in the present chapter the question
proposed at the end of the preceding. Do the gene
mutations observed by the students of genetics constitute the
types of change of genie materials that are required for evo-
lution; and particularly for progressive evolution? Are they
steps in evolutionary progress ?

Here we frequently meet an argument from definition.
Evolution, it is urged, requires, and is finally based on,
changes in the constitution of the genie materials. And any
change in the constitution of the genie materials would be by
definition a gene mutation. Hence it argued that evolution
certainly rests on gene mutations. But such argumentation
has no relation to observed facts; it was equally valid before
any mutations were seen. The question of interest is: Has
genetic science actually before it steps in evolution? Do the
observed gene mutations constitute the type of change that
has resulted in progressive evolution ? It is to this question that
we shall address ourselves. Our discussion relates only to
observed types of change; not to conceivable but unseen types.
What may be a definite answer to these questions is now
emerging before us. On it there are several lines of evidence.
First is the fact, already set forth, that the agents known to
produce gene mutations are definitely destructive agents.
They injure the organism in all its parts and functions,
including its genetic parts and functions, including its germ
plasm. By an astounding paradox, we are asked to believe that

GENE MUTATIONS AND EVOLUTION 59

they also produce the changes on which evolutionary progress
is based.

But more probative than this are two other lines of evi-
dence. One lies in the observed effects of gene mutations on
organisms. The other lies in an experimental analysis of the
nature and causes of mutational change; in the recent uncov-
ering of some of the hidden processes occurring at mutation,
showing just what is happening. We shall look into both
of these.

From the first it has been observed that when carefully
examined most mutations are discovered to be defects. The
organism is the worse for their occurrence. They are not of
such a nature as to yield adaptations, but the reverse. The
changes that they represent are not progress, but the opposite.

These relations are shown in two ways, each of great
importance. First, the specific effect of the mutation on some
particular part or function of the organism is, in most cases,
to bring about a distinct loss, or an abnormality. Certain
mutations cause the body to be distorted, or imperfect in
parts, or to lose certain typical structures. Others make the
limbs abnormal or small, or result in their total loss. By
other mutations the eyes are made imperfect, or very small, or
are totally lost. All these are well known and typical results
of gene mutation : great numbers of such cases could be cited.
In general it is true that a part that has been altered by a
mutation is not so well adjusted to the rest of the organism, or
to the surrounding conditions, as it was before the mutation
occurred. Whether there are any exceptions to this we con-
sider later. But certainly the general impression from a group
of mutated organisms is one of defectiveness and abnormality.

Second, and perhaps still more significant for the nature of
mutations, is the fact that, in addition to their specific effects
on particular parts of the body, mutations usually have a

6o GENETIC VARIATIONS

general effect on the organisms as a whole, and this effect is
injurious. They distinctly weaken the constitution of the
individuals in which they exist. The result is that the
mutated individuals lack resistance to severe conditions, and
have a shorter life and a higher mortality rate than those not
mutated. It is a common observation that individuals show-
ing the mutated characteristics are difficult to keep alive;
specially favorable conditions must be supplied or they perish.
Indeed, a large proportion of the mutations that occur cause
a complete failure of development, unless the mutated gene is
accompanied by an unmutated gene of the same type. Such
are known as lethal mutations. One is surprised, when he
comes to study the occurrence of mutations, to discover that
the commonest types are these lethal mutations; a large pro-
portion of our knowledge is based on these changes that cause
death. Often when one gene of a pair is thus affected by a
lethal mutation, while the other is normal, the individuals
are weak and abnormal in structure and function. If both
genes of the pair are affected by the mutation, the individual
"dies before it is born."

There occur gene mutations whose effects, both special and
general, are so slight that from examination of the individuals
affected by them, it is difficult to determine whether they are
harmful or not, although as a rule the general impression they
give is one of slight abnormality or loss. But here the correct
interpretation is aided by certain extremely ingenious
experiments made possible by the action of radiation on
chromosomes and genes. The results of these experiments are
illuminating; we must examine them.^

As will be recalled, normally there are in the cells of any
individual two genes of each type, forming a pair. That is,
there are present two "doses" of the material represented by
the gene. By the use of radiations it is possible to obtain indi-

GENE MUTATIONS AND EVOLUTION 61

viduals having but one dose of any particular type of gene
material, and others having three doses (Figure 10). An
individual carrying a pair of mutated genes, one in each of
its chromosomes, is subjected to radiation. By this, a small

B

Figure 10. Diagrams of a small portion of the chain of genes,
showing how it is possible to compare the effects of one, two or three
mutated genes. A. Normal condition, two mutated genes (at right).
B. One mutated gene removed. C. Three mutated genes present.

portion of one of the chromosomes, carrying the mutated
gene, can be removed, leaving the organism with but one
dose. Or by radiation a small piece of the chromosome con-
taining the mutated gene is broken off, and later by proper
breeding introduced into an individual that already has the
usual two doses; so that now it contains three. Thus it
becomes possible to compare the effects of one, two, or three
doses of a certain mutated gene.

An example will illustrate the nature of the relations dis-
covered in such comparisons. The eye of the fruit fly is

62 GENETIC VARIATIONS

normally a deep red in color. When a mutation has occurred
in a certain gene that is near the left end of the X-chromo-
some, this causes the eye to change to a lighter color, known
as eosin. The effect of the mutation is thus to lighten the
color of the eye. This lighter color, being recessive, is mani-
fested when two of the mutated genes are present together
(Figure io,A).

Removal of one of these two mutated genes, leaving but
one dose of the mutated material (Figure lo, B) would
naturally be expected to decrease the effect of the mutation,
causing the color not to be lightened so much as when two are
present. But when this was done, it was found that the light-
ening effect was increased instead of diminished; with but
one mutated gene the eyes are lighter in color than with two.
Two mutated genes give a color nearer the normal red than
does one mutated gene. And when a third mutated gene was
added (Figure lo, C), the eye color became still darker,
approaching closely that of the normal red eye.

It thus becomes clear that the effect of the mutation has
been to reduce the normal action of the gene, so that more
doses of the mutated gene are required to produce the deep
color than are required of the unmutated gene. The mutated
gene has the same kind of action as the normal gene, so far
as it goes, but its action is imperfect, incomplete. For it is
found that the presence of only one dose of the mutated gene
leaves the individual very different from the normal unmu-
tated individual. If two are present, the result approaches
more nearly that of the normal gene, while if three of the
mutated genes are present, the individual becomes almost
like one in which there is no mutation.

It will be recalled that this eye gene is one that has been
mutated in many different ways, yielding many different
colors. One of the other mutations, giving a still lighter color.

GENE MUTATIONS AND EVOLUTION 63

known as apricot, has been tested in the same way, with the
same results. One dose gives a very light color; two give a
darker color; three give nearly the normal color. This muta-
tion too is a reduction in the action of the normal gene. The
entire series of mutations giving different grades of color are
different degrees of reduction of action of the normal gene.
Whether it is conceivable and possible that in some of the
mutations one part or feature of the normal activity is
reduced, while in another, another part is reduced, I will not
attempt to say.

In some cases it is not at first obvious that the effect of the
mutation is a reduction in action as compared with the nor-
mal gene, but experiments of the type just described show
that this is indeed the case. The body of Drosophila carries a
large number of simple bristles. By mutation of a certain
gene, some of the bristles are caused to be forked instead of
simple. But this forked condition is most marked in individ-
uals that have but one of the mutated genes (the other gene
of the pair being completely absent). If two of the mutated
genes are present, the forking is still perceptible, but is not
quite so marked. But if three mutated genes are present, the
bristles are much less distinctly forked; they are more like
normal simple ones. It seems that the forking is due to the
fact that certain normal developmental processes are in the
mutated individuals not fully carried out; the effect of the
mutation is to reduce the normal action. The greater the
number of mutated genes present, the more nearly the nor-
mal processes are carried to completion.

Tests of this kind, so far as carried out, indicate that most
mutant genes are of this type. The mutational change con-
sists in a reduction in the activity of the normal gene. So far
as this is true, mutation is process that is not constructive, but
destructive; not progressive, but reductional.

64 GENETIC VARIATIONS

It agrees well with this fact that most mutations are reces-
sive, as compared with the effect of the normal gene; so that
when the mutated gene and the normal gene are present
together, the effect on the characteristics of the organism is
that of the normal gene alone. There do occur, however,
mutations that are dominant; they produce a positive effect
different from that of the unchanged gene; an effect that is
manifested even when the unchanged gene is present. But
such dominant mutations not only as a rule are manifested
as marked imperfections or abnormalities, but also they are
seriously injurious to the constitution of the individual, so
that if two doses of the dominant mutation are present
usually the individual dies or fails to develop at all.

On the whole it is certain that the overwhelming majority
of gene mutations are not adaptive, not progressive, but the
reverse.

But are there no gene mutations that are beneficial to the
organisms in which they occur ? Are there none that leave the
organism more vigorous than before their occurrence? Are
there none that are progressive in character ?

It is possible to collect from the literature of genetics a few
cases in which gene mutations are held to be possibly or
probably beneficial. In the cotton plant there are two varieties,
one having forked or deeply cleft leaves; the other, leaves
that are nearly entire, being but slightly lobed. Horlacher and
Killough" observed that under radiation a mutation from
forked leaves to the lobed leaves was produced. As the lobed
leaves have more leaf surface, Horlacher and Killough hold
that this mutation may have been beneficial.

Again, in the production of mutations by high tempera-
tures, one of the mutations is from the red eye color to white
eye; a common mutation also under radiation, and occurring
too under natural conditions. It is a typical loss mutation,

GENE MUTATIONS AND EVOLUTION 65

resulting in weakening of the constitution, so that under
ordinary conditions the white-eyed flies show higher mortal-
ity than the normal red-eyed. But JoUos^ finds that the white-
eyed flies have a greater resistance to high temperatures than
the normal individuals with unmutated red eyes. This is, so
far as I know, the only case of a mutation held to increase the
resistance of the organism to bad conditions.

Other cases of mutations held to be advantageous could
doubtless be culled from the literature. But they are rare, and
the evidence they present is not impressive. From them a
convincing case cannot be made out for beneficial and adap-
tive mutations.

But with relation to progressiveness, we find it commonly
stated as an established fact that not all mutations are reduc-
tional, although most of them are. Some of them, it is
asserted, are progressive.* But when we look at the examples
of such progressive mutations, what do we find ? The exam-
ples given are what are known as reverse or return mutations.
A normal gene has mutated in the usual reductional manner.
In some cases it later returns to the normal condition, and this
is called a reverse mutation. As before described, a consider-
able number of such reverse mutations have been observed,
some occurring spontaneously, others under radiation. Now,
if the original mutation was reductional, then, it is urged, the
reverse mutation — the return to normal — must be progressive.
This is logical, but what is its relation to the question in
which we are here interested ? A reverse mutation, it is clear,
gives no example of a change producing progressive evolu-
tion. It restores the original condition, but does not carry
things farther. If mutations are essentially reductions, injur-
ies or abnormalities, the return mutation is merely a cure of
the original injury. If there are mutations that yield progres-
sive evolution, they must be of a different character from this.

66 GENETIC VARIATIONS

Pause now for a moment to look back at the history of the
idea that gene mutations furnish the steps in progressive
evolution. You v^^ill recall that for many years, and until
comparatively recently, nothing was known as to the agent
causing mutations; they appeared to occur "spontaneously,"
or from inner causes.

It was observed that most of them — at least most pro-
nounced mutations — are injurious in effect. But, it was set
forth, this is doubtless to be expected. Assuming that these are
undirected changes, it is natural that marked chance altera-
tions occurring in a complicated system such as an organism
should as a rule be harmful. But among the great number
occurring, there will be some few, it was urged, that turn out
to be advantageous; it is these that furnish the steps in pro-
gressive evolution. Moreover, it could be pointed out that
there are great numbers of slight mutations that are not
obviously disadvantageous or reductional. This line of
argument was plausible, was reasonable; it was generally
accepted.

Something of a blow was given to this conception of the
matter when it was found that the agents that produce gene
mutations are powerfully injurious; that they kill most of the
individuals subjected to them, and that they damage the sur-
vivors in all sorts of ways. Particularly remarkable did it
appear that they damage the genetic system, breaking up the
chromosomes, destroying some of the genes and so producing
deficiencies; injuring others so that they will not support
development, and others so that they produce inherited losses
and abnormalities. It seemed a paradox that among ihese
things we should find the progressive changes that give
evolutionary advance. But some paradoxical things are true,
and the theory stood up under the burden thus laid upon it.

GENE MUTATIONS AND EVOLUTION 67

A distinct addition to this burden came when it was shown
that the slight mutations, not obviously disadvantageous, are
in fact reductional in character; that the mutated gene is in
fact a weakened gene, giving effects in the same direction as
the normal one, but less efficient.

Heavy further blows to the theory that the observed types
of gene mutation furnish the steps for progressive evolution
come with the recent discoveries which permit an analysis of
the nature of the changes that are occurring in what are
called mutations. These discoveries lie in two fields. On the
one hand, subjection of organisms to radiations, with the
production of typical mutations, has permitted a detailed
experimental analysis of what is really happening. On the
other hand, discoveries have lately been made that give an
unexpected insight into the structure of the chromosomes,
making it possible to determine what kinds of changes have
occurred after a mutation. We look first at the analysis per-
mitted by the results of radiation.

Under radiation it is found that the number of gene muta-
tions produced is proportional to the energy of the radiation
to which the organisms are subjected; also to the number of
free electrons produced by this amount of radiation. In view
of this fact, it was judged probable that a mutation was the
result of the striking of a gene by a free electron. It was held
that when an electron strikes a gene, this causes a transforma-
tion of its molecular structure, in consequence of which
it acquires new properties and so gives rise to changed
characteristics.

