Institution and
Washington, D.C.
Ltd., New Delh

fan

Pvt

Co.

ish

Published for |
The National Science Foundat

by Amerind Publ

>
=

@ :
: :
>

> №

<

Phenotypic variation is the raw mate-
rial for natural selection, yet a century
after Darwin it is an almost unknown
subject. To a considerabie extent
this may be true because even to see
the significance of studying variation
within populations (rather than taking
it as given) requires an appreciation
of several ordinarily separate discip-
lines : morphology, ecology, genetics,
developmental biology, and to some
extent systematics and paleontology.
Other disciplines are also sometimes
important, but these are central.

Darwin wrote more on variation than
on any other subject. Since his time,
there has been almost no one of
intellectual power in western coun-
tries who has known all the central
subjects adequately to make much
progress. In the USSR, however,
Schmalhausen was such a person.
Some of his insights, such as the dis-
tinction between r and K selection,

(Contd. on back flap)

fabs [efpoom—
| лам |

he } a у
ie Can

ies

м

TT 71-58007

AKADEMIYA Nauk SSSR

Institute Morfologii Zhivotnykh im. A. N. Severtsova

ACADEMY OF SCIENCES, USSR

A. N. Severtsov Institute of Morphology of Animals

VARIABILITY OF

MAMMALS

(Izmenchivost’? Mlekopitayushchikh)
A. V. YABLOKOV

Revised by the author for this edition
Scientific editor of translation : L. Van Valen

Nauka Publishers
Moscow, 1966

Translated from Russian

Published for the Smithsonian Institution and the

National Science Foundation, Washington, D.C.

by Amerind Publishing Co. Pvt. Ltd., New Delhi
1974

© 1974 Amerind Publishing Co. Put. Ltd., New Delhi

Translated and Published for the Smithsonian Institution,
pursuant to an agreement with

The National Science Foundation, Washington, D. C.

by Amerind Publishing Co. Put. Lid.,

66 Janpath, New Delhi 110001

Translator : Dr. Jayant Honmode
General Editor : Пу. Г. 5. Kothekar

Available from the U. 5. Department of Commerce
National Technical Information Service
Spring field, Virginia 22161

Printed at Prem Printing Press, Lucknow, India

FOREWORD

Phenotypic variation is the raw material for natural selection, yet a
century after Darwin it is an almost unknown subject. To a considerable
extent this may be true because even to see the significance of studying varia-
tion within populations (rather than taking it as given) requires an apprecia-
tion of several ordinarily separate disciplines: morphology, ecology, genetics,
developmental biology, and to some extent systematics and paleontology.
Other disciplines are also sometimes important, but these are central.

Darwin wrote more on variation than on any other subject. Since his
time, there has been almost no one of intellectual power in western countries
who has known all the central subjects adequately to make much progress.
In the USSR, however, Schmalhausen (to use the ancestrals pelling of his
name rather than the direct re-transliteration Shmal’gauzen) was such a
person. His major book, Factors of Evolution, is, because of its original ©
treatment of variation, perhaps as important as any book on evolution since
Darwin. Despite a translation in 1949, Schmalhausen’s book is not well
known in western countries, perhaps because of prejudice against Russian
work, perhaps because of Dobzhansky’s introduction to the translation.
This introduction damned the book by praising some minor achievements
rather than the central theme. And, contrary to the introduction, the
translation deleted some of Schmalhausen’s unorthodox insights. Some
of his insights, such as the distinction between r and K selection, have been
rediscovered or absorbed into western science; others have not.

Yablokov’s book documents part of Schmalhausen’s approach with
respect to mammals and makes some general theoretical advances. Taken
together, the books may hopefully be of some importance in re-directing
western research onto the subject of variation. The subject is unexpectedly
recondite when looked at seriously, and its exploration is still mostly by
travellers going elsewhere.

The Russian scientific tradition differs from ours in various ways.
I have tried to retain this as well as the characteristic flavor of Russian
writing. Except for occasional misprints and the like in the Russian version,
I have made no changes from that version. ‘The few apparent mistakes in the
original don’t seriously detract from the book. I caught some important
mistranslations, but as I have not re-translated the book others presumably
remain. Yablokov has gone over the revised translation and suggested
some improvements. ‘The published Russian version and the translation

1V FOREWORD

differs in some matters of substance because of changes by Yablokov in the
interim, but comments on single passages should nevertheless be based on
both versions.

Vaclav Laska, the Slavic bibliographer at the University of Chicago,
went over the entire bibliography with me. He made a number of correc-
tions as well as greatly helping with the references themselves. Katherine
Rylaarsdam also helped with bibliographic matters. Almost all the
references should now be clear, if not easily accessible, to western readers.
Quotations from English sources, or from books translated into English,
are wherever possible given as these versions rather than as re-translations
of the sometimes garbled translations into Russian.

In addition to Yablokov’s book and those of Schmalhausen and Paaver,
in his bibliography, I append here a chronological and probably incomplete
list of book-length treatments of variation after Darwin. I exclude special
topics with large literatures of their own: heredity, mimicry, agricultural
aspects, pathology, and human behavior; and books that also treat other
subjects than variation and so are better known.

Bateson, W. 1894. Materials for the Study of Variation. London:

~ Macmillan. 598 pp.

Pringsheim, H. 1910. Die Variabilitat niederer Organismen. Berlin:

J. Springer. 216 pp.

Deecke, W. 1920. Der Wert der Variationsstatistik fur die Palaontologie.
Ber. Naturf. Ges. Freiburg i Br., 22: 7-218.

Pelseneer, Р. 1920. Les variations et leur heredite chez les Mollusques.
Acad. Roy. Belgique, Classe Sci., Mem. ш-8° (2) 5: 1-826.

ЕГагаттае jf, #5 Jel, 1952. Uber Wachstum und Ruckgang. Leipzig:
Akademische Verlagsgesellschaft. 539 pp.

Robson, С. $., and О. W. Richards, 1936. The Variation of Animals in
Nature. London: Longman and Green. 425 pp.

Colyer, F. 1936. Variations and Diseases of the Teeth of Animals.
London: John Bale, Sons and Danielsson. 750 pp.

Ludwig, W. 1936. Das Rechts-Links Problem im Tierreich und beim
Menschen. Berlin: J. Springer. 496 pp. (Reprinted 1970, Springer-
Verlag).

Williams, В. J. 1956. Biochemical Individuality. New York: J. Wiley.
214 pp.

Olson, Е. С., and В. L. Miller, 1958. Morphological Integration.
Chicago: Univ. Chicago Press. 317 pp.

Berry, В. J. and Н. М. Southern (eds.), 1970. Variation in Mammalian
Populations. Symp. Zool. Soc. London. 26: 1-403.

L. Удм VALEN

“Science begins...at the time mea-
surements were made; no precise science
can be conceived of without measure-
ments.”

D. I. MENDELEEV

INTRODUCTION

Toward the beginning of the 20th century, with the development
of Darwin’s theory, problems confronting the evolutionary studies about
the formation of ways of evolutionary growth of organs could be considered
as understood on general lines (Haeckel, 1894-1896; Huxley, 1858; Gegen-
baur, 1898-1901; Severtsov, 1912, 1914; and many others). But studies
on the mode of growth of individual organs (morphological interrelations)
must serve as only a preface to causal researches on phylogenesis (Severtsov,
ТОЭ р. 58). However, the study of the mechanism of evolution could not
be carried out till the foundation was laid: availability of a large number
of new facts, the development of new concepts and theories in such fields
of biology as the studies of fine and ultra-fine structures in living creatures,
and a composite scientific approach to animal behavior and genetics.

But in the period of the “‘modern synthesis of all these sciences’’ that
followed (Dodson, 1960; Huxley, 1942), the morphological trend in research
was not the dominant one as in the past but played a subsidiary role. As
such, morphology, which was at one time the chief instrument for under-
standing phylogenesis, presently fell aside in deciding the general evolutionary
questions. Each of the new trends in morphology probing the secrets
of structures of living organisms to greater and greater depths (ultraviolet
and electron microscopy, histology and cytology, neurology, etc.) had
only one principle: to know more about everything small.

This fact, in its own way, led to the situation that evolved by the
middle of the 20th century: morphology developed into many independent
lines and became an experimental science, and while aspects of medical |
morphology were attaining foremost importance in their development,
evolutionary morphology and comparative anatomy were relegated more
and more toward secondary importance.

Morphology in the past studied organisms and all of comparative

v1 INTRODUCTION

anatomy was based on these studies. This course was an historic necessity
and facilitated drawing general conclusions of great importance: to establish
the very fact of evolution and to form the general structure of the genealogi-
cal tree of the present animal kingdom. But following this course it was
impossible to understand how evolution occurs.

One of the chief achievements of present theoretical biology is the
development of the concept of population as an elementary evolutionary
unit. This most important fact, substantiated by voluminous data in
genetics (Chetverikov, 1926; Wright, 1931; Dubinin, 1931; Dubinin and
Romashov, 1932; and others), ecology (Sumner, 1909-1932; Elton, 1927;
Pearl, 1939; and others), paleontology and general biology (Schmalhausen,
1939; Simpson, 1941, 1944; Mayr, 1942; Beklemishev, 1960, 1964; and
others), is absolutely inadequate until it takes into account vertebrate
morphology.

This poses highly specific problems because the study of the structure
of a particular specimen is not adequate to characterize in detail the group
of animals under study.

In response to these new requirements a new trend has appeared in:
morphology in the last decade. It is based firstly on comparative anatomy,
secondly on quantitative ecology, and thirdly on population genetics.
This trend is known as “population morphology”.

The basis of this trend is the study of the population as an evolutionary
unit by morphological methods. Here, too, the study of variability, the
material basis of the evolutionary process, attracts more and more attention
from morphologists of all animal groups.

A question arises: Is it correct to study phenotype without studying
genotype? In other words, can a morphologist who, as a rule, is concerned
with the study of phenotypic variability discuss evolutionary processes?
To this question, George Gaylord Simpson (1944) gave a good answer. He
showed that the question of whether a Bien trait is inherited does not
exhaust its significance for evolution:

“Мо morphological character is inherited as such. The ultimate
morphological expression of the same hereditary factors differ markedly.
The particular expression, among several possible expressions, developed
in a given individual does not affect the potentialities passed on to its
descendants. It certainly affects the fate of the individual and therefore
helps to determine whether that individual will have descendants”
(Simpson, 1837; p. 62; 1944, p. 30). For a morphologist, this discussion
has great importance because it allows him to confidently consider that
the morphological and phenotypic variability under study has a direct
relation with the evolution of the group (Cannon, 1959; Yablokov, 1963b).
In the morphological approach to the study of the evolutionary process,
it is also important to note that the so-called “small’’ morphological

INTRODUCTION Vii

variations are usually correlated with significant specializations in the
activity of the organism (Olenoy, 1961) and a variation in an individual
trait usually indicates a variation in the condition of the whole organism
(Sinskaya, 1948). Finally, there is no reason to believe that variability
in one type of organs will be governed by one rule and variability of other
organs by another. Most general properties of variability, as a phenomenon,
will no doubt be similar for the variability of the length of the tail and for
the variability in the number of cells in the cerebral cortex. Of course,
in the living being there are organs with relatively greater or lesser im-
portance, but the relativity of any organ, emerging from the multi-
functionality of organs, diversification in the function of each organ during
ontogeny, and insufficient understanding of the functional importance of
individual organs and their structure (Sokal and Sneath, 1963), brings
all the organs into equality when studying general properties of variability
(in the same way as the general laws of heredity can be studied equally 1 in
viruses, Drosophila, and peas).

It would appear that the brilliant successes achieved by mathematical
and population genetics in the 1930’s have come very close to unfolding the
exact and important aspects of the mechanism of evolution. But the
problem appeared to be much more difficult: the research workers had to
discover the complicated phenomena in the formation of a population as
a dynamic structure, then their relationships in similar populations of
diverse organisms, which together form a self-developing system, a bio-
geocenose. .Only recently have the research workers begun to understand
clearly the different channels or links through which the development of
any population is controlled (Schmalhausen, 1958-1965) ; only now we begin
to actually understand the importance of this differential approach toward
the study of various forms of natural selection (Berg, 1964a, 1964b). It
appears that the general understanding of the complex picture of genetic
variation in populations determined by the action of selection, mutation,
and stochastic (genetico-automatic) processes (Belyaev, 1962; p. 111),
is insufficient to explain almost exclusively on this basis genetical problems
and the role of the so-called ‘‘non-hereditary” variability im evolution
(Kirpichnikov, 1935, 1940; Lukin, 1936; Kamshilov, 1939; Dobzhansky,
1937; and others). It mainly helped to show the inadequacy of only the
genetical approach. Now it is clear that this trend has promising contacts
of a genetical approach to the study of the evolutionary processes with
morphological and ecological points of view. Now, as never before, the
problem of the intermingling of these approaches has become acute; without
it the future fruitful growth of the evolutionary concept appears to be at a
standstill (Blair, 1964).

In the present work, an attempt-has been made to make an advance
in one particular problem against this common background of association

Vili INTRODUCTION

of many sciences in the study of evolution, 1.е., the study of the characteristics
of population morphology in the class Mammalia.

It was considered important to collect every fact about the variability
of different systems of organs and the structure of mammals. This has been
done in Chapter II. On the basis of an analysis of variability at various
taxonomic levels in one specially studied group of Pinnipedia, an attempt

has been made to clarify the general characteristics of variability of a group

at different taxonomic categories (Chapter III). The entire collected data
showed the inadequacy of the existing classifications of variability as a
whole and has permitted us to suggest a new scheme of classification of the
phenomenon of variability (Chapter IV). In the concluding chapters of
this work (Chapters V and VI), an attempt has been made to throw light
on various general questions related to the process of “‘micro-evolution”
operating at the level of a population (a concept introduced by Timofeev-
Ressovskii, 1939, 1958, 1965a, 1965b).

All this taken together indicates the inadequacy of existing studies on
the phenomenon of variability and defines the general structure of the present
work. ‘That is why, because the starting point is the problem of variability
as such, the ecological analysis of evolving structures is given only in a most
general form in the majority of cases.

The importance and the necessity for such formalized (in the initial
stages) study of variability, has been recognized by many research workers
(Alpatov and Boschko-Stepanenko, 1928; Simpson and Roe, 1939; Bader,
1955; Bader and Hall, 1960; Roginskii, 1959; and others), who have noted
many similar situations: these have been dealt with in detail at appropriate
places in the work.

The selection cf mammals and not any other vertebrates or invertebrates
for the present research is not accidental. In this highly developed class
of the animal kingdom, the patterns of evolutionary growth which are of
interest to us must be acting in a most evident way, perhaps because “‘evolu-
tion produces new forms’ (Berg, 1959).

Among mammals, it was of interest to study many ecologically different
forms. In this regard, it was thought necessary to examine critically some
sharply divergent groups, for example groups of aquatic mammals and
Chiroptera, which have undergone substantial morphological adjustments
in the course of adaptation to the environment, and which therefore have
adaptations not characteristic of other mammals. An examination of
these groups, perhaps, will be most fruitful as compared to the “‘typical”’
terrestrial mammals, morphologically nearer to the ancestors for all mamma-
lian forms. The proposed work is a beginning of such analysis and has
been carried out on original data from Pinnipedia and Cetacea of the USSR,
as well as on data obtained by other research workers on all principal groups
of mammals.

INTRODUCTION 1х

Data оп population morphology of mammals are accumulating at a
very fast rate. ‘This situation shows once again the necessity for generaliza-
tion, theoretical comprehension, and the formation of specific working
hypotheses for future planned and intensive studies on populations by
morphological methods.

In biology, as it becomes a more and more perfect science, there arises
a necessity to use clearly defined concepts. Hence, in the beginning it is
important to define exactly as far as possible the basic terminology necessary
for a uniform treatment of the theoretical basis of the problems under study.

1. Variability is sometimes understood as a process of appearance
of qualitative differences as well as the presence of such differences (Simpson,
1948). Specific use of the concept of “variability” should be made only
for stating the fact of qualitative differences. In our work, the definition of
variability has been used as suggested by George Gaylord Simpson: the
presence of differences among individuals within a breeding population.

It is clear from the definition that the variability examined under
such a plan, appears not as a property of the individual organism but as
a property or characteristic of the population.

2. Population in the present work may be understood as a group of
individuals in which panmixis occurs and which is genetically isolated
from similar such groups for more than one generation.

3. Selection (in the most general sense) is understood to be a result
of interaction between individuals of the evolving species and the changing
environment (Schmalhausen, 1939; p. 77. Also, see note in Chapter ТУ
of the present work).

4. Environment is understood as all that surrounds the population
(organism) and related in any form to the continuance of its life functions;
geographical and climatic conditions, food, competitors, behavioral
interactions, etc. When examined in such generality, the organism or the
population can be taken as a part of its own environment (Simpson, 1948).

5. Trait is understood as any characteristic of the organism and of the
population as a whole which can be studied and expressed in quantitative
units.

6. Structure in the present work refers to a component part of an
individual organ or a system of organs; for example, phalanges or fingers
as structures of hands, blood corpuscles as structures of blood, different
groups of cells as structures of individual organs and systems, etc.

A number of researchers, mainly the staff of the Institute of Develop-
mental Biology, Academy of Sciences, USSR (Moscow), took part in
the collection and initial preparation of the material reported in this work.
Professor 5. Е. Kleinenberg was the initiator of many discussions that took
place in the laboratory and which continued during many months of field
work in the Arctic and Pacific Oceans. Dr. G. A. Klevezal’, Dr. V. M.

х INTRODUCTION

Bel’kovich, ang У. I. Borisov rendered me invaluable help in the collection
of the material and V. Ya. Etin in the processing of this material.

Credit in the publication of this work, first of all, goes to those criticisms

and discussions in which the following scientists took part: М. У. Timofeev-
Ressovskii (whose great erudition and personal respect mainly determined
the formation of my viewpoints), N. N. Vorontsov, V. I. Tsalkin, B. S.
Matveev, S. S. Schwartz, members of the laboratory colloquium and the
scientific council of the Institute, members of the seminar on morphology
at the zoological museum of Moscow State University, and the members
of the inter-sectional seminar on evolutionary problems at the Moscow
Society of Naturalists, where, in recent years, various parts of this work
were discussed. ;
: In the process of working on separate chapters, I took advice from
Y. I. Starobogatov, Academician I. I. Schmalhausen, corresponding member
AN 555К О. К. Belyaev, К. К. Chapsku, Т. С. Sarycheva, К. К. Flerov,
Ya. Г.. Glembotskii, and G. О. Polyakov; and critical observations were
made by K. A. Yudin, K. K. Chapskii, corresponding member AN SSSR
G. V. Nikol’skii, A. A. Lyubishchev, N. P. Naumov, V. M. Makushok, and
В. М. Shornikov.

I express my sincere and deep gratitude to all the above-mentioned
research workers.

I dedicate this work to my friend, wife, and constant helper—Eleonora
Dmitrievna Bakulina.

Since I finished the first edition of this work, there have appeared
a large number of new papers on the intraspecific variability of mammals.
In the introduction to the first edition, I mentioned the possibility of having
omitted much work worthy of inclusion, but now I unavoidably have to omit
many interesting studies in this area without even mentioning them. This
is because of the technical unfeasibility of surveying thousands of studies
in one book, and because in some recent studies new (and at times very
interesting) data give us nothing new in principle with respect to the
intraspecific population variability of mammals.

It is quite evident that for the specialist on some relatively small group
of mammals (or on some part of the anatomy), his particular subject will
in many cases be absent from this book. However, I hope that in neighbor-
ing pages he will find information on the variability of other groups of
mammals (or parts of the anatomy), unfamiliar to him in detail, a feature
desirable for broad comparison and deep analysis.

This book has data on fundamental matters: an account of intraspecific
phenomena (more exactly, intrapopulational), phenotypic variability of

INTRODUCTION х1

mammals. The study of the variability of the phenotype is a necessary
step in any kind of deep study of population variability. This work, I hope,
shows some degree of direction in the development of “population morpho-
logy”; it can therefore be of interest to more people than specialists in the
study of mammals. It seems to me that the population-morphological
(or, more broadly, the “‘phenotypic’’) approach permits accurate analysis
of large populations. This approach has been made first to morphological
material, but such populations also occur, now more and more frequently,
in genetical, ecological, and ethological analyses of species, and give a
means of studying current problems of biology.

After the publication in Moscow, in 1966, of the first edition of this
book, I have obtained information from various comments and answers
to requests from many people. I thank these people, and among them
especially P. F. Rokitskii (Minsk), V. S. Smirnov (Sverdlovsk), E. Olson
(Chicago), B. M. Melnikov (Moscow), Ya. I Starobogatov (Leningrad),
Yu. T. Artem’ev (Kazan), C. A. Long (Stevens Point), and others. Many
of these observations were helpful for the present version of the book.
Besides these items, I have added some other data and rather condensed
accounts of some general principles.

I am glad that publication of this work in English will make it more
accessible to workers from many countries, and express my most sincere
and deep gratitude to Professor L. Van Valen for his initiative ш the
translation of this book and editorial work on it.

A. V. YABLOKOV
Moscow Institute of Developmental Biology
November, 1972 Academy of Sciences of the USSR

CONTENTS

Foreword

Introduction

т

II.

ITI.

GENERAL APPROACHES TO THE STUDY OF VARIABILITY a
Method for Complex Analysis of Mammalian Variability ..
Principles of Selection of Traits
Framework of Analysis of Morphological Variability о
Mammals ae we

Methods of Analysis of Data

The Statistical Significance of Differences

Optimum Size of Sample

Accuracy of Measurement and individual Errors of the
Researchers

Coefficient of Variation as an index for the Вора нот
Environment” System

THE NATURE OF VARIABILITY IN DIFFERENT SYSTEMS IN
MAMMALS
Integument
Skeleton
Variability of Other Internal Orsans
Variability of Body Measurements and Weights. :
General Character of Variability of Different Е
and Body Measurements of Mammals

Ам ANALYSIS OF VARIABILITY OF POPULATIONS AT DIFFERENT
Taxonomic LEVELS WITHIN AN ORDER :
Comparison of Variability of Population within: a Species ..
Certain Results from the Study of Variability of Characters
for Different Populations of One Species
Comparison of Variability of Population within a Genus
Comparison of Variability of Populations of Closely Related
Genera
Characteristics of Variability in Ponuladons, of Different
Subfamilies

ill

102
104

109

У

ху

CONTENTS

The Nature of Variability in Populations of Different
Families within an Order .

Certain Characteristics of Population мы, on Dien
ent Taxonomical Levels

IV. CLASSIFICATION OF THE РНЕМОМЕМА ОЕ VARIABILITY

A Short Survey of Certain Classifications of Variability
Types of Variability

Manifestation of Variability

Forms of Variability

Conclusion

V. Some REGULARITIES OF VARIABILITY ..

Elementary Regularities in the Variability of Traits

Relationship of Variability to Dimensionality of Traits

Relationship of Variability to Absolute Size of Trait

Relationship of Absolute and Relative ee” Values
(Indices)

Comparative Variability and its importance in Population
Biology 5

Analysis of Comparative о for Some Mameuls si

VI. VARIABILITY AND THE PROBLEM OF VESTIGIAL ORGANS

A Short Historical Account of Viewpoints on the Problem
of Vestigial Organs

Structure of the Pelvic Girdle in СЯ а

Structure of the Vibrissal Apparatus in Cetaceans

Specialization and Vestigiality

УП. VARIABILITY AS AN ADAPTATION

Relation of Variability to о of ЕО (Accortiins
to Different Authors) a a

Modes of Selection and econ, | BI 5

Directional Selection and Variability :

Prolonged Variability and its es in the Eval.
tionary Process

Adaptive Importance of и

The Possibility of Studying Natural Sélection г. he
Understanding of Variability Ne a

VIII. Conclusion

Summary

124

128

131
131

Wav)

142
174
192

194
194
195
201
207

212
218

Zoi
231
233
255
240
р
250
253
25

256
259

261

267

269

CONTENTS XV

Appendix: The Variability of Body Measurements and Internal Organs
of Mammals a see af 2
Bibliography Bc Cre ae 905

CHAPTER I

GENERAL APPROACHES TO THE STUDY OF VARIABILITY

Method for Complex Analysis of Mammalian Variability

An absence of a general framework for the collection of data to study
population morphology has been increasingly felt in the last few years.
This necessity is all the more felt because until recently data on quantitative
variability of mammals were used only in such zoological disciplines as
taxonomy, zoogeography and genetics. ‘This continued through many
decades with the same traits, usually skull measurements and some body
measurements, being studied. Such a situation 15 of historical significance.

In the 19th or at the beginning of the 20th century, the attention
of researchers was centered on the formation of the family tree of the
organic world in general and, as far as mammals were concerned, to the
construction of a general phylogenetic classification of mammals and an
inventory of the world fauna. Such an inventory was prepared, and by the
end of the 19th century a general classification was formulated. The
researchers in comparative anatomy (which by now had lost much of its
importance in the evolutionary approach) continued and intensified efforts
to bring to light minute details. In the beginning of this century, this
approach to comparative anatomy in phylogenetic research reached its final
expression, through the formation of a theory of organogenesis, in the
works of A. N. Severtsov—a morphologist and evolutionist. Afterwards,
comparative anatomy became closer to ecological morphology, the aims and
objectives of which differed to a considerable degree from the former aims
of classical comparative anatomy.!

1 п this country, this transition is specially noticeable in the growth of morphology
in the past 20-25 years, which was stimulated to a great extent by the spread of Neolamarckism.

2 Variability of Mammals

The methods of comparative anatomy, which initiated quantitative
morphology, now served only as additional working methods in the mixed
fields of biology. As a result of this, comparative anatomy looked
qualitatively impoverished because of a strong restriction to only a few
traits of organisms, but at the same time it was enriched through the
quantitative methods which facilitated deeper studies under a new light.
Such a situation, naturally, was well suited to all the contiguous sciences
and continued for the whole first quarter of our century. That is why even
until the 1950’s not a single broad system of analysis of quantitative
variability was worked out.

Many works on the taxonomy and genetics of mammals, which dealt
with either very few traits or (in taxonomy) a great number of partial
traits not possessing the required quantitative value and not comparable
with other traits in other systems, would exemplify what has been said
above. ‘To obtain sources of data, for example, on the variability of the
intestinal length, quantity of hemoglobin, variability of hair cover, etc.,
one had to look not into works on zoology or on comparative anatomy but
into works on physiology, hematology, commerce, etc., and any accidental
appearance of such information in zoological literature reduced its value
to zero.

These conditions led researchers who conducted studies on the structure
of mammals, to develop a better method for the complex analysis of
variability in the morphological structure of mammals. F. Sumner
(1906-1908) used morphological indices (excluding skull measurements)
for different traits (Sumner, 1908) for characterizing populations of mice;
afterwards, the same author increased the number of traits to twelve
(Sumner, 1919-1932). The research of F. Sumner, and similar works of
other scientists, was carried out in accordance with population genetics.
Afterwards, in an attempt to find out more useful criteria for the similarities
and differences of animal groups, taxonomists began studying not the indivi-
dual traits but their whole complexes (Huxley, 1940; Terent’ev, 1959;
Mayr её al., 1956). Finally, the works of S. S. Schwartz and _ his co-workers
(1954-1965) revealed the possibility of a complex quantitative approach to
the study of ecological and morphological characteristics of mammals
and other vertebrates.

The distinguishing character of all these works (with the exception of
works like that of S. S. Schwartz, based not on morphological but ecological
aspects) happens to be an attempt to obtain a larger number of traits
not for the characterization of the organism as a whole, but for a deeper
genetical study of the population, or a more objective study of the taxonomy
of the group.

At present, we are witnesses to a new era in the development of
quantitative morphology and comparative anatomy. These morphological

General Approaches to Variability Study 3

approaches, in combination with population genetics and quantitative
ecology, seem to be more and more important for solving a number of
evolutionary problems. Population morphology, arising as a result of such
interactions, has the aim of obtaining an objective characterization of
populations as elementary evolutionary units. For this, a wide and complex
calculation of the variability of all the systems of organs of the members
of a population is needed. In a number of cases, it is as yet difficult
to fulfil this condition because of the absence of an objective and a feasible
method of investigation. But there is no doubt that the very posing of
the problem should help in working out such methods.

Principles of Selection of Traits

It is desirable to take the broadest possible sample of variability in traits
of organisms forming a population to study the nature of morphological
variability. This has to be seen from two basic points of view:

1. Study of all traits in population is desirable but impossible; and

2. During the study of the general control of variability, a single
approximation is possible and necessary (reasons for this have been given in
the introduction).

It can logically be assumed that the optimum solution to the problem
of selecting the number of traits to study for population variability, hes
somewhere between the number of organs of animals, and the general number
of functions regulated by these organs. Both limits are highly uncertain
and depend on the level of our knowledge.

George Gaylord Simpson (1948), with reference to studying the rate
of evolution, observed that the most serious difficulties in solving his
problems were as follows:

1. Selecting traits;

2. Presenting them all in a measurable form capable of being expressed
in comparable units; and

3. Imparting to them corresponding weights to obtain a general
mean value (Simpson, p. 43).?

These very difficulties arise during the study of variability.

While measuring the length of intestines in a population of wild
mammals—a quite exact trait—we may obtain insufficiently correct
biological data for the population as a whole. It would appear that the
particular categories studied do not feed in the same manner, as was revealed
from the analysis of food of other members of the same population (the
problem of non-characteristic and ‘‘unclean”’ selection), but on the other
hand the length of intestines will be a good trait for the selection of animals

2 Here and afterwards, this work of G. G. Simpson has been cited according to the
Russian edition, 1948.

4 Variability of Mammals

of the same age and sex from an animal population fed on similar rations
for a long time, for example, for characterization of an experimental popu-
lation of mice in a vivarium. We can see from this example that it is
necessary to decide in advance what types of variability to analyze and
then, depending on the aim of the investigation, to separate the traits showing
the kind of variability to be studied. (For classification of variability,
see Chapter IV).

One of the basic necessities for the study of measurements is that their
calculations and determination should be easy. It is difficult, practically,
to compare the variability of external characters of the body with the
strength of the muscles and speed of movement, but it is easy and valid to
compare wild animals for variability of color, body proportions, and
meristic traits.

It is not proper to study variability in number of eyes or extremities,
but in insects the number of facets in the compound eye may be a good trait
for study; similarly, the number of parapodia in polychaete worms. In
Carnivora or ungulates, it is not proper to study the number of fingers
and the number of phalanges in each finger, but it would be good to study
these traits in cetaceans. At the same time, in ungulates and Carnivora,
it may be useful to study the variability of parameters in individual elements
of extremities, e.g., lmear measurements, weight, form of long bones, and
size of articulated surfaces.

During the selection of traits for comparison, it is necessary to take
into consideration the importance of the function of differently measured
characters of this or that organ, because as seen below (Chapter V) the
character and magnitude of variability depends on the character of a trait.
This means that in order to obtain the most objective estimate of the variabil-
ity of any organ, it is important to consider heterogeneous traits (linear,
weight, meristic, volume, etc.) characterizing this organ from different
angles.
Summarizing in a nutshell the proposed approaches to selecting traits
for comparison, it can be said that it is necessary. to select the following
types of traits:

1. Those which will correspond best to the type, form, and appearance
of the variability under study;

9. Those which may be easily studied and determined;

3. Those related to most different systems of organs;

4, Those characterizing one and the same organ from different sides.

Framework of Analysis of Morphological
Variability of Mammals

On the basis of data obtained on variability in mammals, it seemed

General Approaches to Variability Study 5

possible to make a general arrangement of characters appropriate to most
mammals. The given arrangement should be broadened, corrected, and
detailed in future.

Integument

Hair cover: Number of hairs of different types in a unit area on different
body parts*.? Measurements of hairs of different types on different body
parts.

Vibrissae: Number of vibrissae on different body parts, and within
a part by group*; measurement of vibrissae*.

Skin as such: Number of layers of epithelial cells in the stratum corneum,
stratum granulosum, stratum lucidum, and stratum spinosum of epidermis;
cell measurements; number of dermal papillae and collagenous fibers in
a unit area on different body parts*; number of sweat and sebaceous
glands in a unit area on different body parts; general thickness of individual
layers of skin*; dermatoglyphic ridges on extremities”.

Color: Colorimetric character of body surface*, size of dark and light
parts*, number and disposition of spots*, quantity of pigment in a unit area.

Skeleton

Axial skeleton: Number of vertebrae by divisions (thoracic, lumbar,
sacral, caudal)*. Number of sternal ribs* and asternal ribs*, number of
segments in sternum, number of cartilagenous bones (in cetaceans), linear
measurements of all parts of the skeleton* (scapulae*, vertebrae*, hyoid
Бопе*, etc.).

Extremities: Measurement of skeletal extremities, pectoral and pelvic
girdles, characteristics of disposition and form of metacarpal and metatarsal
bones*, number of phalanges and digits (for cetaceans) *.

