Historic, archived document

Do not assume content reflects current
scientific knowledge, policies, or practices

U. S. DEPARTMENT OF AGRICULTURE.

BUREAU OF PLANT INDUSTRY— BULLETIN NO. 136.

B. T. GALLOWAY. Chief of Bureau.

METHODS AND CAUSES OF EVOLUTION.

O. F. COOK,
BiONomsT, Bureau of Plant Industry.

Issued October 31, 1908.

WASHINGTON:

GOVERNMENT PRINTING OFFICE,

1908.

BUREAU OF PLANT INDUSTRY.

a

Physiologist and Pathologist, and Chief of Bureau, Beverly T. Galloway.

Physiologist and Pathologist, and Assistant Chief of Bureau, Albert F. Woods.

Laboratory of Plant Pathology, Erwin F. Smith, Pathologist in Charge.

Investigations of Diseases of Fruits, Merton B. Waite, Pathologist in Charge.

Laboratory of Forest Pathology, Haven Metcalf, Pathologist in Charge.

Cotton and Truck Diseases and Plant Disease Survey, William A. Orton, Pathologist in Charge.

Plant Life History Investigations, Walter T. Swingle, Physiologist in Charge.

Cotton Breeding Investigations, Archibald D. Shamel and Daniel N. Shoemaker, Physiologists in Charge.

Tobacco Investigations, Archibald D. Shamel, Wightman W. Garner, and Ernest H. Mathewson, in

Charge.
Corn Investigations, Charles P. Hartley, Physiologist in Charge. ♦

Alkali and Drought Resistant Plant Breeding Investigations, Thomas H. Kearney, Physiologist in

Charge.
Soil Bacteriology and Water Purification Investigations, Karl F. Kellerman, Physiologist in Charge.
Bionomic Investigations of Tropical and Subtropical Plants, Orator F. Cook, Bionomist in Charge.
Drug and Poisonous Plant Investigations and Tea Culture Investigations, Rodney H. True, Physiologist

in Charge.
Physical Laboratory, Lyman J. Briggs, Physicist in Charge.

Crop Technology and Fiber Plant Investigations, Nathan A. Cobb, Crop Technologist in Charge.
Taxonomic and Range Investigations, Frederick V. Coville, Botanist in Charge.
Farm Management Investigations, William J. Spillman, Agriculturist in Charge.
Grain Investigations, Mark Alfred Carleton, Cerealist in Charge.
Arlington Experimental Farm, Lee C. Corbett, Horticulturist in Charge.
Vegetable Testing Gardens, William W. Tracy, sr., Superintendent.
Sugar-Beet Investigations, Charles O. Townsend, Pathologist in Charge.
Western Agricultural Extension Investigations, Carl S. Scofield, Agriculturist in Charge.
Dry-Land Agriculture Investigations, E. Channing Chilcott, Agriculturist in Charge.
Pomological Collections, Gustavus B. Brackett, Pomologist in Charge.

Field Investigations in Pomology, William A. Taylor and G. Harold Powell, Pomologists in Charge.
Experimental Gardens and Grounds, Edward M. Byrnes, Superintendent.
Foreign Seed and Plant Introduction, David Fairchild, Agricultural Explorer in Charge.
Forage Crop Investigations, Charles V. Piper, Agrostologist in Charge.
Seed Laboratory, Edgar Brown, Botanist in Charge.
Grain Standardization, John D. Shanahan, Crop Technologist in Charge.
Subtropical Laboratory and Garden, Miami, Fla., Ernst A. Bessey, Pathologist in Charge.
Plant Introduction Garden, Chico, Cal., W. W. Tracy, jr., Assistant Botanist in Charge.
South Texas Garden, Brownsville, Tex., Edward C. Green, Pomologist in Charge.
Farmers' Cooperative Demonstration Work, Seaman A. Knapp, Special Agent in Charge.
Seed Distribution (Directed by Chief of Bureau), Lisle Morrison, Assistant in General Charge.

Editor, J. E. Rockwell.
Chief Clerk, James E. Jones.

Bionomic Investigations of Tropical and Subtropical Plants.
O. F. Cook, Bionomist in Charge.
G. N. Collins and F. L. Lewton, Assistant Botanists.
H. Pittier, J. H. Kinsler, and A. McLachlan, Special Agents.
C. B. Doyle and R. M. Meade, Scientific Assistants.
136

2

LETTER OF TRANSMITTAL.

U. S. Department of Agriculture,

Bureau of Plant Industry,

Office of the Chief,
Washington, D. 0., July 30, 1908.

Sir: I have the honor to transmit herewith a paper entitled
"Methods and Causes of Evolution," by Mr. O. F. Cook, of tins
Bureau, and recommend its publication as Bulletin No. 136 of the
Bureau series.

The doctrine of evolution is now being- made of practical use in the
solution of problems of breeding and acclimatization. The nature of
evolution must be understood before the effects of selection and
environment upon the characters of organisms can be correctly appre-
ciated as a guide in our efforts to improve plants or to avoid the
degeneration which appears in many of our domesticated varieties.
The distinctions made in this paper will assist in the practical appli-
cation of the ascertained facts of evolution, and will help to avoid
confusion from numerous conflicting theories which have been devel-
oped without bringing evolution into relation with the physiology of
plants.

The accompanying report contains a brief statement of the present
status of our knowledge of these important subjects, particularly
on the sides relating to agriculture. It summarizes in direct and
simple form the results of more detailed studies already published by
tins Department and in scientific journals.

The author wishes to acknowledge helpful criticisms from Prof.
Willet M. Hays and from several of his colleagues, Messrs. Walter T.
Swingle, T. H. Kearney, W. R. Maxon, Frederick V. Coville, C. S.
Scorleld, and David Fairchild. The paper has also been read by
Dr. Alexander Graham Bell, who has made the following significant
comments regarding the function of natural selection:

I, too, entertain the feeling that natural selection does not, and can not, produce
new species or varieties, or cause modifications of living organisms to come into exist-
ence. On the contrary, its sole function is to prevent evolution. In its action it is
destructive merely, not constructive — causing death and extinction, not life and pro-
gression. Death can not produce life; and though natural selection may cause the
136 * 3

4 METHODS AND CAUSES OF EVOLUTION.

death of the unfit, it can not produce the fit — far less evolve the fittest. It may per-
mit the fit to survive by not killing them off if they are already in existence; but it
does not bring them into existence or cause improvement in them after they have
once appeared. We must look to other agencies for the causes of evolution. A closed
gate may block a road, but it does not push the traveler into a new path, or. indeed,
cause him to move at all. It is a mere static obstruction, not a dynamic force. In a
similar manner natural selection prevents evolution along certain lines; but it is not
a dynamic force compelling progress along other lines. The motive power of evolution
must be sought elsewhere.

Respectfully, B. T. Galloway,

Chief of Bureau.
Hon. James Wilson,

Secretary of Agriculture.

136

CONTENTS.

Page.

Introduction 7

Evolution proceeds by continuous changes of characters 8

Natural selection unlike artificial selection 9

Natural selection not the cause ot* evolution 10

The significance of useless characters 12

Abrupt variations not incompatible with continuous evolution 14

Reversion, mutation, and Mendelism 14

Adjust ments of characters to external conditions 17

Characters not directly molded by environment 18

Directions of evolution not fixed 19

Diversity a factor in natural evolution 20

Broad breeding a factor in natural evolution 22

Narrow breeding a factor in domesticated varieties 23

Mutual relations of principal factors 24

Evolution problem distinct from descent problem 26

Evolution inherent in species 28

Conclusions 29

Index 33

130 5

B. P. I.— 395.

METHODS AND CAUSES OF EVOLUTION.

INTRODUCTION.

Evolution is the branch of science concerned with the study of
organic change. That species are subject to change is a fact as
general and as fundamental in the organic world as gravitation is in
the inorganic, and no less important in its relations to human progress.
Our knowledge of evolution is far from complete, but it affords our
only insight into many important problems.

The further development of the art of agriculture by which we
live and even the progress of the human race itself are dependent
upon knowledge of the conditions and factors which bring about
changes in organisms. Our attempts to ameliorate varieties or to
avoid degeneration are experiments in evolution. To know how
species have developed in nature affords a necessary basis of judg-
ment in the investigation of methods of improvement of domesti-
cated types.