But besides gene mutations, radiation produces other effects
on the genetic system. It breaks the chromosomes. It destroys
certain parts of them. It sets free small pieces of them. It
causes irregularities of many kinds in the structure and
operation of the genetic system.

68 GENETIC VARIATIONS

And now it has been discovered that there is close relation
between these gross destructive changes and the production
of gene mutations. Two of the chief investigators in this field,
H. J. MuUer and L. J. Stadler, recently published simulta-
neously two papers calling attention to these remarkable and
significant relations.^ The paper of Stadler develops in detail
the fact that gross injuries and gene mutations grade one into
the other, and that there is no sound basis for making a
distinction in kind between them. The paper of Muller goes
into a fuller analysis of the matter; I shall follow it in
the main.

As to the relation between chromosome injuries and gene
mutations, there are a number of facts that are of great sig-
nificance. The number of breaks and injuries produced by
radiation is proportional to the energy of the radiation. So
also, is the number of gene mutations. Indeed, mutations
occur with about the same frequency as the breaks and
injuries. And what is still more significant, gene mutations
commonly occur in the same chromosomes that are broken or
otherwise injured by the radiation. When a chromosome is
broken under the action of the rays, commonly a mutation
occurs in a gene very close to the point of breakage.

Both the mutations and the chromosome breaks are ex-
tremely rare, absolutely considered. They occur in only one
chromosome out of hundreds exposed to the rays. Their
frequent occurrence together therefore cannot be accidental.
They are in some way causally connected. Whatever causes
a break in a chromosome commonly at the same time causes
a mutation. And it causes a mutation at or close to the broken
end of the chromosome.

Now, there is strong evidence as to how chromosome
breaks are caused. And since gene mutations are similarly
caused, we need to examine this evidence. It arises from cer-

GENE MUTATIONS AND EVOLUTION 69

tain curious relations in the occurrence of chromosome breaks
and injuries.

While under radiation, the chromosomes are long threads
bearing the genes at intervals. In well studied organisms, the
order in which the different genes occur on the thread is well
known, so that maps of the genes have been constructed.

When through the action of radiation there occurs in a
chromosome a break at a certain point, it is an extraordinary
fact that there is usually or always in the same cell another
chromosome break, at some distance from the first mentioned
one. This other break may occur at another point in the same
long chromosome, or in another chromosome of the cell. This
is another strange fact, worthy of full consideration.

Chromosome breaks occur in very few of the cells, abso-
lutely considered. The fact that there usually occur two breaks
in one cell cannot be due to chance. Whatever causes one
break causes also another. The chance of a single electron's so
striking the chromosome as to cause two breaks some distance
apart is so minute that it can be excluded.

And now another strange fact. After the two breaks have
occurred, the pieces separated by the breaks usually unite by
their broken ends, often in a different order from before. In
the case of a single chromosome that has broken in two places,
thus yielding three pieces, the three pieces often reunite in
such a way that the middle piece is reversed in position
(Figure 11, 5). That is, the order of its genes is the opposite
of that which prevailed before the break and reunion; this
can be determined by breeding tests. A great number of such
inversions of the middle part of the chromosome are known
— the order of the genes reversed in the middle piece, though
not in the two end pieces. This is an extraordinary phenom-
enon ; it cries out for explanation.

70 GENETIC VARIATIONS

In other cases the middle piece drops out entirely and is
lost. The two end pieces then reunite at their broken ends.
Thus a short chromosome is produced, lacking all the genes
of the middle region, but still carrying those near the two
ends (Figure ii, C). These are known as deletions — the mid-

Aabcdeighij klmtio

B

abcdihgfejklmno

■« • • • [ 9 • • • » I m • • 9 • •-

a bcdj k Imno

-• • • • • • • • •-

D

Figure ii. Diagram of changes in the order or arrangement of the
genes in consequence of breakage of chromosomes. A. Part of a chrom-
osome with genes a to o in the normal order, B. Inversion: the genes
between d and ;' are reversed in their order. C. Deletion: the genes
e to / have dropped out. D. Genes of two chromosomes. E. Transloca-
tion: the two chromosomes of D have exchanged certain portions.

GENE MUTATIONS AND EVOLUTION 71

die part being deleted. A large number of such deletions are
known. The process by which they are produced demands
explanation.

A third strange result is often produced when the chromo-
somes are broken by radiation. If the usual two breaks are in
two different chromosomes of the same cell, then frequently
these two chromosomes exchange parts. The broken end of
one piece from one chromosome unites with the broken end
of one of the pieces from the other. The two other pieces of
the two chromosomes similarly unite. Thus we get two newly
combined chromosomes, each made up of two halves that
formerly belonged to two separate chromosomes (Figure 11,
D and E). Such exchanges, or translocations as they are
called, are known in many cases. Their formation calls
loudly for explanation.

Thus several questions demand answer. First, how does it
happen that in a cell, two chromosome breaks occur instead
of one ? Second, how do the broken ends find each other and
unite, often resulting in a new combination ? And third, how
does it happen that in such reunited chromosomes the order
of the genes in the middle piece is commonly reversed }

There is a single simple and convincing explanation for all
these strange relations; and it is one that throws a sharp light
on the way gene mutations are produced. This explanation,
originally due to Serebrovsky,*' is as follows:

The numerous long thread-like chromosomes in the cells
undergoing radiation lie close together and are more or less
curved, coiled and tangled. They may cross each other and
come in contact, but under ordinary conditions when this
happens they do not stick together.

But under the action of radiation or high temperatures, the
chromosomes change physically, in such a way that when
they come in contact by crossing at a certain point they stick

72 GENETIC VARIATIONS

together. There are indications (as Metz^ has set forth) that
the chromosomes have a transparent sheath, which keeps the
essential parts of the chromosomes — the genes — within the
sheath, from coming in contact. By radiations or heat this
sheath is dissolved or modified in spots, so that the chromo-
somes within it may actually come in contact. When this
occurs they stick together, and through strains set up in the
intracellular processes they may break. The rest of the process
will be best understood from an example (see Figure 12).
Suppose that a chromosome in this adhesive condition be-
comes looped, so that its two limbs cross and come in contact
(Figure 12, A).

The parts in contact stick together, and the chromosome
breaks at the point of adhesion. The chromosome is thus
left in three pieces, a central loop and two end pieces
(Figure i'2, B).

The three broken ends are still in contact, and there occur
at the tips growth processes, so that the broken ends, some or
all, may reunite. The two end pieces may reunite, leaving the
loop free. In that way are produced the "deletions"; short
chromosomes lacking all the middle part. In such cases the
free loop degenerates or is lost.

In other cases the two end pieces may reunite with the ends
of the loop. Sometimes the union may be with the same ends
as before ; then there might be no evidence that any break had
occurred. But in other cases, since the four ends are all close
together, each end piece may unite with the other end of the
loop from that with which it was before united (Figure 12,
C). This produces the inversions. The chromosome now has
its two end parts as before, but in its middle part the order of
the genes is reversed.

If the accidental crossing or contact is between two difTer-
ent chromosomes, then they adhere at the point of crossing.

A

B

GENE MUTATIONS AND EVOLUTION
I J ^ f^ ^ 1^ s t

73

D

o — o-

-o — o

Figure 12. Diagrams of the method by which are produced inver-
sions and translocations of the chromosomes and genes. See the de-
scription in the text.

74 GENETIC VARIATIONS

and both may break there (Figure 12, D). Then the ends, in
becoming reunited, may in some of the cases readily exchange
partners. Thus are produced the well known translocations;
two chromosomes that have exchanged parts (Figure 12, E).

Thus all these remarkable phenomena, so inexplicable
otherwise, find a simple explanation based on conditions that
exist within the cells. No other explanation than one along the
lines suggested appears possible.

The changes in the chromosomes produced in these ways
are of great interest in themselves. They result in producing
new combinations of genie materials, and the study of their
effect on heredity is just at present one of the favorite occupa-
tions of geneticists. But for our present purpose their interest
lies in the light which they throw on the cause of the chromo-
some breaks and of their associated gene mutations. They
make it probable that the breaking of the chromosomes is
produced, not by the striking of an electron against the
chromosome at a certain spot, but by the sticking together of
the chromosomes. What the radiation does, what the heat
does, is to modify or dissolve the chromosomal sheath. When,
in this condition, parts of the chromosomes come in contact,
all the rest follows. The role of the radiation or the heat is to
make a general physical or physiological change in the cell
contents, resulting in ready adhesion and breaking of the
chromosomes. The breaking is not due to the striking of an
electron at the point of breaking; this idea is quite inconsis-
tent with the usual occurrence of two breaks in place of one,
as well as with the fact that heat produces the same effect. The
greater the energy of the radiation, the greater its effect in
making the chromosomes adhesive and breakable; therefore
the greater the number of breaks and injuries produced.

Important in this connection is the fact that breakage and
the accompanying changes are not limited to chromosomes

GENE MUTATIONS AND EVOLUTION 75

that have been subjected to radiation or heat. Under natural
conditions there occur translocations, inversions, and the
other phenomena resulting from chromosome breakage,
though much less frequently than under radiation. Variations
in the physical state of the chromosomes, giving rise to
adhesions and breakage, occurs at times under natural
conditions.

And now come back to the mutations. Mutations, as before
remarked, commonly accompany breaks, occurring at or near
the point of breakage. The mutations must in some way result
from the breaks, or from something that accompanies the
breaks.

Just what is it in the breakage of chromosomes that brings
about the changes called gene mutations ? On the one hand
there is the fact of breakage, the tearing apart of structures
that were together, with its usual connotation of injury. The
changed action of the genes might well be the expression of
small scale injuries, induced at the same time as the greater
injuries that are seen as breaks, deletions and deficiencies. The
usually harmful or reductional effect of the mutations agrees
well with this interpretation.

Another change that usually results from the breakage is
an alteration in the relation of the genes to each other. When
a chromosome breaks, the pieces usually reunite in a changed
position, as we have seen, giving inversions or deletions or
translocations. The result is that genes near the points of
breakage are torn apart from those with which they are
normally in close contact and are brought into close relations
with other genes than those that are originally near them. It
is suggested that possibly the effect that a gene produces
depends on its position with relation to other genes; that is,
presumably on the interaction of genes that are close together.
Thus a gene moved to a new position with relation to others

76 GENETIC VARIATIONS

produces an effect on development and characteristics that is
different from its normal effect; and this changed effect
is what has been called a gene mutation. This is spoken of as
a "position effect." That the "mutations" which accompany
chromosome breakage are such position effects is the view to
which a number of recent investigators of these phenomena
have come.*

What is certain is that when a gene is brought into a new
position as a consequence of the occurrence of a chromosome
break close to the point where it occurs, it often so changes its
action as to produce what has been called a gene mutation. It
is conceivable that the change in relative position is not the
primary feature in producing the changed effect; it may
merely provide the evidence of a break and mark the spot
where it occurred. The change in effect would then be due to
the lesions inseparable from laceration of tissues.

The nature of the changes in gene action that result from
breakage have thus far been most fully studied by Schultz
and Dobzhansky.* They examined thoroughly the effects on
the genes in a case in which the second chromosome of
Drosophila had been broken in two places, one at about the
region of the gene bristle (54.7), the other near the gene
brown (104.5) (see maps of the Drosophila chromosomes in
any good text-book of genetics). That part of the chromosome
between the breaks (nearly half its length) had been inverted
and attached anew to the two end pieces, in the way illus-
trated in Figure 12, at C. At one of the breaks (that near
brown), there was an effect on three of the genes. All three
were weakened, so that they became unstable; in such a way
that during development they went out of action in some of
the cells, while in others they continued to operate, thus
yielding mosaic individuals. In one of the three there was an
additional, dominant, effect, showing itself in a change of the

GENE MUTATIONS AND EVOLUTION 77

color of the eye, which was altered from the usual red to a
mosaic "plum" color. In the region of the other break two
genes were altered. One, the gene "light," was made unstable,
so as to yield mosaic organisms. The other was so altered that
it would no longer support life unless there was present the
other member of the pair in a normal condition (thus a
"lethal mutation"). The fact that genes may thus be weak-
ened by injury so that they go out of action capriciously dur-
ing development is of great interest. The breakage in this case
caused three types of effects on the genes: (i) a color muta-
tion; (2) instability of certain genes; (3) a lethal effect on
a certain gene.

The relation of chromosome breakage to gene mutations
has received further unexpected illumination through the
second of the recent discoveries mentioned earlier. Here we
are dealing with things of the present moment; discoveries
as yet hardly fully announced. In the flies, the organisms on
which our detailed knowledge of genetics is so largely based,
certain of the chromosomes become enormously enlarged, so
that their complex constitution becomes conspicuously visible
(see Figure 13). These are the chromosomes present in the
salivary glands.

Each of these chromosomes is a large rod or thread, of
dimensions in some cases about one hundred times as great
as the corresponding chromosomes of other cells. Each shows
a complex structure of bands and zones, differing in thickness
and in physical structure. This structure is very different from
that of the chromosomes as seen in the other cells of the body.
For a long time it was not recognized that this elaborate
structure in these salivary chromosomes has any relation to
the structure of the chromosomes that has become known
from their genetic effects.