Skull: Condylobasal length, basicranial length, occipital length*,
length of palate*, length of upper toothrow*, mastoid width*, width of
brain cavity*, maximum width of zygomatic arch*, width of nasal passage”,
width of palate*, width of snout near canines*, width of occipital condyles*,
length of auditory meatus*, width of auditory meatus*, length of nasal
bones*, width of nasal bones*, least distance between orbits*, height of
skull*, length of lower jaw*, height of lower jaw behind molars*, number of
teeth in upper and lower jaws*, weight of skull*, volume of skull*, volume
of brain cavity*, diameter of largest tooth, height of canines, length of
molars*, width of molars*. For some orders of mammals, this arrangement

3 The traits used in this work have been asterisked (*).

4 Basic qualities required for the traits are maximum homogeneity and demarcation by
fixed points; this gives an opportunity to obtain similar results even if the measurements
are taken by different researchers. Such an experiment was conducted in our collaborative
work (Yablokoy and Serzhent, 1963) and showed full agreement in the data obtained.

6 Variability of Mammals

will have to be changed, for example for cetaceans (Kleinenberg её al.,
1965).

Digestive system
Intestine: Length as a whole and by parts*, diameter, weight of in-
testine*, weight of stomach*, volume of intestine and stomach*, number of
different types of glands in a unit area of intestinal surface in different
parts*, histological characters of mucus membrane and mesentery™.
Tongue: Histological characters of mucus membrane, number and
arrangement of different types of sensory organs* (papillae).

Respiratory system

Trachea: Number of rings*, measurements.

Lungs: Measurements, number of lobules, number and measurements
of acinus and alveoli.

Urogenital system
Kidneys: Measurements*, size of pyramids, length of ureters, volume
of urinary bladder*, quantitative histological characters of urethra.
Spermatozoa: Measurements and form*, number of spermatozoa in
a unit volume of ejaculation.
Measurements of testis, vas deferens, ovaries, fallopian tubes, uterus.
Glands of internal secretions: Measurements, weight, and quantitative
histological characters of liver, adrenals, thyroid, and parathyroid glands*,
Hypophysis*.

Nervous system

Central nervous system: Measurements of the brain* and its different
parts, form and disposition of sulci and gyri. Histological characters of
gray and white matter (number of cells of different types™, ет of cell
layers in different parts of the brain).

Cranial nerves: Measurements, number of plexuses, number of rootlets,
characters of spinal cord, characters of sense organs”.

Circulatory system

Heart: Measurements and weight*, histological characters.

Arteries and veins: Positions of main © trunks; histological characters
of walls; number of capillaries, arterioles, venules, in different parts of
the organism.

Blood-producing organs:. Measurements of spleen*, quantitative histo-
logical characters of spleen and bone marrow.

Blood: Characters of corpuscles* and plasma.

This is a general list of the traits that can be obtained through modern

General Approaches to Variability Study 7

laboratory and field techniques in the study of qualitative and quantitative
variability of mammalian morphology.

On the whole, it seems possible even now to obtain quantitative data
on approximately 150 different traits of organs of different systems. Even
in a greatly abbreviated case where only the most simple traits are selected
(without histological and anatomical details), the researcher must try to
pinpoint at least 35-45 traits in order to give a far better picture of the
morphology of the population under study than would be possible with
only 10-15 as usual.

Methods of Analysis of Data

The general methods of estimating variability, widely known among
biologists, are used while studying population morphology (Plochinskii,
1960; Rokitskii, 1961; and others). Some of them are briefly described
below:

a) Itis obvious that the simplest measure of the magnitude of variabi-
lity of a sample in a given trait is the observed range (Тат)?. However,
this method happens to be the least perfect, because the extreme values of
a sample depend greatly on the sample size, which is itself subject to sharp
variations; and what is most important, with this measure the characteristics
of the distribution of variation within the sample are not clear. Finally,
the comparison of the range of variability can be carried out only within
one species or in very closely related taxa for which the mean of the trait
under study is similar.

b) The coefficient suggested by R. L. Berg (1964a)—the ratio of
maximum value of a trait of a sample to its minimum value—serves as
the latest modification of this old method of estimation of variability:

Xmax

Amin

The resultant coefficient, which is dimensionless, allows a comparison of
traits of different absolute values and also of heterogeneous traits. How-
ever, even by use of this coefficient, it is not possible to determine the

5 Here, and henceforth, the so-called ‘‘normal’’ distribution, expressed by a Gaussian
curve, has been considered. In nature, the variation of traits cannot always be expressed
by a “normal” distribution. Apart from the ‘‘normal,” some other different types of
distributions could be encountered in nature, of which the binomial distribution and the
Poisson distribution are the foremost. But it is known in biological statistics (Beili, 1964)
that in large samples the binomial distribution and the Poisson distribution approach
normality. This means that evenif the researcher encounters such types of distributions
and at times does not consider them, the usual analysis for the normal distribution allows
proper results in such cases anyway.

8 Variability of Mammals

characteristics of the distribution of variation within the sample and the
effect of sample size is unchanged. It is evident that this coefficient can be
very widely used during some specialized investigations. ‘The simplicity
of its calculation considerably lessens the analysis of the data.

c) A better method of expressing variability is the estimation of the
standard deviation:

Certain researchers, for example, A. A. Lyubishchev (1923), G. V.
Nikol’skii and V. A. Pikuleva (1958) prefer this method of estimation of -
variability to other methods. The standard deviation (sigma) is in fact
capable of exactly characterizing the whole variation of samples by showing
the degree of inclination in the cumulative distribution curve. Comparison
of variability of different traits or traits with different measurement units
is absolutely impossible because sigma is a dimensioned statistic.

d) The “Profile Method” of S. R. Zarapkin is an interesting develop-
ment; in this method, different populations are compared not by the
absolute but by the relative value of sigma (Zarapkin, 1934). This method,
no doubt, is very useful for carrying out comparisons of variability of differ-
ent populations within a species or among closely related types and is worth
developing further. In the present work (Chapter III), an attempt has
been made to modify this method slightly. :

e) The most widely used method for determining variability is the
estimation of the so-called coefficient of variation. This is obtained by
dividing sigma by the arithmetic mean and is expressed in percentages:

с.100

6.0. —= .

The coefficient of variation is considered the best and most widely used
index for characterizing the range of individual variation in one or diverse
traits (Alpatov, 1956, p. 374 ; Haldane, 1955).

The basic advantage of a coefficient of variation lies in the fact that,
in addition to completely characterizing the amount of variation, it happens
to be a dimensionless index and allows a comparison of variability of large
and small organs of different species, variability of completely different
organs, and variability of differently measured traits of one and the same
organ. The relative ease of estimating the coefficient of variation for
original data, as well as for data collected from the majority of detailed
publications related to morphological characterization of a population, is
nonetheless important. It is not proper to consider the coefficient of var-
iation as an ideal index. One of the main shortcomings is its incapability

General Approaches to Variability Study 9

of showing the degree of asymmetry in the distribution under study.
With opposite indices of asymmetry, the coefficient of variation may remain
unchanged in magnitude.

Another shortcoming of a coefficient of variation has been noted by
]. В. $. Haldane (1955). With samples of the usual size investigated,
significantly smaller than the general population, lesser variability is
obtained. This situation arises especially during the investigation of very
small samples, consisting of only a few individuals. ‘These observations
seem completely justified. However, one should remember that in the
estimation of coefficients of variation for different traits of one and the
same population, we obtain completely comparable data.

f) The study of variability is not confined to the methods of its
estimation mentioned above. Analysis of distributions with the help of the
F statistic of Snedecor can be very helpful in morphological investigations.
This method is good for the comparison of different populations with one
species, as shown by B. Sturgen and N. Popovich (1961).

One promising method of estimating variability has been proposed
by Ya. I. Starobogatov. This method consists of a direct ratio of variances
(0?) estimated as with standard deviations. With the use of this method,
as well as with the use of a coefficient of variation, obtaining dimensioned
figures is avoided and at the same time the index obtained expresses the
basic properties of the distribution curve of the trait—range of variation
in the trait and degree of spread of the curve.

g) Experience has shown that for the study of variability of the
polymorphic type (including epigenetic polymorphism), there is a possibility
of successfully utilizing a number of methods based on comparisons of
frequencies of class, without resorting to each statistical analysis of data
in the way generally understood for variation by zoologists. First of all,
it is possible to use the chi-square of Person and the Lambda statistic of
А. №. Kolmogorov and М. У. Smirnov. These methods as used in biological
research have been given in the works of М. A. Plochinskii (1960). It is of
interest to note, as observed by P. F. Rokitskii (1964), that statistics of
this type can be used not only for polymorphic variability but also for the
general case of quantitative variability. These statistics have been investi-
gated by us in one case, in the study of mammalian color (Yablokov and
Etin, 1965), and in addition, prospects of combined use of these statistics
have been put forth in this work (for details, see Chapter IV).

h) Simple determination of the number of aberrant individuals for
a given trait in a population, relative to a given number of individuals,
is the other method of estimating morphological variability in population
morphology (Gershenzon, 1946), for example per 1,000 or 1,000,000
individuals; this is similar to the methods used in genetics for evaluating
the number of mutations per given number of gametes.

10 Variability of Mammals

i) Graphical methods are quite important in the study of variability;
by their use the number of arbitrary assumptions of the statistical*formulas
is reduced to a minimum. The normal curve is based on the assumption
that the number of possible states of variation is practically unlimited and
the curve on both sides of the mean intersects the base line only at infinity.
But all the natural populations, however large they may be, have numerical
end values for variants. As a rule, this assumption does not play a great
role during work on rather small samples from natural populations. But it
is possible to imagine a case where it is necessary to consider this situation
so that the biological significance of the statistical result obtained is
not lost.

Graphical methods are very handy to use because of their considerable
clarity. Transfer of data to a histogram opens the possibility of bringing
to light trends in recurring frequencies of different classes. A scatter
diagram also is a graphical method of expressing quantitative variability.
This has been described by Е. Mayr её ai. (1956) and examples of its use
are given in some biological works (Sarycheva, 1948; Yablokov, 1963b).
When the amount of data is not sufficient for statistical representation, an
opportunity to derive conclusions through comparing the variability of
homogeneous traits in adjoining groups is possible in some cases. Another
advantage of the scatter diagram is that it offers a possibility to compare
not only two characters, but three or more.

In this work, the problem of the study of correlation between traits
has intentionally not been touched, though it is of great importance in the
general analysis of the structure of variability (Terent’ev, 1959; Berg,
1964a, 1964Ъ).

From this short review of the different methods of estimation and
expression of variability, it can be seen that, generally, all usual methods
of variation statistics are used in population morphology. ‘The distinctive
feature of researches in the field of population morphology is not in specific
methods of data analysis but in the very approach to the data, with an aim
of presenting the character of a population as a unity. Mention must be
made of the fact that the use of statistical methods in population biology
is in harmony with the general trend of modern natural science, which is
making the whole of biology an exact science. P. F. Rokitskii (1964) is
right in saying that ‘‘...though difficulties are experienced in the
application of biological principles with these methods (this means
mathematical methods—A. Ya.), their inclusion in the armory of zoologists
is absolutely necessary. Only extensive use of these methods on material
of different types will insure a determination of the practicability of their
use, permit an evaluation of the significance of individual methods, and
reveal the degree of sensitivity of the criteria obtained on their basis”

(pp. 86-87).

General Approaches to Variability Study 11

The Statistical Significance of Differences

The importance of evaluating the degree of significance of differences
between comparable distributions of traits in case of different populations
does not need a special justification. As is generally used in various
investigations, the significance of differences between comparable indices
is estimated with the help of the ¢ statistic according to the formula:

хх

1 №2

t 7
Sq

where, given a sufficiently large sample size (generally more than 25

specimens)
2 2
Ore ee ’
4 Oi, 2p)
and with the comparison of smaller sample sizes

г = (4 —%) + Xp (X2—%2) (oe

(m,—1) + (".—1) Пт
(Rokitskii, 1961; formulas 23 to 25).

It is important to draw attention to the fact that the statistical signi-
ficance of results obtained through mathematical analysis of biological
material cannot always be estimated unconditionally on the basis of these
or any other mathematical formulas. The complexity of the biological
system so much surpasses the statistical methods that I am deeply convinced
there is no possibility of analyzing them through intricate expressions of
probability figures or through Student’s area of the probability curve (¢).
The other exception to the absoluteness of the results of statistical analysis
is in the fact that statistical methods analyze quantitative differences, and
during a comparison of different populations in biology, we want to get
an answer as to the presence or absence of qualitative similarities.

One can completely agree with the opinion of B. Sturgen and
N. Popovich (1961) who say that “unfortunately, we as yet do not make
use of objective criteria in evaluating qualitative differences in the mysteries
of natural life through mathematics” (p. 575).

In the material collected by us on the morphology of Pinnipedia, there
are data which show that significant differences were observed through the
use of statistical methods between certain biologically homogeneous
characters in different samples. One of the reasons for such a situation
may be the considerable deviation of the distribution curve for the given

12. Variability of Mammals

trait under study in a natural population from the “normal”? distribution—
to which alone the usually used statistical calculations are faithful.®

Some mathematicians have come to the conclusion that the laws of
calculations of probability cannot be used for higher forms of movement
of matter without considering the practically infinite quantity of information
available about the state of such higher forms of movement. Explaining
this point of view, Е. Borel’ (1964) writes: “It seems fallacious to use the
laws of estimation of probability for the mystery of formation of crystals
from more or less super-saturated solutions. At least, it is not possible
to examine such problems of probability without estimating certain properties
of matter: properties which facilitate formation of crystals” (p. 104). And,
in the conclusion of the review of the possible uses of the theory of probability
with cosmogonical problems and problems of biology, Borel’ writes: “It
seems that even in this field, the results given by us may not help much
at present’? (Ibid.).

In these excerpts nothing is said about probability and significance
in the evaluation of differences between two populations, but it is important
to pose the question in principle.

It can be assumed that the significance of the differences will be
dependent not only on the magnitude of the given trait but also on those
concrete relations through which this trait influences the life of the popu-
lation. In other words, significance of differences according to the ¢ value
may depend on the force of natural selection for the given trait.

The modern concept of intensity or force of selection does not represent
a regularly applied vector of constant magnitude (Schmalhausen, 1946).
It is already clear that during short time intervals, intensity of selection
will vary in relation to the changes in the state of the organism as part of
the population on the one hand, and of the whole population as a part of
the biogeocenose on the other. Practically, this means that the intensity
of selection changes even at different times of the day, not to mention times
in a year, and in different years, etc.

Another way of approaching the solution to the problem of biological
significance of differences, is the comparative analysis of not one but a
number of paired traits. Examples of such trends will be given in this work.
Here I will cite only one example taken from the work of F. Sumner (1909).
On the basis of the data given by this author, coefficients of variation of
body weights of male and female mice living for a long time in hot or cold
rooms were calculated. A comparison of values shows that, in three
instances, statistically significant results were obtained between females
in hot and cold rooms at the age of six months, and between males and
females of 2.5 months in one series, and in the other series at six months

6 See footnote 5 оп р. 7.

General Approaches to Variability Study 13

in a hot room (respective ¢ values 2.0, 2.1; and 2.7). However, the sum
total of the data obtained contradicts the conclusions about the actual
significant differences in the first group of these animals (already in seven-
month-old females there is no difference at all: t=0.22). All this leads
to the conclusion that the statistically significant results obtained may be
due to some unrecorded conditions of sampling and analysis of the material,
in different years, and in different experimental populations. But the
second conclusion, about the differences in variability between males and
females in the hot room, seems to be convincing, not because the differences
in all series are similar (variability is higher in males) and sufficiently high,
but because the same type of variability (increase in the variability of males
rather than variability of females) is observed in all series of the other room
also. A comparison of these data shows that differences in the variability
of males and females is much more significant in the hot room and less in
the cold room (but biologically significant, though not statistically).

_ While summarizing, it can be underlined that the value of ¢ in biological
experiments may serve only as orientation to the factual material obtained
by the researcher.

Optimum Size of Sample

The question of non-conformity of the “normal” curve with the specific
conditions present in natural populations, has been discussed above. From
this, a practical conclusion having a direct bearing on the collection of
material on variability of morphological structure in populations follows,
namely, a conclusion about the optimal size of sample.

Usually, in zoological works where statistical methods of investigation
are used, it is emphasized that for obtaining objective data, analysis of
sufficiently large series is necessary. The minimum number of individuals
is also usually indicated—25 to 30 or, better, 40 to 100 items (E. Mayr
et al., 1956; P. Е. Rokitskii, 1961; and others). Generally, there is по
mention of the maximum sample size. I. I. Schmalhausen (1935) observed
that 50 measurements are enough. Comparable indications are rarely
found in other works and this leads to the fact that usually the researchers
use hundreds of measurements, hoping to obtain the most exact value of
a simple arithmetic mean, of sigma, and of a coefficient of variation. Because
of this the volume of analytical work considerably increases, which acts
as a hindrance in the use of exact quantitative methods in biology.

What is the accuracy of data obtained from a small or large sample
from a population ? Does any optimum size of sample exist for the investi-
gations of variability; if so, would a numerically large sample be necessary
to obtain useful data, or would a small one be enough in order to avoid
extra analytical work? Let us analyze the data in Table 1 to answer these

14 Variability of Mammals

questions, where statistics characterizing the body length of the white whale
and the Greenland or harp seal are given with reference to sample size. |

Table 1. Comparison of statistics (length of body in cm) of large
and small samples from one population

Mature FEMALES OF GREENLAND SEAL
(harp seal) (Pagophilus groenlandicus)
White Sea (1961)

Mature Mates or WHITE )VHALE
(Delphinapterus leucas }
Okhotsk Sea (1958)

25 447.0-=5.04 5.64--0.81 25 183.2 + 1.42 3.86+0.55
50, 454.44 3.28 5.11+0.51 50, 185.8 -Е 1.07 4.08 -0.41
50, 446.5 + 3.62 5.75-Е0.58 50, 182.7 1.00 3.85 0.39
100 450.5+1.55 3.44+0.34 100 184.1-+0.75 4.08+0.29
200 181.7+0.57 4.4440.22 |
400 183.7-=0.40 4.35-=0.15 ©

The results of this comparison show that:

a) The estimated mean of the body length of the white whale did
not change when the sample changed from 25 to 100 individuals, i.e., during
a four-fold quantitative increase in the observations (and time spent on
the analysis of дада);

b) The value of coefficients of variation in the body length of the
white whale remains the same during three samples—25, 50,, and 50, and
the value of the sample of 100 specimens is statistically significantly lower
(C= 345):

c) The mean body length of the Greenland seal did not vary according
to data collected from 25 and 400 specimens, but shows significant statistical
change between samples of 200 to 400 specimens; and

d) The value of the coefficient of variation of the body length of the
Greenland seal practically does not change during all samples.

It seems possible to derive a general conclusion that for obtaining
sufficiently exact characterization of a population, the sample need not
exceed 50 specimens. ‘This conclusion, naturally, applies only to sufficiently
homogeneous samples (same sex, same stage of growth, obtained at the same
time from the population, and with random sampling), and is open to further
experimental confirmation.

Accuracy of Measurement and Individual
Errors of the Researchers

The question of optimum accuracy during sampling is of great import-
ance in the difficult investigations of large series of material and is ubiquitous

General Approaches to Variability Study 15

in the study of population variability.

As a result of many measurements of different organs and structures,
I arrived at an empirical rule. The measurements should be taken with an
accuracy equal to the individual error of measurements. ‘This individual
error can be determined easily as a result of certain successive measurements.
For example, during measurement of body length of one and the same seal
calf, the following data were obtained: 46.7, 48.3, 47.8 cm. It is clear that
the accuracy of measurement in the given case does not exceed 1.00 cm,
which is due to many reasons not completely controlled by the researcher,
1.е., similarity in stretching the measuring scale, position of hands or of
the animal in relation to the observer, degree of attention, the presence
of side disturbances, etc. All the remaining measurements in this sample
have to be done to the accuracy of 1.00 cm without calculating the tenths
of a centimeter.

Another example: During the measurements of the length of a skull,
the following results were obtained: 351.6, 349.8, 352.3 mm. Та this case,
it is not necessary to measure the tenths of a millimeter in further measure-
ments of skulls of the same type.

Such a conclusion is based not only on the greater simplicity and comfort
while accuracy remains maximal, but also because the measurements
taken with more accuracy are incapable of providing more accurate results
in the final analysis of the data.

It is well known from the investigations of taxonomists that one and
the same measurement, taken on one and the same material by different
individuals, may give different results (P. Jewell and P. Fullagar, 1966).
This fact must be taken into consideration during the study of population
variability. Special experiments conducted by Г. Sumner (1927) some
30-odd years ago have not lost their importance and must be remembered.
The first series of experiments were concerned with comparing the results
of measurements recorded by different investigators.

Three qualified zoologists measured one and the same series consisting
of ten characters of deermice (Peromyscus maniculatus). "The results of twice-
repeated measurements are given in Table 2.

As seen from the data, the differences between results obtained by
different investigators are pronounced and, in addition, it is also observable
that the individual peculiarities of each investigator’s measurements are like-
wise carried through. For example, investigator I obtained, аз а rule, measure-
ments of greater magnitude (six out of ten of his measurements are maxi-
mum) ; investigator III obtained measurements of average magnitude.’

7 It is important to note that in all these cases, the coefficient of variation of the trait
is absolutely the same for all the investigators; this again denotes the objectivity and useful-
ness of this index for comparison of heterogeneous data.

_ 16 Variability of Mammals

Table 2. Results of measurements (in mm) of the same series of Peromysvus
maniculatus by different investigators (Sumner, 1927; Table 1)

Tnvesuantor Body Length Body length Foot = Ear
; length of tail without tail length length
First day I 164.7 74.9 89.8 19.62 16.52-
Il 156.3 70.9 © 85.4 19.65 16.20
Ill 158.6 70.2 88.4 19.70 16.31
Second day I 163.9 74.4 89.5 19.59 16.12
п 163.7 72.2 91.5 19.30 15.70
Ш 164.1 71.1 93.0 20.00 15.94

In another series of experiments, F. Sumner (1927) showed clearly
how important it is to obtain maximum homogeneity in conditions while
obtaining factual data. ‘Table 3 shows how the time that elapsed from the
moment of death of an animal influences the value of measurements of its
body.

It is very clear that while collecting data from the same population,
even if one and the same investigator measures the animal 15 minutes after
death, and in another population after two or more hours, then, even though
measurements of the living animals were absolutely identical, the data
obtained will be different.

Table 3. Results of measurements (in mm) of the same series of mice at
different times after death (Sumner, 1927; Table 2, p. 181)
Time Total Tail Body Foot Kar Body
after death length length length length length weight

EXPERIMENT [

0 166.65 76.95 89.70 20.455 16.950 —
15 minutes 166.70 76.90 89.80 20.440 16.975 —
45 minutes 165.85 77.05 88.80 20.395 16.865 —

2 hours 164.10 75.80 88.30 20.425 16.845 —

р EXPERIMENT I]

0 165.05 72.75 92.30 19.695 17.175 235

26 hours 164.40 72.65 Qed 19.700 16.955 20.97

eee eee Oe eS Ee

Moreover, if some animals of the same population are measured
15 minutes after death and others after some hours and the values obtained
are combined, then the mean value thus obtained will not properly represent
the natural conditions even though each of the measurements carried out
was highly accurate and the individual error of measurement would not
affect the results. It is clear from the example given that the length of an
animal can be measured accurately to 1 to 2 mm and the length of the feet

General Approaches to Variability Study 17

and ears accurately to 0.1 mm; it is unnecessary to measure body weight
with accuracy over 0.5 g.

Another clear example showing the importance of reasonably accurate
measurements is given in the work of M. Alexander (1960). It was observed
that the skull measurements in one and the same population of muskrats
varied according to the time of the year. A significant difference was
observed in the value of measurements of skulls freshly obtained, those
recently obtained by museums, and those preserved for a long time in storage.
The main reason for such changes was also determined. Seasonal changes
altered the humidity level at different times of the year in the storage area
of the American Museum of Natural History. In the final manifestations,
these changes reached up to 1.5 percent and considerably increased the
individual error which, as was observed by us, is nearly 0.5 percent from
the value of the measurements (i.e., for linear measurements). The investi-
gator may derive wrong conclusions during a strict comparison if the
possibility of collecting wrong data is not considered when selecting skulls
to be measured at different times.

Looking at these examples as a whole, an empirical rule could be
formed: The accuracy of measurements should not be higher than possible
errors which may be due to the investigator and also related to the nature
of the material.

Coefficient of Variation as an Index for the
-“Population-Environment” System

The possibility of using different methods for estimating variability
from a formal viewpoint was discussed above with a view to deciding how
far a given method is capable of completely expressing the aspects of
variation. But while studying variability with an aim to understanding
the relationship of processes of variability to processes of micro-evolution,
a question arises inevitably about the extent to which one or another method
expresses the relations arising during the process of growth of a population.

Should the coefficient of variation act as an index indicating the
relationship within the “environment-population” system? А direct
answer to this question was not traced in the literature on variability. Toa
certain extent this can be explained because, as rightly noticed by G. D.
Polyakov (1961), in the majority of cases where variability of some traits
was investigated in zoological works, variability was mentioned only as an
additional character of the trait. Such a situation, of course, is unjust.
Variability, expressed by a coefficient of variation, can and must be used
in zoological works as a separate and important index; but for this, it
first has to be proved that there is a relationship between changes of this
coefficient and changes in the intensity of natural controlling factors.

18 Vanability of Mammals

That such a relationship exists can be proved by arranging data on
changes in coefficients of variation in the changing relationships between
the group of growing organisms and environment to show, for example:

a) The changes in the coefficient of variation in one population in
different seasons, and different years, under common ecological conditions;

b) The changes in the coefficient of variation for one and the same
trait in groups of different growth rates, each with the same relation to the
environment at each stage of growth; and

c) Constancy of coefficients of variation with homogeneous ecological
factors.

Of course, none of the given situations can serve as a final proof of a
relationship between the coefficient of variation and changes in the “‘environ-
ment-population” system, but on the whole, these data should provide
an answer to acceptance or non-acceptance of this coefficient for ecological
investigations of a population. |

Change of coefficient of variation
in different years and seasons

Considerable seasonal variation in coefficients of variation in iodine
concentration of the inner and subcutaneous fat of the muskrat (Ondatra
Zibethica) is known (Schwartz, 1960). А. У. Pokrovskii (1962) showed
clear seasonal change in the coefficients of variation in sexual maturity of
female lemmings (Lagurus lagurus). Analysis of data put forth by L. G.
Krotova (1962) indicates considerable seasonal variation in the coefficients
of variation for traits like absolute weight of adrenals, width of their cortex,
ascorbic acid content in the adrenals, glycogen in the liver, and sugar
in the blood of the water vole (Arvicola terrestris). Data on changes in
coefficients of variation for a number of traits in a population of moles
( Talpa europaea) studied for a number of years in different climatic conditions
are given in the work of K. Ya. Fateev (1962).

Weights of organs vary from year to year and these variations not
only attain high statistical significance (P<0.01) but also show a general
tendency toward reduction of magnitude in the coefficient of variation
from 1958 to 1960. Many examples other than those given above are avail-
able on the differing magnitudes of coefficients of variation for different
traits depending on season and year, and will be given in a large number in
the present work. This may, to my mind, provide a satisfactory basis for
belief in the existence of a specific relationship between magnitude of this
coefficient and interrelationships with the “population-environment”’ system.

Change in the coefficients of variation
for different age groups :
Data on coefficients of variation for tail length of mature and new-born

General Approaches to Variability Study 19

males and females of the Greenland (harp) seal (Pagophilus groenlandicus) found
in the White Sea in the Autumn of 1961 are given below. It seems that the
coefficient of variation for new-borns is somewhat higher than for mature
animals.

Mature New=-born tad-juv, р
Males 4.3+0.31 6.30.80 —2.29 < 0.02
Females 4.44-0.22 7.30.93 — 3.02 <0.01

Similar data on change in coefficients of variation in different age
groups can be found in the work of V. B. Scheffer (1962), where length and
weight of embryos and new-borns of fur seals (Callorhinus ursinus) have
been examined. Measurements and weights every ten days show that at
the end of the embryonic period, the coefficient of variation for length and
weight of body of embryo fur seals regularly declined (Figure 1). More-

over, these changes occurred

с.м. % + simultaneously in males and

females. ‘The latter point in-

creases the significance of this
conclusion.

Model data on changes
of coefficients of variation
with age are given in the
work of Ya. Hromada and
L. Strnad (1962) on variabi-
lity of weight and volume of
skulls for two types of mon-
keys (Macaca rhesus and M.
cynomolgus) ; similar data are
also given in the work of
We Se Siamianor (Gey) = iat

‚ studies of variability in skulls
of arctic foxes (Vulpes lagopus).
Regular changes of coefficients
of variation with age have
been observed in large num-
bers by H. D. King (1923)
during"’studies on the variabi-
lity of body weight in gray
rats, and by John Carmon

201 19.11 301! 181\/ New-born etal. (1963) for deermice

(Peromyscus polionotus) .

Figure 1. Magnitude of coefficient of variation of has
AaSs1ZE

(A) weight, and (B) body length of embryos and Ai в Sie eres :
new-born fur seals—(/) Males, and (2) Females 2 sufficiently clear possi ility
(Scheffer, 1962). that there exists a spccifie

20 Variability of Mammals

relationship between coefficients of variation and change of external
conditions in which a population propagates.

Constancy of coefficient of variation
in adaptively homogeneous traits

The conclusion about the direct relationship of a coefficient of variation
with the “environment-population”’ system is also confirmed by an analysis
of functionally homogeneous traits in a population: Van Valen (1962),
who studied variability of length and breadth of the molar teeth on the
upper and lower jaws of deermice (Peromyscus maniculatus), confirms this
observation on the great constancy of variability in these ecologically
homogeneous structures (Table 4).

Table 4. Variability (s.c.--s...,,) of measurements of Peromyscus
(Van Valen, 1962)

eC

Breadth of tooth
Length of tooth

Table 5. A comparison of coefficients of variation in the number of vibrissae
on the lip group in the Greenland seal and the ringed seal

(Mature females)
Greenland seal (Pagophilus
groenlandicus) Newfoundland

ее о о ое
Right Left t Right Left t

Ist 7.30.99 8.4+1.18 —0.7 8.10.95 9.91.20 —1.1
2nd 4.8-+0.65 5.9+0.81 —1.0 5.6+0.66 6.0--0.70 —0.1
Зга 5.90.81 6.7-Е0.91 —0.7 6.6+0.78 5.5-+0.64 0.8
4th 9.4+ 1.28 6.7-+0.92 1.7 7.60.90 7.10.84 0.4
5th 11.7+1.59 8.4+ 1.14 + 1.7 11.0-Е 1.30 10.7 1.30 0.1
6th 17.62.39 15.22.11 +0.8 17.0--2.00 19.4+-2.29 —0.8
7th 37.3+5.07 35.4+4.91 +0.3 41.2+4.90 37.7+4.40 —0.5
8th 168.2+22.9 184.2+425.1 —0.5 533.3+62.8 65.0: 7.7 7.3
Total 4.9+-0.66 4.8 +-0.67 +0.01 i. 6.00.70 a Teno 40.0.

General Approaches to Variability Study 21

UA comparison. of coefficients of variation in the number of vibrissae
on the lips of the Greenland seal, from the left and right sides, serves as a
second example (Table 5, and Figure 2).

As seen from the data given, the two types of pinnipeds do not differ
in the variation of right and left vibrissae. The 8th line is an exception

in the ringed seal, but even this single exception can be well explained.
The absolute number of whiskers in the 8th line is small—on an average

only 0.03 on the right and 0.18 on the left. With such very uneven

Figure 2. Snout of a young Greenland seal. The location of lip,
nasal, and eye vibrissae is seen clearly. Author’s photograph.

distribution (more than 80 percent do not have vibrissae at all in the 8th
line), the statistical indices used by us in this instance cannot positively
represent the existing condition because distribution of this trait does
not fall under the “normal” distribution. On the whole, the data cited
specifically emphasize that for functionally homogeneous structures in an
organism, the coefficients of variation seem similar.

22 Variability of Mammals

And so, all the three ways examined by us—analysis of coefficients
of variation for functionally homogeneous structures, coefficients of variation
in one population in different seasons and years, and coefficients of varia-
tion in different age groups—permit one to assume a specific relationship
between the magnitude of coefficients of variation and evolutionary factors
functioning in nature. This conclusion allows us further to look into the
coefficient of variation as a sufficiently promising index of relationship in
the “‘environment-population” system.

CHAPTER П

THE NATURE OF VARIABILITY IN DIFFERENT
SYSTEMS IN MAMMALS

In the world literature on the study of mammals, there is a large
quantity of data on the variability of organs and structures. However,
these data are widely scattered, which limits their use for comparison with
newly obtained material. On the other hand, the present need for knowing
the nature of the variability of different organs and systems as a whole
requires us to carry out a broad comparison of all systems and organs among
mammals.

Until recently, no attempts were made to systematize the data on
the variability of mammals or to discover the nature of variability in organs
of different systems of these animals. The work of 5. S. Schwartz (1960)
is an exception; here considerable factual material on the variability of
weight of internal organs (so-called interior traits) for many types of
amphibians, reptiles, birds, and mammals has been examined from an
ecological point of view. Unfortunately, use of the coefficient of variation
for relative values of traits (rather than absolute ones) makes it difficult
to compare these data directly with data on other types of mammals and
also on other systems of organs. :

It is generally known that the variability of functional hard parts of
skeletons in mammals lies between 3 and 10% (Simpson and Roe, 1939).
In 1960, 5. 5. Schwartz showed that variability of “interior traits” (mainly
relative weights of internal organs), though somewhat higher, does not
exceed an average of 10 to 20%.