The problem of evolution is not to be solved without understanding
how species are constituted, since it is only hi species that evolution
is known to take place. The higher plants and animals maintain
then existence in nature not as separate individuals or in small,
isolated groups, but as large groups of freely interbreeding individu-
als— the groups we caU species. The lines of descent of a normal
species are united by free interbreeding into a vast network or fabric
in which progressive variations are accumulated and built up into
new organs and functions. This constructive process also bears the
name evolution.

The members of species are not normally uniform, but are normally
diverse. Continuous interbreeding among the di verse members of a
species is accompanied by continuous changes in the group as a whole,
though without dinnnisliing the normal diversity of descent. The
more advanced the evolution of a group the greater are the diversities
found among the members of the same species.

Natural selection and other agencies of the environment have not
been found to cause the evolution of species. After fifty years of

136

8 METHODS AND CAUSES OF EVOLUTION.

study of evolution it is becoming apparent that constructive evolu-
tionary changes are spontaneously put forth by species, winch tend
to these evolutionary motions as naturally as inanimate objects tend
to remain at rest. Too much emphasis has been laid upon evolution
as a process of formation of new species and too little upon evolution
as a process of change in species already formed.

Species are multiplied by the subdivision of older species, but
evolution goes on whether species are subdivided or not. If two
parts of a species become separated the two groups no longer share
the new characters which may be attained by each, and thus gradually
become different. This occurs even under apparently identical
natural conditions, in neighboring islands, mountains, or valleys.
Facilities for interbreeding are necessary to keep the same characters
distributed throughout a specific group. The formation of more
species is one of the results of evolution, but is not a cause of evolu-
tion. Evolution and species-formation are different problems. a

The recognition of evolution as a property inherent in all species of
plants and animals opens the way to the solution of many practical
problems. It is not necessary that breeders undertake to produce
evolutionary changes, for these take place spontaneously without our
interference. Our power over evolution is to control and direct, to
accelerate, retard, or reverse it. Some of the means of attaining our
purposes are already known to us, and the chances of learning others
are greatly increased by clear conceptions of the evolutionary facts
and relations encountered in the work of the breeder.

Objections which have been urged against mechanical theories of
evolution do not hold against the present interpretation of the facts,
for selection and other external conditions are not reckoned as
primary causes of evolutionary progress. Their functions are
regarded as only secondary and directive. The theory that selection
causes evolution, notwithstanding its wide popularity, seems not to
be well founded. It stands in the way of further progress in evolu-
tionary science and interferes with a full development of the art of
breeding.

EVOLUTION PROCEEDS BY CONTINUOUS CHANGES OF
CHARACTERS.

To maintain a distinction between methods and causes conduces
to clear thinking in evolution. How evolution proceeds is one
question; what actuates the motion is another. By treating these
two questions separately much confusion of issues may be avoided.

« Cook, 0. F. Evolution Not the Origin of Species, Popular Science Monthly,
64:445, 1904; republished in the Annual Report of the Smithsonian Institution for
1904 under the title The Evolutionary Significance of Species. See also Factors of
Species-Formation, Science, n. s., 23: 506.
136

NATURAL SELECTION UNLIKE ARTIFICIAL SELECTION. 9

Evolution may be considered as a fact, and the modes or methods of
evolutionary change may be studied quite apart from any explanation
or theory of the cause of evolution. The regular courses of the
heavenly bodies are taken as evidence of the fact of gravitation,
although the underlying causes still remain unknown.

Darwin ascertained that evolution is accomplished through con-
tinuous, gradual changes in the characters of species. He had a wide
acquaintance with species in nature and with the facts of variation.
He found in the differences between natural groups evidences of
gradual evolutionary divergence. He saw that the evolutionary
courses of species could be traced from their characters, much as
hunters follow animals by their tracks. Those who have a like
acquaintance with species in nature find similar facts and accept
Darwin's conception of the method of evolution.

Studies of the classification and geographical distribution of wild
species have special value in training the judgment in evolutionary
matters. The evidence that evolution is accomplished through
continuous changes of characters can not be appreciated adequately
by those who know plants and animals only as domesticated breeds
and have not become acquainted with the similarities, differences,
and internal diversities of species in nature.

NATURAL SELECTION UNLIKE ARTIFICIAL SELECTION.

In addition to showing that continuous changes represent the
method by which evolution proceeds, Darwin advanced a theory of
the cause of evolution — natural selection by environmental agencies.
He applied to natural species analogies drawn from domesticated
breeds. The selective action of the environment and of the struggle
for existence was supposed to interfere for the preservation of useful
variants in nature, to keep their new characters from being " swamped ' '
by interbreeding, in the same way that artificial selection is used to
preserve the peculiarities of domesticated breeds.

In reality, natural selection is not similar to artificial selection; there
is a complete biological opposition. Natural selection eliminates
deficient and aberrant individuals, but preserves the species as a
great multitude of diverse, broadly interbreeding organisms. Arti-
ficial selection disregards the species, but saves some of the aberrant
forms and restricts their descent to narrow lines of consanguinity.
Nature has been searched in vain for proofs of the origin of new types
from selected individuals or from narrowly restricted varieties.
Natural selection by preserving a broad network of descent permits
a gradual and completely continuous evolution.

Darwin had learned from his studies of classification and geograph-
ical distribution that species in general evolve by gradual subdivision
and divergence, as branches grow on trees, but the analogies of
52641— Bull. 136-08 2

10 METHODS AND CAUSES OF EVOLUTION.

artificial selection are not in accord with this idea of complete con-
tinuity. They do not lead us to think of trees as having inherent
powers of growth, but as growing because they are pruned.0

The breeder cuts off all except the most proficient individuals after
careful selection among small fluctuating differences. He seeks in
this way to obtain the highest and most uniform expression of a
particular character, but this artificial specialization of characters in
narrow-bred varieties is not the same as the free evolutionary progress
of broad-bred natural species.

To keep a few individuals away from the network of descent of
their species is not nature's method of evolution. Narrow artificial
selection turns its proteges into paths of degeneration. Natural
selection does not separate the fittest individuals from others of their
kind. It leaves them to interbreed and share their progress with
millions of the less proficient. Only the unfit suffer from nature's
discrimination. The more prosperous the species the greater the
number of interbreeding individuals. Only when species are verging
on extinction do they approach the condition of narrow breeding
shown in domesticated varieties.

NATURAL SELECTION NOT THE CAUSE OF EVOLUTION.

Darwin retained much of the pre-evolutionary idea that species are
normally stationary, and was accordingly led to infer that changes of
characters must be brought about through changes of external con-
ditions. Species were not supposed to have an inherent power of
making progressive changes without environmental interference or
even in spite of it. Variability was recognized, but variations were
not thought of as resulting in evolution unless selection interposed.
The adaptive fitness of many species to particular environments
seemed to indicate that the causes of evolutionary progress were in
the environment.

It is the office of natural selection to guide the evolutions of species
in directions of fitness to their environments, but not as a cause of
evolution, not through any power to actuate the evolutionary motions
of species. The influences of selection are more often exerted in hold-
ing species back than in carrying them forward. Adaptation is not
the whole of evolution, but only a particular kind, a specially directed

« "Each new variety or species, when formed, will generally take the place of, and
thus exterminate, its less well-fitted parent. This I believe to be the origin of the
classification and affinities of organic beings at all times ; for organic beings always seem
to branch and subbranch like the limbs of a tree from a common trunk, the flourishing
and diverging twigs destroying the less vigorous — the dead and lost branches rudely
representing extinct genera and families." — Darwin, Charles, On the Tendency of
Species to Form Varieties, etc. Proceedings of the Linnean Society of London, 1857,
p. 53.
136

NATURAL SELECTION NOT THE CAUSE OF EVOLUTION. 11

form of evolution. If continued variation is assumed, it is to this that
the evolutionary progress must be ascribed, not to natural selection.
Selection merely limits the directions that progressive variations
may take.a

A logical fallacy, as well as a biological error, is involved in assuming
that the selecting away of a part of a supposedly stationary species
would change the characters of the remainder. Only the statistical
averages of the survivors of such a species would be shifted, not their
evolutionary positions. Selection would leave the surviving members
of a group more uniform, but it would not carry them farther on the
course of development. Progress toward uniformity means that
some of the characters are suppressed instead of new differences being
added. Greater uniformity marks an agricultural improvement in a
domesticated variety, but is not a measure of evolutionary progress
in a natural species.