R

•/

*

S3

78 GENETIC VARIATIONS

jj£^ But recently Painter^ has discovered that the bands

m, and zones of the salivary chromosomes mark the po-
)Si sition of definite genes known from their effects in
^ inheritance. In general, the banded structure cor-
responds to the known arrangement of genes and of
mutations. In the normal flies, not mutated, the bands
^ of any chromosome show a typical order and typical
Q diversities in size and texture, as shown in Figure 13.
c In the different chromosomes of any species the
^ banded structure differs; so that it is possible to recog-
Q nize the particular chromosomes by this structure.
Study of the method of inheritance of the charac-
teristics depending on known genes shows that in
certain particular chromosomes a certain series of
genes are reversed in their order, as compared with

t;rr^

S the usual arrangement (as in Figure 11). When this
tiJ particular chromosome is studied as found in the
^. salivary glands, it is discovered that in a certain region
the order of the bands is correspondingly reversed. It
thus becomes possible to know that the reversed genes
are connected with the reversed bands. Again, in
some cases a few genes are known, from genetic evi-
dence, to have been lost from a certain chromosome;
^ in such cases this chromosome, as found in the sal-
ivary gland, is found to have lost certain bands. By
these and similar methods it has become possible to
determine with certainty that the materials of par-
ticular genes are located in certain particular bands,
t=t' having a definite position and structure. The great
chromosomes of the salivary glands make clearly

*-■■-■,

tv.J

m

:-.-\

r — 1

^

,t=3.

^ Figure /J. The enlarged X-chromosome of Drosophila mel-

anogaster as found in the saHvary gland, after Painter, 1934.

GENE MUTATIONS AND EVOLUTION 79

visible certain relations that a few months ago one would
have expected to remain forever hidden.

These great chromosomes, with their visible structure,
provide a means of determining definitely whether certain
chromosomes have been broken, with the production of
deletions, inversions, translocations, and the like, for these
things are clearly seen in the changed order of the bands.
They make it possible to determine whether known gene
mutations are accompanied by a break in the chromosome at
or near the point where the mutation occurs.

This has already been determined for certain well known
mutations. We have before seen that when a break is known
to have occurred in a chromosome, this break is usually
accompanied by a gene mutation. But many gene mutations
have long been known with no corresponding knowledge of
a break in the chromosome. Examination of the salivary
chromosomes in stocks carrying such known mutations has
shown for certain cases that the mutation is indeed accom-
panied by a break at or near the point of mutation; this break
is presumably what gave rise to the mutation.

This has been demonstrated by H. }. MuUer and his asso-
ciates.^" They have shown that a number of long-known
mutations involve an inversion in the order of the bands of
the corresponding salivary chromosome; including the band
known to carry the material of the mutated gene. Such an
inversion is known to be the result of chromosome breakage,
as described on earlier pages. The inversions are found to be
usually of small extent, often involving but two or three genes
or chromomeres. It is of course important to remember that
at the time the break and inversion occur, the chromosome is
in the elongated, thread-like condition indicated in the dia-
grams of Figures 11 and 12; at such times inversions of small
extent are readily possible. Such inversions of minute extent

8o

GENETIC VARIATIONS

are found to be unexpectedly frequent. In cases where the
same gene has mutated in different ways in different cases
(yielding what are known as multiple allelomorphs), the
inversions are found to be visibly of somewhat different
extent, or of different type, in the different cases.

An example of these relations, as described by MuUer and
his associates, is found in the mutations of the so-called scute
gene. This is a gene that affects the bristles and their distri-
bution on the body, in Drosophila. From genetic evidence it
has long been known that this scute gene is at, or close to, the
tip of the X-chromosome. This would bring it close to the
first or second of the bands at the left or upper end of the
salivary X-chromosome, as seen in Figure 13.

This gene scute has undergone many mutations — some
twenty to thirty — each giving a slightly different effect. Now
comes the essential point. In the stocks showing a mutation
of the scute gene, there is found to be a change in the thick

Figure 1^. Diagrams of changes in the order of the bands in the
salivary X-chromosome of Drosophila, in connection with mutations
in the gene scute. A, B and C represent a portion of the extreme left
or upper tip of the X-chromosome. A. The normal arrangement of the
bands. B. Inversion of bands i and 2, in connection with one of the
mutations. C. Rearrangement of the bands in connection with another
mutation of the gene scute.

GENE MUTATIONS AND EVOLUTION 8i

second band at the tip of the X-chromosome (see Figure 14) ;
a change sometimes affecting other bands near by, also. In
Figure 14, A, B, C, are shown diagrams, based on Muller's
description, of the changes in the bands in certain well known
mutations of the scute genes. Figure 14, A, shows the usual
condition in the normal chromosome, in which there has
been no mutation of the scute gene. Figure 14, B, shows the
situation in one of the well known mutations of this gene,
causing a change in the bristles. The first and second zones
have interchanged places. That is, the chromosome has been
broken close to the tip, and there has been an inversion of
small extent. (Besides the mutation in the scute gene, con-
nected with band 2, there was also another mutation, a lethal
one, probably due to the change in the gene connected with
band i.)

Another mutation of this scute gene was accompanied by
an additional mutation in a gene close to it, affecting the
facets of the eye. In this case there had been a break and
inversion affecting bands 2, 3 and 4; the condition is shown in
our Figure 14, C. A piece has been broken out and inverted.
At the upper end the break has gone through the middle of
the thick second band, leaving part of it in place. At the
lower end the break was between bands 4 and 5. So the order
of the bands now is:

I — half of 2 — 4 — 3 — other half of 2 — 5 — 6 — etc.

In another mutation of the scute gene a part of the thick
second zone was found to have been broken out and attached
in another region of the chromosome.

These large salivary chromosomes are as a matter of fact the
two chromosomes of a pair united lengthwise, the corre-
sponding bands uniting throughout. Traces of the line
separating the two are observable, and in some cases the two

82 GENETIC VARIATIONS

are separate for a short distance. It is not difficult to produce
individuals in which one of the two component chromosomes
is normal (without the mutation), while the other component
has the mutation and hence the changed arrangement of the
bands. These present a striking appearance.

Thus in these cases the mutation, long known, is shown
to be the result of a break in the chromosome; a break whose
occurrence is demonstrated by the changed order of the
bands. (Of course the break and rearrangement have
occurred when the chromosomes were in the extended condi-
tion indicated in Figures ii and 12.)

In sum, it is clear that mutations are in many cases the
result of breakage of the chromosomes, with rearrangement
of some of the genes. In cases in which the first knowledge
is of the occurrence of breaks and rearrangements, these are
found to be accompanied by "gene mutations." And in cases
in which the first knowledge is that a mutation has occurred,
these are found to be accompanied by breaks and rearrange-
ments. Different mutations of the same gene are found to
result from the fact that the breakage occurs in different
regions of the chromosome, in different cases. The relation
of the so-called mutations to breaks and injuries is fully
demonstrated.

Certain further important questions at once arise. Are the
so-called gene mutations all of this type? Or are there two
classes, one occurring under the action of radiation and
resulting from breakage or injury to the chromosomes; the
other of a different type ? Evidence on this will come soon,
from examination of the conditions in the salivary chromo-
somes, in the many stocks bearing diverse gene mutations. In
the meantime, the fact that similar and identical mutations
occur both under radiation and spontaneously, speaks strongly
against the existence of two classes differing in nature.

GENE MUTATIONS AND EVOLUTION 83

Again, are the mutations resulting from chromosome
breakage essentially "position effects," or are they of the same
type as deletions and deficiencies, save that they are of less
extent or degree; injuries that do not result in complete de-
struction of genes, but only in that reduction in their effec-
tiveness commonly seen in gene mutations?

If "gene mutations" should turn out to be simply position
effects, this would require a rewriting of extensive chapters
of genetics. In particular it would obviously have a devastat-
ing effect on the theory that these are the basic factors in
evolutionary change. They would not be changes in the
constitution of individual genes at all. They would be simply
further illustrations of the known fact that the effects of a
particular gene depend on the conditions under which it
operates. A new class of such conditions is added to those
before known; namely interaction with genes that are in
close proximity within the chromosome. A gene removed
from the normal grouping and brought into close relation
with a new set of genes acts in a different manner from
before ; as the facts show, it commonly operates less efficiently,
yielding reduced effects. The so-called multiple allelomorphs,
supposed diverse modifications of the same gene, would be, so
far as due to "mutation," simply the same gene in different
positions with relation to other genes.

The general effect of such changes of gene action with
change of position would be to increase the transformism and
diversification of organisms by giving rise to many weakened
and abnormal types, such as are commonly seen in mutated
stocks. But their relation to progressiveness in evolution
would completely evaporate. They would not supply the new
genes with changed constitution that progressive evolution
requires. They would have the same kind of relation to
progressive evolution as have the other changes in the struc-

84 GENETIC VARIATIONS

ture of the genetic system that were described in our second
chapter.

Indeed, if this be the state of the case the history of these
supposed gene mutations is but a continuation of the earlier
history of supposed mutations, including most of the "muta-
tions" of De Vries. These had a period of flourishing as the
fundamental steps in evolution, but were later found to be
only the result of regrouping and rearrangements among the
chromosomes or parts of the chromosomes/^ And now, if
they turn out to be position effects the "gene mutations" of
Morgan are found to be of the same type ; they are rearrange-
ments of small extent and often of complex character. They
do not supply the material for constructive or progressive
evolution.^"

If on the other hand the mutations are rather injuries than
position effects, the case for their role in progressive evolution
is not better. In this case they are essentially disintegrative in
character, and cannot be expected to give constructive or
progressive results.

Taking the evidence all together, the case is strong for the
essentially disintegrative nature of the observed gene muta-
tions. It weighs heavily against their constructive or progres-
sive character. It argues strongly against the possibility of
explaining by them the adaptive and progressive features of
the evolutionary process. The overwhelming majority of
mutations, it is clear, are certainly not advantageous. Many
of them cause death. Many cause a weakening of the consti-
tution. Many produce obvious defects and abnormalities.
Experimentation shows that most of them are reductional in
their action, while those few that have positive action diverse
from the normal are among the most distinctly injurious of
mutations. And now, to complete the case, comes the demon-
stration that typical "gene mutations" are the result of breaks

GENE MUTATIONS AND EVOLUTION 85

and rearrangements in the genetic system; "position effects"
or expressions of injury and disintegration.

Changes of this sort are not the material from which can
be built progressive evolution. Their occurrence does not
forward evolutionary progress, nay, it retards it. They do
result in diversification, through the production of many
abnormal and imperfect types. But with relation to progres-
sive evolution, these are things to be gotten rid of. The muta-
tions bring about the elimination of the individuals in which
they occur or to which they are transmitted. They demand
an enormous selective elimination that does not forward pro-
gressive evolution; at best it merely keeps evolution from
going backward. When, if ever, the actual steps in progressive
evolution are recognized as they occur, it appears probable
that they will not be the result of action of destructive agents
nor connected with disintegrative changes in the genetic
system, but will rather bear a resemblance to the changes
known as growth.

Whether these conclusions are sound will probably be fully
shown within a few years, for it is clear that genetic science is
on the threshold of an enormous development in relation to
these matters. At the very least, it will be necessary in the
future to make sharper distinction among the types of gene
changes, not lumping together position effects and demon-
strable injuries to the genetic system, along with constructive
changes in the constitution of the genie materials.

In the meantime, in view of the formidable case against the
observed types of mutation as supplying the fundamental
basis for progressive evolution, students of these matters will
be disposed to give more consideration to phenomena and
conditions of other kinds, looking toward other types of
genetic change as the basis for progressive evolution. No other

86 GENETIC VARIATIONS

group of genetic changes is so well worked out as that of
"gene mutations," but seemingly the more fully this is
worked out, the less convincing becomes its role in evolution.

The efforts to find evidence of other types of genetic change
fall into two classes. In one the evidence is indirect; through
examination of the results of evolutionary change, conclusions
are drawn as to its nature. In the other, genetic variations are
directly observed as they occur in living organisms.

The indirect methods of obtaining evidence on the nature
of evolutionary change may be classified in two main groups,
with a third minor group.

I. In the first main group are attempts to trace the evolu-
tionary changes that have occurred in the series of organisms
that have lived in the past.

II. In the second group are studies of the diversities among
organisms now existing, with attempts to judge of the nature
of the changes that must have occurred in producing these
diversities.

III. The minor group concerns changes that have appar-
ently occurred in the genes of natural polyploids.

With these indirect methods of obtaining evidence on evo-
lutionary change it is not the province of these pages to deal.
Yet for comparison with the results of direct observation of
genetic variations, certain typical lines of evidence may be
sketched, with some of the conclusions that have been drawn
from them.

I. To trace the evolutionary changes that have occurred in
the series of organisms, from lower to higher, that have lived
in successive geological ages, study must of course be limited
to the hard parts of organisms, since only these are preserved.
The literature of such study is of course enormous. As an
example there may be mentioned the work of H. F. Osborn,

GENE MUTATIONS AND EVOLUTION 87

who has recently published a number of papers presenting
conclusions drawn from the study of paleontological series/^
The conclusions of Osborn are based on his extensive
studies of the fossil collections of the American Museum of
Natural History, and elsewhere, the details of which are
published in voluminous memoirs. His recent conclusions
have been based largely on study of evolutionary series in the
elephants and their relatives. Some of the essential data and
conclusions may be sketched as follows:

1. Structures ("aristogenes") that in later ages become con-
spicuous and functionally important appear at the beginning
as almost imperceptible features; so slight that their func-
tional use is at this time doubtful. Classical examples of these
things are the cones on the teeth of various vertebrates. They
arise as almost imperceptible prominences (see the figures in
the papers of Osborn above referred to).

2. In succeeding ages these structures increase in promi-
nence and functional value in a continuous way without sud-
den changes. Continuity, not discontinuity, is the characteris-
tic of such evolutionary change.

3. The structures follow a well marked definite course, in
their changes from the imperceptible, doubtfully functional,
condition, to the conspicuous and useful organs that they
finally produce. Such definite change following a course
toward a well defined useful end is spoken of at times as
orthogenesis, or rectigradation.

4. In their later condition the structures produced are useful
in a high degree.

5. The same types of structures (e.g. grinding cones on the
teeth) arise independently in different divisions of the same
group of organisms, so that they are said to be "predetermined
in the germ plasm" of the group. Yet other divisions of the
same group do not produce these structures, but others of a

88 GENETIC VARIATIONS

different kind, so that the predetermination is such only in the
sense that the germ plasm is of a type that can produce these
structures if the conditions are favorable to them.