On the basis of extensive data on the structure of certain skull traits of
horses in the evolutionary line Ayracotherium-Mesohippus-Merychippus-
Neohipparion, С. С. Simpson (1948) concluded that ‘“. . . variation

24 Variability of Mammals

differs surprisingly little from time to time for one trait within a group,
for homologous variates in different groups, and for different analogous
variates” (p. 135; 1944, p. 83. Data on variability of skeleton are also
on the same page). It is possible that within such geological intervals,
during millions and millions of years, variability of the structures of horse
skulls changed insignificantly; but, while studying now living populations,
it has to be conceded that variability of even one and the same trait of small
mammals may change very significantly during some months. But in any
case, this question concerning magnitude of variability is not very clear—as
can be concluded from the quotation from Simpson above.

On the basis of a number of small experiments, В. 5. Bader (1955),
and later Ya. Ya. Roginskii (1959), came to the conclusion that the extent
of difference in variation among non-homologous traits within a species is
higher than the difference of variability among homologous traits between
species; in other words, the difference in the variability of different
organs within a species is higher than the difference in variability of one
and the same organ for other species. But this conclusion was mainly
based upon measurements of body and skeleton. None of the studies known
to me answers the question of whether there is a specific variability for
each large system of organs in mammals. This chapter will attempt to find
an answer to this question on the basis of exact and factual data. To
accomplish that end, every effort has been made to present the data available
on variability of structures of different types of mammals in the form of
coefficients of variation—the most convenient form for comparing diverse
data.

These factual data include:

1. The author’s data obtained specifically at the time of conducting
these investigations (mainly on cetaceans and pinnipeds);

2. The raw data obtained from other researchers (based on variable
populations) and further analyzed by us for standard deviations, means,
standard errors, coefficients of variation and their errors

(FES., 0, С. Е

3. Data obtained аз a result of previous data in the literature, with
analysis of coefficient of variation and error (с. v.+5¢.».), either on the
basis of coefficients of variation already calculated or on the basis of standard
deviations and mean arithmetical errors estimated by the authors.

Any review of this nature would invariably be incomplete. It is
practically impossible to collect the large quantity of work already pub-
lished and still being published on a great scale, on the variability of
different organs.

But in spite of the incompleteness of the data cited, it is proposed
to start such abstracting work now because no such review exists in the

Nature of Variability in Mammals 25

world literature. In order to attain strict systematization, the data
presented herein on variability have been arranged in the following order:
Marsupialia, Cetacea, Carnivora, Pinnipedia, Perissodactyla, Artiodactyla,
and the rest of the orders of mammals.

Integument

Hair

As shown from the data analyzed by E. Z. Kogteva (1963) (Table 2
in the Appendix), the coefficient of variation for the length of summer and
winter fur on different body parts of moles varies from 2.1 to 6.2%. The
variability of summer fur is somewhat higher on the middle back but
variability of winter fur is somewhat higher in the sacral region.. A com-
parison of winter and summer fur shows that whereas the variability of
guard hair, medullated hair, and underfur on the middle back varies
typically in winter and summer (during winter the variability is half as
much), there is little variability of length of different types of hair on the
sacrum.

The variability of summer fur on the sacral region and the shoulder
(scapular) region is higher in summer than in winter. The results show
that in winter the variability of length in all types of hair on various parts’
of the body is, on an average, almost the same even for different types of
hair (adaptive significance ?).

The variability of thickness of hair on different parts of the body
of a seal in different seasons fluctuates from 0.5 (medullated, winter) to
3.1% (underfur, summer). The maximum variability for thickness in all
seasons is characteristic of underfur.

In the work of A. Banerjee (1963) data are given on the variability
of diameter and thickness of human hair. More than 40 series of measure-
ments carried out on the different external traits of hair (straight, curly,
soft, hard, silky, etc.) of different peoples in India, showed that the coeffi-
cient of variation is not directly related to any of the specific characteristics
of hair and varies from 5.8 to 19.8%. (Most values are from 10 to 15%.)

In the work of E. Z. Kogteva (1963) mentioned above, there are data
on the thickness and length of hair of the varying hare for different seasons
of the year. Analysis of these data (Table 2 in the Appendix) shows that
the magnitude of variation of thickness of hair on the sides has characteris-
tically lower values—to 0.6%.

The comparison of coefficients of variation for thickness of hair on
the back, on the sacrum, and in the belly regions in different seasons of
the year shows that the variability of thickness of hair on the back differs
from the variability of hair thickness on the rump and belly (the latter
two being alike in this trait), In some instances the variability of thickness

26 Variability of Mammals

of hair on the back (for medullated hair of I and II categories) is higher;
in others, it is less (for other medullated hair) than the corresponding
values for the rump and belly. ‘The coefficient of variation for thickness of
hair on the back shows large seasonal differences (in three types of medullated
hair) but the variability of hair thickness in the belly region does not differ
by season. ‘These conclusions are confirmed on general lines by the
analysis of data on variability of thickness of hair for varying hare.
Variability of all types of hair on the back is considerably lower on an
average for this trait (2.9%; the range of variation being from 0.6 to
10.9%) than on the sacrum (4.0% on an average, with a range of 1.4 to
9.3%), and on the belly (4.2% average, ranging from 1.8 to 10.1%); the
hair in the border region of the back is different in length from that on
the sacrum and belly. Variability characteristics of these hairs are similar
to each other.

The observations of E. Z. Kogteva (1963) on the variability of length
of hair for different types of summer and winter fur on different body parts
of the varying hare are interesting. There are no overall differences between
different seasons for this trait and the slight differences (regarding length
of directed hair on the rump, length of medullated hair of III category)
can be explained, it appears, on the basis of improper collection of material.
From this example, it can once again be seen that perfection of data on
biological material is not possible through statistical analysis. А mani-
festation of overall trends in the character of structures is immensely more
important than information on their various differences. Because we
lack full details about the collection of the material, many additional factors
which may not have been considered during these morphological investiga-
tions (i.e., time differences in procuring experimental animals, separate
regions and micro-regions of their habitat, absence of information on age
and sex of the populations, etc.), can be presumed to have affected the
data obtained by E. Z. Kogteva (1963). In such cases, many rash con-
clusions based on formal interpretation of factual data can be avoided;
these, in fact, indicate the insufficient “cleanness” of the sampling.

A considerable difference between characteristics of variation of hairs
in winter and summer fur can be noted from the analysis of variability of
length of different hair types on the varying hare. It seems that out of 18
comparisons between different types of hair on different body parts during
the summer, there are no differences in the magnitude of variability in
10 cases. Yet, while comparing the same traits in winter, the number of
clearly defined differences considerably increases. ‘Therefore, it is possible
to make a preliminary conclusion about great differences in coefficients of
variation for length of different types of hair in the winter period. This
conclusion does not confirm the results of investigations on the fur of moles
(see above) but the differences in the ecological peculiarities of these species

Nature of Variability in Mammals 27

must be taken into consideration.

Finally, in the analysis of variability of hair length of different types,
the considerable stability of variability of medullated fibers of I category
and a lower variability of II category of underfur, needs special mention.

Analysis of data put forth by F. Sumner (1909) on mice shows that
the magnitude of the coefficient of variation for the number of hair per
unit area (11.7-17.7%) is considerably lower than the variability of weight
of hair per unit area (23.6-26.8%). The magnitude of variability is such
that a formal determination of differences in variability of animals main-
tained in hot or cold rooms, and likewise between males and females,
cannot be done.

However, an ostensible trend to decreased variability for these para-
meters and a somewhat increased variability for number of hair in females
can be observed in hot rooms. An analysis of the relative value of the
trait—weight of hair in grams/(body length)*—shows the same trend,
that is, the variability is lower in the hot room. ‘The relative variability
occupies an intermediate position (19.5-21.7%) between variability of
number of hair and their absolute weight in respect to the magnitude of
the coefficient of variation.

The data~of R. Huestis (1925) on variability of length of deermice
_ (Peromyscus) show that variability of this trait varies in different subtypes
and their hybrids from 5.5 to 7.0%; the variability of differently colored
portions of hair, from 8.7 to 16.5%. However, the percentage variability
of the number of hair for different types is high.

Hair type Range of variability
A 4..5 to 10.4
B 15.8 to 47.0
C 24.6 to 62.4
D „25.5 to 59.5

The increased variability of number of hair for the last two types
can be easily explained by the elementary relations between magnitude of
a trait and its variability (see Chapter V); hair of these types is very sparse.

Data obtained on the basis of material analyzed from the work of
Е. У. Fadeev (1958) show that in the coypu (Myocastor), the variability
of all types of hair in animals of different growth stages is very similar,
beginning with 1.5 months of growth; it varies from 3 to 12.6% with an
average of about 5 to 6%. (All observations were recorded in the beginning
of December.) The variability for thickness of hair in the same samples is
less pronounced: For the directed and medullated hair of I and IV
categories, it does not go above 4.5% (average about 2%) but for the
medullated hair of V category, it goes up to 10.5%, and for underfur up
6515:7% (Таые 6).

28 Variability of Mammals

Data on variability of parameters of hair of Carnivora obtained from
Kogteva’s work (1963) referred to above, are given in Table 2 of the
Appendix. The coefficient of variation of the raccoon-dog (Nyctereutes
procyonoides) varies from 1.3 to 10% (average, about 5%) and thickness of
hair of different types varies from 0.75 to 11.6% (average, about 3.5%).
Variability of hair on different parts of the body differs: variability of length
of medullated fibers of categories II to ТУ and underfur, and hair thickness
in all underfur and medullated fibers of categories II to III, is different.

Table 6. Variability of length and thickness of hair of Myocastor coypus
(Calculated from the data of Fadeev, 1958)

c. и. of hair length с. и. of hair thickness
ее Ng ce eet fede ee EY Cee ae by ae NN Veter Е APMED ey EDU ИВ И ИИ
40 Days 120 Days 7 Months 40Days 120 Days 7 Months

Guard hair 3.8 Tie 12.6 Lat 1.5 1.1
Medullated I 5.6 10.7 3.6 > 1.8 1.7 Ц:
II 6.6 6.2 4..2 13 1.3 21

ш 8.0 3.0 8.3 3.3 2.5 2.9

IV 4.1 5-7 4515) 4.2 3.5 4.5

У 9.0 5.1 6.2 5 10.5 6.3

Underfur 9.0 12.0 7.3 11.5 5.8 15.7

дж

The results obtained from an analysis of data Бу М. A. Gerasimova
(1958) on variability of hair measurements for sables, and the variability
of length for all types of hair (guard, four categories of medullated, and
underfur) is, on the average, about 5 to 7% (less in the guard hair and
more in the underfur). Variability of thickness of hair for these same
sables is considerably high; on the average, about 17% (maximum for
categories III and ТУ of medullated hair). It is interesting that in popula-
tions transferred to a new environment (from the Barguzin reserve forest
in the Tomsk region), the variability of length for almost all types of hair
increased a little, whereas the variability of thickness did not show such a
regular change.

Variability of wool weight of sheep (yearly clip) on an average was
17.5% from eight different farms (Popova, 1941). According to these
data, variability of length of underfur of sheep averaged 31.3% (range,
20.6 to 49.4%); variability of medullated hair averaged 17.7% (range,
10.0 to 23.3%). It is seen that these results coincide with the observations
of Е. Sumner (1909) and В. Huestis (1925), and are considerably higher
than similar observations on wild mammals investigated by E. Z. Kogteva
(1963). ;

In the same work, E. T. Popova (1941) presents detailed data (based
on an examination of several thousand sheep of different breeds) on a

Nature of Variability in Mammals 29

microscopical analysis of the fineness of different sorts of wool in crossbreeds
of Lincoln sheep and local ones. Analysis of these data shows that the
coefficient of variation for hair thickness of sheep is very much higher than
the measurements obtained for hair thickness of seals and hares by E. Z.
Kogteva (1963). Variability of thickness (fineness) of wool for crossbreeds
averaged 40.9% (range, 26.1 to 50.4%), and variability of the same trait
for local sheep was 52.3% (range, 48.4 to 59.7%).

It would appear that the given data are enough to estimate general
outlines of variability in some parameters of the fur.

Vibrissae

Variability of other hair, like vibrissae, is directly related to the
variability of parameters of the fur. The number of vibrissae on different
mammals has not been studied enough, though, as shown by our work,
vibrissaé could prove to be very sensitive indicators in studies of certain
natural populations. As shown by the investigations of A. V. Yablokov
(1962, 1963a, 1964a), A. V. Yablokov and G. A. Klevezal’ (1964), G. A.
Fedoseev and A. V. Yablokov (1965), the variability of lip, nose, and eye
vibrissae differ characteristically.

On the average, the least variability in number of hair is found in
the lip region, about 5 to 6% (range, 2.7-10.1% for eight types of Pinnipedia).
Two other regions—eyes and nose—have rather similar variability; for
eye hair, the average is 15-20% (range, 1.6 to 37.8%); for nose hair, the
average is about 20% (range, 13.7 to 78%) (Figure 3).

Variability less than 10% is characteristic for all the basic lines of
vibrissae in the lip group, but in the last lines this variability may increase
up to great magnitudes (500-1,000%).1

As the data on variability of vibrissae on different parts of the head
for certain baleen whales (Figure 4) show, the coefficients of variation for
number of vibrissae in three different groups (side of lower jaw, front of
lower jaw, and upper jaw) are rather similar—20.6 to 32.4%.

Color

It will not be out of place to examine certain aspé¢cts of variability
of normal coloration in this section on the structure of the integument
(primarily, just those traits of color which can be quantitatively estimated,
because quantitative variability of coloration and the phenomenon of
epigenetic polymorphism are described in Chapter IV).

1 п a number of cases no variability of nasal hair was observed (с. v. equals 0%).
Hence, the nasal hair and the number of hair in the last line of the lip group are characteris-
’ tic not of the Gaussian distribution, but of Poisson’s distribution; and so, the peculiarities
of variability for these structures in very general terms could be satisfactorily approximated
by using distribution-free methods of analysis on the material.

30 Variability of Mammals

It was determined by the works of S. 5. Schwartz and his co-workers
(Pokrovskii, Smirnov and Schwartz, 1962) that color, along with other

Figure 3. Caspian seal on board a ship. Functional zones of nasal and
eye vibrissae are seen. Author’s photograph.

morphological traits, may be used
successfully for an analysis of vari-
ability both among species and
among subspecies. ‘The index of
variability for the degree of reflec-
tion in red light (measured by
photometer) in 13 series of two
subtypes, was about 3% (range,
2.0 ю 3.7%). Another” ада
coloration, the index of reflection
in white light (whiteness), seems
to be considerably variable—ave-
rage 19.8% (range, 14.9 to 29.9%).

Substantial material on vari-
ability of the same traits in the
red-backed and bank vole (Clethri-
onomys тии; and C. glareolus)
(Bol’shakov, 1962) reveals an ana-
logous pattern. The variability
index of color ranges from 3.3 to
3.6%, and variability of “‘white-

Figure 4. Fin whale. Group of vibrissae at
the end of lower jaw as seen from below.
Kuril Islands (1954). Author’s photograph. ness” from 8.7 to 13.8%.

Nature of Variability in Mammals 31

An analysis of colors for two subspecies of deermice carried out by
Е. Sumner (1924) is interesting. The coloration of the wild populations
was studied under natural conditions and after 7 to 12 generations, and the
color of two other populations was studied in the habitat of a third sub-
species. This experiment offered a possibility of excluding specific biotic
components of natural selection. On the whole, it seems that the variation
in color of each subspecies is stable in the new surroundings even though
a specific change takes place in it. In all cases, variability of different
types of coloration seems to be very slight: black color, 1.4-2.9; white color,
5.9-10%; ratio of red to green, 7.7-18.0%.

In another work, F. Sumner (1923) presents absolute values of
magnitude of variability of coloration in populations of deermice and their
offspring in the first and second generations. From data of ten different
experimental populations it is seen that each color trait has its own para-
meters of variability.

Black has the least variability (average, 2.2%); white comes next
(average, 10.1%); then the ratio of red color to green (average, 14.0%);
the average for all color variations taken together is 19.9%.

These observations of F. Sumner substantiate his own data (1926)
on variability of color of other species of the genus Peromyscus. In any
type, body colors and the coefficients of variation for different traits are
different. Red color varies among individuals of three subspecies from 7.6
to 9.2%, green from 5.3 to 12.1%, and the ratio of these colors varies from
8.0 to 10.8%.

On certain body parts of mammals, measurement of the area which is
covered by one ог the other color seems possible. For example, it is
possible to measure the width of white spots on the tails of rodents. The
coefficient of variation for this trait of Peromyscus maniculatus, estimated
from Sumner’s data (1924), is not large—12.9 to 14.6%.

Variation in the length of color regions of different types of hairs on
agouti-colored mice differs for different types of hair; straight hair 15

25.2 to 29.3% but zigzag hair is 3.8 to 13.3% (Table 7).

Table 7. Variability of size of yellow spots on different types of hair from
different body parts of male mice (Original data from D. Galbraith, 1964) *

Shoulder Middle of back Sacrum

ZIGZAG HairR

0.76+0.01 11.8-+0.9 0.59--0.003 3.80.31 0.60--0.01 ЗЕ
SrraicHuT Hair
1.01 -0.03 25.2--2.1 0.87 0.03 29.3 +-2.4 0.91 --0.03 28.0--2.3

* In this work, additional data on five different types of hair for males and females are given.

32 Variability of Mammals

A simple method of quantitative estimation of coloration, subjectively
dividing all degrees of appearance of different numbers of spots on the skin
of P. maniculatus into four groups (from 0 to 3) has been proposed by L. Dice
(1938). The data obtained on the mean value of variability in this trait
for eight different populations from Arizona show that it strongly varies
between 14.9 and 72%, averaging about 36% (Table 53).

In the same work, the author compares the variability of colors
determined by photometric methods; moreover he has done so not for the
whole skin as Е. Sumner, S. S. Schwartz, and others have, but separately
for the sides, chest, and back. This undoubtedly represents the natural
conditions more exactly (Table 8).

Table 8. Coefficient of variation (с. v.--5¢,,,) of photometric indices _
of body color of P. maniculatus (Estimated from data of Dice, 1938)

Population Red Yellow Green Pale-blue Violet-blue
Back
J 20.8- 1.4 22.3-Е1.5 18.8 1.3 23.1+1.6 22.6+1.5
II 17.4£1.8 20.5+2.1 17.6+1.8 18.8-41.9 21.8+2.2
Ш 18.2-- 1.0 19.7: 1.1 20.8+1.1 21.6-+1.2 23.2+1.2
IV 21.1-=:1.7 20.4-+41.7 22.3--1.8 22.5+1.9 22.2--1.8
№ 26.2-+2.4 25.2+2.3 30.9+2.9 30.8--2.8 31.1+2.9
VI 24.8+3.8 25.443.9 28.6+4.4 29.9-+4.6 32.1+5.0
УП 24,1+1.9 24.9+2.0 28.1--2.2 30.5-L2.4 31.4+42.5
VIII 7.9+1.2 12.7 1.9 12.1--1.8 15.5--2.3 18.0--2.6
SIDES
I 12.3+0.8 13.2-+0.9 13.0+0:9 -17.2+1.2 17.4+1.2
II 13.2+1.3 13.7+1.4 13.3-Е 1.3 16.0-+ 1.6 16.4-+1.7
III 13.2--0.7 12.7 0.7 15.2+0.8 15.0--0.8 15.3-+40.8
IV 12.1+1.0 11.9-+1.0 14.7+1.2 13.8-+1.1 14.7 1.2.
У 16.6 1.5 16.3--1.5 18.3-Е1.7 21.3--2.0 26.1--2.4
УТ 13.8+2.1 14.6-.2.3 13.4--2.1 19.0+2.9 19.2+-3.0
УП 12.85 1.0 14.8 1.16 15.9 1.3 20.5 1.6 20.4 1.6
VIII 11.0-+1.6 11.3+1.7 13.8+2.0 14,.9+2.2 15.6+2.3

An analysis of these data shows that the magnitude of the coefficient
of variation closely approaches the means for variability of colors in
other mammals—11 to 32%. Comparing the variability of all populations
reveals that some populations have comparatively low indices of variability
(12.7 to 12.2%) whereas the variability of coloration in other populations
is higher (24.1 to 36.1%). On the whole, it appears that the variability
of back color is considerably higher than the variability of color
on the sides of the body, and that variability of red, yellow, and
green shades is lower than variability of pale-blue and violet- blue colors.
Without knowing the living conditions of the animal, it is not possible to

Nature of Variability in Mammals 33

understand the adaptive value of such differences, but there is no doubt that
an analysis of variability, as in the given case, would help an ecological
analysis carried out by the same author.

By using the same method of differential estimation of coloration on
different body parts with the photometric method, D. Hayne (1950)
obtained data on the absolute variability for red, green, and bluish-violet
shades in the color of the back for nine populations of Peromyscus polionotus
from Florida and Alabama. ‘The results of analyzing data show that
variability of red. varies in different populations from 16.5 to 66.2%;
variability of green, from 11.1 to 65.3%; and variability of bluish-violet,
from 17.2 to 67.9%. In other words, no significant differences in the
variability of different shades of colors were observed. But it is clear
from the investigations of L. Dice (1938) that a particular magnitude of
variability is characteristic for each population taken separately; thus
the traits of the ninth population are highly variable (65.3 to 67.9%)
while the second population is the least variable (11.1 to 21.2%).

R. Boolootian (1954) presents data on the percent of red pigment in
the color of three subspecies of kangaroo rats (Dipodomys nitratoides). It |
is observed that the variability of quantity of red pigment in different
populations is slight—on an average, 9.9% (8.7 to 11.5%). The variability
of color of feet in deermice seems to be much greater. Calculations made
from the data of F. Sumner (1924) show that variability of this trait in one
subspecies of P. maniculatus is 35.4 to 39.2%, and in another 75.5 to 77.9%.

To conclude this section on the variability of coloration, the data
obtained can be summarized as follows: Notwithstanding the comparatively
high indices of variability of color (up to 60-70%), these indices have
characteristically quite a low mean variability—to the level of 10 to 20%.
At the same time, it is very important that the nature of the variability
of coloration seems to be inherent in the given population as a specific
trait; and for the magnitude of variability of color within the species,
gradations can be discovered from populations with lower values of
variability (5-15%) up to populations with maximum variability (60-
70%). Such behavior in coefficients of variation of color (perhaps deter-
mined by the influence of controlling factors of evolution) makes the study
of variability in color shade a very promising method for the study of finer
differences within a species (see also Chapter V).

Variability of colors expressed not only by shades but by the magnitude
of area occupied by a particular color field (3.8-29.3%) and the number
of spots on a body surface (on an average, 35%) are characterized by
clustered values.

Skin
Data on the magnitude of variability for skin thickness, thickness of

34 Variability of Mammals

epidermis, thickness of the stratum corneum of the epidermis, and thickness
of the reticular layer of skin in sheep of various breeds are included in the
detailed work of M. Tumurzhav (1964). The coefficient of variation for
thickness of the epidermis in six flocks of sheep in Mongolia differs a little
for new-born, six-month-old, and mature animals: new-born, 13.5-18.8%;
mature, 9.07-23.65%. The same picture emerges when comparing
the whole skin: new-born, 5.3-9.2%; six-month-old, 7.9-11.8%; and
more mature, 6.8-10.1%.

The coefficient of variation in thickness of the stratum corneum varies
from 4.3 to 18.7%. But the variability of the reticular layer is somewhat
higher: new-born, 19.2-24.2%; six-month-old, 11.1-20.8%; and mature
ewes, 18.7-27.1%.

These data can hardly be taken as characteristic for mammals. The
possibility remains that these traits of domestic animals differ considerably
from similar animals under natural conditions. However, attention is
drawn to the fact of a considerable constancy in coefficients of variation
in the samples of different age groups, and to the possible use of such aspects
of fine histological structures for characterizing variability of the skin.

The general weight of skin may also serve as a good parameter for
this system. It appears that variations in the variability of skin weight
for the northern fur seal (Callorhinus ursinus) are insignificant for males at
the age of three and four years. The variability within 13 selected animals
of this age (but of different body lengths) varied only from 6.5 to 13.6%
(Scheffer, 1962).

In a different species of marine mammal, the white whale (Delphinap-
terus leucas), variability of skin weight is 14.8%. It is interesting to note
that the variability of skin weight in these two cases seems to be very low,
whereas the reader may remember that traits concerning weights normally
have a very high variability.

Skin derivatives

Original data on the number of plates of the nine-banded armadillo
(Dasypus novemcinctus), are given in the work of Newman (1913). The
variability of this trait in four groups, based on age and sex, was surprisingly
constant:

Xo Sz otSe С. U.ESe.y, n
Adult females 559.0 1.26 14.0+0.89 2.50+0.23 56
Males, their offspring 558.6+0.65 15.4-40.48 2.7440.13 224
Adult females 559.8+. 1.34 15.30.95 2.745 0.25 56
Females, their offspring 559.1 0.57 13.0-Е0.41 2.744 0.13 224

It is known that in this species of armadillo the four offspring are
always monozygous quadruplets.

To study variability of traits in such

Nature of Variability in Mammals 35

cases gives a unique opportunity for evaluating genetical components of a
common. phenotypic variability.

The number of rings of scales on the tails of mice appears to be a
satisfactory trait for genetical investigations (Law, 1938). Calculations
show that the variability of this trait in 28 samples from different populations
varies from 2.2 to 17.8%. It is interesting to note that some populations
clearly differ from others in the magnitude of variability, as in the analysis
of certain color traits. Lower values of variability in the number of rings
on the tail are characteristic of one line (2.2-3.0%) in four series, and
comparatively higher values of variability are characteristic of another
line (10.8-17.8%). Very homogeneous values of the coefficient of
variation in the number of rings on the tail were observed by us for two
populations of coypu (Myocastor coypus) which had acclimatized for two
years in the Caucasus and Central Asia—from 5.9 to 8.6 (for four series)
(Yablokov and Etin, 1968).

It is well known that baleen plates of whales grow as derivatives of the
integument. Variability in the number of fringes on a unit length of the
edge of a baleen plate is similar in different species, 18.1-26.9°% (see
Table 50). Finally, data on variability in the number of dermatoglyphic
ridges are available for man—35.4 to 43.2% (Van Valen, 1963); and
moreover, they are highly correlated with simple interdependence (see
Chapter V). Variability of this trait in males is somewhat less than that
in females.

General conclusion on the variability of integument

Looking at the variability of different organs, and the characteristics
of the traits of the integumentary system, a typical and wide degree of
variability in different structures is noticed. It could be said with confidence
that the variability of parameters of hair in the majority of mammals is
5 to 10%, while the variability of shades of colors determined by photo-
metrical methods averages 10 to 20%. Variability of histological para-
meters of the skin has typically similar figures. But on the other hand,
behind these mean values, the possibility exists of obtaining surprisingly
homogeneous and low variability indices as well as comparatively higher
variability.

It is important that the variability of almost all the traits of skin
structures are characterized by different values in different populations.
This means that the traits related to the skin could be important indicators
in studying variability of environment and genetics for all kinds of popula-
tions. And, on the whole, it can undoubtedly be said that the study of
quantitative morphological peculiarities of the integumentary system is
one field in the general study of population morphology of mammals which
is still very little explored.

36 Variability of Mammals

Skeleton
Skull

It was necessary to select some homogeneous material which could
be used for a comparison with unrelated species from a great number of
studies devoted to the structure of mammalian skulls.

The data obtained are given according to systemic traits in Tables
4 and 5 in the Appendix. It appears that in the majority of mammalian
orders the following measurements have usually been used in the study of
the variability of the skull; condylobasal length of the skull; length of
mastoid; width of the skull at the mastoid and jugal; width between the
eyes; width of rostrum; length of palate; height of skull; length of tooth
row (usually upper) ; and length of mandible. Certain data obtained or cited
in the tables are based on investigations of a large number of populations
within one species; for example, the data of F. Clark (1940) on peromyscus
concerning variability in a number of structures for 30 different populations;
the data of D. Hoffmeister (1951) on 14 populations of the same species;
the data of B. Kurtén and R. Rausch (1959) on 26 different samples of
the grizzly bear (Ursus arctos); the data of V. N. Pavlinin (1962, 1963)
on variability for 11 different geographical and sex groups of sables
(Martes zibellina); the data of V. N. Bol’shakov and S. S. Schwartz (1962)
and S. 5. Schwartz (1962, 1962a) on variability for 10 or more populations
of bank voles (Clethrionomys glareolus), lemmings (Lagurus lagurus), etc.

All the data collected comprise 57 different species; they are concerned
with more than 370 different individual samples (according to age, sex,
geography, etc.) and have been obtained from the following 9 mammalian
orders: Insectivora, Chiroptera, Ungulata, Primates, Lagomorpha, Cetacea,
Carnivora, Pinnipedia, and Rodentia. It is thought that these data as a
whole are sufficient to characterize precisely the nature of the variability
in linear traits of skulls in all groups of mammals.

As seen from the data in Tables 4b and 5 in the Appendix, the varia-
bility of traits of the skull in such orders as Insectivora, Ungulata, and
Chiroptera has a wide range. Comparative data are presented in Tables
4a and 4b. For the class as а whole, the variability of condylobasal length
varies from 0.4 to 6.8%; the mastoid width from 1 to 8.3%; width of
molars, 0.4 to 9.6%; width between eyes, 1.5 to 8.3% ; width of snout, 0.9
to 10.6%; length of tooth row, 1.0 to 10.6%, etc. (Table 5, Appendix).?

2 The data in the Table show that in roughly eight cases of different traits, the variability
was very high, up to 41%. Possibly in these samples errors of calculation have been made,
or the samples were made without due consideration for the necessary requirements. It is
also quite possible that in individual cases the variability of skeletal features can be quite
high. In any case, strictly speaking, we do not have as yet data that preclude such an assump-
tion regarding the possibility of a sharp rise in variability in individual populations under
special conditions of evolutionary factors.

Nature of Variability in Mammals 37

Data for all parameters of variability for all the orders studied, are
summarized in Table 5. It seems that the overall variability of all the
given traits of mammals varies from 0.2 to 12.0%, averaging about 2 to
5%. The magnitude of the linear variability of the skull as determined by
the given data is not related to the size of species nor to its systemic position,
and the range of variability is similar for all the traits of the skull under study.

There are no differences in the range of variability of condylobasal
length, the length of the lower jaw, and the length of the palate. How-
ever differential analysis of variability of different traits is possible if fairly
uniform series of samples, taken from one order or one species, are
available. The data of J. Dynowski (1963) on the skull of the house-
mouse (Mus musculus) show that for the three groups of mature males and
females studied, the least variability occurs for braincase width, width
between eyes, and length of cranial cavity. The most variable traits were
length of palate and condylobasal length of skull.

According to the data of F. Clark (1940) the width of skull in
Peromyscus maniculatus seems to be the least variable measurement; the
most variable was the length of lower jaw. The length of lower jaw
happens to be the most variable trait also for two other species of Pero-
myscus which were studied by F. Clark namely, P. leucopus (eight popula-
tions) and P. eremicus (four populations). However, the length of the skull
seems to be the least variable for P. leucopus, while in P. eremicus (like P.
maniculatus), the width of the skull varies least. Thus it can be said
that width of skull is usually least variable and length of lower jaw most
variable in the genus Peromyscus.

The data of T. Peshev and С. Georgiev (1961) show that Юг ten
populations of field mice (Apodemus sylvaticus) the variability of condylobasal
length of skull is somewhat lower than the width at the jugals. Z.
Gebczynska (1964) shows that in four out of six groups of mature field
voles (Microtus agrestis) the variability of condylobasal length is more than
the variability of width of skull. On the other hand, S. S. Schwartz
(1962) finds that the variability of the condylobasal length in Lagurus
lagurus seems very small. Likewise, У. М. Pavlinin (1962) observed that
low variability of condylobasal length was characteristic of eight samples
of sables; the most variable traits of the measurements for these samples
of sables were length of tooth row and width between eyes.

Data on the variability of Pinnipedia in 17 samples of skulls of different
species confirm the known facts about the least variability of the condylo-
basal length (Table 4c, Appendix). After this come (in ascending order
of coefficients of variation) braincase width, length of lower jaw, height
of skull, length of palate, length of upper tooth row, width of snout, and
width between eyes.

According to the work of V. A, Dolgov (1963), condylobasal length

38 Variability of Mammals

in three species of shrew (Sorex) from the Oksk collection was also the least
variable measurement; after it (in order of increased variability) came
length of tooth row, width of skull, height of skull, and width between eyes
(see Table 4b, Appendix).

Summarizing the results of the analysis of the extensive data (com-
paratively speaking) presented in Tables 4 and 5, it can be noticed that
the general belief that the condylobasal length of the skull is always the
least variable trait among the measurements of the skull, can ‘only be
accepted with great reservation. In fact, in the majority of cases, the
mean values of variation in the condylobasal length in different samples
are a little lower than those of variation in other measurements. But
due to the number of exceptions, this rule deserves guarded acceptance
and cannot be taken as always correct in all conditions.