Some of the most effective evidence of the supposed power of
selection to produce changes of characters has been drawn from the
phenomena of mimicry among butterflies. Harmless species some-
times show very close resemblance in wing colors to members of other
families protected against birds by distasteful secretions. But even
mimicry can be understood by recognizing that selection tends to
restrict evolution better than by supposing that selection has dragged
the mimetic species away from their relatives and painted them with
new colors to resemble unrelated types. Likeness exists before
differences develop. It is only necessary to suppose that selection
has preserved similarities when these were of use, while the related
groups have been free to diverge in colors, as in other characters.
Mimicry affords striking proof that selection can lead two or more
species to evolve in parallel or converging directions, but it does not
show that selection is the motive power of evolutionary change.

If species were not already evolving, natural selection would have
no influence over them: there would be no motion to be resisted and
guided toward greater utility. That selection should not be expected
to originate new characters has often been recognized, but it is equally
unreasonable to suppose that existing characters would be modified
by selection if species were normally stationary. The facts of adapta-
tion abundantly corroborate the Darwinian conception of evolution
as a process of gradual change in the characters of species, but they
afford no proof at all that selection can cause adaptive variations or
carry along the evolutionary progress of species.

Strangely enough, the general popularity of the doctrine of evolu-
tion has been due largely to this theory which made selection appear

« Cook, O. F. Weevil-Resisting Adaptations of the Cotton Plant. Bulletin 88,
Bureau of Plant Industry. U. S. Dept. of Agriculture. 1906.
136

12 METHODS AND CAUSES OF EVOLUTION.

as the cause of evolution, the side of Darwinism which has been most
persistently questioned by students of nature. Many intelligent
people think they can see clearly that natural selection, working
through variability, causes species to evolve, but among biologists
this has continued to appear doubtful. The scientific literature of
the last half century abounds in protests declaring that the alleged
cause is inadequate and the reasoning inconsistent with the facts.

Philosophical writers found in natural selection an extremely
attractive hypothesis which made the organic world a result of its
environment. Those who are not deterred by the biological diffi-
culties begin by accepting natural selection as the cause of evolution,
and then argue that the changes are continuous because they are
results of continuous selection. This is to reverse Darwin's method
of reasoning, for he used the facts of gradual evolution to support
the theory of selection. The withdrawal of the theory of selection
does not weaken the doctrine of continuity, but leaves it the more
firmly established.

Darwin's first concern was not to demonstrate the theory of
natural selection, but to implant in the minds of his readers the
general conception of organic evolution. He made free use of any
facts, suggestions, or theories that seemed likely to aid in establish-
ing the basic idea of evolution. Many of his readers could under-
stand the argument for natural selection as a cause of evolution,
while very few could appreciate the mass of detailed evidence prov-
ing the more fundamental fact that the changes of characters are
gradual and continuous. The theory of selection soon overshadowed
the fact of continuity. Darwinism came to mean merely natural
selection. Advocates and opponents alike have allowed themselves
to forget that Darwin's perception of the method of evolution was
independent of his theory of cause.

THE SIGNIFICANCE OF USELESS CHARACTERS.

Useless characters have more bearing than useful characters upon
the question of evolutionary methods and causes. No interpreta-
tion is adequate that does not provide for the evolution of useless
characters. If useless differences can be evolved without selection,
the same may be expected for useful differences. To suppose that
selection is the cause of evolution no longer appears necessary.

The theory of selection requires us to believe in the utility of all
characters and in the development of each character through the
selective origination of a new species. Darwin appreciated the
difficulties of his hypothesis and continued all his life to seek addi-
tional evidence. Many other naturalists have joined in the search,
and a large literature of adaptations has accumulated. New and

136

THE SIGNIFICANCE OF USELESS CHARACTERS. 13

interesting uses of many characters have been discovered, but the
results do not support the assumption of utility for all characters.
The examination of any group of species or genera shows that the
useless differences vastly outnumber those for which any environ-
mental use can be assigned or even imagined.

Possibilities of former utility in characters now useless may be
admitted, but the preponderance of useless differences is far too
great to be thus explained away. The issue is not so plain among
the complex structures and diverse habits of the higher animals and
plants. Conclusive evidence comes from the lower orders, where
many species and genera have the same habits and live under the
same conditions and yet display endless differences of form and
structure. There may still be doubt regarding the present or past
utility of any particular character, but there can be no question that
diversities of characters are out of all proportion to selective diversities
of environment.0

A useful character has an evolutionary advantage, for it can
develop and spread through the species without being hindered by
the natural selection which discriminates against useless and injurious
characters. Any change not forbidden by the limiting factors of
environment and competition can go on by mere evolutionary pro-
pensity. The evolutionary progress of a useful character is no less
spontaneous than that of a useless character. Selection need not
be thought of as carrying along the useful character, but only as
hindering the development of injurious characters. Selection works
through restriction of descent — restriction of evolution.

A species evolving in the direction of longer hair would be able to
resist increasing degrees of cold, but it is not necessary to think that
the cold causes the longer hair to be evolved. The utility of a new
variation depends upon the environment at hand when the variation
is put forth. Thus, utility is a consequence of evolution; to say
that it is the cause of evolution is to reverse the true relations of the
facts. Instead of selection causing the evolution of the useful
character, it is the advance of evolution that enables the higher
standards of selection to be* applied without destroying the species.
The evolutionary progress of the species is not a result of selection,
but an antecedent condition.

To gain clear perceptions of evolutionary factors we may lay
aside this historical assumption that selection is the motive power
of evolution, the active principle, agency, or cause by which evolu-
tion is brought about. Evolution is to be conceived in the mind as
entirely independent and distinct from selection.

o Cook, 0. F. Evolutionary Inferences from the Diplopoda. Proc. Entomological
Society of "Washington, 5: 84. 1902.
136

14 METHODS AND CAUSES OF EVOLUTION.

ABRUPT VARIATIONS NOT INCOMPATIBLE WITH CONTINUOUS

EVOLUTION.

The failure to distinguish adequately between the theory of selec-
tion and the fact of gradual change has led some writers to suppose
that the idea of continuous evolution must be given up when the
theory of selection is rejected. Hence, they return to the older
notion of fixity of species and make use of the idea of evolution only
to the extent of supposing that new species originate from older
species by abrupt transformations of individuals, instead of being
created altogether anew. Sudden changes of characters often
occur, variously known as "sports," "mutations," or "extraor-
dinary births," and these have been taken as new species and as
true examples of the method of evolution.

Abrupt variations in the characters of individuals do not necessarily
conflict with the method of evolution as a continuous change in the
network of descent of the species. A single sheep may jump the
fence, but this does not prove that a new flock is to be formed, for
the others usually follow. The motion of the flock would still be
gradual, no matter how far the individual sheep might jump. An
individual variation, however wide, would not have evolutionary
value and permanence unless the new character were able to spread
to other members of the species, and thus become established in the
network of descent.

A new character is like a new pattern in a fabric. Substitutions
among patterns already designed are not to be confused with the
working out of new patterns. Sudden alternations among old
characters do not compel us to believe that new characters are made
suddenly. The facts of continuous change shown in natural species
enable us to think of new characters as woven by gradual changes
and combinations of older characters through the interweaving of
the lines of descent of the species.

REVERSION, MUTATION, AND MENDELISM.

A narrow-bred uniform variety shows only one set of characters, so
that the characters of a diverse individual mutant appear to be new.
But if we observe a sufficient series of mutations as they spring from
the same uniform variety, a wide range of individual diversity is
found, like that of the wild species from which our cultivated varieties
are derived. It becomes evident that selective narrow breeding has
not destroyed the ancestral diversities, but only suppressed them.
Though showing only one set of characters in its external visible
form, the narrow-bred variety still transmits a long series of latent
differences. a

°> Cook, O. F. Heredity Related to Memory and Instinct. The Monist, 18: 363.
1908.
136

REVERSION, MUTATION, AND MENDELISM. 15

To ascertain that an individual plant or animal has brought a really
new character into expression it would be necessary to have a com-
plete account of its ancestry, to know all the characters which are
being transmitted along the lines of descent. Such knowledge is not
to be obtained from the pedigrees of a few generations of narrow-bred
individuals. The wild species gives the best idea of the normal
diversity of ancestral characters. When the wild species no longer
exists our only resource is to apply the analogies of diversity shown
in other related wild types.