6. Such new structures as a rule take origin after there has
been a change of environment, or a change of habit on the
part of the organisms, of such a sort that the nev^ structures
in time are useful and "meet the situation" under the new
conditions.

From these and similar data, Osborn draws a number of
important conclusions:

A. The new structures arise in response to a functional
need, in consequence of a changed environment or a change
of habit.

B. But they are not induced during the life of the individ-
ual, like the acquired immunity of an animal. They are not
direct adaptations of the individual animal to conditions, with
inheritance of these adaptations; not "inheritance of acquired
characters." Rather they are, as Osborn expresses it, "secular"
adaptations, requiring ages for their full production.

C. They are not "chance variations," or undirected varia-
tions, it is held, because they follow a definite course, toward
the fulfilment of a particular need. They are from the start
adaptive or "in the direction of future adaptations."

Attempts have been made, on the basis of the modern study
of mutations, to explain these phenomena in the following
way. The origin and farther advance of the new structures
are held to be in fact the result of undirected mutations;
mutations occurring with many diverse effects, as in Droso-
phila. The individuals in which these mutations are harmful,
or are not beneficial, in time die out or cease to propagate.
They therefore do not appear in the line of descent from the
early to the late stages of the given structure. But in some
cases the mutations happen to be useful. These continue to

GENE MUTATIONS AND EVOLUTION 89

propagate; and in some of their descendants again useful
mutations occur, along with many that are not useful. Thus
in the long run, there appear in the line of descent only those
in which one useful mutation has followed another. Thus
genetic variation appears to have followed a definite straight
course; although as a matter of fact it has occurred in many
directions, most of which were not useful and hence are not
found in the pedigree of the final products. The fact that
genetic variations do occur in directions that are useless or
harmful is not in doubt, it may be pointed out, since useless
and harmful mutations actually occur in the genetics of living
organisms.

But in view of what we have seen of gene mutations in
earlier pages, the question remains whether mutations indeed
do supply the constructive and useful changes required for
such evolutionary progress ; or whether progress results from
types of change not yet observed in living organisms.

II. The second of the indirect methods of studying the
nature of evolutionary changes lies in the examination of the
diversities among living organisms, attempting to judge how
such differences must have arisen. On this matter there is of
course a vast literature, dealing with taxonomy, anatomy,
embryological development, geographical distribution and
the like. Much of this was written before the rise of modern
genetics, or by men not familiar with it, so that it is difficult
to interpret in genetic terms. Other parts of it do not suffer
from this difficulty. As an example of such work done in full
awareness of the methods and results of genetics, and indeed
making use of these, some features of the work of F. B.
Sumner^* may be sketched.

Sumner has made extensive studies of varieties and subspe-
cies of certain wild mice — "deer mice" — of the genus Peromys-
cus; with attempts to judge of the nature of the differences

90 GENETIC VARIATIONS

between them. The different types existing in nature have been
bred and intercrossed, to determine the method of inheritance
of the differences, and particularly, to discover w^hether these
differences behave like gene mutations. Some of the chief
results may be summarized as follows:

1. The differences between slightly diverse forms are in the
main heritable, not mere bodily effects of different envi-
ronments.

2. The slight differences between the different types — di-
verse varieties, subspecies or the like — are not like those due
to single mutations; they affect many diverse characteristics.

3. When the different types are crossed, the differences do
not behave like single gene mutations. They do not give sim-
ple Mendelian ratios, like those yielded when a mutant is
crossed with the parent from which it is derived. On the
contrary, the differences yield "blending inheritance," the
offspring being intermediate in most characters between the
two parent types.

"Blending inheritance" is known in many cases to be due
to the dependence of the characters so inherited on many
different genes, that differ in the two parents. This shows
therefore that the related types found in nature differ in many
of their genes; not in a single gene, as a mutant differs from
its parent form.

4. Hence if one of the varieties has arisen from another, or
if the two have arisen from a common ancestral type, there
have been slight genetic changes in many of their genes.

5. Such diverse types (varieties or subspecies) differing
slightly, but in many of the characteristics, commonly occupy
different, usually adjacent, geographical areas. Sometimes a
series of such varieties occupy a series of different areas.

6. Often in such cases there is a relation between the envi-
ronmental differences and the varietal differences, of the

GENE MUTATIONS AND EVOLUTION 91

following character. In a series of varieties occupying a series
of zones, the series of zones show successive gradations of
physical conditions, as climate, or soil, or elevation, from one
extreme to the other; and there is commonly also a series of
correlated gradations in the inherited characteristics of the
varieties that occupy the successive zones. This appears to
indicate that the physical conditions in some way affect the
nature of the genetic variations; either directly, or by selective
action on the variations produced.

On the whole, it appears that such series of varieties might
have been produced through the gradual accumulation of
slight germinal variations, under the selective action of the
environmental conditions. But whether such germinal varia-
tions are mutations of the type observed in experimental
genetics must be doubtful, in view of the evidence presented
in earlier pages of this volume.

III. Further indirect evidence of changes in the germ plasm
may be found in the conditions occurring in so-called poly-
ploid organisms.

As was set forth in Chapter II, at times the number of sets
of chromosomes, and therefore of genes, is increased. Most
organisms have two sets, their chromosomes (and genes)
being in pairs. Sometimes, through irregularities in the
operation of the genetic system, the number of sets is in-
creased to three, four, five, or more. Such organisms are called
polyploids. In such cases the individual genes are likewise
originally in threes, fours, or higher numbers, instead of in
pairs. This changes the method of inheritance. There is, as
before seen, strong evidence that some natural varieties differ
in these ways; the number of sets of chromosomes has been
increased to four, five, six or more. But in such natural poly-
ploids as a rule there appear to be but two of each kind of
gene, just as in other organisms. This shows that the genes in

92 GENETIC VARIATIONS

some of the chromosomes have become altered, so that they
no longer have the same effects as the corresponding genes in
the other chromosomes of the set. The actual occurrence of
such changes has not been observed, and their nature and
causes are unknown. Possibly it is in such polyploids that the
nature of evolutionary changes in the genes is most open for
investigation.

In the direct attempts to observe genetic variations as they
occur in living organisms, one direction that has long been
followed lies in the relation of such variation to environmen-
tal action, and to the reactions of organisms to environmental
conditions. In the lowest organisms, those made up of single
cells, the occurrence of genetic variations has been observed
on a large scale, and of these many seem more readily classi-
fied under the topic of environmental action and reaction
than under that of gene mutation. An account of these will
be given in the remaining chapters.

NOTES ON CHAPTER III

^ Page 60. For a detailed account of such experiments see H. J. MuUer,
"Further Studies on the Nature and Causes of Gene Mutations," Pro-
ceedings of the Sixth International Congress of Genetics, Vol. i, 1932,
pp. 213-55 (see particularly pp. 231-52).

- Page 64. W. R. Horlacher and D. T. Killough, "Progressive Muta-
tions Induced in Gossypium hirsutum by Radiations," American
Naturalist, 1933, Vol. 67, pp. 532-8.

^ Page 65. V. Jollos, "Weitere Untersuchungen iiber die exp>erimentelle
Auslosung erblicher Veranderungen bei Drosophila melanogaster,"
Zeitschrift fiir lndu\tive Abstammungslehre , 1932, Bd. 62, pp. 15-23.

* Page 65. See J. T. Patterson and H. J, MuUer, "Are 'Progressive'
Mutations Produced by X-Rays?," Genetics, 1930, Vol. 15, pp. 495-
578; and further references there given.

^ Page 68. See H. J. Muller, paper referred to in Note i, and L. J.

Stadler, "On the Genetic Nature of Induced Mutations in Plants,"

Proceedings of the Sixth International Congo-ess of Genetics, Vol. i,

1932, pp. 274-94.

^ Page 71. A, S. Serebrovsky, "A General Scheme for the Origin of

Mutations," The American Naturalist, 1929, Vol. 63, pp. 374-8.

■^ Page 72. C. W. Metz, "The Role of the 'Chromosome Sheath' in

Mitosis and Its Possible Relation to the Phenomena of Mutation,"

Proceedings of the National Academy of Sciences, 1934, Vol. 20,

pp. 159-63.

* Page 76. See Jack Schultz and Th. Dobzhansky, "The Relation of
a Dominant Eye Color in Drosophila melanogaster to the Associated
Chromosome Rearrangement," Genetics, 1934, Vol. 19, pp. 344-64. In
this paper will be found references to other papers on this subject.

^ Page 78. See Th. Painter, "Salivary Chromosomes and the Attack on
the Gene," Journal of Heredity, 1934, Vol. 25, pp. 465-76; also other
papers there referred to.

^^ Page 79. See H. J. Muller, A. Prokofyeva; and D. Raffel, "Apparent
Gene Mutations, Due to the Position-Effect of Minute Gene Rear-
rangements," Records of the Genetics Society of America, No. 3, 1934,
p. 48; also in The American Naturalist, 1935, Vol. 69, January-Febru-
ary, p. 72.

94 GENETIC VARIATIONS

'^'^ Page 84. Among the so-called mutations of De Vries it has been the
custom to distinguish certain ones as actual "gene mutations," while
the rest are but rearrangements of chromosomes or parts of chromo-
somes. If the theory of position effects is substantiated, these "gene
mutations" are likewise only rearrangements of parts of chromosomes.
^^ Page 84. This relation of the "position effects" to evolution is recog-
nized by MuUer and his associates. In the first publication giving an
account of the relation of the scute mutations to the change of position
of the zones on the salivary chromosome, Muller, Prokofyeva and
Raffel (see Note 10 above) close with the following remarks:

"These and other cases raise the question: Are most, perhaps practi-
cally all, X-ray 'gene mutations' intergenic position effects rather than
intragenic transformations? And the more momentous question fol-
lows: To what extent may natural mutations be of this type — a type
incapable of providing the material for indefinitely continued
evolution?"

" Page 87. See H. F. Osborn, "Aristogenesis, the Creative Principle in
the Origin of Species," The American Naturalist, 1934, Vol. 68, pp.
193-233; and other references there given.

" Page 89. See F. B. Sumner, "Genetic, Distributional and Evolution-
ary Studies of the Subspecies of Deer Mice (Peromyscus)," Bibliog-
raphia Genetica, 1932, Vol. 9, pp. 1-106; also, "Taxonomic Distinctions
Viewed in the Light of Genetics," The American Naturalist, 1934,
Vol. 68, pp. 137-49.

CHAPTER IV

Genetic Variatiokin Unicellular Organisms, with Special
Relation to the Inheritance of Environmental Effects

and Reactions

THE remaining chapters will be devoted to some of the
types of observed genetic variation in unicellular organ-
isms; particularly certain types that do not fit well any of the
categories of genetic variation thus far considered/ Genetic
variations in unicellular organisms show relations to environ-
mental conditions and to their reactions to these conditions,
that are of interest for any general understanding of the
nature of genetic variation.

Unicellular organisms differ from higher organisms in
several obvious ways that are of great importance for genetic
variation. First, genetic variation, in order to be effective, is
not here restricted to a group of germ cells hidden away in
the interior of a large body, as is the case in the higher organ-
isms; for here the whole body is but one cell. Second, the
entire organism is very small; in some of the unicellular
organisms it is indeed so minute as to be beyond the reach of
the ordinary microscope. In consequence, all of its materials
are in close relation with the physical and chemical conditions
of the environment, and exposed to their action. And third,
parent and offspring are not separated by a long series of cell
generations. In the higher organisms, the offspring begin as
single cells, and a long and complex process of development
occurs, involving thousands or millions of cell divisions, be-
fore the adult condition is reached. All this is lacking in
unicellular organisms. Reproduction here occurs by a single

96 GENETIC VARIATIONS

division of the parental cell, each of the two new cells pro-
duced transforming into an adult individual. Thus the one-
cell organisms show us the nature of heredity and variation
as these occur when there is but a single cell division between
parent and offspring.

Yet it is important to realize that even in this reproduction
by a single division, the individuals of the new generation are
produced by development. Even the unicellular organisms
are complex systems of structures; they are small machines.
We cannot multiply machines merely by cutting them in
two; try that on a typewriter or a gasoline engine! If the
organism is simply divided, the unity and harmony of its
machine-like structures are lost. During division many or
most of the definite organs disappear, by absorption or disso-
lution, and are then after division produced anew; they are
not transferred bodily from parent to offspring. This is true
of all the unicellular organisms, even the simplest, but it is
best seen in the more complex types. Some of the infusoria
bear on their bodies numerous differentiated appendages
serving as organs of locomotion; these are arranged on the
body in a definite pattern. When the animal divides, these
appendages disappear, and as the body separates into two
parts a new set of appendages appears and develops separately
on each half, finally arranging themselves in the typical
pattern (Figure 15). Thus inheritance of the organs and
pattern is not direct transfer from parent to offspring, but is
new production of structures like those which the parent pos-
sessed. Yet even so, reproduction by a single cell division is
relatively simple in comparison with the long process of
development with many successive cell divisions, that occurs
in higher organisms. To this difference is perhaps to be
attributed some of the peculiarities of heredity and variation
in these organisms.

VARIATION IN UNICELLULAR ORGANISMS

97

When we examine unicellular organisms, we find that the
individuals of a species differ among themselves, just as is
true in higher organisms. If one collects a large number of
individuals of some unicellular animal — say Paramecium, or

B

Figure 75. Reproduction and development in the infusorian Stylony-
chia, after Wallengren, 1901. A. Parent before division, showing the
number and distribution of the locomotor appendages. B. Beginning
of division: appearance of two groups of small projections, that are
later to form by enlargement and distribution the appendages of the
two offspring. C. Later stage of division: the two groups of embryonic
appendages are scattering to take their final positions. The old
appendages have not yet disappeared.