The data already available are indicative of the different nature of
variability in different species. This means that the time has come to
pose the question of the necessity of studying the special characteristics
of variability in each species and family, etc; such a study may prove the
necessity for studying the special features of growth in all groups.

And finally, before beginning the analysis of variability in skull
measurement, it must be underlined that however homogeneous these may
be even within one species (for example, during a comparison of variability
of hair), a complex interrelation of parameters of variability of one and
the same trait is found within one population. Sometimes it happens that
the variability of trait sharply differs in males and females, and in young
and mature animals. Such conditions make it difficult to analyze the
variability of a trait and makes us speak with great reservation about the
overall variability of that given trait in a given population. Some
qualifications must be made such as “variability characteristic of aged
males or aged females’; in some cases, details mentioning year and season
of birth of the animal have to be given so as to depict the somewhat
different nature of variability among certain samples.

In complete agreement with the known elementary regularities
(Chapter V), the variability of width of skull considerably exceeds in
magnitude the variability of most other measurements.

Variability of weight of the lower jaw of the muskrat (Ondatra
Zibethica) is 10.5-40.48% (Latimer and Riley, 1934); of human beings,
17.141.18% (Lowrance and Latimer, 1957). The nature of the magnitude
of variability of weight of the whole skull in certain mammals is as follows:

Macaca rhesus 10.4--20.1% (7 groups) Hromada and Strnad (1962)
Macaca cynomolgus 15.7+31.6°% (6 groups) Ss bs
Ondatra zibethica 14.1--0.75% Latimer and Riley (1934)

Homo sapiens 16.6-- 1.15% Lowrance and Latimer (1957)

Nature of Variability in Mammals 39

According to Hromada and Strnad (1962) variability of the volume
of the skull has somewhat lower values (7.3 to 13.7%) for 13 different
groups of two species of Macaca.

An estimation of variability for epigenetic polymorphism is necessary
while studying the skull of mammals, just as it is necessary for the majority
of organs and systems. Many examples of this variability are given in the
works of R. Berry (1963, 1964), R. Berry and A. Searle (1963) and others.
(See the section on “Epigenetic polymorphism” in Chapter IV.) One
example showing prospects for using methods of study of epigenetic poly-
morphism in separate structures of the skull can be cited here.

It is known (Straus, 1962) that a great taxonomic and phylogenetic
value has been attached to the interrelationship of fissure (5. mylohyoideus)
and foramen (/. mandibulare) during the investigations on primates. During
the study on the variability of these traits, it was observed that the fissure
and foramen were found in very different combinations within a popula-
tion of one species and did not characterize either the species or large
systemic groups (Figure 5). But the investigations showed at the same
time that the nature of the distribution for different types of fissure and
foramen in each species may serve as a typical characteristic of a popula-
tion (Table 9).

Table 9. Occurrence (in percentages) of four basic types* of dispositions
of Fissure mylohyoideus in some primates (Straus, 1962)

Type of disposition Type of disposition
RC CUS on eee A ES ee ЕНЕЕ Се
I II ш IV I II III IV

Tupaia 82.1 17.9 — — Papio 73220 6.1 —
ИО. 27.6 34 10:5 58.6 Colobus .. 53.1 40.6 6.2 =
Galago .. 25.0 — — 75.0 Ропро 123.2 11 98.711 156.6 1.4
Tarsius .. 6.3 — 9026 Рап or GO) Эа 7.3
13:6 24. 55.2 6.9 СИИ ве (MS. 180 9.2 1.2
Аве .. 94 56.3 6.2 28.1 ;

* For character of types, see Figure 5.

Unfortunately, in the cited investigations the characteristics of species
and not of populations have been obtained, but there is no doubt what-
soever that clear-cut differences can be found in the frequency of occurrence
of different types of fissure structures even in different populations of one
species.

Dental system
Variability in the parameters of the dental system is close to the

40 Variability of Mammals

Figure 5. Inner surface of lower jaw of mature orangutans
with different varieties of relationship between fissure (s. mylohyoideus)
and foramen (f. mandibulare) (Straus, 1962).

variability of the structure and measurements of the skull. A review of
different investigations carried out for 17 species of mammals shows that
characters of the dental system like measurements of the incisors, canines,
premolars, and molars of the upper and lower jaw, and similarly the
number of teeth in cetaceans (and some pinnipeds), have characteristically
rather low variability values (Table 10).

As shown in the data presented in the work of L. Freedman (1963),
variability of different teeth differs considerably in magnitude. (Length
and breadth of teeth in male and female baboons in three different popula-
tions were studied.) Canines appeared to be most variable in the upper
jaw (mean coefficient of variation, 12%), and the molars were the least
variable (mean coefficient of variation, 4.8%); linear variability of

Nature of Variability in Mammals 41

incisors is a little higher than the variability of premolar measurements.
Variability of corresponding teeth in the lower jaw is in the same order.
It is interesting to note that while the magnitude of variation of molars,
premolars, and incisors is almost identical between the upper and lower
jaws, the variability of the canine of the upper jaw is somewhat higher
than for the one on the lower jaw. But the differences among the types
of teeth in their coefficients of variation is, perhaps, not characteristic
of mammals in general.

Table 10. Summarized data on variability (Lim, , ) of linear measurements
and number of elements of the dental system in some mammals*

— Se OO EO ne Eee Ee eS ee eee eee ee

Incisors Canines Premolars Molars No. of teetht
Upper Jaw 6.2-14.6 5.9-26.5 2.3-14.8 2.0-9.5
(4) (6) (40) (60)
3.9-17.4
(22)
Lower Jaw 5.2-10.4 4.8-19.8 4..1-10.1 2.9-17.0
(16) (6) (23) (68)

* Primates, Lagomorpha, Carnivora, Cetacea, Pinnipedia, Chiroptera, Rodentia,
Ungulata. Number of samples studied are given in parentheses.
Т For walrus and dolphins.

In the work of R. S. Bader (1955), carried out on a large number of
fossil ungulates belonging to ten species of the subfamilies Мегусйутае
and. Merycochoerinae, it has been shown that the coefficients of variation
of measurements of premolars and molars are not different. This fact
confirms the general conclusion that in each case the value of the coefficient
of variation is controlled mainly by ecological and not by stochastic
determinants. The differences in the magnitude of variability of tooth
measurements for two ecologically different species of Black Sea dolphins
(Table 11) could hardly be explained by accidental reasons.

Table 11. Variability (с. v.+-5,,,.%) of length of teeth of
adult male and female dolphins of the Black Sea
(Analyzed from the data of Kleinenberg, 1956)

eee a CC

Genus Male Female
Delphinus delphis an 8.441.38 7.9-+ 1.02
Tursiops truncatus м 12.4+2.77 14.82.23
te aon by —1.29 —2.87

It is natural to assume that diverse modes of feeding (there is direct
information on this) and different methods of using teeth bring about
such variability through selection.

42 Variability of Mammals

On the whole, variability of the dental system completely corresponds
to the variability of skull measurements. It may be said that, as a rule,
the variability of measurements for individual teeth (not canines) is in
the range of 4-10%, though considerable deviations are possible. Perhaps
the study of such variable traits in the dental system may point to specific
conditions occurring during such important functions as the capture of food,
defence from enemies, etc. in the life of species in the process of
evolution.

The study of the dental system is also necessary in the context of
epigenetic polymorphism. Examples of such work can be cited for various
orders of mammals. Zejda’s work (1960) presents data on the simple
form of M3 of the bank vole (Clethrionomys glareolus) and some interesting
aspects of its distribution in Czechoslovakia (Figure 6). Determination
of the distribution of color on the molars of the common shrew (Sorex
araneus) may serve as a good way to characterize individual populations of
this species (Markov, 1957). It was possible to distinguish populations of
martens and sables (Pavlinin, 1963) according to the appearance of addi-
tional teeth, protuberances, grooves, and the form of the external and
internal sides of М3. Definite population differences have been observed
(Stein, 1963) in the occurrence of different abnormalities in the dental
system in moles (Talpa europaea).

- Моге than
60 specimens

@® ‘40-60 specimens

Figure 6. Frequency of occurrence of “‘simple” (black) and “complex” (white) form
of the third upper molar of the bank vole in different populations in
Czechoslovakia (J. Zejda, 1960).

Nature of Variability in Mammals 43

These examples show how much unexpected and interesting information
can be had from the study of variability in seemingly well-studied and
simple dental systems.

Post-cranial skeleton

Unfortunately, because of a number of methodological and technical
difficulties in the study of the anatomy of mammals, the main part of the
skeleton, as such, receives little attention.

In analyzing the linear variability of the post-cranial skeleton, it
may be noted that the latter is distinguished by an extremely low amount
of variability, chiefly lying in the range of 3 to 5%. This is especially
noticeable when compared with the linear measurements of the skull, which
are more labile. So far, this fact has not been examined by investigators who,
as a rule, consider theoretical possibilities for measurements of the skull only.

D. S. Webb’s data (1965) on the variability of the post-cranial
skeleton (linear measurements of vertebrae and the pectoral and pelvic
girdles) in Camelops hesternus (a North American Pleistocene species) are
noteworthy. Out of 101 coefficients of variation estimated in this study,
51 were not above 4.8%, 92 did not exceed 8.5%; and only 9 coefficients
of variation seemed to be in the range of 9.0-14.9%.

The number of ribs seem to be a good trait for characterizing
different populations. The analysis of data by A. Shaw (1929) shows that
the variability in number of ribs for a population of domestic swine—male
and female—is 4.0--0.24% to 4.2-++-0.26%. Results obtained by us on
variability in number of ribs for different pinnipeds are given in Table 12.

Table 12. Magnitude of variability (с. 0.--5,.,,; Lim,,,,) of number of ribs
: Ribs
Species Sternal Asternal

Harp seal (Pagophilus groenlandicus) о: 0-8.7(10)* 3.8-21.4(9)
Bearded seal (Erignathus barbatus) ve 0.0 2.2+0.38
Hooded seal (Cystophora cristata) bes 0-4.9(8) 7.8-9.3(8)
Common seal (Рйоса vitulina) ae 1.4-L0.14 17.5+2.48
Ringed seal (Pusa hispida) ae 0-1.4(5) 0-4.2(5)
Baikal seal (Р. sibirica) 0.0 5.6-Е0.66
Соури (Myocastor coypus) 0-4.3(4:) 5.9-6.4(4)

* Number of samples carried out are given in parentheses. In this and other tables on
samples of Pinnipedia, animals of the same sex and as far as possible of the same age,
caught during one season (i.e., 2 to 3 months) are included.

Variability of the number of vertebrae according to different divisions
of the vertebral column is a very good trait for expressing the variability

44 Variability of Mammals

of the post-cranial skeleton for different populations and species.

Data obtained as a result of analyzing the material of L. Law (1938)
of variability of caudal vertebrae in four lines of laboratory mice, show
that variability of this trait ranges from 2.7--0.42 to 6.2+0.93% and
is unrelated to the variability of tail length. In another group of
laboratory mice of other phenotypes, the variability of this trait seems to
be more pronounced—from 2.9+-0.30 to 11.7+1.06%. It is interesting to
note that out of eight lines, the variability of number of caudal vertebrae
in males was lower than that in females in seven lines.

Hamburgh and Lynn’s data (1964) on the variability of the number of -
caudal vertebrae in experimental populations of white rats are interesting.
At the age of five days, the variability in number of vertebrae was considera-
bly lower in the normal group than it was in those animals in which the
growth of thyroid gland was artificially restricted (hypothyroidism).
The values were, respectively: 5.00.80 and 10.141.55%. In older
animals, the situation was reversed: 9.0+1.64% in control, and 9.9-
2.9% in the hypothyroid animals.

Data on the variability in the number of vertebrae in certain mammals
are given in Table 13.

Through the wide researches carried out by Latimer (1936-1959),
it is clear that variability in the weight of the skeleton and the individual
parts is considerable. From this work, the variability in general weight
of skeleton is as follows:

Species Number of
samples

Rabbit .. 6.9-17.3% р 6

Cat о 1332130 ей S2

Мани loos a 1

Variability in weight of individual elements of the skeleton is far
more pronounced. In man the weight of parts (in descending order of
magnitude of variability) like the hyoid bone, sternum, metacarpus,
calvicle, patella, scapulae, radius, all ribs, and metatarsus, was observed
to be particularly variable. The variability of the above-mentioned
organs is 43.6 to 19.5% (data of Lowrance and Latimer, 1957).

While concluding this section about the variability of the post-cranial
skeleton, mention of two specific interests in the study of organs of this
system may be made. Firstly, the surprisingly small scope for variation
of linear measurements in homogeneous samples. ‘This reveals the cons-
tancy of skeletal traits and the possible potential of their use for the study of

3 This observation is apparently not confirmed by the analysis of number of vertebrae
in different types of tailless cats (Howell and Spiegel, 1966).

Nature of Variability in Mammals 45

Table 13. Variability of the number of caudal and thoracic

vertebrae of certain mammals (Number of samples

used is given in parentheses)

Species ea я о Remarks
Caudal

Greenland seal (P. groenlandicus) ; Own data

White Sea 13.4-14.0 — 7.8-10.2(3)* ey St

Region of Yan-Maien 13.2-13.6 6.2-8.7(5) 4

Newfoundland Region 12.6-13.0 3.1-4.5(4) И
Hooded seal (Oystophora cristata) 13.4-14.4 6.0-8.6(9) a fe
Common seal (Phoca vitulina) 12.8--0.18 7.8-+0.98 Adult females
Bearded seal (Erignathus barbatus) 11.9-12.3 6.7-8.0(2) г a
Ringed seal (Pusa hispida) 12.9-13.4 1.5-8.6(3) 3 43

Baikal seal (P. sibirica)

Caspian seal (P. caspica)

13.00.09 4.4+0.53
12.1+0.28 9.3--0.58

Males and females
Adult males

White mice (Mus musculus albicans) — 31.6-32.1 2.3-2.8(2) Sumner (1909), males
ее a3 us ae > — 2.9-11.7(16) Law (1938), different

lines

White rat (Ratius norvegicus alb.) 15.8-26.6 1.3-16.9(9) Hamburgh, Lynn (1964)

Coypu (Myocastor coypus) 22.8-23.8 3.3-4.1(4) Own data

Thoracic and Lumbar

Orangutan (Simia satyrus) 16.0+0.05 3.2+0.23 Males and females
Schultz (1917)

Man (Homo sapiens) 17.00.02 1.2+0.08 Men and women
Schultz (1917)

Pig (Sus scrofa) 20.5-21.3 1.9-2.3(2) Males and females

Freedman (1939)

—— a = _ _ _ _ _ . _ ———_. —

precise problems in the field of population morphology. Secondly, the
study of these parts of the skeleton is promising also for studies concerning
epigenetic polymorphism; this has been well shown by Griineberg (1950-
1963) and Berry (1963-1964) on rodents. Other researchers (Dolgov,
1961; Ashton and Oxnard, 1964; Sikorska-Pivovska, 1965; and others)
agree with this. In Figure 7 some traits which are used in the study of
variability of the skeleton are clearly seen.

We see that the variability of linear measurements for the post-
cranial skeleton usually varies from 3 to 5%. Variability of discrete
traits is, on the average, higher (5 to 8%); and variability of characters
comprising weight of organs is still higher (on the average, about 10 to

15%).

46 Variability of Mammals

Figure 7. Examples of epigenetic polymorphism in the skeleton of :
a—mouse (Mus musculus) ; b—guinea pig (Cavia porcellus) ; c—squirrel
(Sciurus carolinensis) (Berry and Searle, 1963).

Variability of Other Internal Organs

Masculature

Fewer data are available on variability of other internal organs of-
mammals than is available for the skeletal system as a whole.

Data on variability of weight (absolute and relative) for all muscula-
ture of rabbits of different genetical lines are presented in the work of
Latimer and Sawin (1955). It was observed that variability of absolute
weight of the musculature of males varies less than that of females:
14.7 to 17.2% in males and 11.7 to 21.5% in females, averaging about
15%. Variability of the relative weight of muscles (ratio of muscle weight
to body weight) has values three times lower for males and females (3.9
to 5.5%); moreover, the variability of relative weight of musculature
seems to be more a variable trait in males than in females: 3.9 to 5.5%
in males, 5 to 5.4% in females. In one line out of three, females have
higher coefficients of variation for absolute and relative weight. In one
instance, these coefficients were higher for males and in another instance,
variability of the absolute weight was higher in males but the variability
of the relative weight was higher in females.

Characteristics of the musculature of the common cat have been
examined in detail by Latimer (1944), who evaluated variability of the
locomotor group of muscles separately for males and females (Table 14).

The variability of weight for the whole musculature of cats is a

Nature of Variability in Mammals 47

little higher than the variability of the locomotor group of muscles, and
the variability of relative weights is appreciable—3 to 4 percent, more
than the similar values for rabbits. Among the lines studied (sufficiently
large in number, homogeneous, and hence completely representative), a
clearly greater variability of the absolute weight of muscles is observed
in males, whereas the variability of relative weights is higher in females.

Table 14. Comparison of variability (с. v.--S¢,y,) Of absolute and relative
indices in weight of skeleton and muscles of cats (From Latimer, 1944)

Se Te SS SS Abso- Каа-
Organ Absolute weight Relative weight lute tive
es ee Se ee t & 1 <
Маез Females Males Females Q 2
Skeleton 21.30-Е1.50 13.33-+0.93 15.44--1.06 16.2641.15 +6.06 —1.59

All musculature 27.93+2.03 23.95+1.74 7.02+0.48 8.114056 +3.23 — 0.13

Locomotor system 24,95+1.75 21.62 + 1.46 4.44--0.29 5.214+0.35 ++4.84 —1.61

Based on the examples given above, it could be concluded that the
relative weight of the musculature may be a good and constant index of a
population since its variability in all instances is appreciably less than
10%, falling in line with the variability values of organs like the skull.
But such a conclusion appears to be not absolutely correct. Other data
available on the relative weight of musculature for other mammals (Tribe
and Peel, 1963) show a considerable degree of variation for this trait.
In six species of ungulates (sheep, gazelle, swine, etc.), variability of
relative weight of musculature varies from 8.1 to 88.1%. The fact that the
variability of relative weight in domestic ungulates appears to be directly
proportional to the value of relative weight, attracts attention to these
data. Regarding relative weight of musculature, Bos indicus js highest
(32.5+2.0%). For this species, the highest variability of relative weight
is also found. Sheep, which have the lowest relative weight of musculature
(2.9-+-0.6%), also have the least variation for this trait (8.1+1.58%).
In the same work of Tribe and Peel (1963), it was shown that in kangaroos
(Macropus sp.) the relative weight of musculature is the least variable
(3.4+0.68%) of any species studied.

On the whole, while characterizing the variability of weight of the
musculature in mammals, it can be assumed that the variability can be
20% (ranging from 11.7 to 27.9%) and is considerably less in homoge-
neous samples for relative weight (up to 10%).

48 Variability of Mammals

Only muscle weight has been used as a standard method of determin-
ing quantitative morphology in the study of muscular systems, while
physiological traits have been used to a considerably less degree. Thus
the methods used for the study of quantitative variability in epigenetic
polymorphism appear to be necessary. References for such examples
can be found in the works of S. Basu and S. Hasary (1960) on the study
of muscles of anterior extremities in a large sample.

Brain and eye

Data are presented in the work of St. Reinis (1964) on variability
in number of glial and nerve cells in different layers of the cortex of the
brain in rats of different ages. An analysis of these data shows (Figure
8) that the variability of nerve cells in rats from 1 to 25 days old
adult animals, varies from 6.0 to 25%; the variability in number of glial
cells varies from 12.7 to 44.5%. It is interesting that the variability in
number of cells in the glia of the fourth layer takes a reversible trend
with age. Variability in number of glial cells considerably increased

C.v.%

15.61.89 14 17 2021 25
Age in days

Figure 8. Variability (с. о. 25...) of (1) glial and (2) nerve cells in the
fourth zone of the cortex of rat brain (from Reinis , 1964).

Nature of Variability in Mammals 49

while variability of nerve cells decreased. Sharp peaks and depressions
in the curve are seen in Figure 8, which may be explained perhaps by the
unsatisfactory method of obtaining cross-sections exactly from one and
the same part of the brain, and also perhaps by the fact that in different
genotypes, developmental processes may proceed at different rates.
However, the general conclusion is not changed. Such decrease in the
variability of number of nerve cells in the brain cortex of adult mammals
is well understood from the functional point of view. The control of
morphological structures which facilitate correct decisions in all situations,
must be very strict.

The variability in number of ganglia in the nerve plexuses ranged
from 12.8--0.6% to 20.6+0.9% according to an analysis of the data of
Sauer and Ruble, 1946.

Variability in the weight of the brain is comparatively not large.
According to Brown, Pearce, and Van Allen (1926), the variability of the
absolute weight of the brain in rabbits is 7.5-8.4% (three samples), and
according to Latimer and Sawin (1955) it is 5.0-7.9% (six samples).
Ohne interesting fact, in the opinion of these authors, is that the variability
of the relative weight of the brain (in proportion to body weight) appears
to be considerably higher than the variability of the absolute weight (for
all nine samples, 8.2 to 16.9%, average about 14%). Variability of the
absolute weight of the brain of Peromyscus maniculatus (King, 1965) Юг”
one sample is 6.3%; data on the variability of brain weight for 27 different
samples are given in the same work.

Data on the high variability of the relative weight of the brain are
confirmed by material recently obtained on the brain of the pigmy ground
squirrel (Citellus pygmaeus), where the variability of the relative weight
of the brain varied from 12 to 42% (Schwartz, 1960). |

According to A. Kramer (1964) the variability of the absolute weight
of the brain in the gerbil (Meriones unguiculatus) is also quite low—3.8 to
5.0% (two samples); whereas the variability of relative weight (in mg/
100g body weight) is considerably higher—15.7 to 23.6%.

The variability of parameters of the spinal cord of laboratory rabbits
is characterized by moderate values. Variability of the absolute weight is
8.3 to 11.2% (six samples) and variability of the length is 3.2 to 4.9%
(four samples); variability of corresponding relative values is 8.1 to
12.9% and 2.6 to 3.7% (Latimer and Sawin, 1955).

Thus the variability of nerve cells in the cortex of the brain in adult
animals is less than 10%; variability of the absolute weight of the brain
averages about 6 to 7%; and variability of the relative weight of the brain
appears to be considerably larger, from 12 to 20 and up to 42%.

Variability of the absolute weight of the eyeball appears to be like
that of the entire brain, i.e. 5.5 to 10.5% (seven samples of rabbits and

50 = Variability of Mammals

guinea pigs) (Latimer, 1951, 1955). The variability of the relative
weight of the eye, like the variability of the weight of the whole brain,
appears to be quite high (8.5 to 14.8%) for the same samples.

The variability of the crystalline lens of the eye in deermice appears
to be very high—up to oe (King, 1965).

Glands

All the data on the variability of glands are for their absolute and
relative weights.

Salivary glands. According to the data of Tribe and Peel (1963)
the variability of the submandibular salivary glands in Е (Macropus
sp.) 13 24.1%.

According to the data of А. Buchalczyk (1961), the variability of

relative weight of salivary glands in the common shrew (Sorex araneus)
varies from 3.6 to 28.7% (20 samples); higher values of the coefficients
of variation are characteristics of the submandibular gland (18.0 to 28.7%),
and for the parotid gland it is less (3.6 to 19.2%). Variability of the
weight of these glands in relation to each other varies little, 1.3 to 2.6%
(10 samples). Variability of absolute weight of the salivary gland of
Meriones unguiculatus is 11.2 to 2.5% (2 samples) according to Kramer
(1964), and the variability of the relative weight for this organ is 18.8 to
20.5%.
Pancreas. The variability of absolute weight for the pancreas in
captive foxes (Vulpes vulpes) ranges from 10.4 to 34.3% (4 samples, Fateev
et al., 1961). The variability of relative weights of these glands in the
same samples is, on the average, lower: 11.7 to 27.0%. The variability
of absolute weight of the pancreas in rabbits is rather high but is main-
tained on the same level, 22.1 to 30% (6 samples, Latimer and Sawin,
1955). Relative weight variability for the same gland in the same samples
13 19.4 ю 25.95%.

The variability of weight Юг the pancreas of Meriones unguiculatus
ranges from 27.4 to 29.2% according to Kramer (1964), whereas the
variability of relative weight is somewhat less, 26.3 to 26.4%.

Relative weight variability for the pancreas of the pigmy ground
squirrel (Citellus pygmaeus), according to data by S. 5. Schwartz (1960),
varies widely—in the range of 13.7 to 46.8% (6 samples). In the Caspian
seal (Pusa caspica), the variability of absolute weight of this gland is
26.1% but for the white whale (Delphinapterus leucas) it is only 1.7%.

On the whole, the variability of absolute weight of the pancreas for
different mammals ranges from 10.4 (not considering the data on the white
whale) to 34.3%, averaging about 25%. Variability of the relative weight
of this organ is somewhat less, 11.7 to 46.8%, averaging about 20%, but
the scope of variation in relative variability is greater.

Nature of Variability in Mammals 51

Liver. In insectivores, the variability of absolute weight of the
liver is quite high compared to the variability of weight of other organs.
In the European mole (Talpa europaea) the variability of absolute weight is
13.7 to 32.5% (four samples); the relative weight varies to a lesser degree,
15.4 to 22.4% (six samples) (Fateev, 1962); for three lines of the pigmy
ground squirrel, 15.4 to 21.4%; for the common shrew (Sorex araneus),
10.8%; and for Neomys fodiens, 12.3% (Schwartz, 1960).

Similar values of variability for absolute and relative weights of the
liver are observed in other mammals: 5.0 to 24.2% for absolute weight
in foxes, and 6.3 to 24.1% for relative weight in the same four samples
(Fateev et al., 1961); 22.1 to 31.8% for absolute weight in rabbits, and
20.4 to 27.9% for relative weight in the same nine samples (Brown её al.,
1926; Latimer and Sawin, 1955); 10.4 to 41.9% for absolute weight in
15 samples of six species of marine mammals; 17.5 to 21.8% relative
weight in pigmy ground squirrels (Schwartz, 1960); 9.7, 8.9, and 12.4%
respectively for Microtus oeconomicus, red-backed voles (Clethrionomys rutilus),
and water voles (Arvicola terrestris) (Schwartz, 1960); and finally 21.2 to
27.3% for other samples of muskrat (Ondatra zibethica) (Schwartz, 1962).

The variability of absolute weight of the liver in Meriones unguiculatus,
according to A. Kramer (1964), is 28.8 to 32.3%, and the variability of
relative weight is a little less, 21.0 to 23.3%.

On the whole, the range of variability for absolute weight of the
liver is 5.0 to 41.9% (average, 25 to 30%) and that of relative weight is
6.3 to 27.9% (average, 15 to 20%).

Thyroid gland. In kangaroos, according to Tribe and Peel (1963),
the variability of absolute weight of the thyroid gland is 24.1% and that
of relative weight is 21.2%. The variability of the thyroid gland of Neomys
fodiens varies from month to month (Table 15).

Table 15. Variability of the weight of the thyroid gland in
Neomys fodiens in various months (Calculated from the
data of Bazan, 1956)

Sex, Month R--sx 6. Ve-tSe.y, п Sex, Month KES% 6. 9. 5с.0. 2
Male Female

July 75.1+ 7.4 50.2+ 7.0 26 July 76.1411.1 56.44-10.3 15

August 65.9-1:17.7 71.0419.0 7 August TINE 9 156165 9.2, 719

September 63.4+11.6 60.6--12.9 11 September 63.2+10.4 — 3

о

Similar values of variation for this trait in rabbits are presented by Brown
et. al., (1926) namely, 56.2 to 70.3%. The relative weight of the thyroid
gland in the latter case for three samples was 61.4 to 62.3%. Data presented
by Latimer and Sawin (1955) for rabbits (six samples) differ considerably

52 Variability of Mammals

from the foregoing observation; they obtained an absolute weight variability
_ of 22.3 to 40.6%, and for relative weight, 18.1 to 27.9%. The values of
absolute and relative variability of weight of the thyroid gland in the
guinea pig (Сала porcellus) are close to the last mentioned values—29.9
and 31.5% respectively (Latimer, 1951).

The last example differs from previous ones in that the variability
of relative weight of the thyroid gland appeared somewhat greater than
the variability of absolute weight. It is difficult to say why it should be
so but the fact that in the samples under study (C. porcellus) the relative
weight variability for the adrenals and for the hypophysis was also higher,
attracts attention {see below).

Variability of weight for the thyroid gland in Mertones unguiculatus
was given by Kramer (1964) as 23.8 to 24.8% and the variability of relative
weight as 20.4 to 38.2% (for two samples).

On the whole, the variability of absolute and relative weight for
thyroid glands in mammals is exceedingly high and is и. 22.0
to 71.0% and 18.1 to 62.3%.

Spleen. Variability of absolute weight of the spleen in moles (Fateev,
1962) ranges from 25.8 to 43.2% and the variability of relative weight
_ щ the same four samples is from 25.3 to 30.2%. Variability of absolute
and relative weights of the spleen in rabbits seems to be close to these
figures: respectively 30.8 to 51.5% and 26.3 to 55.5% (for nine samples
of rabbits) (Brown et al., 1926; Latimer and Sawin, 1955). A still
greater range of variability for absolute weight of the spleen is character-
istic for our pinnipeds and cetaceans: from 28.1 to 84.2% (ten samples,
six species) (Table 7, Appendix). Finally, the data of Newson and
Chitty (1962) show a similarly large range of variability in absolute values
for the field vole (Microtus agrestis), 50.5 to 80.3% (Table 16).

Table 16. Coefficient of variation (с. v.5;¢,y,) of weight
of spleen of Microtus agrestis (Calculated from the
data of Newson and Chitty, 1962)

Population Population
Sex I Il t
Males a 60.3-- 9.5 50.5+ 8.9 —1.40 (1)
Females Be 80.3 10.7 76.9-+11.4 —1.84 (II)

Variability of weight of the spleen in Meriones unguiculatus is very
high, 58.7 to 65.5%; variability of relative weight is 68.1 to 78.7%.

From the data presented, it is seen that the variability of the spleen
is exceedingly high, 1.е., 25.8 to 84.2% for abselute weight, and 25.3 to
78.7% for relative weight. In the majority of the species studied the

Nature of Variability in Mammals 53

average variability of absolute weight of the spleen is close to 35 to 45%.

Adrenals. In the above-mentioned nine samples of rabbits, the
variability of absolute weight of the adrenals varies from 29.6 to 43.3%
and quite clearly differs from the material presented by Brown её al.
(1926) and Latimer and Sawin (1955). The variability of these samples
appeared to be comparatively high (40.1 to 43.3%) in the case of the first
author and quite low (29.6 to 34.1% for six samples) in the case of the
other authors. Variability of the relative weight of the adrenals in rabbits
in the second work was 35.9 to 40.7%, and in the third, 20.0 to 36.7%;
on the average, this is less than the variability of absolute weight in
both cases.

According to Latimer (1951), the variability of absolute weight of
the adrenals for porpoises is 29.4 and for the relative weight is 31.2%.
The variability of absolute weight of right and left adrenals of chinchillas
is 22.7 and 24.7%. L. С. Korotova’s findings (1962) on the variability of
weight and fine structure of the adrenals in the water vole (Arvicola terrestris)
is interesting. The variability of absolute weight for the adrenals of
males and females appeared to be very similar—25.0 and 25.5%; but
the variability of relative weight in the same groups of animals was from
15.5 to 28.6%. In the same data, the variability of the thickness of
the adrenal cortex was 18.2%, thickness of the zona fasciculata, 34.3%,
and thickness of the zona glomerulosa, 32.5%.

Adrenal weight of Meriones unguiculatus varied from 18.3 to 22.9%
according to Kramer (1964) and 26.1 to 31.5% (relative weight). Data
on the weight of the adrenals of M. hurrianoes follow these values closely
(Prakash, 1964).

McKeever and Quentin (1963) showed that in the mongoose Her-
pestes auropunctatus the variability of the absolute width of the zona glomerulosa
ranged from 26 to 80.8%; the zona fasciculata, 39.9 to 49.6%, and the zona
reticularis from 80.8 to 122.9%.4 The variability of relative thickness
(in relation to the thickness of the cortex) varied in three groups of
animals as follows : zona glomerulosa, 43 to 94%, zona fasciculata, 28 to
33%, and zona reticularis, 57 to 63%.

The average variability for absolute weight of the adrenals is a
little less than the variability for weight of the spleen, liver, and thyroid
glands : 18.3 to 43.3% (average, 25 to 30% for absolute weight, and 20
to 30% for relative).

Hypophysis. The variability of absolute weight of the hypophysis of
rabbits ranges from 18.5 to 34.1% (nine samples); the variability of
relative weight is from 19.8 to 26.9% (Brown et al., 1926; Latimer and
Sawin, 1955). The variability of absolute and relative weights of the

4 These figures were obtained by the author from graphs and do not represent great
accuracy.

54 Variability of Mammals

hypophysis in guinea pigs ranges from 17.2 to 18.1 (H. Latimer, 1951).
Finally, the variability of absolute and relative weights of the hypophysis
in the chinchilla (Chinchilla laniger) is 24.6 to 40.9% and 26.4 to 39.9%
respectively.