The ranges of diversity shown in wild species are not equal, but
they are generally very much wider than in narrow-bred domesticated
varieties. Thus, aboriginal cotton varieties from Central America,
though often as well isolated and as distinct from each other as natu-
ral species, have an individual diversity corresponding to that of our
whole series of varieties of Upland cottons and even greater. The
mutative variations of our Upland varieties often show some of the
forms of diversity which are regularly present in the Central American
series, but are not represented in any of the Upland varieties now
cultivated in the United States.

The sudden mutative changes in our Upland cottons which fall out-
side the range of diversity of the Central American types are in the
nature of abnormalities, such as the shortening of joints, fasciation of
branches and flowers, and partial or complete absence of lint. Such
a degeneration does not constitute a new character in any truly evo-
lutionary sense. It is merely a failure to bring an old character into
normal expression.

By hybridizing two narrow-bred varieties we can often recall to
expression many of the diverse characters of the aboriginal species —
characters which may have been suppressed for so many generations
that they appear to us as entirely new. Thus, the characters of mu-
tations and of hybrids may have the same origin in the store of
latent diversities which even the most uniform varieties continue to
transmit.

Hybridization serves our purposes best in plants grown by vegeta-
tive propagation, but mutations are preferable in species grown from
seed, because the change of expression is more definitely "fixed"
from the first, without the need of the series of selections used for
fixing the characters of hybrids, sometimes inefTectuallv.

Thus, characters which have been supposed to originate suddenly
in mutative variations can be viewed with better reason as reversions
or as degenerations: often they partake of the nature of both. That
the great majority of mutations are obviously abnormal and inferior
in fertility, even to the degenerating stocks from which they spring,
is a sufficient reason for refusing to regard them as true examples of
a constructive evolutionary progress.

136

16 METHODS AND CAUSES OF EVOLUTION.

Many instances of abrupt variations, or u sports/' were known to
Darwin, but he considered that sudden changes of characters indicate
abnormality. He was too familiar with the evidences of continuous
change in natural species to look with favor upon the idea of evolu-
tion by sudden jumps from one species to another. The following
passages may serve to show his general conclusion regarding the
group of facts now commonly described as mutation and Mendelism :

The ordinary result of a cross is the production of a blended or intermediate form;
but in certain cases some of the offspring take closely after one parent-form, and some
after the other. This is especially apt to occur when the parents differ in characters
which first appeared as sudden variations or monstrosities/'

When two breeds are crossed their characters usually become intimately fused
together; but some characters refuse to blend and are transmitted in an unmodified
state either from both parents or from one.

From this fact, and from the several slight, aggregated differences which distinguish
domestic races and species from one another, not being liable to this peculiar form of
transmission, we may conclude that it is in some way connected with the sudden
appearance of the characters in question.

Some few characters, however, are incapable of fusion, but these are unimportant,
as they are often of a semimonstrous nature and have appeared suddenly.^

Although the sudden changes and the "peculiar form of transmis-
sion" shown in mutative varieties are often connected with abnor-
malities, other forms of alternative expression of characters are
general in nature and must be reckoned as entirely normal. Sexual
differences represent a form of alternative expression of characters
quite as definite as in the so-called "Mendelian hybrids" in which
contrasted parental characters " refuse to blend."

It is a most familiar fact that sexually diverse parents produce
sexually diverse offspring, and in equal numbers when averages are
considered. Darwin and others have shown that the characters
expressed in only one sex are transmitted by both sexes.. There is
much additional evidence that sexual diversity arises from the une-
qual expression of alternative characters, not from unequal transmis-
sion, and the same interpretation can be applied to all the facts of
Mendelism. Considered as a phenomenon of alternative expression,
Mendelism finds a place in a connected series of phenomena as one
of many forms of diversity which may exist among the members of
the same species. The assumptions used to explain Mendelism as a
phenomenon of alternative transmission are too complicated to secure

« Darwin, Charles. The Descent of Man and Selection in Relation to Sex, New
York edition, p. 172. 1896.

& Darwin, Charles. The Variation of Animals and Plants Under Domestication,
New York edition, pp. 69, 72, and 157. 1897.
136

ADJUSTMENTS OF CHARACTERS TO EXTERNAL CONDITIONS. 17

credence when the possibility of a simpler interpretation is once
appreciated."

ADJUSTMENTS OF CHARACTERS TO EXTERNAL CONDITIONS.

Another form of alternating expression is found in the changes of
characters, often very abrupt, by winch organisms are able to adjust
themselves to different conditions of existence. Before the possi-
bilities of alternative expression were realized, these environmental
changes were supposed to represent a power of the environment to
form new characters in organisms, a supposition no longer required.
The alternative characters of the same species are like the two or
more forms winch the same chemical substance may take in crystal-
lization.

Environmental changes, like mutations, are to be considered as
alternative expressions of characters already developed; the}' do not
represent the progressive changes by which new characters are origi-
nated. One alternative character need not be considered more
normal than another, seeing that freedom of substitution among
many alternative characters is a normal condition in species.

The finely divided leaves winch some amphibious plants put forth
when growing in water are as truly characters of the species as the
entire leaves shown by plants growing on dry soil. Even the same
individual when transplanted from one habitat to the other changes
the character of its leaves. The environment merely calls one
alternative or another into expression, just as the pupils of our eyes
open in the darkness and contract in the light. It is not necessary
to suppose that characters which are changed with changes of
environment are products of environment or that such changes add
anything to the evolutionary progress of the species.

How the external influence works upon the internal adjustments
so as to call forth a substitution of one alternative character for
another is still unknown, but there is no lack of evidence that such
relations exist. To take the adjustments of such characters as
examples of evolution is another way of confusing the result with
the cause.

a Cook, O. F. Mendelism and Other Methods of Descent. Proc. Washington
Academy of Sciences. 9: 189. 1907.

A Mendelian method of inheritance, by the alternative transmission of contrasted
characters in pure germ cells, could have no direct relation to evolution. It has been
shown by Mr. G. H. Hardy (Science, n. s., 28: 49) that a character inherited in the
Mendelian fashion would not tend to spread through a species. The proportion in
which such a character was represented in a group would not increase with successive
generations, but would preserve a constant average after the second generation, if
there were no selection or other interference.
52641— Bull. 136—08 3

18 METHODS AND CAUSES OF EVOLUTION".

The power of external conditions to influence expression is not
limited to the substitution of one environmental character for
another. The phenomena of acclimatization, no less than those of
mutation, require to be brought into relation with the facts of normal
diversity. Transfers of cotton and other plants to new conditions
are often followed by striking changes of characters far beyond the
range of ordinary environmental accommodations. Sometimes all
the individuals of the same planting change in the same way,° and
sometimes a wonderful individual diversity is aroused.

Instead of rare mutations of a few individuals of an otherwise
uniform variety, there is a promiscuous mutation of all the indi-
viduals at once, each in a different direction. To believe that the
same environment makes all these diverse characters anew is obviously
unreasonable. But when the normal diversity of the members of
species is recognized, and the fact that such diversities are trans-
mitted, it becomes possible to understand that an unwonted condition
may recall many ancestral diversities into expression by disturbing
the internal relations which govern the expressions of the characters.
The result is the same as though the plants were making experiments
with all the various degrees and combinations of their ancestral
characters to see which might be best adapted to the strange sur-
roundings. Thus, the planting of a variety under new conditions
may serve the same purpose as hybridization in recalling diverse
characters to expression.

CHARACTERS NOT DIRECTLY MOLDED BY ENVIRONMENT.

In the century before Darwin man}T students of nature interpreted
facts of adaptation as indications of evolution. But instead of
api^reciating evolution as progressive change in species, most of the
pre-Darwinian evolutionists dwelt upon the idea that the organisms
were directly molded by their environments into adaptive conformity.
They assumed that all of the diversities found among members of the
same species are due to inequalities of environment, and pointed to
the results of environmental limitations and adjustments as true
examples of evolution. This idea of evolution as a direct result of
environmental influences has continued to find supporters. Many
attempts have been made to substitute the teaching of Lamarck for
that of Darwin, or to supplement natural selection by alleged environ-
mental factors.