Difflugia, or Arcella — one usually finds diversities of many
kinds among them. In the shelled rhizopod Difflugia corona
(Figure 16), an animal closely related to Amoeba, one finds
the individuals differing greatly in size, form and structure of
their shells, as well as in many physiological ways. Many

98

GENETIC VARIATIONS

of these differences are inherited ; when the individuals repro-
duce, their descendants show characteristics like those of their
parents. Again, in the well known infusorian Paramecium,
one finds a large number of diverse stocks, differing in size.

M8

%iSS.

Figure i6. Inherited differences in Difflugia corona. Six races or
biotypes differing in size, number and length of the spines, and in
other respects. Each horizontal row is a series of individuals all derived
by successive divisions from the individual at the left end of the row.
After Jennings, 1916.

and in other respects. Within each stock there are size differ-
ences due to diversity of age and nutrition; but the mean
differences are inherited ; they are handed on from generation
to generation.

VARIATION IN UNICELLULAR ORGANISMS 99

Such inherited diversities one finds in any unicellular or-
ganism, if it is studied with sufficient thoroughness. And
there are similar variations in less conspicuous matters. There
are differences in vigor, resistance, vitality. Under poor con-
ditions, some races die quickly, others resist long. Some are
adapted to one set of conditions, some to another. Some stand
higher temperatures than do others. Some are resistant to a
chemical to w^hich others quickly succumb. Some multiply
rapidly, others slow^ly. Some races of Paramecium conjugate
frequently, others rarely. Differences of all these types are
inherited.

What is the origin of these inherited diversities? This is
the problem of genetic variation.

In the unicellular organisms, as in those composed of many
cells, several different types of genetic variation occur, some of
them show^ing direct correlations vv^ith environmental condi-
tions such as have not been demonstrated in higher organisms.
It is these that are of special interest for our present purposes.
But before examining these, certain other types of genetic
variation must be briefly presented.

In the unicellular organisms, as in higher organisms, bipa-
rental reproduction is a source of genetic variation. In such
ciliate infusoria as Paramecium, at times tw^o individuals unite
temporarily in conjugation and exchange parts of their
nuclei, then separate and continue reproduction by fission.
Conjugation among the members of a uniform stock pro-
duces among the descendants races or stocks differing greatly
in their inherited characteristics. They differ in size and in
other physical features. This has been show^n by the recent
work of Sonneborn and Lynch, and by that of DeGaris.Thus,
De Garis produced repeated conjugations between the mem-
bers of a uniform stock of Paramecium caudatum, the indi-
viduals of which, when full grown, had a mean length of 172

100 GENETIC VARIATIONS

microns (De Garis' race 75). From four pairs of this race
there were produced four stocks showing respectively the
following mean sizes: 165 microns, 133 microns, 184 microns,
136 microns. From his race 3, having a mean length of 198
microns, there were two pairs, giving rise to two stocks of
having respectively the mean lengths of 70 microns and 203
microns. From two ex-conjugant pairs of race 17, having a
mean length of 153 microns, were produced two stocks of the
following mean sizes: 145 microns, 170 microns. Again, when
two races of diverse mean size are repeatedly crossed, stocks of
diverse sizes are produced. De Garis made several crosses of
his race 39, of 203 microns mean length, with members of his
race 11, of 81 microns mean length. Four pairs gave stocks of
the following mean lengths, respectively : 144 microns, 191 mi-
crons, 207 microns, 91 microns. Crosses between the race 56
(mean length 191 microns) and race 83 (mean length 75
microns) gave from three pairs stocks of the following mean
lengths, respectively: 65 microns, 140 microns, iii microns.

The different stocks produced by conjugation within a
single uniform stock differ greatly in vigor, in rate of multi-
plication, and in resistance to unfavorable condition. This
has been conspicuous in the investigations of myself and
associates on Paramecium aurelia and Paramecium caudatum.

This production of inherited diversities at conjugation is
presumably due to the fact that conjugation, like biparental
reproduction in higher forms, is a method of forming many
diverse combinations of genes or genie substances. The indi-
viduals having the different combinations show very different
characteristics, as to size, fission rate, vigor and resistance.

Furthermore, it can be shown that even during reproduc-
tion by fission there may occur at times irregularities of
division that result finally in the production and isolation of
diverse stocks, differing in hereditary characteristics. For by

VARIATION IN UNICELLULAR ORGANISMS loi

long-continued careful selection of differing individuals
throughout many generations, in a stock that is multiplying
by fission, inherited diversities may be produced. It is true
that such selection may be practised for long periods with no
apparent results; but if it is persisted in for a very long time,
and a suitable basis for selection is employed, there may
finally result inherited differences. In DifHugia corona there
were produced in this way, in my own work, races slightly
diverse in form and size. Middleton and Parker found that
by long-continued selection in infusoria, races differing in
inherited rate of multiplication could be isolated.

Just what processes underlie these changes is not known. It
appears possible that there occur at times irregularities of
nuclear division, analogous to some of those described in
Chapter II for higher organisms, resulting in unequal dis-
tribution of the genie materials in the two individuals pro-
duced, so that they have diverse inherited characteristics. The
inherited differences produced in this way have thus far been
but slight.

In addition to these genetic variations resulting from the
production of new combinations of genie materials in bipa-
rental reproduction, and those presumably due to irregularities
in nuclear division, there are others that occur under the
action of environmental conditions, and that show special
relations to the conditions that induce them. These are of
special interest; to them we now turn our attention.

It is easy to change the characteristics manifested by uni-
cellular organisms, by subjecting them to changed environ-
ments. In DifHugia the size, form, number of spines, and
length of the spines are readily altered by subjecting the
animals to certain conditions. By subjecting Paramecium to
certain chemicals, they can be altered greatly, in size, form.

102 GENETIC VARIATIONS

rate of multiplication, and in other respects. Other Protozoa
can be readily altered in similar ways.

But usually, when the animals are restored to the original
conditions, in the offspring there produced the same old
form, structure and functioning return. I have myself made
extensive experimental studies on this, in Paramecium and in
DifHugia, by subjecting them to diverse chemical and other
agents, which changed them greatly. But on removal from
the chemical, in the next generation the animals returned to
their original state; the acquired characters were not inherited.
This is the common result of such experiments. The sub-
stances of the organisms have not been modified in constitu-
tion; it is only the superficial characteristics that have been
changed. The effects are like those of changing the shape of
a crystal, by filing off its angles or carving it into new shapes.
If such changed crystals are dissolved and recrystallized, the
original crystal form returns, for the chemical constitution
has not been changed.

Most effects of the environment are of this kind. They do
not change the constitution of the fundamental materials of
the organism; therefore they produce no genetic variation; no
lasting effect on its characteristics.

But in the unicellular organisms, not all environmental
action is of this sort. Sometimes, by environmental action, the
underlying substances are changed in constitution, so that a
genetic variation is induced, a variation that is carried on to
later generations.

The difference in effect appears to be connected with the
length of time that the effective agent continues to act. If the
organisms are subjected for generation after generation suc-
cessively to the special conditions, so that these act for a very
long time, then the constitution may become changed; there
is produced an alteration in the inherited characteristics, a

VARIATION IN UNICELLULAR ORGANISMS 103

change that is passed on to later generations. This factor of
time, or of something connected with the time, of action,
appears of great interest.

In unicellular organisms such results have been produced,
not by one or two external agents only, but by many different
types of environmental conditions; by conditions that operate
on the organisms in their normal life. Many different types
of inherited alterations have thus been produced. Some of
them are degenerative ; they leave the organism the worse for
their action. Others are distinctly to the advantage of the
organism. They are what can be called adaptive changes in
the hereditary constitution. Changes produced by a particular
environmental agent may have a definite specific relation to
the agent that produces them, fitting the organism to survive
and thrive under the action of that agent. We shall examine
here the main different types.

The first type of inherited environmental effect brought to
light consisted of inherited injuries to the constitution of the
organism, weakening it, inducing a degenerate condition. It
has often been suggested that in man, life for many genera-
tions under evil conditions may produce inherited degenera-
tion. In unicellular organisms this is indeed true. It was dis-
covered through the classical work on life histories, depres-
sion, degeneration and rejuvenescence in Protozoa, begun
fifty to sixty years ago by Balbiani and Maupas, and continued
by many investigators from that time to this.

Everyone who has had experience in this kind of work
knows that if one cultivates an infusorian for generation after
generation in isolation, on glass slides, or under other condi-
tions differing much from nature, as a rule the organisms
begin after a long time to "run down." They divide less fre-
quendy, they do not feed well; they lose resistance to bad
conditions; they have digestive and assimilative troubles;

104 GENETIC VARIATIONS

they show unfavorable symptoms of many kinds. The prog-
ress of these changes is well shown by plotting a curve of the
frequency of division of the animals (Figure 17). This curve
of the rate of reproduction gradually "runs down hill,"

10 20 30 40 50 60 70 80 90 100 110 120 150 140 150 160 170 180
Figure ly. Partly smoothed curve showing the gradual decline in
the rate of reproduction in Stylonychia pustulata, when subjected to
a high temperature (28 to 32 degrees). The vertical scale shows the
mean number of fissions per day; the horizontal scale shows the num-
ber of days from the beginning of the high temperature. Constructed
from the data of Middleton, 1918.

though with various fluctuations. There is on the whole a
rather steady downward progress, sometimes to a final cessa-
tion of reproduction, and death of all the organisms. The
slowing of reproduction is accompanied by degenerative
changes in the other functions above mentioned. The degen-
erative changes are not perceptible till many generations have
elapsed, and many more generations are required for the slow
descent to deep depression or death.

As to the exact nature of the changes that are occurring,
there have been differences of opinion. Some have believed

VARIATION IN UNICELLULAR ORGANISMS 105

them to be in essence of the same sort that occurs in growing
old, in the case of higher organisms. Others think of them
rather as damage to tiie constitution resulting from bad
living conditions. The essential facts for our purposes are,
first, that the changes are alterations in the hereditary con-
stitution; in the basic materials of the organism. And second,
that the environmental conditions play an important role in
bringing on these changes. In some species, indeed, if not in
all, a bad environment is the exclusive factor in bringing about
the degenerative changes; they do not occur if the animals
are kept in a favorable environment, but do occur if the
conditions are not quite those fitted to the organism. Such
decline of vitality, or depression, may be produced by sub-
jecting the animals for many generations to unfit food, to
scanty food, to too high a temperature, to the action of various
chemicals, and to various other unfavorable conditions.

Thus the material of these depressed organisms has in the
course of many generations become gradually altered, de-
pressed, damaged. But what is of the greatest interest, this
damaged material continues to assimilate, continues to pro-
duce more of its own damaged material ; it grows and repro-
duces in the damaged condition, producing new generations
showing the depressed characters. For if we compare side by
side under the same conditions a stock that has been under
bad conditions with one that has not, it is found that both
reproduce, both multiply, but the difference between them
continues. Under the same conditions one multiplies rapidly,
at a high level of vigor, with high resistance, and good
digestion and assimilation; the other slowly, at a low level,
with poor resistance, poor digestion and assimilation. This
may continue for twenty, fifty, one hundred, five hundred
generations or more. While the materials of the normal indi-
viduals multiply themselves in the normal condition, the

io6 GENETIC VARIATIONS

damaged materials of the depressed individuals multiply
themselves in the damaged condition. In merely ten genera-
tions the damaged material has produced more than a thou-
sand times its original amount; in a hundred generations it
has multiplied itself by billions.

Thus in these unicellular organisms, the condition of the
individuals of the present generation as to health and vitality
depends on the conditions in v^hich their ancestors have
lived; it may depend on the ancestral environment a hundred
generations ago. If their ancestors a hundred generations back
have been living for many generations under bad conditions,
the present individuals are weak and depressed. If their
ancestors of a hundred generations ago have been living for
many generations under good conditions the present genera-
tion is strong and vigorous. We have then in these organisms
the "inheritance of acquired characters"; the inheritance of
environmental effects, at least insofar as damage to the genetic
constitution through bad living conditions is concerned.

But such inheritance is not limited to injurious effects of
the environment. It occurs too when the action of the environ-
ment is such as to change the organisms for the better. The
environment may produce in unicellular organisms changes
that are adaptations to the conditions that produce the
changes. When subjected for long periods — for hundreds or
thousands of generations — to gradually increasing tempera-
tures, infusoria may slowly acquire an increased resistance to
high temperatures. The increase may be very great; Dallinger
and Drysdale in seven years increased the temperature which
certain species of infusoria could resist, from about 25 degrees
centigrade up to 70 degrees; or expressed on the Fahrenheit
scale, from about 80 degrees to 150 degrees. Before the experi-
ments, none of the organisms could endure these high tem-
peratures; at the end all live and thrive in such conditions.