The magnitude of variability of weight of the hypophysis of Meriones
unguiculatus is close to these figures: 23.4 to 29.4% for absolute weight,
and 26.1 to 27.4% for relative weight (Kramer, 1964).

All the data presented on variability of the hypophysis show the
similarity of the magnitude of variability of absolute and relative weights
of the hypophysis—on an average, about 25%.

Other glands. Data are available to show that the variability of
absolute and relative weights of the pineal body (epiphysis) in rabbits
practically coincide; they are respectively 26.7 to 32.5% and 27.7 to
34.0% (Brown её al., 1926). Similarly, the variability of absolute and
relative weights of the thymus in rabbits coincide with these figures, being
35.7 to 38.4% and 35.6 to 36.6% (according to Brown). The variability
in length of the metatarsal gland in 12 samples of Odocoileus hemionus ranged
from 3.2 to 22.2%, averaging about 11% (Anderson, её al., 1964).

For the Harderian gland in rabbits, the variability of absolute weight
ranged from 12.7 to 31.0% and of relative weight from 14.9 to 29.4%
(for six samples; Latimer and Sawin, 1955). !

Bhatnagar and Taneja (1960), and similar data in other works,
revealed that the variability of milk quantity in one lactation of a cow
does not vary much in different seasons of the year, 2.7 to 3.7% (five
seasons). This characterizes the activity of the mammary gland; data on
the variability of secretions in other glands do not exist.

Digestive tract

A number of works present material on the variability of number of
vallate and foliate papillae on the tongue of elephants and certain other
mammals. А partial reference appears in the work of Kubota Kinziro
(1967); unfortunately, the data presented indicate only the presence of
variability, so an exact analysis of its parameters is not possible.

Length of whole digestive tract. The variability of the length of the
whole intestinal tube varies within narrow limits in different species:
in moles (Fateev, 1962), from 8.1 to 18.7%; in rabbits (Latimer and Sawin,
1955), from 6.0 to 11.8%; in rats, from 11.6 to 12.8%; in white whales
(our data; see Table 8 in the Appendix), 4.2%; in captive foxes (Fateev,
1961) from 6.5 to 27.8% and (per Sokolov, 1941) from 5.4 to 8.6%. In
minks (Mustela vison) the variability of gut length ranges from 11.2 to
13.6% and, in certain pinnipeds, from 5.3 to 16.0% (from our data, Table
8, Appendix). ‘The variability of length of the.gut in gerbils ranges from
7.7 to 8.3% (Kramer, 1964).

Nature of Variability in Mammals 55

It can be concluded that on an average, the variability of length of
the whole digestive tract in the majority of mammals comes to about 10
to 12% (ranging to 4.2 to 27.8%).

The relative value of variability of the gut varies within close limits—
from 2.3 to 28.4%. The maximum value of variability is in female foxes
(Fateev, 1961), and the minimum is in pigmy ground squirrels (Schwartz,
1960). The relative variability of the gut in insectivores is seen from
that of the shrews Sorex araneus and Neomys fodiens, which have 11.1 and
12.9% respectively (Schwartz, 1960).

c.v.% ° The variability of length of
the small intestine in kangaroos
(Tribe and Peel, 1963) is 25.8%;
the length of this in female foxes
varies from 6.7 to 9.4%. The
variability of length of the small

‚ intestine in three species of pinni-
peds—Baikal seal, Greenland seal,
and hooded seal (total, eight
samples)—ranges from 2.0 to
16.4%. The length of the small
intestine and the total length of

Ay в А: B, the intestines of both new-born

and immature animals have the

Figure 9. Value of coefficients of variation of Pepe all ee ats
intestinal length in (A) new-born, and (B) а м и

adult males and females of hooded seal the samples of pinnipeds studied.
Cystophora cristata(from the Greenland Sea, 1962). It is clearly seen from Fig. 9
that variability of length of the

intestines of the hooded seal (Cystophora cristata) considerably decreases with
age. The variability of length of the small intestines in gerbils is moderate,
9.8 to 10.59% (Kramer, 1964). т

The variability in length of the large intestine is rather high in the
above-mentioned species of pinnipeds; the variability in the same seven
samples was 15.0 to 58.0%; in kangaroos, 29%; in foxes 9.3 to 9.7%. It
may be generally accepted that the variability in length of the large
intestine is about 12 to 15%, whereas the variability of length of the
small intestine is close to the value of variability for the length of the
whole digestive tract, 8.0 to 12.0%.

The variability in length of the large intestine in ое is high:
10.1 to 22.2% (Kramer, 1964).

The length of the esophagus in three species of pinnipeds varies
from 4.4 to 13.1% (Table 8, Appendix).

The length of the stomach is more variable in the same three species
of pinnipeds—8.3 to 15.7%.

56 Variability of Mammals

_ The weight of intestines appears to be quite a variable trait, ranging
from 11.8 to 21.3%. It is interesting to note that in rabbits the relative
weight of intestines is a more variable trait than the absolute weight.

The weight of the stomach is characterized by still higher values of
coefficients of variation—from 10.7% in foxes to 24.8% in white whales,
33°% in Histriophoca fasciata, and up to 51.8% in moles (Table 8, Appendix).

Data obtained by E. A. Sokolov (1941) on the volume of the stomach
and intestines in mice can be presented while concluding this section on
the variability in parameters of the gut. Variability of volume of the
stomach in minks is 6.1 to 12.9% and in foxes 9.1 to 16.6% ; the volume of
the whole gut in minks is 9.7 to 10.5% and in foxes 12.5 to 12.6%. These
data show the relatively small variation in the coefficient of variation Ш.
traits like volume of intestines. It must be noted that these data differ
somewhat, however, from those of Tribe and L. Peel (1963) on varia-
bility of weight of the stomach and large, small, and blind sections of the
intestine (26.8 to 56.8%). But, of course, a direct comparison should
not be made between these two sources as it is likely that both represent
natural conditions.

The following list is a characterization of the average values of
variability in parameters of the intestinal tract of mammals :

Mean value Lime.v.

Length of: 6. 9.

Total gut ое 8-12 4.2-27.8

Small intestines ue 8-12 4.5-25.8

Large intestines aN 12-15 9.3-58.0

Esophagus EAs about 10 4.4-13.1

Stomach a 11-13 8.3-15.7
Weight of intestine #0 about 25 _ 11.8-46.1
Weight of stomach ve about 30 18.2-51.8
Volume of stomach she — 6.1-56.8
Volume of intestine Re 15-25 9.7-46.0

Of course, it is not possible to take the data presented as final. How-
ever, it can be specifically said that insofar as the mean values of linear
variability are concerned, the intestine is like organs of other systems
(such as the nervous system and the skeleton); 1.е., it has a characteris-
tically low variability and hence may be widely used in the investigations
concerning population.morphology.

Lungs and trachea

A comparison of values of variability characterizing the weight of
lungs in different mammals (five species of pinnipeds: white whale, fox,
rabbit, mole, and gerbil) in all 27 samples given in Table 9 of the Appendix,
shows that the variability of this trait varies considerably from 15.5%

Nature of Variability in Mammals 57

(for one sample of rabbit) to 63.7% (for the sample of the Caspian seal).
However, the majority of the coefficients of variation fall between 20
and 30%. The relative weight of lungs (percent of body weight) varies
from 8.3 to 48.7%.

A convenient trait for studying population morphology of mammals
is the number of rings in the trachea. The choice of this trait is based on
the fact that it has a very low variability (Table 17).

Table 17. Variability in number of rings in the trachea
(Adult males)

eS eee eee ie

Greenland seal (Pagophilus groenlandicus),

Yan-Maien region ae 42.7+0.47 9.4+0.78 72
Common seal (Phoca vitulina) ay 69.4-++0.54 3.8+0.55 24
Hooded seal (Cystophora cristata) 50 38.2+0.52 9.9--0.96 53
Ringed seal (Pusa hispida), Pechora Sea oe 91.5-=0.87 5.80.67 37
Baikal seal (P. sibirica) We 78.7 £0.55 4.4--0.86 13
Caspian seal (Р. caspica) 50 89.4+1,12 5.8+0.89 21
Coypu (Myocastor coypus) 30 88.9--0.28 6.3-40.7 42

a ee SS ee Sh ee

In 36 samples of pinnipeds, the variability of this trait did not
exceed 10.4% (3.2 to 10.4%) and averaged from 5 to 7% (Table 9,
- Appendix).

So the variability of generally used traits of the respiratory system
is characterized by the following values :

Mean values

Trait 6. и. Lime.y,
Weight of lungs we 20-30 15.5-63.7
Trachea : number of rings pe 5-7 3.2-10.4

Blood

It is generally assumed that the morphological picture of the blood
may serve as a good index for the general condition of animals. Un-
fortunately, there are few data on the variability of blood parameters as
such. From those available, it may be concluded that the variability of
the quantity of hemoglobin in the blood may differ considerably from
species to species; for example, from 1.6% in ewes (Kushner, 1941) to
26.1% in water voles (Schwartz, 1960). However, differences in varia-
bility are considerably restricted within a species (Table 18).

Variability of hemoglobin content of the blood in three groups of
ewes appears to be close, 1.6 to 2.3% (Kushner, 1941) and in three groups
of deermice (Peromyscus leucopus) 4.5 to 7.5% (Salander, 1961).

The same is true with variability in the number of erythrocytes.

58 Variability of Mammals

Table 18. Coefficient of variation (c.v.+-5,,.) for quantity of
hemoglobin in males of field voles (Microtus agrestis) in two
nearby populations in different years (Calculated
from the data of Newson and Chitty, 1962)

Males Females
Data ——
*“Forest”’ “Road” ‘*Forest”’ “Road”
1956
December 5.7+1.03 — 4.5--0.87 —
1957
March 5.7-0.78 -4,.6-+0.82 - 6.7--0.98 5.8-L0.98
5.0--0.98
Мау 5.1-- 0.85 8.9-- 1.63 6.9-- 1.06 8.3-- 1.38,
September 8.2-++0.90 11.8+2.15 8.1-- 1.08 —
December 4,3-0.75 6.91.22 6.5-- 1.23 11.9=- 1.8
1958
March 9.44 1.84 5.3--0.95 7.4+1.45 7.11.59

— жж een ec eee a ee ae Ce Ce

In the pigmy ground squirrel (Citellus pygmaeus) the variability of this
trait ranges from 16.2 to 22.6% (three samples), and in muskrat (Ondatra
Zibethica), it is 20.6%, according to S. $. Schwartz (1960); but it is only
2.3 to 3.1% (three samples) in ewes according to Kushner (1941).

For the most part, the variability of number of erythrocytes and the
quantity of hemoglobin in the blood varies in a limited range from 22.6
to 26.1%. This shows that after quick methods of determination for these
traits in natural populations have been developed, they could prove to be
good morphological indices for the study of peculiarities of animals in a
population.

Variability in the total number of leucocytes varies in a more pro-
nounced way [V. P. Kozakevich (1967) “Асе and Seasonal Variability
in Seven Samples of Susliks,’”” С. К. Chai (1957) “‘Generational Variability
in the Number of Leucocytes in Various Lines of Laboratory Rats,” and
others].

Unfortunately, until recently physiologists gave very little time to the
study of variability of blood characters. Hence, many quantitative
differences in blood among different species, families, and orders reported
previously (review by P. A. Korzhuev, 1964) have to be examined in the
future on a more factual quantitative basis.

Heart

There are a large number of works on the variability of the weight
of the heart. Generally, variability of the absolute weight of the heart
ranges from 7.8% (in one of the six groups of rabbits studied by Latimer
and Sawin, 1955) to 44.3% (in one of the groups of moles studied by

Nature of Variability in Mammals 59

Fateev, 1962) and even up to 70.9% in the Greenland seal (Pagophilus
groenlandicus) (Table 10, Appendix).

Variability of the absolute weight of the heart in adult dogs was
24.9% (Latimer, 1961) and variability of the relative weight was 18.9%.
According to the data of the same author, variability of the weight of the
right ventricle was 25.5% and of the left ventricle was 25.0%; variability
of the ratio of right ventricle to left ventricle was 11.2%. The relative
weight of the ventricles (in relation to the weight of the whole heart) is
the least variable (of the right, 7.5% and of the left, 6.5%).

Variability of the absolute weight of the heart in gerbils is 15.9 to
21.7% (Kramer, 1964) and the variability of the relative weight of the
auricles (in relation to the whole heart) is 14.6 to 21.5%.

It can be said that the variability of the absolute weight of the heart
for the majority of mammals is about 17 to 22%. Variability of the
relative weight of the heart in all instances is a little less than the
variability of the absolute weight. For example, the variability of absolute
weight for the heart of moles is 16.7 to 44.3% and the variability of relative
weight is only 14.4 to 26.5% (in the same population and in the same four
groups). Similarly, variability of the weight of the heart in six groups
of rabbits ranged from 7.8 to 19.4%, and variability of the relative weight
in the same groups was 7.2 to 12.1%.

On the whole, it is clear that the variability of the relative weight
of the heart is, on the average, about 13 to 16% (ranging from 6.7 to
20.99)

As shown Бу the work of James (1960), characters of the structure
of the heart and the arrangement of its blood vessels can be successfully
studied also from the point of view of epigenetic polymorphism.

Urogenital system

The weight of the kidneys varies greatly and its variability is close
to that of the weight for glands. The lowest known variability of kidney
weight occurs in one of the groups of the hooded seal (Cystophora cristata),
with a value of 11.9%; the highest known variability was 44.4% in His-
triophoca fasciata of the Bering Sea and 43.1% in Ondatra (Schwartz, 1962).
Within a population and species considerable differences are observed in
the magnitude of variability, 1.е., 15.9 to 29.0% in five samples of Green-
land seals; 24.2 to 43.1% in two lines of muskrats (Schwartz, 1962); and
11.4 to 21.7% in five groups of hooded seals (Table 10, Appendix).

On an average, the variability of absolute weight of the kidneys is
20. 25%. р

Variability of relative weight of the kidneys is a little less. This is
clearly seen in cases where a comparison of a single trait for one and the
same population is possible. . Variability of absolute weight of kidneys in

60 Variability of Mammals

rabbits (Latimer and Sawin, 1955) was 17.4 to 22.0%; variability of
relative weight of kidneys in the same group was 11.3 to 19.5%. Varia-
bility of absolute weight of kidneys in moles was 13.3 to 34.0%; their
relative weight variability was 15.1 to 19.3% (Fateev, 1962).

The same tendency is seen in the weight of kidneys for gerbils;
14.8 to 17.3% was the variability of absolute weight, and 13.1 to 16.5%
the variability of relative weight (Kramer, 1964). On an average, the
variablity of relative weight of kidneys is 15% (ranging from 11.4 to
25.4%).

There are some data available on the variability of kidney measure-
ments. In large animals like cetaceans, it is difficult to weigh such organs
as kidneys or lungs. Hence, we proposed a method of investigating the
linear characteristics of these organs (Kleinenberg, Yablokov её al., 1964).
It appeared that the measurements of length and width of kidneys of white
whales have a very stable variability (Table 19).

Table 19. Variability of measurements of kidneys of white whales
(Delphinapterus leucas) in adult males —

Length of Kidneys:

Right A bie 44.0 +0.86 7.3 -Е 1.37 14
Left BE ke 43.6 +0.79 6.8 + 1,28 14
Width of kidneys:
Right be oh 16.6 -+0.41 9.3-Е1.76 14
Гей С < 17.2 0.45 9.7- 1.83 14
~ Length of body ae Be 374.6 +7.1 7.141.35 14
Weight of kidneys: a: 4,73 +0.42 38.9-+6.3 19
Right us Le, 2.33+.0.20 38.2 +6.2 19
Left a me 2.38--0.22 39.9 6.5 19
Length of body ah a 397.6 +8.0 8.8+ 1.42 = ©

Variability of the weight of the ovaries in chinchillas (Chinchilla laniger)
and rabbits (Latimer and Sawin, 1955) is 38.2 and 56 to 62.7% respec-
tively. Variability of the relative weight of the ovaries in rabbits appears
to be somewhat greater, 50 to 63.8%.

Variability of testis weight in the same populations is 29.5 and 31.4
to 38.7%, respectively, and that of relative weight in rabbits, 26.5 to 29.7%.

The variability of absolute weight of the ovaries in adult gerbils of
the same age, according to Kramer (1964) is 29.5+5.9%, and variability
of relative weight of the testis in two samples of gerbils is 9.3 and 15.1%.
Variability of absolute weight of the testis in different age groups of house
mice (Mus musculus) ranges from 9 to 30% (eight samples, Suroviak, 1965):

From the data obtained during investigations on white whales, the
variability of the width of the urinary bladder was 8.8%; of the length

Nature of Variability in Mammals 61

of the urinary bladder, 16.1% ; of the length of the urethra, 14.8%; of the
length of the ureter, 11.6 to 13.6%. On the whole, the variability of linear
measurements of the urinary tract of white whales is near 11 to 13%.

The variability of measurements for the urinary bladder in gerbils
ranges from 20.2 to 20.9% Юг length, and from 25.7 to 25.8% for width
(Kramer, 1964).

The variability of the weight of the gall bladder in rabbits ranges
from 17.4 to 31.5% (absolute weight) and 13.3 to 22.3% (relative weight).

The measurements of reproductive organs of white whales have shown
that the variability of the length of the vagina is 20.3°%; of the length of
the uterine cervix, 31.1%; and of the length of the penis, 11.6%. These
values are appreciably higher than usual ones for the linear measurements
of organs, which possibly is explained by varying physiological states of
the reproductive organs of the females at the time of investigation. This
assumption is supported by the fact that the measurements of the penis
agree with other values of linear measurements for the urogenital system.

The variability of the weight of the urinary bladder in rabbits (Lati-
mer and Sawin, 1955) ranges from 30.3 to 43.0% for absolute weight,
and 29.2 to 31.0% for relative weight.

Some general conclusions on the values of variability for parameters
of the urogenital system in mammals are given below :

Measurements cent Lime.y.

Weight of kidneys si 20-25 11.4-44.4
Relative weight of kidneys ig 15 11.34-25.14
Size of kidneys ie 7-8 6.3-9.7
Weight of ovaries ge 40 29.5-62.7
Weight of testes see 30 93.0-38.7
Size of urinary pathways ae 15 8.8-25.8
Size of genital organs fe

Females as — 20.3-31.1

Males s Ae — 11.6
Weight of uterus зи — 30.3-43.0

From the summarized data presented, it can be seen that the varia-
bility of weights of organs of the urogenital system is like that of the glands,
while the variability of linear measurements is close to the variability of
organs in the skeletal and nervous system. Unfortunately, weights, which
are characterized by very high values of variability, are usually used in
the study of the urogenital system of mammals; this poses a difficulty for the
correct evaluation of the significance of variability for these or other organs.

The fact that all possibilities of obtaining an exact quantitative analysis
of this system have not yet been exhausted, is shown by the study of the
morphology of spermatozoa by Hughes (1964). According to his data,

62 Variability of Mammals

the variability of the length of the head of spermatozoa in rat kangaroos
(Potorous tridactylus) is only 4.4-0.4%.

General conclusions on the magnitude of
variability for internal organs

In summarizing internal organs, the following picture is obtained
(Table 20).

Table 20. Mean values of variability for internal organs and structures

a a ee Se ee i ee

о Variability of Variability of linear
rgans :
weight measurements

Muscles ze 20 (11.7-27.9) —

Brain sis 6-7 Less than 10 (number
Pancreas, hypophysis te 25 of nerve cells)
Liver оо 25-30

Adrenal ae 25-30

Spleen nee 35-45

Thyroid 50 30-50

Intestine ae 15 8-12
Intestinal volume as 10

Lungs ei 20-30

Trachea ть — 5-7 (number of rings)
Hemoglobin .. Usually about 10

Number of erythrocytes i — 10-15

Heart BG 17-22

Kidneys a 20-25 7-8

Ovaries, testes a 30-40

Urinary and genital systems ite — 15

As seen from the figures given, the minimum values of variability of
weights are those of the nervous system and sense organs (5 to 10%), fol-
lowed by the intestine (10 to 15%). The greatest variability of weight
is characteristic of the thyroid gland, spleen, and testis and ovaries, 30 to

50%; variability of weight of muscles, and the majority of cane in the
mammalian body, falls in the range of 20 to 30%.

The variability of linear measurements is more homogeneous. Varia-
bility of number of rings in the trachea, measurements of kidneys, and
number of nerve cells in the cerebral cortex of adults is less than 10%;
while length of intestine, number of erythrocytes, and measurements of
the urogenital system give values from 10 to 15%.

These conclusions provide good material for researches in the field
of population morphology for mammals, pointing to the potential prospect
of their utilization for characterizing. groups of, linear measurements and
weights of the intestine and parts of the nervous system.

Nature of Variability in Mammals 63

Variability of Body Measurements and Weights

While comparing the variability of general measurements of the body,
it is necessary to select some of the comparable measurements generally
used during the investigations of the majority of mammals. This task
has been made somewhat easy because most of the data given below are
on rodents or insectivores, and their measurements generally comprise
body length, tail length, length of foot, and length of ear. These very
measurements are usually used during studies on other mammals. The
pinnipeds are an exception; they have been compared among themselves
according to specific measurements. The weight of the body has also
been included in the general table (Table 1, Appendix).

The variability of body length in insectivores (Table la, Appendix)
varies from 1.3% for Sorex isodon to 11.8% for Sorex minutus (Schwartz,
1960) and even up to 18.8% for Neomys anomalurus (St. Borowski and Dehnel,
1953). However, the average scope of variability is quite low—from 2 to
3% to 7 to 8%. Extensive data on certain species of insectivores allows
a comparison of different groups belonging to one and the same popula-
tion. Thus, an analysis of the work of St. Borowski and A. Dehnel (1953)
and 74. Pucek (1956) shows that the variability of body length for 30 (!)
groups of the common shrew S. araneus varies from 3.1 to 8.4% (Table 21);
the variability of body length of S. minutus varies from 2.4 to 6.6% (data
of 26 groups, Table 22); the variability of body length for 11 groups of
Neomys fodiens ranges from 3.1 to 6.7% (Table 23).

On the whole, it is clear that the range of mean variability of body
length in Insectivora is from 3 to 7%.

The variability of tail length in insectivores is, as a rule, higher than
the variability of body length and ranges from 0.8 to 13% with an average
of 5 to 8%; Sorex araneus is a single и from the data of Schwartz,
1960-1962 (Table 1, Appendix).

The variability of length of feet appears to be less than the variability
for general body length in Sorex araneus (Schwartz, 1960), Sorex minutus
(Dolgov, 1963; Schwartz, 1962), and the common hedgehog (Markov,
1957). The range of variation for the measurement in the groups investi-
gated is from 1.7 to 8%, averaging about 3 to 4% (Table 1, Appendix).

The variability of ear length in Insectivora is appreciably high; 9.6%
in the hedgehog according to G. Markov, 1957.

The variability of body weight for insectivores is much greater. Mean
values and their range are considerably higher than for any body measure-
ments—from 3.4 to 28.2% in Sorex araneus. On the average, this varia-
bility is higher than 10% and is usually about 12 to 15%, though in indivi-
dual species the variability of weight is less than 10%; (for example, in
three samples of the lesser shrew and two samples of the heterodont shrew,

64 Variability of Mammals

Table 21. Variability of body length for males and females of
Sorex araneus in different years (Calculated from the
data of Dehnel 1953, and Pucek, 1956)

Month KE Sx б 6. 0. Е 5с.0 п Remarks
April 71.0+0.44 3.03 4.30.43 48
Мау 75.4-+- 0.68 2.70 3.6--0.63 16
June 74.4 0.68 3.61 4.90.65 28
1 76 76 о
July en ae Nene caught in 1947.
August 76.5+0.55 3.06 4,0-40.51 31
September 75.30.57 3.87 5.1--0.53 46
October 73.0--0.75 3.61 4.9--0.73 23
April 67.3-40.50 4.58 6.80.53 84
Мау 73.30.63 3.71 5.1--0.60 35
June 75.3--0.30 2.89 3.8-40.28 94 jase
: Born in i
July 75.3-+0.44 3.32 4.4+-0.41 57 caueht anton!
August 75.8+0.39 3.45 4.6 --0.36 79
September 75.6+-0.37 3.17 4,2+0.35 73
October 75.80.46 3.41 | 454043 54
June 78.3+0.51 3.86 4.90.48 58
July 77.4£0.41 3.04 3.9--0.65 56, ра 1948,
August 76.0--0.54 4.62 6.07--0.28 74 ~~ caught in 1949.
September 74.80.53 4.17 5.6--0.50 62
July 68.4+1.31 5.56 8.4+ 1.36 19
Born in 1949,
August 70.7+0.75 3.10 4.40.91 17 caught an) 1050
September 70.7 1.21 4.54 6.4 1.21 14
June 64.7--0.30 2.33 3.60.38 62
July 62.8--0.25 2.80 4.50.28 а 1950,
August 63.4+0.25 2.62 4.1--0.38 108 caught in 1951.
September 63.1+0.38 2.60 4.10.42 47
66.0--0.38 .03 .1-0.
June + 2 3.10.40 1954 (17)
December 64.4+1.10 4.78 7441.20 19
4, 83 4.54 ь :
May 74.70 6.1--0.78 30 1955 (18)
September 72.9--0.69 3.77 5.17 0.67 30

Table 22.

July

September
October

August

AF NAG

August

September

Nature of Variability in Mammals 65

Variability of body length of Sorex minutus (Calculated from
the data of Borowski and Dehnel, 1953)

57.40.86
58.1-40.58
58.6--0.68

48.4 -+-0.69
49.0-++0.39

58.5+0.19
58.50.28
58.60.36
58.80.47

51.4-Е0.49
52.00.24
50.70.32
51.2+0.51

59.10.45
58.3+0.41
59.10.41

51.40.21
51.4+0.20
51.7+0.25
51.5+0.27

52.40.34

51.9+0.24 -

51.6+0.21
50.60.30

6.4-+ 1.06
5.20.71
5.0-=0.80

3.4+0.23
3.60.34
3.8+0.42
3.5-Е0.56

3.60.67
2.4+0.32
3.7+0.45
4.4+0.71

5.41 0:54
4.60.50
4.3 + 0.49

4.1--0.29
4.4--0.27
4.6--0.34
4.40.38

3.90.33
3.90.29
3.7+0.41

Born in 1946,
caught in 1947.

Born in 1947,
caught in 1947.

Born in 1947,
caught in 1948.

Born in 1948,
caught in 1948.

Born in 1948,
caught in 1949.

Born in 1949,
caught in 1949.

Born in 1949,
caught in 1950.

Born in 1950,
caught in 1950.

66 Variability of Mammals

and similarly, four samples of the common shrew studied by the same
author, V. A. Dolgov—all from the Oksk collection).

In almost all instances when samples from six or more populations
were investigated, the variability of body weight appeared to be of a very
different character: from 8.7 to 28.2% in 23 samples of the common shrew
(Table 24), from 9.0 to 14.9% in 13 samples of Neomys fodiens (Table 25),
and from 5.7 to 23.8% in six samples of moles (Fateev, 1962).

On the whole, the variability of the body of insectivores is characterized
by the following values :

Length Micra Lim.,.y,
Body aN 3-7 1.3-18.8
Tail aie By 5-8 0.8-13
Foot an 3-4 1.7-8.0
Еаг ae a — 9.6
Body weight 12-15 3.4-28.2 -

Table 23. Variability of body length of Neomys fodiens in different generations
(Calculated from the data of Borowski and Dehnel, 1953)

i i i a ss ee ee Se а о

Month X+s% (oy 6.V.+S¢,y n Remarks
June 86.11.04 5.09 5.90.85 24
July 88.8+0.71 4.09 4.60.57 33 ~— Born in 1948,
August 88.4 1.12 4.46 5.10.89 16 caught in 1949.
September 86.40.76 3.41 4.00.62 20
August 80.80.67 2.92 3.6--0.60 18
September 80.140.49 2.45 3.14044 24 Born in 1949,

caught in 1949.

October 79.2 +1.00 4.13 5.2+0.89 17
July 80.40.88 4:58 5.70.78 27
August 80.61.05 5.44 6.70.92 27 Born in 1950,
September 80.3-40.91 4.57 5.70.80 25 caught in 1950.
October 81.9-L0.75 4.47 5.5-Е0.64 36

Data on the variability of body measurements Юг lagomorphs show
that of the body measurements, the variability of body length is least
(3.3 to 6.8%), and variability of ear length is greatest (13.8 to 15.1%)
(Table 26). Re)

Extensive research on the variability of body weight (in 47 samples |
of rabbits; Castle and Reed, 1936), allows one to determine that the upper
limit of variability is similar in all groups investigated (about 16 to 20%),
and the lower limit of variability ranges from 4.8% to 13.6%.

Nature of Variability in Mammals 67

Table 24. Variability of body weight of Sorex araneus in different
years in one population (Calculated from the data of
Borowski and Dehnel, 1953, and Pucek, 1956)

Е 6. U.+Se.y n Remarks
August 7.05+0.04 0.78 11.10.44 322
September 7.06+-0.07 0.89 12,640.74 145 вогп in 1947,
October 7.01-0.11 0.95 13.6+1.11 74 caught in 1948.
November Gi8G=5 0413) 0.73 106—157. 30
June NOT ON 153 14,741.35. 59
July = ЛО 9.80.93 56
August 11.02-Е0.15 1.29 11.70.96 US Born in 1948,
September 9.85-0.15 1.21 12.3-+1.10 62 caught in 1949,
October 9.22+0.21 0.89 9.7 1.61 18
June 6.29--0.05 0.67 10.740.56 178
July 6.57--0.04 0.85 12.9--0.41 504
August 6.9340.04 0.90 13,040.40 532 Born in 1949,
caught in 1950,
September 6.98 0.09 0,93 13.30.90 110
October 6.42 --0.15 0.86 13.4 1.65 33
June 9.92 E0525 2-26 22.8+3.70 19
July 9.16+0.59 2.58 28.24.57 19
August 9.82+40.24 0.98 10,041.71 17 Bornin 1950,
caught in 1951.
September 9.45 {0.27 1.25 13.2 -- 1.99 22
October SiON ema 101 619 14
June 6.1 £0.10 0.53 8.71.44 29 о.
|
December 6.4 +0.02 0.66 10.3-+- 1.72 18
May OP SLO 17 14.4+1.86 30 те
September Of O25 1 BOL а Е

68 Variability of Mammals

Table 25. Variability of body weight of Neomys fodiens in different generations
(Calculated from the data of Borowski and Dehnel, 1953) |

Month Я 5х б Cr О.Е п Remarks
June 16.77 0.36 72 10.3 1.48 24
July 16.98 0.39 2.22 13.1-= 1.61 33 Born in 1948,
August 16.97 0.38 1.52 9.0-Е 1.53 16 found in 1949.
September 16.00--0.40 1.79 11.2-+1.77 19
August 12.64--0.28 1.23 9.7-Е1.62 18
Заре 12251037 183 М6 22 = oo
found in 1949.
October 12,710.39 1.62 12.82.79 17
June 11.38+0.29 1.31 11.55 1.78 21
July 11.24+0.29 1.49 13.34. 1.80 27
August 12.19+0.30 1.57 12.9+1.75 27 Born in 1950,
September 12.54+0.31 1.51 12.0-- 1.74 24 found in 1950.
October 12.49+.0.30 1.80 14.4+1.70 36
November 13.32 + 0.23 1.41 10.61.21 38

Table 26. Variability of weights and body measurements of the black-tailed
jack rabbit (Lepus californicus) (Original data by Bronson, 1958)

Males Females
Wri, ee Se ee

XE SX Фо Ел Я 5х 6. 0-Е 5са
Body length 533.9-Е 1.5 4.20.20 551.9 1.7 4,540.22
Ear length 107.0+1.0 13.80.66 106.4-- 1.1 15.10.73
Foot length 125.6+1.0 11.8+0.56 126.5+1.1 12.7-L0.61
Tail length 73.141.2 8.4+1.16 78.1+2.3 12.8+2.1
Body weight 90.9+0.6 9.8+0.47 102.8+0.8 11.3+0.55

Detailed data on the variability of body measurements for the order
Rodentia are available (Table 1, Appendix). In this table, data on the
variability of body measurements and weight for more than four hundred
different samples of 21 species are presented. Of course, the number of
species studied is but a small part of the order, but these data no doubt |
reflect the real characteristics of variability for the group.

The range of variability of body length in rodents is great—from
0:5% in field voles (Microtus agrestis) to 12.9% in yellow-necked mice
(Apodemusflavicollis) and even 21.0% for one of the 24 lines of laboratory mice.

Nature of Variability in Mammals 69

However, in the majority of cases, the variability of body length ranges in
different species of rodent from 3 or 4% to 6 or 8%, approaching 5%
on the average (Table 1, Appendix).