Darwin's superior insight enabled him to take an important step
in advance of his predecessors. He agreed with them that adapt a-

« Dr. C. A. White, of Washington, D. C, has described such changes in the tomato
and calls them aggregate mutations. The Mutations of Lycopersicum, Popular
Science Monthly, 47: 151.
136

DIRECTIONS OF EVOLUTION NOT FIXED. 19

tion shows relation to the environment, but judged that the relation
is not direct. He recognized also that evolution pertains to species,
not merely to organisms in mass, and he found in natural selection
the way by which evolutionary changes in species are influenced
indirectly by their environments. Those who have rejected selection
and returned to Lamarck's theory of adaptive evolution through a
direct molding of characters by the environment have not made
good their case. To find that selection does not produce new charac-
ters does not prove that they arc produced by the environment. It
has not been shown that the environment has any active constructive
influence in descent, either in individual organisms or in species at
large.

The environment often appears to change the characters of organ-
isms, but there is no indication that these changes represent j)ro-
gressive evolution. All forms of environmental influence thus far
discovered can be interpreted in one or two ways, either as limitations
of existing characters or as substitutions. Mutilations, diseases, and
similar abnormalities imposed by the environment often destroy
organisms or weaken their progeny, but are not known to contribute
anything to evolutionary progress. The substitution of one character
for another as following an environmental change or interference
can not be accepted as the formation of a new character, now that
organisms are known to inherit and transmit characters in latent
form and to bring one or another into expression with or without
relation to external conditions.

DIRECTIONS OF EVOLUTION NOT FIXED.

Other writers have turned away from the external environment to
seek the causes of evolution in internal substances or mechanisms of
heredity. From minutely detailed studies of species of Hieracium,
Naegeli gained a conception of evolutionary motion as tending to carry
species in definite directions, and incorporated this idea into a theory
of the structure of protoplasm. He refused to accept Darwin's
opinion that species are variable indefinitely in all directions, but
declared that they vary in single definite directions.

The facts are again matters of observation. The members of a
normal sexual species have many kinds of diversity, showing that
many evolutionary changes can go on simultaneously in the same
species. It has been found that Hieracium is parthenogenetic; that
is. the flowers frequently develop seed without fertilization. This
explains the unusual constancy of characters, for parthenogenesis is
like vegetative propagation in being exempt from the indiscriminate
diversity shown in normally sexual groups. The variations of par-
thenogenetic plants are not shared among the different lines of

13G

20 METHODS AND CAUSES OF EVOLUTION.

descent, but follow separate courses and gradually diverge in definite
directions. Nevertheless, these more definite changes are not to be
accepted as examples of normal evolution, for they lead toward
sterility and extinction if sexual reproduction does not intervene.

Xaegeli's conception of internal causes of evolution through proto-
plasmic mechanisms of heredity has been borrowed and combined
with many other theories, sometimes even with the idea of evolution
by environmental causation. Weismann denied the inheritance of
characters imposed on the bodies of organisms by the environment,
but conjectured that heritable characters might be imposed by the
environment directly upon the germ cells. This made it possible to
think of evolution as carried forward by environmentally directed
variations, so that adaptation would be explained without natural
selection. A return toward Lamarck's doctrine became possible for
those who wished to reject natural selection because it did not explain
variation.

That natural selection is not responsible for all that has been
ascribed to it does not justify a neglect to appreciate its true power
of leading the evolutions of species toward adaptation. If the direc-
tions of evolution were fixed in advance or controlled by internal
mechanisms, the preservation of the species would depend altogether
upon coincidence. Selection would have no power to influence the
direction of evolution.

Adaptations are too numerous and too finely specialized to be
explained as mere coincidences; they are joint products of evolution
and selection. Though not adequate to prove that the evolutionary
motions of species are caused by the selective agency of the environ-
ment, the testimony of adaptation is ample to show that the evolu-
tionary directions of species are not definitely determined by internal
agencies or mechanisms.

DIVERSITY A FACTOR IN NATURAL EVOLUTION.

The problem of evolution always leads back to the discussion of
variation, because this term is used to cover all the differences found
among the members of a species. Somewhere among the internal
diversities of the species are characters which at some future time
will become the common property of the whole group and serve
to distinguish it from other species in which these characters do
not develop. The problem is to fix upon the truly evolutionary
differences.

Several kinds of differences have been used as examples of evolution,
as shown in the previous pages. Some writers have looked upon
the fluctuating individual differences of narrow-bred varieties as the
means of evolutionary progress, while others have considered that

136

DIVERSITY A FACTOR IN NATURAL EVOLUTION. 21

new species are formed by sudden mutations. A third doctrine
treats of environmental variations as representing evolution; a
fourth would illustrate "evolution by the smaller and more definite
changes found hi parthenogenetic and self-fertilized varieties.

There is still another type of diversity which seems to have more
significance in evolution than any of the preceding. This is the nor-
mal diversity of natural broad-bred species, the differences found in
species before descent is restricted — differences manifest even when
the conditions are so uniform as to call forth no environmental
variation.

To distinguish this diversity of normal descent from fluctuating
and environmental variations, the word "heterism" has been used.
Heterism can be eliminated by restricting descent to narrow groups
or to individual lines, and can thus be distinguished from the fluctu-
ating differences that are always to be detected, even in the most
uniform of our narrow-bred domesticated varieties.

Fluctuating differences have been subjected to careful statistical
study and have been found to conform to mathematical curves of
chance distribution, like curves made from the records of a marksman
shooting at a target. This conception of chance deviation from a
fixed standard of heredity evidently does not apply to sexuality and
other normal forms of diversity among the members of species.
Fluctuations continue to be shown alike on all the lines of descent
into which a species may be separated, but heteristic differences are
separately expressed on different lines of descent when a species is
subdivided by selective narrow breeding. The fact that all charac-
ters fluctuate leaves the fluctuating variations without special sig-
nificance in evolution, though they are of interest hi showing that
uniformity is an artificial ideal not to be attained in practice, even
under experimental conditions.

Sexual differences represent a very highly specialized form of
heterism, but all gradations of diversity are found between sexual
differences and mere environmental variations and fluctuations.
Though evolution is a gradual process it is not necessary to suppose
that advance takes place only by the selective accumulation of
infinitesimal fluctuating differences. Xormal diversity of descent
provides for free substitutions and alternations in the expression of a
much wider range of characters.

In diversities of descent rather than environmental or fluctuating
variations are to be found causes of evolution, factors of change in
the characters of species. Heterism does not serve the purposes
of evolution alone. It has a physiological value in reproduction.
High-grade organisms are the offspring of diverse parents. The liigher

136

22 METHODS AND CAUSES OF EVOLUTION.

we go in the scale of life the more numerous and complex are the pro-
visions for securing this diversity of parentage. Heterism thus con-
tributes to evolution in two ways, by maintaining the vitality of the
species and by furnishing the characters by which the species can
change.

BROAD BREEDING A FACTOR IN NATURAL EVOLUTION.

Another cause of evolution is free interbreeding among the mem-
bers of a species to maintain a broad network of descent. A species
is an evolutionary unit because its members travel together along
the path of development. Unless new characters were distributed
through the species by broad interbreeding there would be no such
evolution as that shown in nature. Characters would remain the
exclusive property of single lines of descent, as in varieties of. plants
propagated from cuttings.

Complex structures are not developed without binding the species
into a broad, continuous network. Among all the higher types of
plants and animals the processes of descent are normally sexual and
thus insure the weaving of a network of descent. Constructive
evolution ceases whenever the network is cut into narrow strands
or individual lines of descent; changes continue to take place, but
leading toward degeneration.

The theory that natural selection is the cause of evolution led
Darwin and many later writers to regard free interbreeding among
the members of a species as an obstacle or hindrance to evolution.
Broad breeding would hinder evolution if progress were dependent
upon the selective restriction of descent, as in domesticated varieties.
Broad breeding does not conflict with evolutionary changes in the
network of descent of a natural species; such changes are facilitated
by the distribution and combination of the characters of the component
organisms.