VARIATION IN UNICELLULAR ORGANISMS 107

Also, if certain Protozoa are subjected for long periods to
solutions of certain chemicals, their resistance to these chemi-
cals may after many generations become increased. Such a
result, it may be remarked, is not readily and easily reached;
many investigators who have attempted to produce it have
failed. But with some Protozoa, under proper conditions,
striking results are reached ; the resistance is greatly increased.
The animals become acclimatized or immunized to different
chemicals, just as higher animals become immunized to
certain organic poisons and bacterial products, so that they
can later live and thrive under conditions which earlier would
have killed them. In these respects the relations are the same
in unicellular organisms and in higher animals, save that in
the unicellular forms many generations are required for the
production of immunity. —

These matters have been studied most fully in the bacteria
and in the parasitic Protozoa.^ In the trypanosomes, flagellate
Protozoa that live in the blood of various mammals, including
man, the relations found are these : When the vertebrate host
— for example, the rat — is infected, the parasites multiply
rapidly for a time by fission, till very large numbers are pres-
ent in the blood. After a time, almost all of these suddenly
die. This is due to the production by the host of an "anti-
body," which is poisonous to the trypanosomes. But not all
the parasites are killed, for later a large number of them
reappear, inducing a "relapse." Investigation shows that this
recrudescence of the parasites is not due to the disappearance
of the antibody from the blood; it is still present and it is
still fatal to parasites of the type that were originally present.
What has happened is that a small proportion of the trypan-
osomes have acquired resistance to the antibody; they can
now live in its presence. These immunized individuals then
multiply, so that the host again becomes filled with parasites.

io8 GENETIC VARIATIONS

now resistant to the antibody. This has been found to occur
even when the original infection was by means of a single
trypanosome, so that it is clear that resistance has been
acquired by individuals of the same strain that were before
non-resistant. In a similar way, resistance has been found to
be induced toward drugs introduced into the body for the
purpose of killing the parasites.

The resistance once acquired lasts for a long time, for
hundreds or thousands of generations. Ehrlich found that
resistance thus induced by the arsenious drug atoxyl con-
tinued for three years, even though the parasites had been
transferred to animals in which the drug was not present.

Great numbers of cases of this type are known in the para-
sitic Protozoa. This acquisition of new and adaptive features,
inherited in reproduction by fission, is a characteristic feature
in the life of these parasitic organisms; it is a matter of great
practical importance in the treatment of the disorders caused
by the parasites.

Knowledge of these matters in the free-living Protozoa is
less extensive; there is much need here for further work. But
a number of the features that are of importance for genetics
have been brought to light, principally by the work of Jollos,
and of Neuschloss. By long subjection of the free-living proto-
zoan Paramecium caudatum to certain chemicals, resistance
to these chemicals has been greatly increased. This has been
accomplished with certain compounds of arsenic and anti-
mony, and with certain organic compounds, notably quinine,
methylene blue, trypan blue, and fuchsin.

It was shown by Neuschloss that the acquirement of re-
sistance to such materials is due to an induced change in the
intimate chemical processes occurring in the animals; a
change in their metabolism. All the substances named above
are injurious to the organisms. The Paramecia acquire

VARIATION IN UNICELLULAR ORGANISMS 109

through long subjection to a weak solution of them a power
to transform the harmful substance into something else that
is innocuous; practically to destroy it. It was found that each
of the above named organic substances induces a different
change in the animals, so that infusoria that have acquired the
power to resist and destroy quinine do not resist and destroy
methylene blue, and vice versa; and similarly for the other
substances. The acquired resistance is specific for each sub-
stance. Thus the materials of which the organism is composed
— or certain of these materials, at least — are changed in a
different way by each of the chemicals to which resistance is
acquired.

But what is most important for our purposes, such acquired
resistance is inherited, in the free-living Protozoa as well as
in the parasitic ones. The organisms are removed from the
chemical that has induced the resistance. They continue to
multiply by fission, and their descendants for hundreds of
generations continue to possess the resistance acquired by their
ancestors. In the course of a great number of generations
under conditions lacking the chemical, the increased resis-
tance may gradually decrease and finally disappear; to this
fact we return later.

But the positive fact of inheritance for hundreds of gener-
ations shows clearly that the organisms, or certain material
of the organisms, have become changed under the action of
external conditions, and that this modified material con-
tinues to assimilate, increase, reproduce, in its modified con-
dition. This modified, more resistant substance increases
itself a thousandfold in ten generations; many millions of
times in a hundred generations. It is clear therefore that the
environment has produced a genetic variation; an inherited
variation.

no GENETIC VARIATIONS

Such changes are particularly striking in the bacteria; here
they are of great practical importance, as well as of great
interest for the general properties of living matter. The bac-
teria are perhaps the lowest organisms for which we can hope
to have a knowledge of genetics and variation. They are
extremely minute; certain of the common forms are spheres
about one micron, or one-thousandth of a millimeter in
diameter. Each has the volume of less than one cubic micron.
Taking the volume at that figure, calculation shows that it
requires fifteen thousand billion of such bacteria to yield the
volume of a cubic inch. Thus all parts of the bacterial body
are in extremely close relation to the surrounding conditions;
all parts are within half a micron, or one fifty-thousandth of
an inch, of the enveloping conditions, chemical and physical.
There is no part of the bacterial body that is protected from
the action of the surroundings by being lodged in the interior
of a large body, as is the case in higher organisms. This is a
great and important difference.

We know that the minute bacterial body is the seat of
chemical activity of a high potency, manifesting itself as fer-
mentations, putrefactions, production of diseases and the like.
All this is occurring in closest contact with surrounding con-
ditions. Reproduction occurs by division of the body into two.
When this division occurs under uniform and favorable con-
ditions, usually the two parts produced have the same con-
stitution and characteristics as the parent, and this continues
indefinitely, so that the stock remains uniform, without
genetic variations. If we have side by side two different types
of bacteria, when they divide the diversity continues; the
characteristics of each type are inherited ; the two types remain
constant. For many years there was an effort to show that
this is always the case in bacterial reproduction. The uniform

VARIATION IN UNICELLULAR ORGANISMS iii

and favorable kinds of culture media commonly employed
seem to have favored this effort; under the conditions sup-
plied there was little genetic variation.

Yet even so, many phenomena were observed that did not
agree with this interpretation. In a supposedly pure culture of
a single type, there appeared types that were diverse in form,
structure or physiology. It was believed that in most cases such
phenomena were due to contamination. The supposedly pure
culture was held not to be pure. Mingled with the prevailing
type were presumably a few of another type, undetected.
Under certain conditions these did not multiply, so that they
remained inconspicuous. But under changed conditions these
hidden bacteria multiplied and became prominent. Thus
there appeared to have been a genetic variation, a change of
type that was inherited. But it was held that all that had
really happened was the substitution of one type for another.

There were, however, certain changes, particularly in dis-
ease-producing bacteria that were very extensively studied,
which could not be explained in that way. In old cultures, or
under changed conditions, the forms, structures and physio-
logical action of the cells were greatly altered, and these
changes persisted even when the organisms were returned to
the original conditions, so that genetic variation had occurred.
These were characterized as abnormal or involution forms,
and their significance therefore slighted; although for the
problems of genetic variation such changes are of high
interest.

In recent years these changes have been studied systemati-
cally and thoroughly for their own interest. The effects of
changed environmental conditions have been extensively
studied. Improved methods have been devised, for beginning
a culture with a single bacterial cell, so that apparent changes

112 GENETIC VARIATIONS

that occurred could not be accounted for as due to contamina-
tion with another bacterial species.

The result of these intensive studies has been to show that
genetic variations of many types do indeed occur in bacteria,
and particularly that such are brought about by changed
environmental conditions. They show that many well known
changes in bacterial cultures are due, not to contamination,
but to inherited changes produced under the action of the
environment.^ Some of the commoner types of change are the
following:

Attenuation, or decrease in virulence: Disease-producing
bacteria are kept for a long time under unfavorable condi-
tions, and these cause them gradually to lose their power to
produce disease. This decreased virulence is hereditary.

Increase of virulence: A change, the reverse of that just
mentioned, is brought about in many disease-producing bac-
teria by subjecting them to appropriate conditions. By inocu-
lating them successively into animals of higher and higher
resistance, they are caused to acquire by degrees a higher
virulence, and this acquired virulence is inherited.

Another type of inherited change induced by environmen-
tal conditions consists in alterations in the secretions or excre-
tions of the bacteria. Bacteria that secrete mucous coverings
are caused to secrete less mucus, or none at all, and the
changed condition is inherited. Some bacteria secrete colored
materials; by subjection to certain conditions, they are caused
to cease producing the colors; and the change is inherited.

Most important perhaps are the acquirement of adaptive
changes in metabolism. There are certain kinds of bacteria
that can split up particular kinds of sugar, such as lactose,
getting nutriment from them. Certain other related races can-
not. If cultivated in the presence of this kind of sugar, these

VARIATION IN UNICELLULAR ORGANISMS 113

latter make no use of it, for several or many generations. But
in the course of time some of them develop the power to split
up this sugar and to thus obtain nutriment from it. They
thereupon begin to multiply and flourish greatly. And such
individuals transmit this nevi^ly acquired power to their
descendants for a great number of generations, even though
the descendants may live where there is none of the sugar
that induced the change.

Many examples of this type of change could be cited ; and
many other phenomena of extraordinary interest for genetic
variation are known in the bacteria. Instead of being patterns
of uniformity and constancy, as was supposed, it is now found
that bacteria develop genetic variations, inherited changes,
with great frequency and in great variety. The technical
worker who endeavors to keep his conditions constant by
employing cultures of a single strain of bacteria is dismayed
to find that his single strain has broken into two or more, with
diverse physiological properties. Particularly are the bacteria
genetically sensitive to changed environments; these induce
new inherited characteristics; so that the "inheritance of
acquired characteristics" is a characteristic feature of genetics
in these organisms.

It must be said that the phenomena of genetics and genetic
variation are by no means fully worked out in the bacteria. At
present there is much difference of interpretation as to some
of the phenomena. In some quarters we read of free bacterial
genes that circulate about in nature and that may become
associated with various bacterial stocks, changing their heri-
table characteristics; and other adventurous views are put
forth.* When these matters are fully cleared up, they will
beyond doubt yield a most important chapter in genetic
science, and particularly in genetic variation. In any case, the

114 GENETIC VARIATIONS

great role of environmental conditions in producing inherited
variations is not in doubt.

Thus in the unicellular organisms in general — bacteria and
Protozoa — we find that environmental conditions may pro-
duce genetic variations of many sorts. We may classify them
under two heads. On the one hand they may produce injury.
The materials of the organisms are damaged, but continue to
multiply in the damaged condition, so that a long series of
depressed or degenerate generations is produced. On the other
hand, the environment may produce adaptations; it may
result, not in damage, but improvement. The organism under
the influence of the environment acquires power to do certain
things that it could not before do; to bring about chemical
reactions of which it was before incapable, and so to live
under conditions under which it formerly could not live.
These acquired powers are inherited for many generations
after removal from the conditions which induced them.

In producing such results, there are certain important rela-
tions between the length of time that the special environmen-
tal condition has acted, and the intensity and duration in
generations of the genetic variation that it produces. In many
cases the degree or intensity of the inherited effect is propor-
tional to the length of time that the effective environmental
condition acts. This is notably the case with the inherited
damage resulting from living under bad conditions; the dam-
age does not appreciably begin till the conditions have existed
for many generations; it increases in amount as the bad condi-
tions last longer, and finally may result in destruction of the
stock. The constitution is progressively more modified the
longer the conditions last. The same relations are found in
many cases in the production of inherited resistance by the

VARIATION IN UNICELLULAR ORGANISMS 115

action of heat or chemicals ; the longer the action, the greater
the resistance produced. The action of the environmental
conditions on the materials of the organism is progressive.
There is in many cases a similar relation between the length
of time the effective agent acts, and the length of time meas-
ured in generations, that the induced genetic variation con-
tinues to be inherited, after removal from the conditions
that produce it. When the environmental condition — the heat
or the special chemical or the like — has acted but a short
time, usually its effect is not inherited at all. The constitution
has not been modified. If, however, the special environment
has operated for a longer period, its effect is inherited for a
number of generations, but finally disappears. If the action of
the special environment has continued for a very long period,
through many generations, the inherited effect continues for
a great number of generations after removal from the condi-
tions that induce it. An example of these relations is seen in
the inheritance of an abnormality in the shell of the rhizopod
Arcella, as studied by JoUos. The abnormal form of the shell,
known as double-lobing, is partly produced by the accumula-
tion of waste products in the fluid in which the animals
live. If after the abnormality has persisted for many genera-
tions, the individuals that are least abnormal are continually
selected for propagation, gradually the population becomes
normal. When the animals had lived for four months in the
condition tending to produce the abnormality, it required
two months to bring them back to normal; when they had
lived for seven and one-half months in such conditions it
required three months to bring them back to normal; when
they had lived for twelve months in such conditions, four
months were required to bring them back to normal. In these
cases there was about one generation or more a day, so that

ii6 GENETIC VARIATIONS

the number of generations in which the abnormaUty was
inherited was very great.^

Thus the action of the environmental conditions on the
genie substances is gradual and progressive. The longer the
environmental condition acts the greater is the effect pro-
duced. And the longer the environmental condition acts, the
longer does the acquired character persist; that is, the greater
is the number of generations through which it is inherited,
after removal from the conditions that induced it. The modi-
fied organic substance multiplies itself a millionfold. But
finally, as a rule, if the organisms live sufficiently long in an
environment that does not contain the effective condition, it
loses the acquired character. Thus under new conditions, new
inherited characters are acquired, but only after many gener-
ations. On restoration to the old conditions, the old character-
istics are restored, but again only after many generations.

In a general view, therefore, we find that the condition and
capabilities of the unicellular individual depend on the way
its ancestors have lived. If the ancestors have lived for many
generations under conditions that gradually injure their genie
substances, the injury is handed onto the children, not merely
to the third or fourth generation, but to the hundredth or
thousandth generation, causing them to be weak and degen-
erate. If the parents have lived in such conditions as to acquire
new powers, these new powers too are handed on to the hun-
dredth generation and later. Their welfare and capabilities
depend largely on how their ancestors have lived. The general
situation appears to be that which is implied by the terms
"Lamarckian inheritance," and "the inheritance of acquired
characters."

Lamarckian inheritance has long been one of the great
problems of biology; no subject in this field has been more
disputed. One might say that all theory has been for it, all

VARIATION IN UNICELLULAR ORGANISMS 117

experience against it. The observed occurrence of such inheri-
tance in the lower organisms is of great interest. It is worthy
of careful examination. In the next chapter we shall under-
take such an examination. We shall inquire into the processes
that underlie these changes, and their significance for the
general problems of evolution.