Species with many samples can exhibit a large range of variability for
body length even in one population of one species; (data on yellow-necked
and forest mice of the genus Peromyscus and especially on mice of different
populations under laboratory conditions). At the same time, natural
data obtained on a number of samples in a species, show little scattering
in values of coefficients of variation: 5.7 to 9.8% for field mice (Apodemus
sylvaticus, eight samples); 3.7 to 5.6% for deermice (Peromyscus leucopus,
Clark, 1940; eight samples); and 3.1 to 5.2% for nine samples of P. mani-
culatus (Table 27). :

It is interesting to compare the three genera represented by the largest
quantity of material on the variability of body length—Apodemus, Pero-
myscus, and Mus. It can be said confidently that the coefficients of varia-
tion for body length in the genus Peromyscus are the most stable and also
the lowest—3.1 to 7.6% for 83 samples (Table 1, Appendix). The genus
Apodemus takes second place in stability and value of variability, 5.1 to 12.9%
for 34 samples. The genus Mus is characterized by the most scattered
coefficients of variation—0.7 to 21.0% for 56 samples; very small as well
as very high values of this index were found in the three families. Strictly
speaking, such a comparison on the available material can only be condi-
tional because most of the data on the-genus Mus were based on experi-
mental populations in laboratories; whereas the data for the other two
genera were obtained from natural populations. But the differences
between the genera Peromyscus and Apodemus are sufficiently clear so that
a general conclusion can be drawn about the possible range of variability
in the generic structure. In perusing the data on the variability of different
taxa of pinnipeds (Chapter IV), such specificity in coefficients of variation
is not observed. However, the data on pinnipeds are, at best, composed
of 25 to 30 samples from each genus; hence, it is possible that if more exten-
sive and homogeneous material could be studied, some weak patterns
could be detected.

The values of variability for body dimensions of rodents approach
5% on the average, but the variability of tail length in all groups is, as a
rule, high. The overall range of variability is also high: from 0.8% in
one of the six samples of field voles studied, up to 26.5% in one of the samples
of laboratory mice, and even up to 31.5% in one of the eight samples of
field mice (Table 1, Appendix). On the average, it can be said that the
magnitude of variability of tail length ranges from 8 to 10%. While
analyzing the data on variability of tail length, it is interesting to see that
the whole genus Peromyscus differs appreciably in having both constant
and very low values of variability : 3.2 to 8.9%, averaging about 5%.

70 Variability of Mammals

Table 27. Variability of body length for some rodents

ee

ЖЕ 6. V.AS, y n Remarks
Apodemus flavicollis
(Original data from К, А. Adamczewska, 1959)

August Be 100.0-L 0.88 9.0--0.63 103 Males

97.00.73 7.60.53 102 Females
September 7. . 102.0-L0.48 6.70.27 248 Males

99.0-+0.40 6.8+0.29 279 Females
October ae 103.6--0.95 6.30.65 46 Males

96.6-| 1.44 6.8-L0.83 33 Females
November te 105.1 -L0.64 5.7+0.43 86 Males

98.2 0.70 6.4£0.51 80 Females

Peromyscus maniculatus
(Clark, 1940)

Populations
I — 4,1-.0.46 — Males and females of the same
II — 4.20.30 — age produced in the labora-
Ill — 6.20.98 — tory from one or two pairs
ТУ — 4.20.23 — taken from each natural
У — 4.5+0.21 — population
VI — 5.1--0.42 —
УП — 4.5--0.24 —
УП — 4.20.49 —
Р. maniculatus
(Original data according to Dice, 1938)
Populations

I 94.6--0.38 4.30.28 116 Males and females of about
II 94.60.50 3.70.37 49 опе уеаг оГазе
Ill 93.60.50 5.20.38 96
IV 96.90.41 3.60.30 74
У 93.3--0.37 3.1--0.28 61
УТ 93.50.67 3.30.51 22
УП ‘95,840.55 5.30.41 84
VIII 91.8-L1.26 4.6--0.97 11
IX 91.40.75 3.9+0.58 23

The variability of absolute and relative tail length can be compared
in two instances. The variability of absolute tail length in P. maniculatus
(16 samples, Sumner, 1920) is 3.9 to 5.2%, while the variability of relative
tail length is ostensibly lower, 3.2 to 4.8%. In the second instance, with

Nature of Variability in Mammals 71

house mice, the situation seems to be the reverse. The variability of
absolute length of the tail in two samples ranges from 4.1 to 7.2%, whereas
the variability of relative tail length in the same samples varies from 5.3
to 7.5% (calculated from Sumner, 1909).

During the analysis of data on variability of foot length in insectivores,
a tendency was observed toward a lower variability for this trait than for
body length. This feature is confirmed by more extensive data on rodents.
The variability of foot length was not more than the variability of body
length in a single group under study. On the whole, the variability of
foot length ranges from 1.5 (for Peromyscus maniculatus) to 11.49 (for one
out of five samples of bank voles, Clethrionomys glareolus, Table 1b, Appendix)
and approaches 3 to 4% on the average. Even in the groups with many
samples, a surprising similarity in variability of this trait in different samples
is observed : 2.7 to 5.8% for 12 samples of yellow-necked mice (Apodemus
flavicollis); 3.9 to 7.1% for another 12 samples of the same species; 3.5 to
6.9% for seven samples of red-backed voles (Clethrionomys rutilus); 2.5 to
3.6% for eight samples of Peromyscus leucopus (Clark, 1940); 2.5 to 4.1%
for 22 samples of P. maniculatus; 1.5 to 3.1% and 2.5 to 4.0% for 16 and
30 samples respectively of the same species (Sumner, 1920; Clark, 1940).

Even in the genus Mus, where linear parameters vary so greatly, the
variability of length of foot is very low and concentrated: 2.3 to 5%
for four samples and 2.9 to 4.0% for six other samples (Table 1, Appendix).

A comparison of values of variability for ear length with values of varia-
bility for other body measurements seems to place the values for ear length
between the low variability of foot length and the high variability of body
length. The range of variability for this trait is quite restricted, from
2.7% (Peromyscus maniculatus) to 11.1% (Clethrionomys glareolus), with 5 to
6% as an average. In some cases, variability of ears is considerably less
than variability of body length—for example, in some samples of deer-
mice, Peromyscus maniculatus, and the house mouse (Table 1, Appendix),
In other cases, variability of ear length is definitely higher than that of body
length: common susliks (Citellus citellus), bank voles (Clethrionomys glareolus),
and red-backed voles (Clethrionomys rutilus) (Table 1, Appendix). The
genus Peromyscus has a low variability of ear length and other measurements.

The available data show that the variability for relative indices of body
measurements of rodents is considerably higher than that of the similar
absolute indices. Thus the variability of ear length in the house mouse
varies from 2.7 to 4.4% and the variability of relative length of ear in the
same sample varies from 4.9 to 6.1% (calculated from Sumner’s data,
- 1909). А similar situation arises in the variability of foot length in the
same samples; the variability of relative values appears to be considerably
higher than the variability of absolute values.

As a rule, the variability of body weight in rodents is always higher

72 Variability of Mammals

than the variability of linear measurements. There is considerable varia-
tion in this trait in comparable samples and specimens : from 1.4% in
lemmings (Lagurus lagurus) (calculations from the data of А. I. Kryl’tsov’a,
1957) to 30.9% in a natural population of Mus musculus (estimated from
the data of J. Dynowskii, 1963). The average values, inasmuch as they
can be determined from quite different types of data, approach 13 to 14%.
According to the data of К. Adamczewska (1959), J. L. Carmon её al.
(1963), Е. Sumner (1909), and W. Е. Castle её al. (1936), the variability among
individual groups belonging to a species is quite compact; for example,
in yellow-necked mice, 15.1 to 24.6% for eight samples; in house mice,
6.3 to 8.5% for eight samples, and 9.8 to 12.0% for another eight samples.
In other cases, within the limits of one species, very scattered values of
coefficients of variation for body weights are encountered: for example,
4.1 to 9.7% in 14 samples of field voles (Microtus agrestis) according to John
Newson and D. Chitty (1962); 8.9 to 28.2% for 84 samples of rats of diffe-
rent ages (King, 1923); 5.6 to 22.7% for 24 samples of house mice under
laboratory conditions, calculated from data by L. Law (1938); 1.4 to 12.0% |
‘for nine samples of lemming (Г. lagurus), calculated from data by A. Г.
Kryltsov’a (1957).

On the whole, variability of measurements and body weight in rodents
can be defined according to the following mean values :

Koy, Lime»,
Body length ie 5 5 0.5-21.0
Tail length ti | we 8-10 0.8-26.5
Foot length we bi 3-4 1.5-11.4
Ear length И Me 5-6 2.7-11.1
Body weight ae ss 1-15 1.4-30.9_

The last order for which there are extensive data on variability of
body measurements is the order Pinnipedia (Table 1, Appendix). Data
on eight species, comprising more than 95 samples, showed that the уама-,
bility is similar to that of the orders Rodentia and Insectivora; it is thought
that such tendencies are characteristic for each order as a whole.

Very detailed work is available for seals on the variability of body
length. From the data presented in Table 1, it is seen that the variability
of body length varies quite widely, from 1.4 in one of the samples of Pusa
hispida to 18.7% in one of the two samples of the Baikal seal (Pusa sibirica).

Considerable variations in values of variability are not only found
among species, but even within species among different samples: 8.1 to
18.7% in Baikal seals, 3.1 to 7.4% in Greenland seals, 2.4 to 8.4% in Cas-
pian seals, etc. A group of four samples of Pusa-hispida in the eastern seas
forms ап. exception, with 7.5 to 9.3%. It is possible that such heterogeneous

Nature of Variability in Mammals 73

coefficients of variation for length of pinnipeds result from insufficient
homogeneity of age in the samples (see Chapter IV, the section on varia-
bility by age), a characteristic failing in the study of all large mammals
with high longevity.

All other body measurements of pinnipeds differ from the measurements
of terrestrial mammals and cannot be compared with them. It may be
noted that the variability of all the rest of the body measurements studied
in pinnipeds, appears to be higher than the corresponding values for body
length—a situation which is similar to the one present in other orders of
mammals.

The variability of body weight in Pinnipedia fluctuates greatly, from
5.0 to 11.3%; the last figure concerns the variability of weight of embryos
in fur seals (Scheffer, 1962) and cannot characterize the variability of adult
animals. Some data on adult animals show that the variability of body
weight for Greenland seals is nearly 10%.

The variability of body length of ungulates and carnivores appears
to be close to the variability for other orders studied, averaging about
3 to 5% (Table 1b). The variability of body length in a comparatively
large group of sables (Martes zibellina) ranges from 1.7 to 3.8% in ten samples
(calculated from the data of Kuznetsov, 1941а). Similarly, the variability
of captive foxes is also low although it is for various types: from 2.3 to 7.9%
(Fateev её al., 1961). The variability of body measurements for mule
deer (Odocoileus hemionus) appears to be low, 1.7 to 5.5% in 13 samples of
different ages (Anderson её al., 1965; Klein, 1964). The variability of
body length for horned cattle is also not very great, 2.4 to 4.0% (Tsalkin,
1960).

The variability of body length of white whales (Delphinapterus leucas)
does not differ from that of other mammals, ranging from 3.4 to 12.4% in
14 samples.

The weight of sheep, even under artificial conditions of rearing, does
not exceed 20.7% in 39 samples (Popova, 1941). The variability of body
_ weight for two populations of Odocoileus hemionus in 12 samples which con-
trolled age and sex, varied from 8.9 to 34.7% (averaging about 16.5%)
(Anderson её al., 1965), and from 10.4 to 12.8% in a red deer population
(Cervus elaphus) (Brna, 1964): This speaks for a considerable constancy
of this trait in the main representatives of this group. The variability of
body weight in captive foxes varied from 5.4 to 19.4% (Fateev et al., 1961).

Before summarizing the data obtained on the variability of body
weight for primates, it is to be remembered that the variability of body
measurements probably ranges within the same limits. Instead of dwelling
on the vast amount of anthropological data, I will turn to the recently
published material of D. A. Zhdanov and B. A. Nikityuk (1964) on the
variability of body height in a large number of Muscovites of similar age;

74 Variability of Mammals

for 17 groups, the variability is very stable, from 3.0 to 3.8%. On the
other hand, it is known (Schultz, 1926) that the variability of tail length
in 11 species of higher primates ranged from 6.6 (in Callithrix jacchus) to
30.2% (in Symphalangus syndactylus).

The data presented on the variability of body parameters in various
orders of mammals confirm some of the general findings of the researches
on variability in the best studied orders, insectivores and rodents. These
results are given below :

A. The variability of foot length in terrestrial mammals has the
lowest value (body length in aquatic mammals) and then, in the ascending
order of variability—body length, tail length, and finally, ear length.®
The variability of body weight is higher than the variability of any linear
body measurement;

B. Differences in variability are observed in all traits among different
species and genera and, similarly, among different populations within one
species. However, in a number of cases a specific magnitude of variability
can be detected for a genus;

С. As a rule, the variability for relative values appears to be not
lower, and in a number of cases higher, than the variability of absolute
values for the same) traits;

D. It seems that for all terrestrial mammals, the following values
of variability, which are absolute values of variability (in percentages)
in a sufficiently homogeneous sample, are characteristic:

Body lengh a se со 4-6
Tail length We Бо a 5-10
Foot length te го aye 3-4
Ear length 50 со os 5-6
Body weight be 56 Se 12-15

Е. A slightly higher variability of body measurements is charac-
teristic of Pinnipedia and Cetacea.

General Character of Variability of Different Systems
and Body Measurements of Mammals

After an inevitably short perusal of the variability of the basic system
of organs and body measurements for different mammals, an attempt to

5 Though variability of ear length appears to be one of the most variable of linear
measurements, it appears that the range of variation in the variability value of this trait is
quite constant. It is found that in many instances the minimum coefficient of variation is
twice as small as the maximum one: for body length, it is 20 times; for body weight, 12 times;
for tail length 9 times; for foot length, 7 times; and for ear length, only 2 times!

Nature of Variability in Mammals 75

define the variability of different systems of individual organs is possible
(Table 28).

Table 28. Average values of variability of structures,
organs and body measurements of mammals

Mean values of coefficients of
Organ, structure, variation
measurement SS SS SS SS SS SS Remarks

eee, —=n eS eee eee ee

Hair Bi 3-8 15 20

Vibrissae кН = 5-10 — In cetaceans, nearly 20

Skin . a 10-20 — 10-15

Skull за 4-6 — 15

Teeth ae 5-10 10 — Higher in molars and

‘ canines

Post-cranial skeleton .. 3-5 1 10-15

Muscles al — — 20 Relative weight is lower

Brain ay — 10 6-7

Glands Bes — — 20-30 Spleen, thyroid, and ovary

up to 50%

Gut Ба 10 — 15

Lungs, trachea Ne — 5-7 20-30

Blood, heart os 10 15 20 Relative weight of heart, 15
Нето- Erythro- Heart
globin cytes

Urogenital Me 10-15 — 20-35

Body measurements .. 5-10 — 12-15

ee

It is important to emphasize again that the data presented in the
Table must be looked upon as guidelines (greater details of the systems
are given above). ‘These guidelines show that the variability for organs
of different systems and different measurable parameters within a system
of organs differ considerably from each other. This is a trivial result. How-
ever, such a pattern of variability gives rise to conclusions which will
change some widely-held viewpoints.

All organs can be divided provisionally into large groups in regard
to variability.

The variability of organs in the first group is equal to, or less than,
10%; that for the second group is 10 to 15%; and that of the third group
is at least 15%. The first group comprises (in order of ascending
variability) :

1. Linear measurements of the post-cranial skeleton,
2. Linear measurements of the skull,

3. Number of rings in the trachea,

4. Absolute weight of the brain,

76 ~~ Variability of Mammals

5. Linear measurements of hair,
6. Linear measurements of teeth,
7. Linear measurements of the body,
8. Number of vibrissae.
The second group® consists of:
1. Linear measurements of the digestive tract,
2. Quantity of hemoglobin in the blood,
3. Number of elements in the post-cranial skeleton,
4. Number of elements in the nervous system.
The third group comprises:
1. Weight of the skin,
Weight of the post-cranial skeleton,
Body weight,
Linear measurements of the urogenital] system,
Number of hairs,
Number of erythrocytes,
Linear measurements of the skin structures,
Weight of the skull,
Weight of muscles and heart, and similarly, weight of hair,
Weight and volume of the digestive tract,
Weight of lungs, stomach, the majority of glands, parts of the
urogenital system,
12. Weight of the salivary gland, the thyroid gland, and the ovary.
The list of organs in the first group should help in reconsidering to
some extent the field and laboratory methods of researches on the variability
of mammals. Out of all traits given in the first list, usually only two
groups are studied—measurements of the skull and the body. The constant
and sufficiently simple data like number of rings in the trachea, characters
of the hair, and above all, the characters of the post-cranial skeleton
(which are easily analyzed) have not yet attracted enough attention.
Further, one of the objects of study in the field of population
morphology of mammals is to collect data for determining limits and
parameters of variability of all groups of systems for a single purpose.
It is possible that in the near future it will be feasible to determine the
degree of ecological plasticity ina species and the magnitude of its specia-
lization with the help of a “Мар of Variability” for one or another group of
mammals. But, for the time being, it is important to carry out the study
of animal ecology and the micro-evolutionary process of a population,
with the help of an exact determination of the place of the population
under study according to the location of an. appropriate organ on the
scale of variability.

eS: OS OPN ages COL NS)

— —

6 The order of placement of traits in the second and third groups is affected at times by
insufficiency of data.

CHAPTER III

AN ANALYSIS OF VARIABILITY OF POPULATIONS AT
DIFFERENT TAXONOMIC LEVELS WITHIN AN ORDER

For an evaluation of variability from the viewpoint of evolution,
- it is important to know the characteristics of variability for the same
organs in representatives of different phylogenetically compact groups
as well as in representatives of ecologically compact and diverse sets of
species. During such an investigation, the necessity arises for a gradual
comparison of variability for different populations within a species, within
a genus, within a family, etc.

_ Data on the variability of traits in three populations of Greenland
seals (Pagophilus groenlandicus) in the East Atlantic allow one to examine
the question of variability for populations within a species. This same
approach is useful in an examination of data on the morphology of ringed
seals (Pusa hispida).

In the genus Pusa, a comparison of variability for the Baikal, Caspian,
and ringed seals permits the comparison of variability within a genus.

A comparison of variability of traits for the Greenland seal and
Fistriophoca fasciata allows one to look into the problem of variability in
related genera, and a comparison of variability between the genus of
common seals (Phoca) and that of ringed seals (Pusa) helps to survey the
problems of variability between less similar genera.

A comparison of variability of common seals and Greenland seals
allows one to examine the variability of different groups in one subfamily,
whereas a comparison of variability of the hooded seal (Cystophora cristata),
bearded seal (Erignathus barbatus), and common seal permits one to examine
the variability of different subfamilies. Finally, from the data on fur
seals (Callorhinus), it is possible to compare two basic trunks in the order

78 Variability of Mammals

Pinnipedia—the earless and the eared seals.

Thus, from this example of pinnipeds, a comparison of variability
for groups within an order on different taxonomic levels seems
possible. i

Comparison of Variability of Populations within a Species _

Variability of traits for three populations
of Greenland seals of the East Atlantic

Greenland seals formed the very basis of the hunting industry in
the East until recently. Three different places of hunting (White Sea,
Yan-Maien region, and the Newfoundland region; Figure 23) facilitated
the idea that three distinct populations of this animal existed; however,
the degree of inter-dependence among these populations, and their origin,
remained unknown until recently (Chapskii, 1961; Yablokov, 1962, 1963,
1963c; Sergeant, 1963). At the same time, it is necessary to know the
degree of inter-dependence exactly for a rational utilization of stocks
because all the populations intermix at places of summer migration and
constitution of each group is uncertain. Hence, the industrial exploita- —
tion of each of the three regions should perhaps be different.

It must be noted that the variability of morphological structures
did not attract the attention of researchers working on Greenland seals
until our work was carried out. A comparison of absolute indices was done
before [Khuzin, 1963; Yablokov, 1963; Yablokov and Serzhent (Sergeant),
1963; Yablokov and Klevezal’, 1964; Yablokov and Etin, 1965; Kleinenberg
et al., 1965; and others], and hence instead of discussing these aspects here
again, a direct analysis of indices of variability like the standard deviation
and coefficient of variation may be made.

Comparison of the standard deviation. The standard deviation of a
population (sigma) shows the degree of inclination for portions of a
cumulative distribution curve characterizing the value of the trait and,
it being a dimensioned number, allows a comparison of only those traits
with similar absolute measurements (Chapter I). ‘This situation creates
an interest in comparing these indices for different populations with sex
and age controlled.

The first stage of such a comparison is to compare the absolute values
of the standard deviation for different populations. A graphical rep-
resentation of the values of the standard deviation for a number of traits
other than the skull, and also for the skull, in adult males and females
of different populations, is given in Figures 10 and 11 respectively.

The values of the standard deviation are very close to each other
in all populations for all the traits studied; individual variation does not
exhibit a particular tendency.

Variability of Populations 79

Considering that the disposition of traits on the abscissa is arbitrary,
we can continue to combine these traits within each group arbitrarily.
(It is because the disposition of traits on the abscissa is arbitrary that it
is not possible at times to join the points indicating value of sigmas; a
histogram would have been the only solution. But, in view of the great

Females

123 4 5 6 7 8 9 10111213 141516 1718 19

Figure 10. А comparison of absolute values of the standard deviation for traits other than
the skull in three populations of the Greenland seal (Pagophilus groenlandicus) :

On the abscissa: 1—Number of sternal ribs; 2—Number of asternal ribs; 3—Number of
caudal vertebrae; 4—Number of tracheal rings; 5—Body length; 6—Distance from the anus to
the navel; 7—Distance from the anus to the opening of the urethra; 8-16—Number of vibrissae of
lip according to rows; 17—Usual number of lip vibrissae; 18—Number of vibrissae of eyes ;

19—Number of vibrissae of the nose.

80 Variability of Mammals

ease for examining the situation which the curve affords, and considering
the conventional conditions of its construction, it is appropriate to use it.)
One of the methods for combining traits, where a gradual decrease
of the standard deviation value for skull traits of the White Sea popula-
tion has been taken as a basis, is shown in Figure 12. Specific conclusions
can be derived from the analysis of standard deviations for males in this
figure (Figure 12).
o~
8

Males

Females

12345 7 В 1112 13 1415 18 19 20.2324 25 26

Figure 11. А comparison of absolute values of the standard deviation of skull
measurements for three populations of the Greenland seal:

On the abscissa: 1—Condylobasal length of skull; 2—Length of base of skull; 3—Length
from occiput to nose; 4—Length of palate; 5—Length of upper tooth row; 7—Wiéidth of mastoids ;
8—Width of brain cavity; 11—Width of palate; 12—Width of snout; 13—Width of occipital
condyles; 14—Length of right auditory bulla; 15— ИП of right auditory bulla; 18—Length
of right nasal bone; 19—Wéidth of nasal bone; 20—Width between eyes; 23—Length of right о

lower jaw; 24—Height of right lower jaw; 25—Length of tooth row of lower jaw ;
26—Height of lower jaw behind molars.

Variability of Populations 81

Males

Females

Figure 12. A comparison of absolute values of the standard deviation for skull measure-
ments in three populations of the Greenland seal; skull measurements of the White Sea
_ population are given along the abscissa in order of decreasing values of sigma.

1. The general features of the values of the standard deviation
coincide in all three populations; . ;

2. There is a similarity of behavior in the values of the standard
deviation for the populations from the Yan-Maien region and the New-
foundland region but less when comparing either of these populations
with the White Sea population. It is seen from Figure 12 that the change
of the value of the standard deviation in the regions of Yan-Maien and
Newfoundland is absolutely synchronized from the second trait to the
ninth; and ae

3. The absolute value of standard deviation for the Newfoundland
population is, as a rule, less.

An analysis of the transformed graph of the sigma values for skull
measurements of females, leads to additional observations. ‘The absolute
values for the population of the White Sea exceed those of the Yan-Maien

82 Variability of Mammals

region (in 13 measurements out of 19). The main similarities between
the Yan-Maien and Newfoundland populations in the configuration of
the curves are from the third trait to the fifteenth inclusively.

And 50, on the basis of an analysis of absolute values of the standard
deviation for the traits under study, a comment should be made on:
1) the similarity in the variability of traits for all populations, 2) the
great similarity between the populations of Yan-Maien and Newfound-
land region, and 3) the comparatively lower variability for a number of
traits in the Newfoundland population than for the other two populations.

For а further analysis of the standard deviation, let us use the
method proposed by 5. В. Zarapkin (1934-1939) to analyze the differences
among populations of insectivores. The Zarapkin method consists of a
comparison of the relative values of standard deviation. Any population
can be taken as the standard: in the present researches, the White Sea
population has been taken as the standard:

с White Sea с White Sea
с Yan-Maien Region anc в Newfoundland Region

No specific regularity for either males or females is observed when
comparing the standard deviation of absolute values, nor during the
comparison of relative values of sigma obtained from the initial perusal of
the curves (Figures 13 and 14). Ш such cases, one main factor is out-
standing and that is a great common character in all populations. It
is clearly seen that the majority of the ratios are above one; this reveals
the comparatively greater magnitude of sigmas for the White Sea popula-
tion than for similar populations.

The relative values of sigma can be arranged in a different way.
This method helps to compare sigmas not only of the organs within a
system but even organs of any measurements, because the values with which
we are dealing now are concrete.

So it is possible to propose one additional step in the formalization
which would lead to better future comparisons. While comparing relative
as well as absolute values of the standard deviation, a leveling of all
traits takes place!; hence it is possible that it just does not matter whether

1 Dissenting voices (among morphologists) are generally raised against mathematical
analysis of data, especially since the biological value of a trait is removed. Of course, normal
functioning of the nervous system or even only the brain, is more important for an animal
than completeness of dental system or having the full number of phalanges in fingers, etc.;
secondly, selection in nature proceeds on “important’’ as well as “unimportant” organs
and structures and the main process in effective selection will probably be the same in all
instances. It has again to be understood that the concept of importance of organ is to a
great extent relative and, as а rule, the majority of “unimportant” organs become vitally
important at a particular moment of life.

Variability of Populations

2.5 \
IZ
5 il [=
2.0 3 |! ТЕ
ЩЕ И ГЕ
Pei NS i | Г |2
%, => | ioe
1.5
|

0.5

Figure 13.

A comparison of relative values of the standard deviation for traits in
females of the Greenland seal for three populations (Sigma values of the

White Sea population have been taken as standard).

Figure 14. А comparison of relative values of the standard deviation for traits in
males of the Greenland seal for three populations (Sigma values of the
White Sea population have been taken as standard).

sigma of the Newfoundland region coincides on the graph with the position
of sigma for the same trait from the region of Yan-Maien. And if this is
so, there is a possibility of formulating two independent courses, gradually

declining in a sigma value, in relation to the White Sea population whose
sigma is taken as a unit.

83

84 Variability of Mammals

An analysis of such curves as seen in Figure 15 is not difficult and
the conclusions are very specific. The population in the Yan-Maien
region (II) is much nearer to the White Sea population, and the New-
foundland population (1) is significantly farther away in respect to standard
deviation values for the traits studied.

Figure 15. A comparison of relative values of the standard deviation
in three populations of the Greenland seal.
(Explanation given in the text).

An analysis of the distribution of the standard deviation values,
unlike the analysis of absolute indices characterizing this or that organ
or structure, does not lead to any ecological conclusions, but makes the
conclusions about the degree of closeness of populations in respect to the
traits studied, more understandable and sound. ‘This conclusion is based
not on the variability value of one trait or of a small group of functionally
homogeneous traits, but of all the characteristics of morphological structures.

Comparison of coefficients of variation. ‘The second stage of the analysis
is to analyze the variability of different populations expressed in coeffi-
cients of variation.

As a start, let us see how the magnitude of the coefficient of variation
is distributed for different groups of the traits studied (Figures 16 and
17). In a number of cases, the values of the coefficient of variation may
differ significantly, but in fact they are determined by very accidental
reasons, as, for example, during the comparison of variability in the last
(eighth and ninth) lines of vibrissae on the lip region. When the arith-
metic means for the traits are very small, an increase of a tenth or even a
hundredth changes the value of the coefficient of variation. Naturally,

—-

a ee ee за

Variability of Populations 85

a great biological importance could not be attached to such deviations.’

It is more important to take into consideration any general tendency for

‘псгсазе or decrease of variability in the given trait.
When approaching the data on variability of adult males and females

c.v.%

4 5

10

1234567891011 121314 151617 18

15
Females

10

1234567 89 1011 121314 15 1617 18

Figure 16. Coefficient of variation for number of vibrissae (1-11), body measurements
(12-14), number of tracheal rings (15), caudal vertebrae (16), and ribs (17-18) for three
populations of the Greenland seal (Pagophilus groenlandicus) and (* --”? symbols)

the ribbon seal (Histriophoca fasciata).

2 In the given example, it happens because the nature of the distribution of number of
yibrissae in the last lines does not coincide with the formula of the normal distribution.

86 Variability of Mammals

in the previously studied populations of the Greenland seal with such
criteria, it becomes necessary first of all to note the great homogeneity
in the behavior of the coefficient of variation in a comparison of the
variability of vibrissae, body, and skull measurements, and other traits.

Some traits, whose variability deviates from the common tendencies,
come to light at once on the above background. In males these traits
are low variability of number of vibrissae in the first lip line of the New-
foundland population, reduction of variability in the measurements of
“anal orifice to opening of urethra,” reduction of variability in number
of rings of the trachea from White Sea males, and the interrelations of
variability for length and width of nasal bone in the Newfoundland region
(measurements 18-19). In females, the deviations are expressed by a
sharply increased variability in the number of caudal vertebrae for the
populations of the White Sea and Yan-Maien regions, and similarly, by
increased variability of mastoid width and braincase width of the skull
(measurements 7-8) in the White Sea population.

C.v.%
20

Males

C.V.%

Females

i jean т
12 34 5 7 8 1142 13'14 15 18 19 2023 24 25 26

Figure 17. Coefficients of variation for skull measurements of three populations
of the Greenland seal. For measurement legend, see Figure 11.

Variability of Populations 87

An analysis of the factual data shows that the majority of the devia-
tions noted can be explained. The differences regarding the variability
of vibrissae need not be considered because the sampling error of the co-
efficients of variation generally masks the discrepancies. The fact that
during the collection of White Sea data, observations of adult males were
totally lacking, and only a mixed catch (consisting mainly of new-borns)
was available for comparison, could have caused the lower variability
observed for the number of tracheal rings. The reported tendency in
arrangement of traits for other populations with regard to variability
(Figure 16) allows one to assume that in an adequate sample, the varia-
bility in number of vibrissae for the White Sea population would be about
10%. The difference in the magnitude of skull measurements, determin-
ing the deviation of those coefficients of variation which affect the general
trends, also seems to be quite insignificant when expressed in qualitative
terms. As a result, the sharp decline in variability for only one trait in
the body of malesin the White Sea remains as real. This difference can-
not be explained on the basis of insufficiently representative sampling and
small differences in the variability, and thus it attains a special importance
if it is compared with less detectable similar situations arising during the
comparison of variability of these very traits in females (Figure 15).

A comparatively sharp increase in variability of number of caudal
vertebrae over the variability of rings in the trachea, is observed in the
White Sea females. This is unlike the situation for females from the New-
foundland region and, to a lesser extent, from the Yan-Maien region.
Data on females are sufficiently large and are trustworthy. In other words,
during the analysis of variability of males, instead of a comparatively in-
creased variability of number of tracheal rings, now a large variability of
number of caudal vertebrae in comparison with the variability of number
of tracheal rings draws attention. Whatever the case may be, the differ-
ence in variability for these traits in the populations studied is evident, and
the coincidence in behavior of coefficients of variation in the samples of
males and females makes this difference more convincing.

The next step in the analysis of variability is to analyze the factual
results obtained in conformity with the appearance of simple relationships
among parameters (Chapter V). This method offers a possibility of posing
questions about the specific effect of evolutionary factors on the structure
under study in the event of an occurrence of sharp deviations from the
usual relationships.

A good correlation between the number of vibrissae and the coefficient
of variation was observed in all the groups where lip vibrissae were studied.
Minimum variability is characteristic for the first four lines of lip vibrissae,
when the number of vibrissae is maximum. Maximum variability is
characteristic of the vibrissae in the last lines, where the number of vibrissae

88 Variability of Mammals

is minimum (Table 29). The maximum number of vibrissae is in the
lip group, from 45 to 49 (in different samples, Table 3, Appendix). Vari-
ability in number of vibrissae in this group is quite low, from 5 to 8%, in
different groups according to growth and sex for all populations. The
eye group averages three vibrissae and variability ranges from 26 to 37%;
compared to the variability of number of vibrissae in the lip group, it is
sharply increased (which corresponds with the already determined ele-
mentary regularity—Chapter V). In the last group—nasal vibrissae—
there are generally two vibrissae and the coefficient of variation ranges
from 14 to 35%; this is, in six cases out of nine, less than for the eye group.
On the whole, it seems that the group of nasal vibrissae is controlled by
factors of different intensity than the group of eye vibrissae.