The evolutionary prosperity of a normal species means large num-
bers, wide distribution, and abundant individual diversity. It con-
trasts at every point with the results of narrow selection, and warns
us that the systems by which we breed our domestic animals and plants
are contrary to the conditions of normal evolution.

Evolution is still to be denned as a process of change in the charac-
ters of species, but several kinds of changes in the characters of species
have to be distinguished from the genuinely evolutionary changes.
The paring away of species by selection, the dividing of species by
isolation, the adjustments of characters to external conditions, and
the suppression of characters by narrow breeding are processes as

136

NARROW BREEDING IX DOMESTICATED VARIETIES. 23

different from evolution as they are from each other. They can all
be studied from the evolutionary standpoint, but they do not repre-
sent evolution itself. They are not the progressive, constructive
changes through which the world of organisms has come into existence.
They show some of the consequences of evolution, but do not reveal its
true nature or its motive power.

NARROW BREEDING A FACTOR IN DOMESTICATED VARIETIES.

Changes occur under conditions of narrow breeding which would
not take place if descent remained unrestricted, but these changes do
not represent a true evolution. Xarrow breeding induces degenera-
tive variations, but is at the same time our readiest means of making
such variations serve our economic purposes. Any type is degenerate
which is no longer on the pathway of constructive, progressive
evolution.

The practical importance of our agricultural improvements of plants
and animals is not lessened by the fact that they are not evolutionary
improvements, but include many kinds and stages of degeneration.
Xor is it any less a part of evolutionary science to study these valuable
degenerations and ascertain the best methods of producing them.
Breeders of specialized varieties have the problem of increasing desir-
able degenerations as much as possible, with as little as possible of
injurious degeneration, such as weakness or sterility.

It is here that the belief in selection as the cause of evolution brings
confusion and discord into the art of breeding. Darwin concluded,
after weighing much evidence, that persistent narrow breeding always
results in degeneration, because it violates a general law of nature.
He makes the following statement:

These two great classes of facts, namely, the good derived from crossing and the evil
from close interbreeding, with the consideration of the innumerable adaptations
throughout nature for compelling, or favouring, or at least permitting, the occasional
union of distinct individuals, taken together, lead to the conclusion that it is a law
of nature that organic beings shall not fertilize themselves for perpetuity.0

As long as selection is represented as the chief agency of evolution,
the teaching of science will appear to favor the indiscriminate appli-
cation of narrow breeding instead of warning against its dangers and
limitations. The idea that selective restriction of descent is the cause
of evolution is not in accord with a law of nature against close breed-
ing. To hold that organic progress depends on selection and isolation
is to place evolution in conflict with itself. It amounts to saying that

« Darwin. Charles. Variation of Animals and Plants Under Domestication. New
York edition. 2: 159. 1897.

24 METHODS AND CAUSES OF EVOLUTION.

evolution is everywhere working to prevent evolution, since every
adaptation to facilitate interbreeding is at the same time an obstacle
to the restriction of descent required by the theory that selection
causes evolution.

When diversity of descent and broad breeding are recognized as the
conditions of normal reproduction and evolutionary progress, this
contradiction disappears and the meaning of the law against perpet-
ual narrow breeding becomes plain. It no longer appears as a merely
arbitrary regulation applied to some species and not to others, but
reveals essential requirements of organic existence and evolutionary
progress. Broad networks of descent are necessary to maintain
efficiency of reproduction and organic vigor in the species. To
destroy the network of descent removes the support of the organic
structured

Degeneration comes to narrow-bred organisms as surely as clocks
run down when they are no longer wound. Some clocks run only a
day; others may not need rewinding for a week, a month, or even a
year. Species differ widely in their ability to live by vegetative
propagation, parthenogenesis, self-fertilization, or other forms of
descent in narrow lines. Nonsexual methods of reproduction are
often very useful to supplement normal sexuality, but exclusive
dependence on restricted descent evidently brings extinction to wild
species as well as to domesticated varieties. Nonsexual methods are
not found in consistent use in whole species, genera, and families
of plants, but in a few narrowly restricted forms closely related to
normally sexual types.

Deterioration is slower with vegetative propagation and other sim-
ilar processes of descent in individual lines than with less discriminate
forms of narrow breeding, but all forms of restricted descent lead
ultimately to infertility and degeneration. Line breeding is superior
to narrow breeding, but not superior to broad breeding. The elabo-
rate contrivances of plants to secure cross-pollination show at once
how difficult and precarious are the processes of sexual reproduction,
and at the same time how indispensable.

MUTUAL RELATIONS OF PRINCIPAL FACTORS.

Diversity and broad breeding are not merely causes of evolution,
but are necessary to normal processes of descent among all higher
types of organisms. Evolution not only results from diversity and
interbreeding, but at the same time brings increased diversity and
improved adaptations for still broader breeding.

a Cook, O. F., and Swingle, W. T. Evolution of Cellular Structures. Bulletin 81,
Bureau of Plant Industry, U. S. Dept. of Agriculture.. 1905.
136

MUTUAL RELATIONS OF PRINCIPAL FACTORS. 25

The facts are concrete and their relations obvious. The members
of each species are diverse (heterism) . They freely interbreed among
themselves (symbasis). They produce successive generations (de-
scent) . Later generations differ from the earlier (evolution) . With-
out diversity, interbreeding has no significance, and descent is not
accompanied by evolution. Heterism, symbasis, descent, and evolu-
tion form a continuous series, a complete chain of phenomena. For
some purposes the four processes may be described as distinct, but
physiologically they are mere phases or aspects of organic existence,
each at once requiring and assisting the others.

Thus, the constitutions of species and the methods by which they
are propagated show that they are in motion. We no longer seek to
prove that species evolve; we see that they do and understand why
they must continue to do so as long as they remain groups of
normally diverse, broadly interbreeding organisms.

Diversity and interbreeding do not appear as causes of evolution
in the same way that natural selection was supposed to be the cause.
They are rather the general conditions or means of evolution. An
actuating cause of evolution, in the narrow sense of an agency which
takes hold of normally stationary species and changes their charac-
ters, it is not necessary to seek, since species do not tend to remain
stationary in nature, nor do domesticated varieties continue constant
even when breeders strive for uniformity.

The search for causes never ends. There is always something
unknown behind the facts we are able to understand. The most that
we can hope to do is to maintain the right direction in our inquiries »
To have known that the causes are in the species and not in the
environment would have saved many wanderings.

The only relations thus far established between evolution and envi-
ronment are negative ; the environment often hinders or limits evolu-
tion, but has not been found to cause any constructive change. On
the other hand, normal diversity and free interbreeding have direct
relations with evolution, for they represent essential conditions of the
existence of species. Without them our species would not have the
characters they now possess.

The true nature of evolution as a progressive change in the network
of descent of a species, accomplished through diversity and inter-
breeding, was understood bv^ Goethe a century ago. The statement
of his evolutionary conclusions in metrical form may have caused
them to be overlooked as scientific discoveries, but it is evident that
he perceived the essential relations more clearly than the later writers
who have ascribed evolution to environmental agencies or to restric-
tion of descent. Goethe was an active investigator of natural phe-

136

26 METHODS AND CAUSES OF EVOLUTION.

nomena and represents himself as having studied and solved the
problem of evolution, as shown in such passages as these:

Years ago it was a pleasure
Of the eager spirit's days,
Searching for the course and measure
Of creative Nature's ways.
Lo, a unity eternal
Stands in manifold revealed;
Large or small kinds, great or little,
Point one law in every field.
Always changing, yet enduring,
Far and near, by land or main,
Forming now, and now transforming —
Wondrous goal that I attain !

Radiant this truth has shown,
Vital more than any:
Nothing living lives alone,
Always there are many.

Behold we now the primal art,
The ancient Weaver-woman's part;
One tread uplifts a thousand lines
While back and forth the shuttles go,
And meeting strands together flow,
Each stroke a thousand unions twines;
Her web by no mere chance is made,
Forever has her warp been laid,
The Master-workman's one behoof
Content for aye to throw the woof.

EVOLUTION PROBLEM DISTINCT FROM DESCENT PROBLEM.