NOTES ON CHAPTER IV

^ Page 95. For an account of our knowledge of the genetics of unicel-
lular animals up to 1929, see the present author's "Genetics of the
Protozoa," Bibliographia Genetica, 1929, Vol. 5, pp. 105-330. In this
work will be found an extensive bibliography. References there given
will not be repeated in these pages. For detailed accounts of investiga-
tions briefly referred to herein, reference may be made to the above
work. Only in the case of investigations made since 1929 will reference
be made to the original papers.

^ Page 107, For a detailed account of the conditions in some parasitic
Protozoa, see W. H. Taliaferro, "Host Resistance and Types of Infec-
tions in Trypanosomiasis and Malaria," Quarterly Review of Biology,
1926, Vol. I, pp. 246-69. See also by the same author, The Immunology
of Parasitic Infections, 414 pp., New York, 1929.
^ Page 112. For an extensive review and discussion of the variations in
bacteria under the action of the environmental conditions, see Philip
Hadley, "Microbic Dissociation. The Instability of Bacterial Species
with Special Reference to Active Dissociation and Transmissible
Autolysis," Journal of Infectious Diseases, 1927, Vol. 40, pp. 1-312.
* Page 113. See the recent review by W. H. Man waring, "Environmen-
tal Transformation of Bacteria," Science, 1934, ^^1- 79» PP- 466-70.
' Page 116. A somewhat detailed and illustrated account of this work,
with references to the original, is given in Genetics of the Protozoa
(Note I, above), pp. 261-8.

CHAPTER V

Genetic Variation in Unicellular Organisms, Continued.
What Is the Method of Operation of the Genetic System
in Bringing about Genetic Variations under the Long-
Continued Action of the Environmental Conditions?

THE preceding chapter set forth that in unicellular organ-
isms, changed environmental conditions eventually
cause corresponding changes in the characteristics; and these
changed characteristics may be inherited in later generations.

The occurrence of the "inheritance of acquired characters"
in any group of organisms is of great interest, and is w^orthy
of careful examination as to its method and significance. Such
an examination w^e undertake in the present chapter. We
ask: How does the organic system operate in bringing about
genetic variations under the long-continued action of the
environmental conditions? How does it happen that a
changed environment, although it finally alters the inherited
characteristics, does not give an inherited effect until it has
acted for many generations ? How does it happen that resto-
ration to the original conditions finally causes a return to the
earlier characteristics, but not till after many generations ?

Something is known as to the nature of the organic system
that operates in this way; something is known as to its
method of operation. Examination of these matters will com-
pel us to consider some details in the structure and physiology
of the organisms ; it will illustrate to what lengths nature goes
in making difficult a correct interpretation of her activities.

New inherited characteristics acquired under the action of
the environment are essentially modifications of the organic

120 GENETIC VARIATIONS

materials on which the characteristics depend. If the acquired
characters are indeed inherited, this means that the modified
organic materials increase and multiply in their modified
condition.

The materials which determine characteristics and their
inheritance are, as we know, in the nucleus; they are what we
have called the genie materials or genes. Less is known of the
genie materials in the Protozoa than in the higher organisms.
But it is known that in the Protozoa there are genie materials
in the nucleus, and that these control in large measure the
characteristics of the individuals. By changing the genie
materials of the nucleus, and leaving the rest of the bodily
material the same, the inherited characteristics of the entire
individual are changed. This has been shown to be true in
unicellular organisms, through the processes that occur at
conjugation. In conjugation, two individuals unite and ex-
change halves of their nuclei (see Figure i8). Just before
conjugation, the single nucleus in each individual divides
into two halves. Then one-half of the nucleus of individual A
passes into B', and one-half of the nucleus of B passes into A.
The two halves in each then unite to form a single nucleus.
The two individuals then separate.

Thus after conjugation the individual A retains its original
cytoplasmic body, and half its original nucleus, but contains
in addition a half nucleus from 5; and a corresponding
change has taken place in B. Now both A and B begin to
divide and produce descendants.

Now, the characteristics shown by these descendants prove
that the transfer of half a nucleus from B carries with it the
power to produce in A the characteristics of B. For after
conjugation, the descendants of A are found to be, in the long
run, as much like B as they are like A. The half nucleus from
B produces a very great change in the characteristics of the

METHOD OF OPERATION

121

1

a/ ♦ oU

Figure i8. Diagram of the exchange of nuclear materials at conjuga-
tion, in Paramecium.

descendants of A. It is clear therefore that the nucleus con-
tains materials which affect deeply the development and
characteristics of the individual containing them, just as is
true in higher organisms.

122 GENETIC VARIATIONS

After conjugation, the individual A contains a nucleus that
was half derived from its mate B) and this half nucleus from
B changes the characteristics of the descendants of A. But to
make this change may require a very considerable time, dur-
ing vv^hich the individual A and its descendants divide many
times, giving rise to many successive generations of offspring.
All this has been demonstrated for the infusorian Paramecium
by the researches of Sonneborn and Lynch,^ and those of
De Garis." To follow through a typical case of size inheritance
will give the best conception of what happens. We will take
a case described by De Garis. He crossed an individual A of
a very large race of Paramecium caudatum, having an aver-
age length of 198 microns, with an individual 5 of a very
small race, having an average length of but 73 microns (see
Figure 19). The two exchanged half nuclei and separated,
each retaining its original length. Each began to divide, about
once a day. As the generations passed, the descendants of A
began to grow smaller, the descendants of B began to grow
larger. Our figure shows the successive changes in size for
two-day periods, to the twenty-second day after conjugation.
By this time the two had reached practically the same size. A
similar course was followed in other cases of crosses between
individuals of different size, save that in some cases an even
longer time was required to reach the final size, equal in the
descendants of the two mates. In the most extreme cases,
thirty-six days were required to reach the ultimate size. In the
inheritance of fission rates, the new nuclear combination pro-
duces its final effect much more quickly. In some cases the
fission rate was changed at once after conjugation; within the
first five-day period the slow race had begun to divide more
rapidly, the fast race more slowly, so that the two were equal.
In other cases there is apparently a lag like that which occurs
in the inheritance of size, but not lasting so long.

METHOD OF OPERATION

123

A B

A B

A B

A B

A B

A B

A B

A B

A B

Figure ig. Change in size of the two members of a pair, A and B,
at conjugation, and at two-day intervals after conjugation. At the
upper left are shown the relative mean sizes of the two races before

124 GENETIC VARIATIONS

and at conjugation; the successive figures from left to right show the
changes as time passes (see text). Diagram constructed from the
measurements made by De Garis.

Thus after conjugation the new nucleus, formed half from
one of the mates, half from the other, begins to take control
of the body, gradually changing its characteristics. At the end
of a period of three or four weeks the descendants of the two
mates are alike in their characteristics. All this shows that the
inherited characteristics finally depend on the nature of the
genie materials contained by the nucleus.

Now let us look at the relation of these things to the inher-
ited characteristics acquired under the action of the environ-
mental conditions. Have the environmental conditions
changed the nature of the genie materials present in the
nucleus, thus giving rise to new inherited characters ?

To get evidence on this question certain further details
must be examined as to the conditions affecting genie mate-
rials in these organisms. In such Protozoa as Paramecium we
find that the genie materials are separated into active and
reserve portions. The active and reserve parts are separated
into two distinct nuclei, though these are produced by the
division of an original single one.

Immediately after conjugation, as we have seen, each
infusorian has a single nucleus, formed by union of halves
from the two individuals that have conjugated. This nucleus
now divides into two, one of which remains as a small reserve
of genie materials, while the other grows large, and its genie
materials become active in the life of the individual. The
small reserve nucleus is known as the micronucleus ; the
large active one as the macronucleus. From the macronucleus,
genie materials pass out into the rest of the cell, interacting

METHOD OF OPERATION 125

with the cytoplasm to produce the structures and functions
shown in daily life (Figure 20).

Thus in the infusoria the genie substances are separable into
three parts. There is the reserve part in the micronucleus

Figure 20. Diagram showing the separation of the genie substances
into three parts, in Paramecium. A. Micronucleus (reserve). B. Macro-
nucleus (active). C. Genie materials diffusing into the cytoplasm.

(Figure 20, A); there is the active part in the macronucleus
{B) ; and there are the much modified parts given off by the
macronucleus into the cytoplasm (C).

Just before conjugation occurs this large active nucleus, the
macronucleus, breaks up and disappears (Figure 21). Pieces
of it are seen in the cytoplasm; these gradually become
smaller and are absorbed. This absorption of pieces of the
nucleus by the cytoplasm presumably affects the nature of the
cytoplasm. There is then left but the single micronucleus.
This then divides and the two conjugating individuals ex-
change halves in the way before described.

Look now at the relation between the characters acquired
through the action of changed environmental conditions, and
these genie materials. The acquired characters are the result
of modifications of genie materials. But of what genie mate-
rials ? Of the reserves, in the micronucleus } Or of the active
ones, in the macronucleus ? Or of the altered genie products

126

GENETIC VARIATIONS

that have passed out into the cytoplasm and there interact
with each other and with other things ?

On this point the investigator Jollos, to whom is due so
much of our knowledge on the genetics of these organisms,

B

Figure 21. Diagram of the disintegration of the macronucleus in
Paramecium. At A are seen both micronucleus and macronucleus. At
B, the macronucleus has broken into pieces, which are being absorbed
by the cytoplasm.

has promulgated a widely known theory, according to which
none of these environmental modifications affect the reserve
genie materials in the micronucleus; they are held to be
limited to the genie materials of the macronucleus or of the
cytoplasm. For he assumes that any modifications of the genie
materials of the micronucleus would be permanent, since the
micronucleus is never lost. Since the acquired characters are
as a rule not completely permanent, but disappear or are
exchanged for others if the animals are kept for a sufficiently

METHOD OF OPERATION 127

great number of generations in conditions other than those
that produced the modifications, JoUos holds that they must
have their seat in parts of the cell that are not permanent;
that is, in the cytoplasm or the macronucleus. The environ-
mental modifications he holds therefore to be essentially
transitory; he calls them therefore Dauermodifikationen^
which may be translated "long-standing modifications."
These he holds "are only changes forced upon the organism
from the outside, which do not alter its potentialities, and
though they do for some time prevent these potentialities
from being realized, the latter finally overcome them."^

And since these modifications affect, as he believes, only
transitory parts of the organism, leaving the permanent genie
materials unchanged, he holds that they "are without signifi-
cance for the question of the heritable transformation of a
line of descent; for the question of the origin of species."

This conclusion rests upon the assumption that the genie
materials of the micronucleus must retain permanently any
modifications that they undergo. This, however, is not a
demonstrated fact. It is certain from the fact of the inheri-
tance of environmental characters that genie materials some-
where in the organism are modified by the environmental
conditions. All genie materials come finally from the micro-
nucleus; if they are modifiable in one location they may be
modifiable in another. And if they can be modified in one
direction by certain conditions, they may well be modifiable
in a reverse direction by opposing conditions. The recent
discoveries as to inheritance in conjugation taken in connec-
tion with the facts as to modification by the environment,
certainly suggest an alternative theory to that of Jollos. They
suggest strongly that genie materials in the two nuclei may
become modified, as well as those elsewhere.

128 GENETIC VARIATIONS

The facts of inheritance at conjugation show with cer-
tainty that the genie materials in both the reserve nucleus and
the active nucleus help to control the characteristics, and they
strongly indicate that the genie products in the cytoplasm play
a part in this also. To see this, follow in the cross between a
Paramecium of a large race and one of a small race what
happens in the descendants of one of the individuals. For
this we may select the small individual B in Figure 19, above.
After conjugation it has a small body, composed of cytoplasm
containing genie substances that came from the nucleus of
the small parent B before it had conjugated. Its cytoplasm
likewise contains many dissolving pieces of the old macro-
nucleus of the parent B. All these parts tend to produce a
body that is small. It also contains a nucleus, half derived
from the large parent /4,half from 5. Incidentally, this nucleus
becomes later separated into its two parts, the minute reserve or
micronucleus and the large active macronucleus. This nucleus
tends to make the body large; in the long run it produces a
body of twice the length, and therefore eight times the bulk,
of this original B. But during the intervening three weeks,
while at least twenty generations are produced, the body does
not reach this size. It is intermediate between the size that
the new nucleus tends to produce, and the size that the
materials of the old body tend to produce. It is clear therefore
that the materials of the old body help during this period to
control the body size. But as generations pass the new nucleus
more and more dominates the body, until finally a size is
reached which is that due to the new nucleus; a size inter-
mediate between the sizes of the two parents.

During this period of change, the organisms are regularly
dividing; some twenty or more generations are produced
before the final size is reached. At every division the original
material of the body is diluted to half its volume. After ten

METHOD OF OPERATION 129

generations, any individual contains less than one-thousandth
of the original body material ; all the rest is new. After fifteen
generations the amount of old material remaining in any
individual is less than one thirty-two-thousandth of the origi-
nal body. Yet at these times there is still a marked distinctive
effect of the original body material; the size is not yet that
which the nucleus tends to produce. Clearly, either this
original body material has a great effect even in enormously
high dilutions; or what seems rather the case, some part of it
has the power of reproducing itself true to type during divi-
sion, just as does the genie material of the nucleus, so that we
must call it also genie material. This genie material of the
body is either the cytoplasm itself, or some remnants of the
old macronucleus. But this original body material gradually
loses its distinctive effect, through coming under the influence
of the nucleus. After three or four weeks or twenty to thirty
generations, it is completely dominated by the new nucleus.

With these facts in mind, return to our question: Are the
inherited environmental characters the result of modification
of the reserve genie materials of the micronucleus, of the
active genie materials of the macronucleus, or of the genie
products present in the cytoplasmic body ?