In all the female samples, the distribution of magnitude of variability
for body measurements corresponds well with the elementary regularity :

Table 29. Variability in number of vibrissae of new-born males of
the Greenland seal from the Newfoundland region (1963)

Se о

ee

Lip vibrissae

Ist From the right ae he 7.16+0.07 0.37 5.1-- 0.66 30
From the left ses ong 7.30+0.09 0.47 6.4+ 0.85 29
2nd From the right an ae 9.7 0.12 0.64 6.6+ 0.85 30
From the left .. о 08) BOs 05 3.7+ 0.49 29
Зта From the right и т 9.5 0.09 0.50 5.3+ 0.68 30
From the left AAS ie 9.4 +0,09 0.50 5.3+ 0.69 30
4th From the right 5 dhe 8.5 +0.12 0.67 7.9+ 1.02 30
From the left ae ae 8.8. 0.11 0.60 6.8+ 0.88 30
5th From the right an ne 6.3 +0.10 0.54 8.6+ 1.11 30
From the left ae ae 6.5 +0.11 0.62 95+ 1.23 30
6th From the right oe os 3.7 +0.15 0.82 © 22.2+ 2.86 30
From the left an she 4.1 +0.10 0.54 13.2+ 1.70 30
7th From the right as HA 2.0 +0.12 0.63 31.5+ 4.07 30
From the left nd ae 1.9 +0.14 0.79 41.64 5.37 30
8th From the right ue ae 0.40 --0.09 0.52 130.0 16.78 30
From the left a ave 0.30 +0.08 0.46 153.3+19,79 30
All rows : right a .. 47.6 +0.51 2.81 5.9+ 0.76 30
From the left a _.. 48.2 40.51 2.73 5.7+ 0.74 30
Eye vibrissae
Right and left vs a 2.60-+0.10 0.77 29.6-- 2.80 56

Nose vibrissae
Right and left oe 5 1.39-L0.07 "0.49 35.3 +3.33 56

О м

ee ЧЕРВИ

Variability of Populations 89

the greater the magnitude of all measurements, the less the variability.
This situation is true for the sample of males from the Newfoundland and
Yan-Maien regions but not for the sample from the White Sea region
(see Figure 16).

In all the samples, the variability of the number of asternal ribs (mean
number of ribs, 4-5 on eash side) is considerably higher than the variability
of the number of sternal ribs (mean number, ten pairs) without excep-
tion; this corresponds well with the elementary regularity.

The interrelations of magnitude of variability for different skull mea-
surements have been examined in detail in Figure 18, where all the mea-
surements have been arranged on the abscissa in decreasing order of ab-
solute value. If there is conformity with the elementary regularity, the
coefficients of variation for these measurements must increase monotically
from left to right. It may be said that overall, in both males and females,
such a tendency appears sufficiently clear though many deviations make
the picture more complex. As seen from Figure 18a, the first nine mea-
surements of the skull in males quite exactly follow the general regularity;
the coefficient of variation for these measurements increases from 3 to 5%
in all the three populations (the deviations in the curves are insignificant
and may be attributed to errors of measurements). Measurement 24
(height of lower jaw) sharply increases the variability in the population
of the White Sea and measurement 25 (length of tooth row of the lower
jaw) has decreased variability in all the three populations. In the first
instance, the differences between the White Sea population and the other
two populations have been expressed markedly; this is confirmed by the
fact that data for right and left measurements of the White Sea population
give exactly the same picture. Similarly, the decline of variability in the
length of the tooth row of the lower jaw in all groups of males undoubtedly
exists. It can be proposed with confidence that this phenomenon is depen-
dent on the adaptive value of the given trait, i.e., length of tooth row is a
very important trait, which can determine the possibility of catching, hold-
ing, and killing the prey (Yablokov, 1958). It is significant that the vari-
ability of the length of the tooth row of the lower jaw practically coincides
with the variability of the length of the tooth row of the upper jaw
measurement 5).

For a further analysis of variability of skull traits, let us use the con-
cept of ‘‘variability drift”, which is discussed in Chapter У.

A sharp rise in variability as a result of such analysis is seen for mea-
surement 18 in all populations (length of nasal bone), measurement 9
(width of jugal bone) for the White Sea sample, and to a lesser degree for
measurement 26 (height of lower jaw near the base of canines) for the
Newfoundland sample. A considerably reduced variability seems to be
characteristic of width of palate (measurement 11), length and width of

90 Variability of Mammals

c.v.%
20

©. №
154-

10

12 3 287 8 4 5 22241325 18 1115 1412 10 26 9 19 20

Figure 18. Magnitude of ев of variation for skull measurements of different
populations of the Greenland seal, arranged in order of decreasing absolute values of the
measurements. For the identity of the measurements, see Figure 11:
A—males ; B—females ; 1\—limit of the variability “ат”.

right auditory bulla (measurements 14-15) for all populations, width of
nares (measurement 10), and width of jugal bone (measurement 9) for
the Yan-Maien sample. Although, in the absence of detailed functional
analysis, it is difficult to say anything about the functional significance
of the palate (though it is undoubtedly true that this trait must have a
specific relation to the feeding habits), for measurements of the auditory

Variability of Populations 91

bulla such analysis directly shows that the growth of this structure is directly
relevant to hearing. The obvious decline in variability of measurements
of the bulla in all populations cannot be compared with the absence of
sexual dimorphism, especially in regard to the same trait noted during the
investigations of absolute values characterizing the skull of Greenland
seals (Yablokov and Serzhent, 1963). The decrease in variability of the
width of the nostrils is more characteristic of the populations in the Yan-
Maien region than of the other populations, though there is a tendency
in the White Sea population also (Figure 15a). Finally, the sharp dec-
line in variability values for the width of the jugal bone in the Yan-Maien
sample cannot be explained yet, but it is suspicious that conflicting ten-
dencies in the appearance of this trait arise in different populations. Does
this situation represent the genuinely different vectors of selection in diffe-
rent groups ?

The last measurement to exhibit deviation in variability from the
mean values is the width of the nasal bones (19); it behaves similarly in
all the comparable populations. It seems impossible to correlate the low
variability of this trait with a great functional importance; it is known
that the nasal bone is a sufficiently variable trait—in respect of form and
of asymmetry—and these variations can be examined adequately by the
methods of study of epigenetic polymorphism. But a comparison of the
variability of the width of the nasal bones with the variability of their
length allows one to put forward another explanation of the low variability
of the first trait. The variability of the width of the nasal bones is con-
trolled by factors similar in scope to those which are active during the
regulation of the character of primary importance. Other instances of
similarity in variability for two quantitatively different but functionally
homogeneous traits (length of upper and lower tooth rows in the skull of
males), have already been cited in this chapter and Chapter I.

While commencing the analysis of variability of traits in the skull
of females (Figure 18b), the common tendencies of variability for the sample
of both sexes may be noted in the beginning: the general dependence of
variability on regulation of characters of primary importance, resulting in
an increase in variability with a reduction in size of the trait. On this
background, a number of deviations fall in the framework of ‘‘variability
drift” on one side; they correspond to those that have already been
indicated in the variability. of traits for the skull of males, and on the other
hand, there are some traits specific for females.

In the first eight measurements (1, 2, 3, 23, 7, 8, 4, 5) some differences
in the values of variability for different populations could be observed:
the variability of all these traits in the White Sea population is somewhat
higher while the variability of the same traits in the population in the
Newfoundland region is lower. The variability values of the Yan-Maien

92 Variability of Mammals

region are between those of the other two populations in the sample of males,
which confirms the above conclusion to a certain extent (Figure 18a). A
somewhat higher level of variability for all traits is observed in the White
Sea samples; it continues almost up to the midpoint of the curve (up to
measurement 24 inclusive). In this case, if we consider the absolute values,
the White Sea population is quite close to the rest of the populations; any
individual instance does not matter, but the general trend is important.

In females, a sharp decline in variability of the length of the tooth
row of the lower jaw (measurement 25) in all populations is observed
exactly as in males. ‘The variability is surprisingly similar for all the popu-
lations under investigation. As increase in the variability for length of
the nasal bones (measurement 18) and a decrease in the variability of width
of the nasal bones (measurement 19), which brings the variability values
of these traits quite close (despite the large difference in their absolute
measurements), is observed here in the same way as in males. ‘The varia-
bility of the auditory bulla is reduced in females just as it is in males. In
the sample of females, the variability of width of the nostrils (measurement
10) also seems to be decreased; this tendency was likewise observed in
males (Figure 18a). Among the differences noticed in the variability of
females are the homogeneous behavior of the coefficient of variation for
jugal bones (measurement 9) and (from the traits specific for individual
populations) the increase in variability of width of the lower jaw behind
the canines (measurement 26) for the sample made in the White Sea region,
and generally scattered coefficients of variation for different traits of the
Yan-Maien population.

Thus, as a result of analyzing the coefficient of variation, data charac-
terizing the unity of the general characteristics of variability for all three
populations emerge on the one hand, while the existence of specific diffe-
rences among all three of the population groups studied is revealed on the
other. This has already been confirmed by the observations made during
the analysis of concrete traits and standard deviations. However, the most
significant result of the study of variability of structures in this case is the
possibility of an analysis of micro-evolution. The study of variability
allows one to investigate comparable populations not only from a static
but also from a dynamic point of view, while making an effort to under-
stand the peculiarities in the development of their morphological nature.

A separation in this way of the most and of the least variable traits
may seem to be the key to understanding recondite ecological character-
istics of animals with regard to specific relations with the environment.
This is a factor overlooked in other methods of investigation.

Variability of Populations 93

Variability of traits for three groups of
ringed seals from the Okhotsk sea

Comparison of standard deviation values.2 With the experience obtained
from a comparison of the absolute values of standard deviation for various
traits in different populations of the Greenland seal, let us examine the
interrelationship of sigma values for skull measurements of male ringed
seals (Pusa hispida) from three regions where they are hunted. Аз seen
from Figure 19, the values of the standard deviation for all three popula-
tions are very close to each other. For a visual appraisal, this figure can
be compared with Figure 12, in which the interrelationships of absolute
values of sigma for three populations of Greenland seal in the North At-
lantic have been depicted through the same methods. ‘There is no doubt
that all the three samples of Okhotsk seals are considerably closer to each
other than the samples from different populations of the Greenland seal.

But further analysis of the data on the Okhotsk seal indicates that
there is a greater similarity between samples from the Gizhigin Inlet and
the region of Shantar Islands, than between these samples and the sample
from the region of the Taui Inlet. This similarity is especially evident

oO

|
11136219 4 8 515 9 18 141721 10 7 12 20 13 22 16

Figure 19. А comparison of absolute values of the standard deviation for skull
measurements of males of the ringed seal (Pusa hispida) from three
regions of the Okhotsk Sea. Measured traits are arranged in order of
decreasing value of sigma for the Tauisk sample.

in the beginning of the curve (traits 1, 11, and 6) and at the end of the
curve (traits 12, 20, 13, 22, and 16). In these traits not only the mode
of behavior but also the absolute values of sigma coincide. In addition,
it can be seen that the values of sigma in the Taui sample are as a rule
somewhat higher than the corresponding values of sigma for other samples
(with the exception of four traits for the Shantar sample and five traits
for the Gizhigin sample).

3 The works of С. A. Fedoseev and A. У. Yablokov (1965) and С. A. Fedoseev (1964,
1965) have been devoted to a characterization of the ringed seals of the Okhotsk Sea and of
other Far East seas.

94 Variability of Mammals

A comparison of the samples regarding relative values of sigma allows
one to confirm the already derived conclusions about the relatively great
closeness between the Shantar and Gizhigin samples (Figure 20).

Figure 20. A comparison of relative values of the standard deviation for skull measure-
ments of males of the ringed seal from three regions of the Okhotsk Sea (The sigma
value of the Taui sample has been taken as standard):
1—Gizhigin sample; 2—Shantar sample.

In addition, this comparison opens a possibility (through determining
the area occupied by the curves) to conclude that the Shantar sample is
somewhat dissimilar to the Taui sample (but the similarity between the
Shantar and Gizhigin samples is, nevertheless, considerably higher than
that between Taui and any other sample).

Comparison of coefficients of variation. Absence of data on traits other
than the skull makes a comparison of coefficients of variation for these
samples rather easy, reducing it to a determination of the magnitude of
variability in accordance with absolute size. Let us analyze the varia-
bility drift by arranging the skull traits in decreasing order of their absolute
values (Figure 21). Our samples are small in number (9, 12, and 30
individuals) and hence, despite some corrections made through a special
formula in the calculations of statistics for such small samples (Rokitskii,
1961), it is not worth relying upon some of the values of the coefficients
of variation thus obtained. During analysis, attention should be given
mainly to the places where the curves coincide; in other words, it is more
promising to carry out an ecological analysis of the data than to show the
degree of similarity or difference in the magnitude of variability among
samples.

The increase in the variability of measurement 11 (length of nasal
bones) ‘in all three samples is immediately noticeable. It is interesting
that, as in the Greenland seal, the variability for length of the nasal bones
corresponds well with the variability for the width of these bones

Variability of Populations 95

Gave

10

123198 9 4 1514 6 212011181715 7 10 22 16 12 13

Figures 21-22. Variability of skull measurements for (A) males, and (B) females of the
ringed seal from different regions of the Okhotsk Sea. Traits have been arranged
in decreasing order of the absolute value of the measurements:

1—Condylobasal length; 2—Occiput-to-nose length; 3—Snout-basicranium length; 4—Length
of palate; 5—Width of palate; 6—Length of upper tooth row; 7—Width of snout; 8—Mastoid
width; 9—Width of braincase; 10—Wéidth of nasals; 11—Length of nasa! bones; 12—Width of
nasal bones; 13—Width between orbits; 14—Condyle width; 15—Height of skull; 16—Width
of jugal bone; 17—Length of auditory bulla; 18—Wiéidth of auditory bulla; 19—Length of lower
jaw ; 20—Height of lower jaw ; 21—Length of lower tooth row ;
22—Height of lower jaw at the end of the tooth row.

96 Variability of Mammals

(measurement 12); this provides a basis for concluding that this pair of
traits is not dominated by other characters of primary importance.

In the Okhotsk seal, the variability of measurements of the auditory
bulla and length of the lower tooth row (measurements 17, 18, and 21
respectively), condyle width (14), and height of skull (15), seem to be low
in all three samples. A tendency toward low variability of brain width
(9) is also observed. It is interesting that the low variability for parameters
of the auditory bullae and the length of tooth rows was likewise observed
in the Greenland seal.

On the whole, the variability of the sample from the Taui region
seems to be higher and characterized by sharper fluctuation between the
upper and lower boundaries of the drift.

If the values of variability characteristic of females are transferred
to a graph, then it will be observed that all the basic peaks of variability
correspond in males and females (Figure 22). Some other tendencies not
noticed before appear, 1.е., an increase in the variability of measurement 22
(height of lower jaw at the end of the tooth row); increases in variability
of length of palate (measurement 4) and length of lower jaw (19); and
similarly, a decrease in variability of occiput-to-nose length (2). Such
coincidence in basic tendencies in the variability of traits in males and
females and between males of different groups is quite significant. It
confirms the correctness of concluding that there is a relatively fixed amount
of variability, from an evolutionary point of view, and allows one to use
this material for a functional analysis of structure.

An analysis of population morphology for three different groups of
ringed seals from the Okhotsk Sea (Fedoseev and Yablokov, 1965) showed
a similarity among populations in absolute measurements for almost all
the traits studied, as well as in the standard deviations and in coefficients
of variation. If the results obtained are compared, then a specific conclu-
sion about the similar underlying morphological processes between these
three groups of Okhotsk seals and three populations of Greenland seals
can be drawn.

At the same time, the results obtained throw light on the presence
of specific differences between comparable samples. These differences
give, in general, a basis for recognizing a similarity between samples from
the Gizhigin Inlet and the region of the Shantarsk Islands, and the rela-
tive independence of the Taui sample.

4 Such sharp fluctuations in the magnitude of variability correspond well with a few
characteristics observed especially in the Taui region, and show on the other hand that the
general characteristics in variability of traits can be determined even on very small samples.

end,

Variability of Populations

wo
|

wo
WH

09 05 О 0
b Sa ——

т ie
——

|

|
hing
4

bas)

ПЕНА
J
- N oO
|

|

|
|
|

|

ll

ZS
=
С

1 |

= (i)

{J
| ИЕН
хи.

98 Variability of Mammals

Variability of ringed seals of the northern
and far-east seas of the USSR

A comparison of different groups of ringed seals from the Okhotsk
Sea revealed the heterogeneity of the morphological characters of the major
populations of seals in that region. But some large populations of ringed
seals, isolated to a greater extent than the ones studied in the Okhotsk
Sea, live in almost all the northern and far-east seas of the USSR. It
is interesting to compare these populations, geographically separated from
each other by thousands of kilometers, existing in quite different climatic
conditions, and characterized by ecological differences (Fedoseev, 1965).
Using data on the ringed seal from Okhotsk (three groups), Bering, Chukchi,?®
Kara, and Pechora Seas (Figure 23), let us examine first the characters
of the territorially close populations in the Kara, Pechora, Chukchi, and
Bering Seas, and then carry out a comparison of all populations.

Unfortunately, there are not sufficient comparable data from the
region of the Kara Sea in our collection; all that we have is a good group
of skulls, whereas data from the Pechora Sea have been specially collected
for the purpose.

The standard deviation. First of all, let us compare the absolute values
of the standard deviation for skull-size measurements in the ringed seal
from different populations (Figure 24) by arranging all the values in des-
cending order for one population. The population from the Pechora
Sea has been taken as the standard in Figure 24.

It is clear from the figure that there are many general tendencies in
the distribution of standard deviations for skull measurements of females
from different populations. If the concept of “‘variability drift” is applied
to these curves, then it seems that the data on the Pechora seal give the
mean line of this drift, especially in the second part of the whole figure
beginning from measurements 4 and 15. It may also be noticed that the
absolute value of sigma for the Taui, Chukchi, and Kara samples is consi-
derably less than for the Pechora sample. A considerable similarity can
also be noticed in the relative and absolute values for measurements 6, 11,
4, 19, 3, 18, and 22 in the samples from the Bering and Chukchi Seas and
in the two samples from the Okhotsk Sea. The distribution of standard
deviations in the majority of cases for the Kara sample similarly coincide
with the rest of the samples.

The next step is an analysis of the relative values of the standard devia-
tions. The Chukchi population has been taken as a standard for this
analysis because it has a central geographical situation (see Figure 23).

The distribution of the relative values of sigma for traits in the order

5 The values of standard deviations are given by С. A. Fedoseev for the Chukchi and
Bering Seas,

Variability of Populations 99

10

123198 9 4 15 14 6 2120 11 18 17 15 7 10 22 16 12 13

Figure 24. А comparison of the standard deviations for skull measurements of
females of the ringed seal from different populations:

1-—Pechora population. Explanation in the text.

arranged in the table of skull measurements, does not allow one to analyze
the picture obtained because of its complexity (Figure 25). Hence, let
us analyze the standard deviation values by the method proposed above,
using the proper order.

It is seen from Figure 26 that the relative values of sigma for two
populations, namely the Bering Sea (minimum) and the Таш! region (тахи-
mum), differ clearly from the others. However, the sigma values for the
Taui population are closest to those of the Chukchi population, which has
been taken as a standard in the graph. If the degree of deviation for diffe-
rent populations from the Chukchi sample is determined simply by mea-
suring the area which is occupied by the sigma values of the populations
studied, it seems that our samples arrange themselves in the following
order: Chukchi, Taui, Shantarsk, Pechora, Kara, and Bering Seas.

For further analysis, it should be noted that not all the natural popu-
lations of seals on the shores of Europe and Asia have been investigated.
There are no data on the East Siberian, Laptev, or Kamchatka seals.
Because of this, it is not possible to draw any final conclusions, but the data
obtained allow one to review the interrelationship of neighboring popu-
lations.

100 Variability of Mammals

a=

“gag ley
4S4 2 ueyus

1.0

0.5

‘Chukchi Sea

Figures 25-26. A comparison of relative values of the standard deviation for skull
measurements for females of the ringed seal from different populations [Sigma value of
the Chukchi population has been taken as a basis in the (A) usual and
(B) regulated manner. See text]:

1—Chukchi Sea; 2—-Taui; 3—Shantarsk; 4—Pechora Sea; 5—Kara Sea; 6—Bering Sea.

Variability of Populations 101

An analysis of the relative values of sigma (using the graph) points
to the similarity of the characters in the Taui Inlet and Shantar popula-
tions, to the similarity of the characters in the Kara and Pechora popula-
tions, and to the sharp differences between the Bering Sea and Chukchi

populations.
A comparison of the arithmetic means calculated for all the relative

values of sigma for the skull gives similar results (78s) :

Sp Sp
Bering Sea а we 0.69+0.17
Pechora Sea ; ae К 0.79-=0.07
Shantar region it ats 0.95--0.06
Кага Sea RP ae iM 0.98--0.11
Taui region Be 1.07+0.21

tering Sea—Taui region =1.44

It is not difficult to notice that with regard to the mean values of
relative sigma our samples could be arranged in an order close to the one
in the graph: Bering Sea, Pechora Sea, Shantar/Kara, Chukchi, and
Taui. as ;

The coefficient of variation. A comparison of variability of traits for
different populations of the ringed seal is depicted graphically in Figure 27,
where dependence of the magnitude of coefficients of variation on the

C.V.%

15

10

132198 9 4 6 1514 21201117185 7 10 2216 1213

Figure 27. А comparison of values of coefficients of variation for skull measurements
for males of Shantar, Kara, Taui, Pechora populations of the ringed seal.
For identification of measurements, see Figure 22.

102 Variability of Mammals

absolute value of the trait has been examined. Some specific tendencies
are brought out more clearly in the usual “variability drift’? than during the
analysis of only Okhotsk seals; these are as follows : -

А very high variability of length of the nasal bones(11), a high varia-
bility of length of the palate (4) and length of the lower tooth row (21);

A low variability in measurements of the auditory bulla (17), width
of palate (5), height of skull (14), and mastoid width (6).

There are some local characteristics in the variability of measure-
ments for certain samples : increased variability for large measurements
of the skull in the Kara sample, sharp fluctuations in the magnitude of
variability in the Taui sample, and a low variability for traits having small
absolute values in the Pechora sample. :

The variability of body measurements coincides well with the elemen-
tary regularities in three populations, but it is not so in the Pechora sample.
This may be “chance,” because the sample of females shows a sharp
increase in variability with a decrease in the value of the measurement
(Table 30).

The Pechora sample stands out because of the strange reduction in
the magnitude of variability in the number of caudal vertebrae (which
can be explained, perhaps, by an insufficiently exact analysis of material
due to the difficult conditions of winter trade in seals).

The ecological significance of many of the above-mentioned devia-
tions from the mean values of variability, has already been discussed.

Table 30. Variability of body measurements of Pechora ringed seal, 1962

Adult males

Body length .. 118.5+1.44 10.8 9.1+0.86 56 +2.92
Measurement : Anus to navel .. 33.1+0.38 2,92 8.8+0.84 55 +3,12
Measurement : Anus to penis .. 18.2+0.22 1.61 8.8--1.55 54 —

j Adult females

Body length .. = 112,141.65 9.5 8.5+1.04 33 —
Measurement : Anus to navel .. 30.7--0.66 3.81 12.31.51 33 —
Measurement between nipples .. 8.0+0.36 2.13 26.6+3 — =

Certain Results from the Study of Variability of Characters
for Different Populations of One Species

The comparison of morphological statistics for different samples
(of different age and sex) allowed us not only to establish differences for
specific traits between individual samples (for мс the usual methods of
comparative anatomy would also do), but also to determine the degree of

Variability of Populations 103

these differences. This method made it possible to determine that sexual
dimorphism is expressed in different degrees among young and adult animals
and in a characteristic manner in each population of the species [Yablokov,
1963; Yablokov and Serzhent (Sergeant), 1963; Fedoseev, 1965; and
others].

While studying the characters themselves of traits in different samples
and determining the general characteristics of the groups of different sexes
and ages, as well as for the individual populations as a whole, it was possible
on the basis of knowing the quantitative characters of traits to make hypo-
theses also on differences in the ecology of the groups under study or to
determine prospective lines of ecological research.

It seemed possible to draw conclusions from the magnitudes of differ-
ences about the similarities or dissimilarities of populations as a whole,
though this conclusion is based only on the absolute values of traits and
can only be made in a most general form (Fedoseev and Yablokov, 1965).

The second method of study especially examined aboye—the study of
regularities in the distribution of standard deviations of traits—permitted
us to obtain data directly characterizing the degree of similarity between
the populations under study and the sub-samples according to age and
sex. It is important to emphasize that the use of the relative value of
sigma allowed us to compare populations not by individual traits but by
a complex of quite heterogeneous traits, which is close to the question
raised by R. Sokal and P. Sneath (1963). The third way of studying
population morphology examined in this work—the study of variability
expressed in coefficients of variation—opens up a new possibility of
obtaining the ecological significance of a trait, by briefly characterizing
the degree of dependence and the relationship of the trait to effective
evolutionary (ecological and genetical) factors. The study of coefficients
of variation confirms the existence of a well-developed sexual dimorphism
in various structures, and shows varying degrees of dimorphism in various
populations; here is a similarity. with the study of the absolute magnitudes
of traits. However, this similarity is superficial because variability ex-
pressed in coefficients of variation seems to be an index of rather common
peculiarities, regularities, and interrelations among various age-sex groups
within a population, and also an index of the relations by which the whole
population is situated in the biogeocenose.

Finally, on the basis of analyzing variability of traits expressed in
coefficients of variation, and also analyzing standard deviations, it seems
possible to say a little (based on a comparison of a number of traits) about
the degree of difference and similarity of different populations where
age and sex have been considered.

All the three methods used, while characterizing a population from
different points of view, allow its consideration as a dynamic unit

104 Variability of Mammals

consisting of complex and at times conflicting parameters (in different
age and sex groups) and differing, on the whole, from other populations
of the species.

It is important to emphasize the possibility of conflicting tendencies
arising in the degree of differences among populations for individual
traits. For example, the greatest differences in color are observed between
the White Sea and the Yan-Maien populations of the Greenland seal, and
for many other traits, maximum differences are found between the New-
foundland and White Sea populations. Such divergence in the degree of
dissimilarity for different traits may indicate their evolutionary indepen-
dence and the presence of specific differential methods for the effectiveness
of evolutionary factors. This conclusion is confirmed by a comparative
analysis of variability for certain traits in different populations of the
ringed seal.

In reviewing the study of population morphology for different popula-
tions within a species, the following statements can be advanced:

Different populations within a species exhibit a high degree of
similarity with regard to absolute values for traits as well as in the nature
of variability for individual structures and groups of traits;

On the background of the similarities noted, it can be seen that real
differences in variability exist between individual age-sex groups on the
one hand and between individual populations on the other;

The range of characters for the morphological structures of the
populations studied seems to be always insufficiently wide and differen-
tiated to reliably distinguish populations on the basis of the absolute
magnitude of individual morphological traits. For this, it is better to
consider the general trends in the variability of the traits (by the method
of comparing the values of standard deviation and coefficients of variation).

Comparison of Variability of Population within a Genus

While studying the variability of a single population, attention has
to be given (to a greater degree than during the study of variability of
different populations within a species), not to the absolute values charac-
terizing the various traits of each of the species, but to indices of variability
(like standard deviation and coefficient of variation).

A comparison of variability of traits within a genus has been carried
out below, taking a genus of seals (Pusa) as an example, including the
Caspian (P. caspica), Baikal (P. sibirica), and ringed (P. hispida) seals.

Comparison of the standard deviations. As observed from the comparison
of standard deviation (Figure 28), a great similarity was noticed for the
respective values of different species, and it is difficult to determine the
degree of their divergence. The comparison of relative values of standard

Variability of Populations 105

Figure 28. Comparison of standard
deviations for certain traits of three
species of seals of the genus Pusa and
also Phoca vitulina:
1-3—Body measurements; 4—Number of
tracheal rings; 5—Number of caudal verte-
brae; 6-8—Number of vibrissae in different
areas; 9—Length of digestive tract ;

-++-—Phoca vitulina larga.
1 e 5 +

1.0 Pechora population

Figure 29. Comparison of relative values of standard deviations for certain traits
of three species of seals of the genus Pusa and (a) Phoca vitulina.
For identification of measurements, see Figure 28.

106 Variability of Mammals

deviations with the values for the seal population from the Pechora Sea
taken as a standard, seems somewhat more divergent (Figure 29).

It seems that the Baikal seal differs more from the ringed seal than _
from the Caspian seal. It is seen from the figure that a sharply deviating
value of sigma for gut length provides the greatest difference. Other
comparisons of relative values of standard deviations, comprising 30 traits
(mainly skull measurements) for three species of seals, are depicted in
Figure 30.

It is well seen in the figure (30) that the sigma values of the Pechora
seal (taken as a unit) are rather close to the sigma values of other popula-
tions of the same species, i.e., the ringed seal from the Okhotsk Sea, and
the sigma values of the Caspian and Baikal seals are more different—
especially seen in the upper half of the graph. However, it is difficult
to say anything concrete about the relative degree of difference between
the Baikal, ringed, and Caspian seals from this figure because the curves
of the Baikal and Caspian seals pass very close to each other.

Comparison of coefficients of variation. A comparison of coefficients of
variation for non-skull traits is shown in Figure 31. Notwithstanding
the rather small number of traits studied, the following specific conclusions
can be drawn from the data already available :

The variability for all traits is similar in all samples;

The Baikal and Caspian seals seem to be similar in variability,
differing only in variability of the number of caudal vertebrae, whereas
the Pechora population of the ringed seal has a usually lower variability
(in six out of ten traits) ;

The variability in number of vibrissae is the only single measurable
trait which corresponds completely with the elementary regularities. For
body measurements, these regularities are disrupted (rather considerably
in the ringed seal);

With regard to the variability for number of caudal vertebrae, all
the species differ from each other, but these differences do not exceed
those which were observed during the comparison of variability for different
populations within a species (See Figures 16 and 26).

The last situation appears to be very important in determining the
place of variability analysis in the general framework of population
morphology. If the range of variability among populations within a
species does not differ from the range of variability among populations
of different species, then this means that the analysis of variability expressed
in coefficients of variation with an aim to determining the degree of
differences among populations seems to be considerably less valuable
for comparisons among species. Let us add a conclusion from the
analysis of standard deviations: while comparing the standard
deviations of different species, the results obtained seem to be less

Variability of Populations

Figure 30. Comparison of relative values of standard deviations for
certain traits of the seal Pusa. Explanation in the text.

C.v.%

О Sly ign ЧИТЕР

Figure 31. Comparison of variation coefficients of non-skull measurements
within the genera of (Pusa), Phoca (1) and Histriophoca (2):
Legend to measurements as in Figure 28. 10—asternal ribs.

107

108 Variability of Mammals

significant than when populations are compared within a species.
However, this conclusion holds good only with regard to an analysis
of variability done with the aim of determining the degree of differences
among populations and species. In studying biological characteristics of
populations, the analysis of coefficients of variation seems to be useful
even while comparing different species. For an illustration of this situa-
tion, let us analyze the magnitude of coefficients of variation for skull
measurements of different populations within a genus of seals (Figure 32).

C.V.%
20

15 |

10 2 А 3 И

: A ~V: Sey, : М /
ей Lf ioe oe Noe

123198 9 4 15 14 6 21201118 17 15 7 10 22 16 12 13.

Figure 32. A comparison of the magnitude of coefficients of variation for
skull measurements in females within the seal genus Pusa:
1—Taui; 2—Shantar ; 3—Baikal ; 4—Caspian.

For identification of measurements, see Figure 22.

During the comparison of variability for these measurements, it must
be remembered that the samples: from the Caspian and Baikal are very
small and the probability of obtaining insufficiently characteristic values is
great. Hence, in the analysis, attention must be paid to the general
character of the curves and not to the sharp fluctuations in the variability
value of one or the other trait. The following features come to light:

1. An increase in variability for length of palate (measurement 4)
is correlated in a surprisingly homogeneous manner with regard to
coefficients of variation for all the samples compared;

2. A reduction in the variability of the ae of the с auditory bulla
(measurement 17);_

Variability of Populations 109

3. A reduction in the variability of the width of the nostrils (measure-
ment 10), except for the Taui sample of the ringed seal.

The several agreements in the variability of traits provide а basis
for a conclusion about the magnitude of the variation.

A study of the coefficients of variation with regard to differences
among groups seems to be rather ineffective for a comparison of different
species within a genus because the coefficients of variation for different
species may not differ more than those of different populations of the
same species. ‘The species as a whole, perhaps, does not possess charac-
teristic values of coefficients of variation; these appear to be very mobile
traits, easily changed depending upon the conditions of existence for a
population in the biogeocenose.

On the other hand, an analysis of coefficients of variation in a study
of different species, carried out not to determine their degree of systemic
differences but to understand the regulation of their development in
definite conditions of existence, seems not only promising but also the
only effective means of analysis at present. An analysis of coefficients
of variation for three species of seals showed their rather great similarity
in all the habitats of the genus. Such a situation is perhaps determined
mainly by a sufficiently similar relationship of each species to evolutionary
factors. In all instances where coefficients of variation of similar value
are observed, the evolutionary control acts with approximately similar
force on homologous traits.

An analysis of variability for traits seems to be even more important
than the direct analysis of morphological characters for the study of ten-
dencies in the evolution of each species. By knowing that the number of
vibrissae in one species of seal is a little more than in another, we can only
make an assumption about the role of specific conditions existing in the
past which may have brought out such differences. But the fact of
similar variability in these structures for various species shows that at
present these structures have an identical relationship to evolutionary factors.