Many investigators have thought to reach the problem of evolution
through the problem of descent. By learning how characters pass
from one generation to another it was hoped to learn also how charac-
ters are changed during series of generations. The problem of
descent was supposed to include the problem of evolution.

It now appears that evolution may throw light upon . descent
rather than descent upon evolution. Descent is carried on through
individual organisms and conjugations of germ cells, but evolution is
conducted through species. Until we recognize diversity and inter-
breeding as of physiological value, the elaborate complexities of
descent by sexual reproduction appear wasteful and meaningless.

A species is a mutual physiological organization for the improve-
ment of descent, instead of being a merely arbitrary assemblage of
organisms fenced in by hostile environments and compelled to
compete with each other in a struggle for existence. Adaptations
for strengthening the network of descent by diversity and inter-
breeding exceed environmental adaptations in number and complex-
ity. Adaptations for interbreeding are a permanent gain for the

136

EVOLUTION PROBLEM DISTINCT FROM DESCENT PROBLEM. 27

species, while environmental adaptations may have only temporary
advantage. The more perfect the adaptation to special conditions
the smaller are the chances that the species can survive a change of
environment .

Geology repeats in many different groups the lesson of the earlier
extinction of the more specialized types and the preservation and
continued progress of the less specialized, those able to maintain
themselves under a wide range of conditions. The new types which
appear to have arisen suddenly in remote geologic ages were already
widely distributed. Instead of being confined to limited areas where
interbreeding would be restricted, the species must have contained
vast numbers of individuals, more than in the later times when these
same groups become subdivided in their turn into many genera and
species.

The higher we go in the scale of existence the greater is the need
that we study cells and organisms in their social or collective relations.
In the lower groups reproduction can be accomplished by simple
splitting of single cells, but in the highest groups it requires the
cooperation of two extremely complex organisms, each composed of
millions of cells. In man and in many other social species, numerous
individuals contribute to the nurture of the young, to bring them to
fully developed maturity, and thus complete the process of descent.
To maintain evolutionary progress a still wider cooperation is required,
a huge assemblage of interbreeding individuals which constitutes a
species.

Gait on considered the system of double parentage as necessary to
sustained efficiency of reproduction among the higher plants and ani-
mals, but this conception falls short of the full indications of evolu-
tionary facts. Xot merely double parentage, but a highly multiple
ancestry is required to maintain a succession of complex organisms.
The species rather than the individual is the true unit of organic
existence and of evolutionary progress. Evolution may prove to be
as essential a factor in descent as descent is in evolution. There is a
general evolutionary physiology of species and varieties, as well as a
special physiology of descent shown in the reproduction of individual
organisms.

CeUs may be thought of as containing mechanisms of descent, but
species are mechanisms of evolution. The problem of evolution is
much more accessible to our observation than the problem of descent.
and comes first on the normal route of inquiry from the known to the
unknown. Though closely allied, the two problems lie on different
planes and can be studied separately. In bringing them into relation
we should be careful not to bring them into confusion. We do not
need to seek in the cells for the functions performed by the species.

136

28 METHODS AND CAUSES OF EVOLUTION.

To defer the recognition of evolutionary facts because our knowl-
edge of descent remains incomplete would be much like refusing to
walk or to practice manual arts until we should have a complete
explanation of the mechanism of contraction in the cells of muscles,
on which our bodily movements depend. It may be long before we
can gain any complete understanding of the physiology of descent,
of the activities and relations of cells in organisms, but this need not
prevent our gaining knowledge of the physiology of evolution, the
relations of lines of descent in species.

EVOLUTION INHERENT IN SPECIES.

A normally constituted species can carry forward changes of
characters without environmental or selective causation. Under
natural conditions of free interbreeding, new variations and combi-
nations of characters are to be thought of as tending to spread and to
transform the species or to accentuate the normal and desirable
diversity of descent inside the species. The evolutionary develop-
ment of a species is as spontaneous as the growth of the individual
plants and animals of which the species is composed. External
conditions influence the evolution of species, just as they influence
the development of individual organisms, but both forms of growth
are manifestations of life, not effects of inanimate surroundings.

Environmental utility influences the rapidity of extension and
uniformity of expression of characters, but useless characters can
also spread and serve the species physiologically by ministering to
a healthful diversity of descent. The state of free interbreeding
found in natural species is most favorable for evolutionary progress;
no rare or exceptional conditions have to be imagined.

Evolution is analogous to history. The importance of events is
determined by the conditions and the sequences in which they occur.
Species are not carried along by their environments, nor compelled by
internal clockworks to change at particular times or to go in definite
directions. Nevertheless, the timely putting forth of a new char-
acter may have a profound influence upon the subsequent history of a
species. Variation is genetic, for species tend to evolve, but it is not
definitely directed or determined (orthogenetic) . A character once
formed tends to further development, and thus finds the range of
greatest utility or of most harmonious combination.

Though evolution is often described as a doctrine of transforma-
tion, the earlier theories really taught only the selective substitution
of one species or variety for another; they did not provide for the
transformation of species. They represent evolution as possible oidy
when descent is restricted by selection or other means of isolation, and
require at every step highly improbable coincidences between varia-

136

CONCLUSIONS. 29

tion and isolation. An advantageous new character, if not pre-
potent, would be preserved and extended only when opportunely sepa-
rated from the unmodified type, but a prepotent variation, whether
useful or not, can gradually pervade and transform the species. There
is no need that it be isolated from the parental stock or that it wage
a selective warfare of extermination against its nearest relatives.

CONCLUSIONS.

Relation of evolution to species. — The evolution of the higher types of
plants and animals is an essentially complex process, not to be accom-
plished without the association of organisms in species and not to be
understood without recognizing that the association of organisms into
species has a concrete physiological significance. Evolution is change
in species. Unless we learn how species are constituted we do not
become acquainted with the results of evolution or with the conditions
under which it takes place.

The constitution of species. — Species in a normal condition of evolu-
tionary progress are large groups of diverse, freely interbreeding organ-
isms. The joining of the lines of descent of each specific assemblage
into a huge, continuous network provides for endless individual
diversity by free combinations and substitutions among gradually
changing characters. New characters must combine harmoniously
with old characters and must spread through the network of descent
of the species.

The method of evolution. — A comparison of the diversities found
among members of the same species with the diversities of separate
species shows that evolutionary changes in the characters of species
are completely gradual and continuous. Species are found in nature
in all stages of division and divergence. Evolution does not depend
on occasional originations of new species, nor upon abrupt changes of
characters in individual organisms, but it is shown in gradual changes
of the characters of the organisms of which species are composed.
Species may be differentiated without any new characters being
added, or new characters may arise in a species though no subdivision
takes place.

Causes of evolution. — The differentiation of species and the attain-
ment of new characters and functions are results of evolution made
possible by the facts that the members of species are normally diverse,
and that there is free interbreeding in the networks of descent. Evo-
lution is not caused by the struggle for existence, nor limited to char-
acters of environmental fitness. Harmless and even harmful char-
acters may be acquired by species in the same way as beneficial
adaptations.

13G

30 METHODS AND CAUSES OF EVOLUTION.

Expression of characters distinct from transmission. — The number of
characters that organisms inherit from their ancestors and transmit to
their descendants is much greater than can be brought to visible
expression in any individual. Keversions show that characters may-
be transmitted in latent form for many generations and then suddenly
reappear. On the other hand, characters long expressed with regu-
larity may suddenly disappear, as often occurs in mutations. Many
of the differences which have been used as examples of evolution rep-
resent mere substitutions and alternations among characters already
evolved by the species, instead of evolutionary progress toward new
characters.

Influence of external conditions on expression of characters . — The con-
ditions of existence often influence the expression of characters in
individual organisms, calling some characters into expression and
leaving others latent. In different environments the same species, or
even the same individual organism, may put forth different characters.

By varying expressions of their characters, organisms are able to
adjust themselves to wider ranges of environmental conditions, but
this versatility does not indicate that the environment is responsible
for the formation of new characters or for the evolutionary progress of
species. Doctrines of environmental causation of evolution tend to
deceive us with the false hope of making over inferior breeds by
environmental manipulation, instead of leading us to appreciate the
primary importance of normal descent.