If the environmental modifications are in the genie mate-
rials of the cytoplasm only, we should expect the acquired
characters to disappear in twenty to thirty days, since we
know from what happens at conjugation that by the end of
such a period the nucleus has taken full control. Again, if the
acquired characters are in the active macronucleus only, they
would disappear at once when the macronucleus disappears.
If they are due to modifications of the micronucleus, they
would disappear only when some new influence modifies this
micronucleus.

130 GENETIC VARIATIONS

It is clear that the acquired characters do not disappear in
any such brief time as thirty days; on the contrary they may
persist for many months. This appears to be strong evidence
against the limitation of the environmental effect to the
cytoplasmic genie materials.

At certain times in the course of generations, as w^e have
seen, the large macronucleus disappears; it breaks into
pieces, and these are absorbed, though they may remain for
a few days in the cytoplasm. This leaves the micronucleus as
the only nucleus. If the modifications resulting from the
action of the environment are limited to the active genie
substances of the macronucleus, they should disappear w^hen
the macronucleus disappears; or at least a little later, after its
pieces in the cytoplasm have disappeared. At this time the
acquired characters should be lost.

Sometimes the acquired characters are indeed lost when
the macronucleus is lost. At the time of conjugation the
macronucleus is lost in both individuals, and is replaced from
the new nucleus formed from the union of the two half nuclei
— one half from each of the individuals of the pair.

And at conjugation certain of the inherited modifications
partly or entirely disappear. This is true to a large extent of
the injurious modifications resulting from bad living condi-
tions. They often almost or quite entirely disappear when con-
jugation occurs, so that conjugation has been said to cause
"rejuvenescence." The animal after conjugation regains its
original high level of vitality, even though it remains in the
conditions which brought about the injuries. It is also true
that the increased resistance produced by long subjection to
certain chemicals, or by heat, often disappears at conjugation.
Since the macronucleus disappears at the same time, this
could be held to indicate that it is the genie substances of the
macronucleus that have been modified, while those of the

METHOD OF OPERATION 131

micronucleus were unaffected. When the macronucleus
disappeared, therefore, the acquired modifications also
disappeared.

But disappearance of the macronucleus and its replacement
from the micronucleus are not all that occurs at conjugation.
There is also the exchange of half nuclei; so that each individ-
ual has a new combination of genie materials. This might
well give a new combination of characteristics, so that the old
characteristics, however acquired, might be no longer mani-
fested. The seeming loss of the acquired characters might be
due to this formation of a new combination of genie mate-
rials, rather than to the loss of the macronucleus.

The difficulty here presented was overcome in the infuso-
rian Uroleptus by an experiment of Calkins. In a conjugating
pair from a depressed stock, by an operation Calkins pre-
vented the exchange of the half nuclei. The macronucleus
disappeared as usual, but the new one was formed from the
individual's own micronucleus, without aid from that of its
mate ; so that there was no new combination of genie mate-
rials. Yet the descendants of the individual in which this
occurred had lost their depression. With the disappearance
of the macronucleus the acquired modification likewise
disappeared.

The same thing is shown to be true for the infusorian
Didinium by Calkins and by Beers. When the animals be-
come depressed through the action of bad conditions, they
become encysted, and while in the cyst the macronucleus
seemingly disappears and is replaced from the micronucleus.
At the same time the depressed condition of the organism
also disappears.

In these cases then it appears probable that the modifica-
tion produced by the environment was only in the genie
materials of the macronucleus. The macronucleus is a transi-

132 GENETIC VARIATIONS

tory body; it disappears at intervals, and when it disappears,
the acquired characteristics dependent on it must disappear
also. Such acquired characters then are only transitory. They
play an important role in the life of the organisms; they last
for generations, but they do not permanently alter the charac-
teristics of the species. For such modifications JoUos' term of
Dauermodifikationen is an appropriate term; they are long-
standing modifications, but destined to disappear when the
macronucleus disappears.

But it is certain that not all environmental modifications
are limited to the macronucleus. In Paramecium, the organ-
ism most studied with respect to these matters, it was unex-
pectedly found, after much work had been done, that the
macronucleus disappears at regular rather short intervals,
being replaced in each case from the micronucleus. This is
the process known as endomixis. In Paramecium aurelia this
disappearance and replacement occurs every three or four
weeks; in Paramecium caudatum the interval is about twice
as great.

Yet most of the acquired modifications do not disappear
at these periods. Though the macronucleus is lost, they re-
main. It is clear therefore that they are not modifications
limited to the genie materials of the macronucleus. This is
true of the depression, or injurious modifications, induced by
poor living conditions. Paramecium continues to "run
down," becoming more and more depressed, in spite of
repeated replacements of the macronucleus from the micro-
nucleus. This is true also in many cases of the inherited
resistance induced by subjection to heat or to chemicals; this
often does not disappear when the macronucleus disappears.
On the contrary, it continues to be inherited for many gener-
ations, during which there are repeated disappearances and
replacements of the macronucleus.

METHOD OF OPERATION 133

All this shows that other genie materials besides those in
the macronucleus may become modified by environmental
conditions. We have already given reasons for holding that
the cytoplasmic genie substances cannot be the seat of the
environmental modifications; for if they were, the modifi-
cations would be expected to disappear inside of about four
or five weeks. There remains only the genie materials of the
micronucleus itself. These reserve genie materials would
doubtless be modified less readily than the active materials in
the macronucleus; but there is no reason to reject absolutely
the possibility of modification for them, for they are literally
genie materials of the same type as those in the macronucleus,
since the two are halves of the same original nucleus.

Furthermore, in certain cases a second-degree modification
is induced, of such a character as to suggest strongly that its
seat is in the micronucleus. In Uroleptus and in Spathidium
it has been shown that when the depression induced by bad
conditions is moderate in degree, it is completely lost at
conjugation when the macronucleus is replaced. But if the
depression has become very severe, it does not fully disappear
at conjugation, even though the macronucleus is then re-
moved. The condition of the animals is improved, but a con-
siderable degree of depression remains and continues to be
inherited. This is what we should expect if the environmen-
tal conditions first modify the genie materials of the macro-
nucleus, then if severe and long continued, modify the
reserves in the micronucleus.

The facts therefore appear to point to the following situa-
tion. Environmental conditions, so far as they induce inher-
ited modifications, first alter the active genie materials of the
cytoplasm; such modifications last but a short time. Later,
after acting longer, they alter the genie materials of the
macronucleus; such modifications disappear with the disap-

134 GENETIC VARIATIONS

pearance of the macronucleus. Finally, after severe and long-
continued action, the environmental conditions alter the
reserve genie materials in the micronucleus. Such modifica-
tions last either until new^ environmental conditions again
alter the micronucleus; or until through conjugation a new-
combination of genie substances is made that yields new^
characteristics.

A final test of whether these intense modifications do in-
deed affect the micronucleus must come through crossing a
deeply modified stock with one that is not modified, thus
determining whether the modification is transmitted with
the micronucleus. By the methods recently devised by Sonne-
born and by De Garis, such a test may come soon.

We may attempt now a summary of the established facts
as to the inheritance of acquired environmental characters in
unicellular animals. Some of the inherited characteristics
manifested by these organisms are consequences of the type
of environment under which their ancestors have lived; in
some cases ancestors hundreds of generations earlier. A
changed environment induces a change in the genie materials
that persists long after removal from the conditions that in-
duced the change. And this altered genie material, as it
assimilates, grows and multiplies, produces, not the original
type of material, but the modified type. This continues for
many generations.

Neither the original characteristics nor the modified char-
acteristics are fully stable. The original characteristics are
found to become modified as the conditions alter. In this
respect they are like the modified characters; these too become
altered again when the original conditions are restored.
There is then a return to the original characteristics, and this
return is to be expected on the same ground that the modifi-

METHOD OF OPERATION 135

cations were to be expected. One set of conditions produces
(if it acts for a sufficiently long period) the original charac-
teristics; another set (by long-continued action) produces the
modified characters. The striking feature of these changes in
characteristics is that they lag far behind the conditions that
finally induce them. When the organisms are cultivated in
such a chemical as arsenic, it may be many generations before
their resistance to this chemical is increased ; when they are re-
turned to "normal" conditions, it may again be many gener-
ations before the "normal" resistance is again produced. In
either case, the characteristics shown by the organism depend
largely on the conditions in which its ancestors of long past
generations have lived.

How are to be explained the adaptive features of the genetic
variations induced by environment ? How is to be explained
the acquirement, under the action of particular agents, of
inherited acclimatization or immunity; increased powers of
resistance to these agents ? To attribute such results to a gen-
eral quality of adaptiveness does help, for in each example of
adaptation the result is always a specific relation between a
particular organism and a particular situation or condition;
and the question for answer is: How is this specific relation
brought about? How is to be explained the acquirement,
under the action of a particular chemical, of increased powers
of resistance to that particular chemical ? When the organisms
are placed in contact with nutritive material which they can-
not use, how is to be explained the acquirement of a changed
method of metabolism, enabling them to use precisely that
material ?

These questions are identical with those that arise with
relation to the powers of direct acclimatization or immuniza-
tion that are found in the somatic tissues of higher organisms.
There is as yet no assured answer to the questions. In an

136 GENETIC VARIATIONS

earlier work* the present author has developed the view that
such adaptive action must be due fundamentally to selective
action occurring within the tissues of the organism. If there is
such action, it occurs as well in the genie materials of unicel-
lular organisms as in the somatic tissues of higher forms.

Does the fact that environmental effects and reactions are
inherited in Protozoa, giving rise to adaptive and progressive
inherited characteristics, suggest that we should eventually
expect to find this method of operation in the higher
organisms ?

Perhaps the most important feature in the genetics of the
Protozoa is the demonstration that the genie materials pos-
sess here the same powers of direct acclimatization or immu-
nization to injurious agents that are possessed by the somatic
tissues of higher organisms. Genie materials are materials
that assimilate and reproduce, each true to its type. The facts
in the Protozoa show that after the genie materials have
undergone an adaptive change, they may assimilate and
reproduce in the changed condition, resulting in inheritance
of the change. There is thus no incompatibility between the
acquirement of adaptive changes and the capability of assim-
ilation and reproduction in the changed condition; for these
things occur in the same material in lower organisms. There
is no inherent impossibility in this method of operation.
There is thus no general reason why it should not occur in
higher organisms; no a priori reason why the germ cells of
higher animals should not thus acquire certain adaptive
characteristics, and hand them on to descendants. Whether
they do or do not is simply a question of fact, to be deter-
mined by observation. If the higher organisms differ from
the lower ones in this matter, it is because of special condi-
tions operating in higher organisms.

METHOD OF OPERATION 137

There are indeed such special conditions working against
the inheritance of environmental effects in multicellular
organisms. As compared with unicellular organisms, the
dijfiferences that are significant in this respect are mainly the
following. The unicellular organisms are minute, their genie
materials are in close relation with environmental agents. The
higher organisms are relatively enormous. Only the genie
materials of their germ cells can affect the characteristics of
descendants, and the germ cells are hidden deeply within the
body, protected from the action of most outer agents. Be-
tween parent and offspring a long series of cell generations
intervenes, during which there is opportunity for loss of any
modifications induced in the germ cells of the parent. These
conditions operate against the inheritance of the effects of
environmental agents in higher organisms.

Weighing more strongly than any of these considerations
perhaps is the fact of the great weight of negative evidence
from observations and experiments directed toward testing
the inheritance of environmental acquirements. No subject
in biology has been worked on more than this. The over-
whelming majority of the results are negative. Sometimes the
discovery of the inheritance of acquired characters is an-
nounced; usually it is discovered later that an error under-
lies the interpretation. As yet outstanding are the results of
William McDougalP on the inheritance of the effects of
training on the rat. These if accepted at face value would
demonstrate the inheritance of acquired characters. In view
of the history of many such investigations in the past, one can
hardly do otherwise than expect that some error of interpre-
tation will be found to underlie these results; and that they
will be found not to demonstrate the inheritance of the effects
of training. It may turn out that the inheritance of environ-
mental adaptations is limited to unicellular organisms.

138 GENETIC VARIATIONS

If this is the situation, where are we left as to the source of
the variations in genie materials that must underlie evolution
in higher organisms?

Some investigators have expressed the opinion that such
variation is not required; that mere change of combination
and grouping, without alteration in the nature of any gene,
can in some way give rise to progressive evolution; can have
produced the seeming differentiation among genes now
existing. To most, however, this appears unintelligible.

Possibly a type of gene change that is in the nature of
development rather than in that of disintegration and abnor-
mality will eventually be detected. When this is demon-
strated, the road for progressive evolution will be open ; until
this is demonstrated genetic science appears to be left with
the task that Osborn has called The Search for the Unknown
Factor in Evolution.

NOTES ON CHAPTER V

^ Page 122. T. M. Sonneborn and R. S. Lynch, "Hybridization and
Segregation in Paramecium aurelia," Journal of Experimental Zoology,
1934, Vol. 67, pp. 1-72.

^ Page 122. C. F, De Garis, "Heritable Effects of Conjugation between
Free Individuals and Double Monsters in Diverse Races of Para-
mecium caudatum." To appear in the Journal of Experimental
Zoology.

^ Page 127. V. JoUos, "Experimentelle Vererbungsversuchen an In-
fusorien," Zeitschrijt fur inductive Abstammungslehre, 1920, Vol. 25,
p. 89. See discussion of this matter in Jennings, Genetics of the
Protozoa, p. 299.

* Page 136. H. S. Jennings, Behavior of the Lower Organisms (New
York, 1906), Chapter XXI.

^ Page 137. See J. B. Rhine and William McDougall, "Third Report on
a Lamarckian Experiment," British Journal of Psychology, Vol. 24,
1933, pp. 213-24.