Influence of selection on expression of characters. — Changes induced
by the narrow artificial selection of domesticated plants and animals *•
are not true examples or equivalents of natural evolution. The
removal from a group of all the individuals which show a certain
character tends to suppress this character. The selective suppression
of characters renders the members of a group more uniform, but it
does not make new characters. Selection works by eliminating differ-
ences ; evolution cOmes by introducing them. Selection is an analytic
process, while evolution is synthetic. Evolution can be guided
through selection, but the guidance must not be too narrow or condi-
tions essential to the continued existence of organisms are destroyed.
The teaching that evolution is caused by the selective action of the
environment and hindered by interbreeding among the members of
the species reverses the facts of nature. The environment permits
some changes and forbids others, and can thus limit the directions in
which the changes proceed, but the changes are put forth by the
species, instead of being actuated by the environment.

Status of domesticated varieties. — Uniform domesticated varieties are
small groups in which the normal diversity has been suppressed by
selection. The greater uniformity is secured at the expense of normal

136

CONCLUSIONS. 31

diversity and free interbreeding, the primary conditions of constructive
natural evolution.

The idea that uniformity is to be sought for its own sake or that
"pure-bred" uniform types are essentially superior to those showing
greater individual diversity finds no warrant in the general facts of
evolution. It is misleading and dangerous to advise that all domes-
ticated types of plants be made uniform by breeding from single
individuals. To the extent that it is necessary to obtain uniform com-
merical products, restriction of descent to individual lines may be
justified, but not on the ground that it represents a normal or truly
evolutionary condition.

Evolution must continue to furnish the materials for breeding. It
is vain to expect that the work of breeding in any economic group can
be finished by securing narrow strains, however superior. Narrowly
selected breeds must continue in the future, as in the past, to follow
in a gradual succession, the older and weaker being discarded for
the newer and more vigorous, more recently derived from broadly
interbreeding stocks.

136

INDEX.

Page.

Acclimatization in relation to normal diversities 18

Adaptations, discussion 18-20

for interbreeding contrasted -with environmental adaptations 26-27

Breeding, broad , a factor in natural evolution 22-23

condition of normal reproduction 24

line, superior to narrow breeding 24

narrow, a factor in domesticated varieties 23-24. 31

approximation in nature 10

danger 22, 23

inferior to line breeding 24

results in degeneration 24

Cells, the mechanism of descent 27

Characters, adjustments to external conditions 17-18, 30

alternation, distinct from new characters 14

alternative expression an explanation of Mendelism 16

ancestral, latent in narrow-bred varieties 14-15

changes, continuous, means by which evolution proceeds 8-9

kinds 22

disappearance 30

diversity, range in wild species compared with domesticated varie-
ties 15

expression, distinct from transmission 30

influence of external conditions 30

selection 30

inheritance and transmission 19

Mendelian, failure to spread 17

not directly molded by environment 18-19

useless, significance 12-13

Conclusions of bulletin 29

Cotton, acclimatization 18

Central American, diversity 15

variation, mutative 15

Darwin familiar with facts of Mendelism 16

statement on narrow breeding 23

summary of results of study of evolution 9

Degeneration, result of narrow breeding 23. 24

Descent, diversities, a factor in evolution 21

problem distinct from evolution problem 26-28

relation to symbasis, heterism, and evolution 25

Diversities, latent, transmission 15

Diversity a factor in natural evolution 20-22. 24. 25

range in wild and cultivated forms 15

suppression by narrow breeding 14

136 33

34 METHODS AND CAUSES OF EVOLUTION.

Page.

Environment, adjustments of characters to external conditions 17-18

changes, as alternative expression of characters 17

influence on expression of characters 18-1 9, 30

relation to evolution 25

Evolution analogous to history 28

causes ,, 25, 29

contrasted with species formation - 8

continuous, not incompatible with abrupt variations 14

defined 7, 22

directions not fixed 19-20

distinctions between methods and causes 8

early theories 9, 10, 18, 25, 28

inherent in species 28-29

method 29

natural, broad breeding a factor 22-23

diversity a factor 20-22

not the result of environmental influences 18

natural selection 10-12

objections to mechanical theories 8

problem distinct from descent problem 2(3-28

relation to agricultural development 7

heterism, sym basis, and descent 25

species 29

relations, mutual, of principal factors 24-26

spontaneous nature 8, 28

through continuous changes of characters 8-9

Factors of evolution, principal, mutual relations 24-26

Galton, views on double parentage 27

Goethe, view of evolution 25-26

Hardy, G. H., note on Mendelism 17

Heterism defined 21

physiological value 21

relation to symbasis, descent, and evolution 25

Hieracium, parthenogenesis and constancy of characters 19

Hybridization, value in vegetative propagation 15

Hybrids, comparison with mutations 15

Improvement, agricultural, not evolutionary improvement. 23

Interbreeding, means of evolution 25

wide, of early types. 27

Introduction to bulletin 7-8

Lamarck's theory of evolution 18

Mendelism considered as alternative expression 16

facts known to Darwin 16

reversion, and mutation, discussion 14-17

Mimicry, discussion 11

Mutation, reversion, and Mendelism, discussion 14-17

Mutations, aggregate 18

contrasted with hybrids 15

not evolutionary progress 15

value in propagation by seed 15

Naegeli's theory of evolution 19

136

INDEX. 35

Page.

Parentage, double, views of Gallon 27

Parthenogenesis compared with vegetative propagation 19

Fr< tpagation, vegetative, compared with parthenogenesis 19

rate of deterioration. 24

Reproduction, method in simple and complex organisms 27

sexual, precarious nature 24

Reversion, mutation, and Mendelism, discussion 14-17

Selection and change of characters "11

elimination from evolutionary factors 12

influence on expression of characters 30

mark of agricultural and not evolutionary progress 11

natural, contrasted with artificial 9-10

importance 20

not the cause of evolution 10-12

restrictive, not in accord with law of nature 23

Sex, differences, a form of alternative expression 16

heterism 21

Species an organization for the improvement of descent 26

constitution 29

definition 7

evolution inherent 28-29

relation of evolution 29

Symbasis, relation to heterism. descent, and evolution 25

Variations, abrupt, not incompatible with continuous evolution 14

different kinds 20

direction not determined 28

fluctuating, significance 21

Varieties, domesticated, danger from uniformity 31

narrow breeding a factor 23-24

status • 30-31

Weismann on inheritance of characters : 20

13G

o

»*AO

> ^> 3:

3 3 jm

^P^> ^>3 3

-3S )»3 --.37

33

33 ^
^»

3 33 ^
3> 3>

3 '5 33^

>> >>^
3 3 > > >^r

> > 33

^#

3 333 3 33

3 > 3 3 )r>1

33 3 333 3

3 3 3333 3
33 X333 3

33 ^33 3^>

33 33 3 >
33 )3 > 3;

€ >?> «
3> 333 >^>

33 ^>o >3

3> 333 >3

^> »> 3D

3> >3 3 > >

> 33 )3> >3

> 33 OJ 3>>

3 % >> V3
• \ > 3 3 >3

) > 3> >5

3 > >y 3 >

> 3 > >3 3>3

N3 3 >> > j>

> > .>
\V* 3> 3 >-

! 3>

\ 3>:>

> a

£>^*>
a3> 3

>3Z> 3

Mi

3>3> 3

> 3 3

3 3 >

3.^ T

3 3 ~
>3

>, 33

> 3 3
* ^P
>> TO
}3
3J>

) 3 >

> 3 >:

^ 3 "•'

» OJ»3 3

* 3JKO 3

> 3>

^ > >3
3 >3

3^3
3^>

y >

)3

~> 3
3,3
3 >
>3 > 3 3
33 > >3
33 3> 3 3
►33 3> 3 3
>"33 3 »3

>3> > >3

> H> > >_>

> 3> > 3 >

> > 3 >J>

3 3 > >J>

■-> 3 >

■ » ^

>

) >>

• ^3

> »
3 »

3^ 3 3
>^ 3 3
>^> 3 3
3>3 3 3

3 3 3

^> ^>
>3 33

^> > 33
> >>
»3 3^
J» > 3?>
»3 "»
»3
>3 > ^

> ■> 3

3> > >

33 >

3> 7-

>> :

33

>3

> 3
3 >

> >3

3>
>3

33

>3 :

3 >