Comparison of Variability of Populations
of Closely Related Genera

The characteristics of variability of species belonging to different
but closely related genera, could be seen while comparing representatives
of the seal genus Pusa and of the common seal (Phoca vitulina), which is
systemically close to it; for a long time both groups of seals were placed
in one genus. Interest in such a comparison is greater because the
morphological characteristics of the Okhotsk seal, common seal, and
other species and populations of seals living in the same body of water
(Okhotsk Sea), as well as those living in the very different conditions of

110 Variability of Mammals

other waters like the Arctic and Pacific Oceans, can be compared from
available data.

The standard deviations for a number of traits of the common seal
(Phoca vitulina) with homologous traits of Pusa are shown in Figure 28. In
accordance with the differences in absolute value of traits, the values of
the standard deviation show considerable differences for the various traits.
The ordering of standard deviations in this case is close to that for the
mean values of the traits. 2

In the comparison of relative values of the standard deviation and
coefficients of variation, the matter is different. The relative value of
sigmas for the traits of the common seal (Phoca vitulina) has been given in
Figure 29. If the degree of deviation is expressed in the form of area
under the curve obtained (the one from the data on the Pechora popula-
tion of the ringed seal being taken as a straight base line), it is observed
that the Caspian seal deviates (with regard to four traits, taken from the
beginning of the curve) by approximately 14 to 15 cm?, the Baikal seal by
27 to 28 cm?, and the common seal by 39 to 40 cm?; in other words, the
sample from a population belonging to another genus differs sharply from
the samples from the populations of species within one genus.

The comparison of values of coefficients of variation for traits of
Pusa and the Phoca (Figure 31) leads.to some interesting conclusions. As
seen from the coefficient of variation for the common seal (Phoca vitulina),
there are no differences (with the exception of a somewhat lower varia-
bility for lip vibrissae). An interesting trend becomes apparent. The
values of coefficients of variation for the ringed seal of the Okhotsk Sea
are closer to the coefficients of variation for the common seal (P. vitulina)
of Okhotsk than to the coefficients of variation for other species and
populations of seals. It seems possible that the variability of traits for
the common seal and the ringed seal are close to each other because of
the similar demands made by natural selection in the same body of water.

Such an hypothesis is appealing, but it must be noted that a number
of other factors are needed for its confirmation. The fact that a small
number of traits have been studied does not speak in favor of this assump-
tion. A detailed additional examination of such a proposal and a compa-
rison of the variability of organs of the common and ringed seals of the
Okhotsk Sea with the variability of other species of pinnipeds living in the
Okhotsk Sea is needed.

The species Pagophilus groenlandicus and Histriophoca fasciata can be
examined as representatives of closely related genera (Chapsku, 1955a;
Scheffer, 1958; Shustov and Yablokov, 1967). The results of analyzing
the morphological material obtained are presented in a table of the Appen-
dix. A comparison of the morphological characteristics of the Greenland
seal and Histriophoca fasciata as related genera (genera in the family of true

Variability of Populations 111

seals) shows that there is a considerable morphological difference in many
traits between the two (Table 31).

Table 31. Degree of difference (t) of morphological traits of the banded
seal (Histriophoca fasciata) and Greenland seal (Pagophilus groenlandicus)
(From data of Shustov and Yablokov, 1967)

Traits ty to ty
Body length aa 5% — 11.50 — 8.12 — 6.37
Number of tracheal rings ie a6 + 2.72 + 1,67 + 1.90
Eye vibrissae BG oe — 4,94 — 6.15 — 5.4]
Lip vibrissae nie 37.0 +26.7 +34.5

Nore : ¢,—banded seal, White Sea population of Greenland seal; t,—banded seal,
Yan-Maien population of Greenland seal; t;—banded seal, Newfoundland population
of Greenland seal.

A comparison of absolute values of standard deviations shows a great
similarity in the disposition of all values for three populations of the Green-
land seal and noticeable differences in the corresponding values for His-
triophoca fasciata (Figure 33). A comparison of relative values of sigma
(with the population of the Greenland seal from the White Sea taken as a

standard), leads us to conclude that the degree of differences in comparable
traits is lowest in the population from Newfoundland. ‘The population
of the Yan-Maien region differs considerably and the curve corresponding
to the sigma values of Histriophoca fasciata shows these differences
(Figure 34).

A comparison of the coefficients of variation for the traits studied
in the Greenland seal and Науторйоса fasciata leads us to conclude that
there are no significant differences in the magnitude of variability in the
latter (Figure 16). The variability of number of vibrissae in various lines
is almost identical in the Greenland seal. The variability of the last line,
for which there are always a larger number of vibrissae in Histriophoca fasciata
than in the Greenland seal, appears to be reduced. ‘The variability of the
total number of vibrissae in the lip group of H. fasciata appears to be some-
what increased, which corresponds with the reduced number of vibrissae
in this group. The variability in number of eye vibrissae is similar, which
is rather strange since the number of these vibrissae in Н. fasciata is twice
as large. The variability of body length appears to be somewhat higher;
yet, on the other hand, the variability of number of tracheal rings appears
to be lower. But all these deviations and situations observed from different
populations of the Greenland seal are not fundamentally characteristic.

6 This may be explained as due to a similar action of the controlling factors of this trait.

112 Variability of Mammals

1 2 3

45678

Figure 33. А comparison of absolute

values of standard deviations for certain

traits of two seal genera: (a) Pagophilus
and (6) Histriophoca:

1—Body length ; 2—Number of tracheal rings ;

3-4—Number of lip and eye vibrissae ;

5-8— Weight of heart, liver, kidneys and spleen.

Figure 34.. A comparison of relative
values of standard deviations for a
number of traits (see Figure 33) of two
seals: (1) Pagophilus and (2) Histriophoca.
Sigma of the White Sea population of the
Greenland seal has been taken as a
base.

On the whole, it may be said that the variability for the traits studied in
Н. fasciata is close to that of the Greenland seal.

Now let us compare the characters of these and other genera of the
subfamily Phocinae which are more distant taxonomically (Figure 35).
For the comparison, we shall first take the Greenland seal on one side and
two populations of the ringed seal (Okhotsk and Pechora Seas), the Baikal
seal, and the Caspian seal on the other side. After that all these seals
will be compared with Н. fasciata, and finally the common seal with the
Greenland seal. The comparison of values of the standard deviation has
been intentionally left out because it has been specifically shown above
that such a comparison of relative values of sigma above the generic level
reveals just a formal significance, and a comparison of absolute values of
the standard deviation does not add significantly new information in the
analysis of morphological differences. :

The comparison of coefficients of variation for traits in the. species
of different genera shows (Table 32) a picture opposite to the one that is
observed during the comparison of absolute values of the traits (Table 33).
A considerable number of comparable pairs of traits do not show signifi-
cant differences in their coefficients of variation. During the comparison
of absolute values, out of 145 comparisons only 21 were indistinguishable,

Variability of Populations 113

Gtaria
Eumetopias

Wariidae Zalophus
— Neophoca
Arctocephalus
Callorhinus

Qdobaendae
eas /ИОФОРЛИУ

Phoca
= usa

Pinnipedia pee Histriophocd
Phocinae os Pagophilus
Halichoerus

ИЕ Erignathus

Monaches
Vhocidae
А (060002
Ommatopheca
Hydrurga
Leptonychotes

Cystophora

lLystophorinae \. Mirounga

Figure 35. Taxonomy of Pinnipedia (Scheffer, 1958). The comparison
carried out in the present work has_been shown by a double line.

but in the comparison of coefficients of variation, there are 74 indistin-
guishable pairs. The distribution of differences in coefficients of varia-
tion exhibits certain features :

The variability of H. fasciata is more similar to that of the other seals
than is that of the Greenland seal;

The greatest similarity in variability is observed between H. fasczata
and the Okhotsk seal and between Н. fasciata and the Baikal seal. The
variability of the Pechora seal and the Caspian seal is not much more
different than is that of the Okhotsk seal or the Baikal seal;

During the comparisons of Pusa with the Greenland seal, the pair of
the Greenland seal and the Okhotsk seal showed the fewest differences,
followed by the Baikal seal (and here the Pechora and Caspian seals differ
from the first two representatives of this genus).

It is clearly seen that a tendency for a greater variability of body
length is observed in comparison with the same trait in the Greenland seal,
but traits like number of tracheal rings, number of ribs, and the total number

114 Varability of Mammals

Table 32. A comparison of coefficients of variation of certain
traits for different genera of seal*

Trait ее.
Body length a ~~ + + + Nil + Nil Nil Nil Nil ++
Anus-navel measurement .. Nil Nil ++. ,;;
Anus-penis measurement Е М,
Number of :
Tracheal rings Hh op) eS SE ee т Nil Nil Nil Nil +
Caudal vertebrae ie .. +t Nil Nil +. Nil
Sternal ribs и О HH Е +
Asternal ribs Pate Sioa abe +4 Nil
Nasal vibrissae Be А м о Nall
Eye vibrissae Ey: we t+ ,. ++ ++ Nil ++ Nil
All the lip vibrissae .. МХ Ni Е , 4++4+4+ 44 +
Ist line .. Е НЫ а о
2nd line Be . Ni , ++ Nil Nil Ni ++ Ni + +
3rd_ line ae а о ING ances A ом м
4th line ee ОЕ + 4s 2 5 + +
5th line He .. ++ Nil Nil Nil ,, BS Hs > ++
6th line, ie -. Ni ++ 44+ 44 ,, a aa A + +
7th line ass ee ss Nil +4+ 4+4+ 44+ , ЧЕН 4 +
8th line a a5 ча IL + 4+ Nil + м
9th line + + + +

* For ease of reading, the following abbreviations have been used in the table : t1.96;
difference exists (-- -- when t= +1.96 or + when t= —1.96); when #13 less, there
are no detectable difference (Nil).

** For abbreviations ¢t, to ty), see Table 33.

of vibrissae, show a tendency towards a lower variability in all the species
of Pusa. С

If these trends with regard to body length, number of tracheal rings,
and number of vibrissae could be well understood in the language of ele-
mentary regularities, then the increased variability for number of sternal
and asternal ribs could not have another explanation, such as changes in
certain biological relations in the populations of these genera.

In the majority of traits in the Greenland seal and H. fasciata, the varia-
bility is greater than in the common seal (Tables 32 and 33) and this at-
tracts attention; moreover, in the first comparison, the lower variability
of all vibrissae in the common ‘seal (Phoca vitulina) is remarkable because
their number is less. Similarly, a somewhat lower variability in number
of vibrissae of the lip group for the Caspian seal, whose absolute number

Variability of Populations 115

Table 33. A comparison of absolute values of traits for species of the genera
Pagophilus, Pusa, Histriophoca, and. Phoca*

Trait by ly bs ty t; the lig tg ty tio
Body length sou bec SRSR SEAR SESE SESE GES Sea cae: ede INES ice
Anus-navel measurement О +4+ ++ ++
Anus-penis measurement - ++ + ++ ++
Number of :
Tracheal rings .. go SPS SPSP APSF чая Se SP SR Sp aPae ae
Caudal vertebrae Be . t+ Nil ++ Nil a5
Sternal ribs ne р Par Nil
Asternal ribs AY ye ae ae + ae fe
Nasal vibrissae Se oe tH а а
Eye vibrissae os . t+ Мм + + tt м Nil
All the lip vibrissae .. ООО + + М +
Ist line но 00 AR SR SR SR OS aR SR Sp ges JNM
2nd line м.
Зг4 line у и м fer aii eR Siar
4th line и м
5th line Nay J ph TE at Ese TA STR ella yl SOA TERI eet ee fle
6th line ai gas Mile i TUS My ati ets Peis ely i AUST faa apes ee
7th line fe a + Ni + + + + + +; +
8th line Be ANIL A Мм
9th line a a th

* For ease of reading, the following abbreviations have been used in the table: #5 1.96;
difference exists (++ or ++); when ¢ is less, there are no detectable differences
between comparable traits (Nil):

-t,;—Greenland seal, Pechora seal ; tp—Greenland seal, Okhotsk seal; ts—Greenland seal, Caspian
seal; t,—Greenland seal, Baikal seal; t;—H. fasciata, Pechora seal; t;—H. fasciata, Okhotsk
seal; t,—H. fasciata, Caspian seal; t,—H. fasciata, Baikal seal; t,—Common seal,
Н. fasciata; t,>—Common seal, Greenland seal.

of vibrissae is the greatest of all the species studied, cannot be explained
on the basis of elementary regularities. Apparently, the controlling factors
acting naturally on this trait in the Caspian Sea differ considerably from
those in the biogeocenose in which other pinnipeds are included.

One of the conclusions which can be derived from the above compari-
sons of variability among species belonging to different close and distant
genera, coincides with the conclusion arrived at earlier from the compa-
rison of variability between individual species within a genus : the value
of a direct quantitative comparison of morphological traits appeared to
be of little significance for comparing different species, because such species
usually differ clearly and specifically not by quantitative peculiarities

116 Variability of Mammals

(brought to light only through refined methods of population morphology)
but by qualitative differences (brought to light by the usual methods of
comparative anatomy).

Although it is necessary to have a large sample and an exact statistical
analysis of quantitative data for a comparative characterization of some
population with another population of the same species, there is no neces-
_sity for using methods of population morphology for comparative charac-
terization of different genera. There is no sense in using quantitative
analysis of refined peculiarities where qualitative differences exist. Of
course, some special problems may show up in the analysis of individual
homologous structures in a number of cases, but as a rule there is no need
for wide usage of quantitative methods to find differences between species
of different genera.

Thus, a quantitative analysis of traits becomes less important during
the analysis of differences between species of different genera. But this
analysis happens to be the basis for a further, a more detailed, and deeper
study of the processes taking place in a population and in the biogeocenose,
and the analysis of quantitative features of morphological structures of
different species begins to play a leading role in the study of evolutionary
processes.

In this analysis at the level of closely related genera, analysis and
inference with regard to variability is of primary importance, because
it can relate structures to the process of evolution in populations. Methods
of comparing the variability of ecologically different structures in different
species provide the possibility of discovering the trend of evolution in one
population as compared with another population belonging to a different
species and sex. A comparison of values of a trait in different species gives
the possibility of detecting larger or smaller differences, but the compa-
rison of variability of the traits allows one to judge the very first drifts in
the mode of morphological changes in a population.

The analysis of variability of structures in different species within
closely related genera is not less important in presenting specific problems
for ecological research (by drawing attention to structures existing in
specific relationships to evolutionary factors).

One general conclusion can be drawn from the data of this chapter :
Any trait can change independently in different populations. ‘This con-
clusion was drawn up in principle by H. F. Osborn (1915), A. N. Severtsov
(1939), and R. Goldschmidt (1935); it is now widely used in the “new
systematics” (Mayr, 1947, 1962; Beregovoi, 1963; Krivosheev and Rosso-
limo, 1964; Schwartz, 1965; and many others). But it has significance
in principle for population morphology by asserting the importance of
studying many individual traits of an organism.even in cases where we
do not know their adaptive significan te as yet.

Variability of Populations 117

Characteristics of Variability in Populations
of Different Subfamilies

Two subfamilies (Figure 35) can be distinguished in the family of
true seals (Phocidae) : the true seals themselves (Phocinae) and the sea
elephants (Cystophorinae).

There are a number of references in this work to the morphological
features of Cystophora cristata, which is the only northern representative
of the subfamily of sea elephants (Cystophorinae).

While comparing the absolute values of traits typical of pinnipeds,
only one very general conclusion can be drawn: АЦП the species differ
from each other and the sea elephant differs quite considerably from others
(Table 34). The situation when comparing the coefficients of variation
(Table 35) is altogether different : for absolute values of morphological
traits, the sea elephant differs from others by 83%, whereas in the соеЙ-
cients of variation it differs only by 50% (in 73 comparisons out of 145).

No specific trends in the distribution of variability of the traits are
observed when other pinnipeds are compared with Cystophora cristata and
Erignathus barbatus. It is possible that the incompleteness of the data pre-
sented leads to such a conclusion as might accrue from the complexity
of the phenomenon itself along with the various factors which determine
the coefficient of variation in each specific case. ‘That is why the tendency
toward no difference in the coefficients of variation in the total number
of lip vibrissae is significantly expressed during the comparison of almost
all types (with the exception of the comparisons of Cystophora cristata with
Phoca vitulina and Erignathus barbatus with all the rest of the seals). On
the other hand, in every case (except when comparing (С. cristata with the
Pechora seal), the variability in number of nasal vibrissae appears to be
different.

To examine the variability of C. cristata with a view to evaluating
general environmental factors, is inevitably to compare the variability of
traits of С. cristata and the Greenland seal living in the same region. Hence,
a comparison of (. cristata with the Greenland seal has been carried out
in detail by presenting data in Table 35 on all the three populations of
this seal studied thus far. ‘The results obtained deserve a special discussion.

There are clear differences in the variability of body length between
С. cristata and the Greenland seal. It is interesting to observe that С. cristata
does not differ from other types in the variability of body length (except
Histriophoca fasciata, which is the nearest relative of the Greenland seal).
On the other hand, none of the three populations of the Greenland seal
differ from the population of (С. cristata in variability of number of tracheal
rings, though the rest of the seals differ with regard to this trait. С. cristata
and the Greenland seal are also close in variability of number of asternal

Variability of Mammals

118

Ритие ро ‘57724404 "Е S ns0ydorajsipy 57729424 "7—1
[295 [eyreg ‘57129424 `— 19 ‘yeas ueidsery “57124479 Я - 017 [29$ 1530ЧЗО ‘57429424 “q—%) ‘(uorSo1 иэтей-ие А) [eas ризиээле) ‘57724409 -“q—*}
squeydaja wag ‘57724429 snyyousiug—t, грицпие э20у ueydaja vag—» 5001250 вооудощяа ‘queydaja vag—) ‘yeas peyreg “ueydora
vog— ‘yeas uvidsen “ueydoja eag—*, ‘Теэ$ елочоэа ‘yueydoja eag—*, ‘yeas рие[аээле) ueydela vag—y ‘sjoquids ло} GE эт, 209,

ats ar ar ET ae ay oul] Ч3/
IN ee ee 291 439
И НЕ яя ap ale oul] Ч3с
НЕ НЯ Als oul] 43$
ai tN НЕ Я Sip ar sul], Pag
ЯР ЕЕ als ap ЯР aE oul] pug
ИМ а ЯР oul] 351
о | | SeSSLIQIA АН эЧз Пу
се TIN “ a SESSLIQIA [eSeNy
Jb ele + gs + ele Е ENG а а эезтлатл эАЯ
OG $ —“ if [IN 2 ar + se зал [291935
TIN ENG SUNS UN EN = IN IN м м TN зайл yeusais
ara Е Е ЕЕ Неа PS = Pa = ae ae SEALER BLMS (4149)
fraqe gear ео ae a oF ap a = Зи теэчоеа,
: JO Joquinyy
++ +4 44 44 ap us yUsWaINsvoUE зтаэ4-зпи\у
| + + + + + + + + oe eZ JUIUTIINSeIU [э^леч-5пи\у
О В о .. .. Ч38иэ1 Apog
о ae = Oy oe = = a iy = Ра т: ей % “7 “a WLI,

4oeprpoyg Апихеу oy} UIYIIM 5елз JO зэптел эзитозае Jo uostzvdurory у “FE эает,

119

Variability of Populations

“ри vooyqorysypy ‘51124404 “q—-8"7 Spurpnzia vo0yg ‘571424424 “q—""} ‘Теэз [eyreg ‘57124424 “q—"y ‘yeas
uevidsen “57729424 “q—*y 129$ 1530430 ‘57729424 "Я— у ‹(иотЗэл uarey-uex) [eos риериээле) ‘5712440 “q—"y 557779404 snyywusiug ‘улеча
-э[э еэ6—7 50и1]пд20 0204 ‘заечЯзр vag—*y : 21171250} 220440149] “yueydats eag—) ‘yeas [eyreg ‘yueydoja еэ9—7 ‘yeas uerdsen ‘улечаз
vag— {1295 ys10yyxO “ueydoja 29—47 ‘pees елоЧоза ‘ueydaja vag —) $(иотЗэл puv[punojmon) [ves рие[аээлю) ‘лиечаэ]э eag—*) (96
a1 AA) [Bas риз[иээле) ° (2115242 7) ) yueydoyo vag-—*) ‘(uolser иэтеу-иел) [eas риз[аээлю) ‘21115142 DLoydojsd7y—"y SsjoquIks ло} ZE Э1ЧЕЛ, Sy

dbs “dese Usps aeae. Spar Зарар (sear. orar- ara Be oul] Yt
о : s Setar Ne [ENT ats aur] 139
apa Х = Е 3 JUNG WING о ore oul] 436
о out] Ч
А cos м :. ze а oo oul] раб .
м ree ee oul] руб
et ee eee eee А oul] 4S]
cs fea ARSE (UN ara ЯРаг Go TNE Ве IND = SRE Sea м ese 8S aesstiqia dif 24} ПУ
SEY HES ST ote Cie ee eee eo эезетлатл [eSeN]
UN ++ Е ++ 44+ ++ + ++ м № EN м [IN [IN “ - в эеззматл эАЧ
a ae о НЕ qpap see UN зад 1202935
и: Е ++ а" затх [291935
Е а М м Е WINGS ane ae = py 2
м м вил [eoyoesy,
a jo Joaquin}
tp fe t+ +4 TN Paro ARSE -+ - quowioansvoul stuod-snuy
sf es ar a ++ TIN ++ °‘* queuomsvou jeavu-snuy
NE ee IN UNG ор аа yisusy Apog
ее зы ет.

„эертооча Арпитеу э9з UIYITM попехел Jo ззиэтоэоэ Jo позтлеЧилоо у “CE PIGUET,

120 Variability of Mammals

ribs and caudal vertebrae. All this taken together gives a basis for saying
that the Greenland seal shows a specific similarity with C. cristata in the
variability of the traits mentioned above, a rather closer relation than
with any of the other seals studied.

It is interesting that in comparing Е. barbatus with other seals the
greatest similarity in variability was observed between it and the Okhotsk
ringed seal, which is geographically rather close to it. The data are meager
for deriving a specific conclusion, but one cannot rule out the possibility
that here lies a chance to decipher the effect of natural selection on varia-
bility, through a comparison (using a sufficiently large number of traits)
of species inhabiting the same regions. It is quite possible that such a
comparison may in the future open the possibility of differentiating varia-
bility (brought about by general environmental factors with non-specific
relation to each species) and that resulting from a specific interrelation of
a given population in its own biogeocenose.

The last stage in the analysis of variability for C. cristata is a comparison
of variability with absolute values of traits in order to seek out deviations
from the elementary regularities of variability. Such a comparison can
be done, for example, by comparing variability values of skull traits. In
Figure 36 the coefficients of variation for different traits have been depicted —

C.v,%

mm. 15°

200

150

100

123198 9 4 15 6 2014215 18 7 11 1710 13 16 22 12

Figure 36. The variability (coefficients of variation) of skull measurements for
females of Cystophora cristata as compared to the absolute values of the trait:
a—Border of variability drift. For identification of the traits, see Figure 22.

Variability of Populations 121

as usual by placing them on the abscissa in order of declining absolute
values of the trait.

A comparison of the curves and variability drift in the male and female
C. cristata shows that, as with all seals, the variability of skull traits generally
increases with decreasing absolute values. But a number of values consi-
derably deviate from the mean drift in both males and females. More-
over, the trends coincide in both samples in certain instances : the varia-
bility for both males and females is low in relation to the mean for the
following traits : braincase width, height of skull, condyle width, width
of auditory bulla, length of auditory bulla, width of zygomatic arch. ‘The
width of palate and width of nostrils for both male and female C. cristata
are characterized by relatively high variability.

Variability of the following traits differs most sharply in males and
females: length and height of lower jaw, length of nasal bones, and width
of nasal bones. The differences in variability of major measurements
of the lower jaw, like its length and height, are inexplicable as yet. It
must be noted that the height of the lower jaw is the only trait for which
the variability is higher in females than in males; the variability of all the
rest of the traits, except for the auditory bullae, is greater in males than
in females. But the remaining two traits for which a considerable differ-
ence in variability exists between males and females, can be explained
specifically. The variability of length and breadth of nasal bones in males
of C. cristata is higher because the nasal bones carry a hypertrophied carti-
laginous partition serving as a base for the large (up to one meter or more
in expanded form) nasal pouch, which is a secondary sex characteristic
of the animal and which gives it its name; the pouch does not grow in
females.

The increased variability of the measurement “‘width of palate” might
have been influenced by the fact that in С. cristata, because of its feeding
on cephalopods (which invariably leads to a reduction in number of teeth :
Tomilin, 1954; Yablokov, 1958), a process leading to an early and frequent
loss of the last cheek teeth takes place. But according to the standard
method the width of the palate is measured between the borders of the
alveoli of the last cheek teeth. In some instances a full set of cheek teeth
exists in the upper jaw, but in others the last cheek teeth may be absent,
which brings about a sharp reduction in the measurement (although the
skull width in this region may have no reduction).

It is not clear how to explain the rather high variability of width
of nostrils, a trait which usually has low variability in other pinnipeds.
The relatively high or low variability of certain traits of C. cristata coincides
with analogical deviations for Pusa and Pagophilus (Figure 37). This is
apparent in the greater variability of the palate, lower variability of brain-
case width, greater variability of the width of the snout, and especially

122 Variability of Mammals

in the lower variability of length and width of the auditory bullae. The
characteristics of variability noted must be dependent on some factors
which are common to the whole family of true seals. Further ecological

research would show what these factors are.

C.V.% | i

423198 4 9 6 20 1552114 7 1110181317 12 2216

Figure 37. The coefficients of variation for skull traits of (а) male Gystophora, (b) Pagophilue,
and (с) Pusa hispida. For identification of the traits, see Figure 22.
A comparison of variability for other traits is given in Figure 38.

C.V.%
150

]

]

30.

20

10

123456789 10111213141516171819

Figure 38. Coefficients of variation for some traits of Cystophora :
1-7—Body measurements (dotted line—absolute magnitude of measurements) ; 8—.Number
of caudal vertebrae; 9—Number of tracheal rings ; 10-12— Total number of vibrissae

in different regions; 13-19—Number of lip vibrissae by rows.

Variability of Populations 123

The variability of number of rings in the trachea is lower than the varia-
bility of caudal vertebrae, as in many other seals. The behavior of the
coefficients of variation is somewhat unusual in the case of body measure-
ments. The least variability is not of the body length with fins but of
the girth. There is a disproportion also in the variability of other values
like body length measured along projections and curves, the distance from
nose to navel, and the distance from anal orifice to navel.

The comparison of variability for morphological traits of different
genera within one large and varied family of seals (the earless) does not
contradict the general conclusions derived from the comparison of species
within a genus, close genera among themselves, and different genera within
one subfamily. Morphological traits expressed in absolute values can
yield little information on the processes of micro-evolution. It is reason-
able to think that the features once formed under the influence of those
micro-evolutionary processes which still change the face of a population,
are now largely beyond the control of these processes. For example,
ecological divergence in feeding habits leads to a restricted diet in Condy-
lophora (deep-water cephalopods), in Erignathus (benthonic invertebrates),
and in Pagophilus (fish). And the adaptations to these diets (a reduction
of teeth, an improvement of eyesight, a capacity for deep submersion under
water, etc., in Condylophora; a growth of large numbers of lip vibrissae in
Erignathus ; a development of dental specializations together with a uniform
development of the vibrissae in Pagophilus, etc.), at the present moment
of evolution, partly determined the variability of such traits in each species.
Thus a delicate and soft trachea with many tracheal rings does not occur
in Condylophora as in Erignathus; and in the latter, all the results of a low
number of vibrissae, which are incapable of getting food from the ocean
floor, will necessarily be eliminated. A more general conclusion seems
possible : Without a considerable change in the external environment,
all the individual features of the pinnipeds now alive should develop in
the direction of further specialization, the general trend of which was
determined long before the present-day occurrence of the group.

Therefore knowledge of specific morphological features is essential
only for orientation within the boundaries of that variability which is cha-
racteristic of a given population as an evolutionary unit in its present deve-
lopment. The variability itself, expressed rather exactly and significantly
by the coefficient of variation, is capable of answering quite concrete ques-
tions on relatively recent micro-evolutionary processes and on effects of
general and partial controlling factors.

By comparing certain populations within a species, some species
within a genus and some close genera, it was not possible to find those
traits controlled by common environmental factors and those traits con-
trolled by specific conditions; a broad comparison of species within the

{

124 Variability of Mammals

boundary of the whole family, and perhaps such a comparison of specific
traits, permits one to do such an analysis. In all the seals studied in the
family, the variability in skull measurements appeared to have a similar
trend: relatively large variability of length of palate, relatively low varia-
bility of braincase width, relatively high variability of width of snout, rela-
tively low variability of parameters of the auditory bullae, etc. ‘There
can be only one conclusion on this level: that specific factors present in
all instances, influence all species and all populations. Even one such
conclusion helps in separating the ‘“‘doubtful factors’’ out of thousands of
factors which might have an influence on the variability of these traits.
But the analysis only begins at this point. The possibility arises of selecting
different organisms ecologically and geographically and, as it were, “asking”
nature in each specific instance to sort out the analogically influencing factors by select-
ing traits which are comparable to a maximum extent with regard to variability.

One of the possible ways of making such investigations is to compare
the variability of species which are considerably separated taxonomically
from each other but living in one biogeocenose. In the pinnipeds, these
will be Erignathus barbatus, Phoca vitulina, Pusa hispida, and Histriophoca fasciata
in the Okhotsk Sea, the Bering Sea, etc.; I am confident that the results
of such an analysis will let us arrive at conclusions of ecological as well as
of evolutionary importance. Even now, the comparison of variability
in a number of genera looks promising in spite of the limited data, for
Erignathus barbatus and Pusa hispida in the Okhotsk Sea and for Pagophilus
groenlandicus and Cystophora cristata in the North Atlantic. Naturally, such
an approach should prove itself to be very fruitful in studying other animals
also. But for the time being, while continuing the study of variability
within the boundaries of an order, it is essential to investigate the general
trends which appear at the level of a family.

The Nature of Variability in Populations of
Different Families within an Order

The amount of data available for a comparison of the three living
families of pinnipeds is not as much as was available for a comparison of
species within genera, or genera within the family of earless seals. But
I think that it would be improper not to attempt the comparison, though
in a most general way, with the limited data already available.

The three living families of pinnipeds are represented by different
numbers of species. We have studied the family of earless seals, which
constitutes about 60% of the living species of pinnipeds.’ The other two

?According to У. В. Scheffer (1958), the earless seal family (Phocidae) has 18 species;
the family of eared seal (Otariidae) 12 species; and in the family of the walrus (Odobenidae)
there is one species.

-

Variability of Populations 125
_ families of pinnipeds—eared seals (Otariidae) and walruses (Odobenidae)—
have a smaller number of species. The walrus family is represented by
a single genus and species (Odobenus rosmarus), and the family of eared seals
has a comparatively large number of species of fur seals, and sea lions
(Scheffer, 1958).

In the investigations carried out to date, there are practically no
data on population variability of walruses and few on the variability of
fur seals (Callorhinus ursinus) (Scheffer, 1958; Bychkov, 1965). It is possible
to compare variability for skull traits of earless seals with that of the nor-
thern fur seals (Callorhinus ursinus) of the commercially exploited group,
because V. F. Muzhchinkin has kindly made the analysis possible by pro-
viding unpublished data. ‘These measurements, taken for earless seals,
are presented in Figure 39.

As seen from this figure (where
the variability of 15 skull traits of
two different age and sex groups
of fur seals is given), a common
trend is observed in a number of
traits. They appear to be interesting
and significant in the given compa-

-rison, thus allowing one to bring
forth some traits typical not only for
the species but, perhaps, for the
whole family Otariidae. These traits
are as follows: braincase width (very
similar in all groups with a rather

c.v.%

Cystophora

12198 9 4 15 146 21201817 5 10

Figure 39. A comparison of coefficients of
variation for some skull measurements in a
fur seal (Callorhinus ursinus), from the data
of У. Е. Muzhchinkin, in dotted lines;

low variability); length of palate
(showing rather high variability
for all groups) ; height of skull (vari-
ability is low); width of occipital
condyles (variability is high) ; height

of lower jaw (variability similarly
high); and width of bullae (vari-
ability low).

The coefficients of variation for the same traits of the ringed seal (Pusa
hispida) and the hooded seal (Cystophora cristata) have been put on the same
graph (Figure 39). A comparison of these data with the data for the
fur seal shows that a high variability of length of palate and a low varia-
bility in height of skull and of width of bullae is characteristic of pinnipeds.
Of these three traits, only the low variability of the auditory bullae can
be explained on a functional basis. (They undoubtedly play a major
role in auditory reception in the case of aquatic mammals.)

Considering the data of V. A. Bychkov (1965), a gentle slope of the

Cystophora and Pusa are in solid lines.
For identification of traits, see Figure 22.

126 Variability of Mammals

ceiling of variability appears to be the general character for the sample
of northern fur seals. The coefficients of variation of measurements for
traits of the smallest absolute size, do not differ considerably from the
coefficients of variation for traits of the largest absolute size (condylobasal
length, etc.). It is difficult to say now to what extent this peculiarity
is inherent in the eared seals, though such a possibility is not ruled out.

In finishing the discussion of features of the population morphology
for different families of pinnipeds, let us note that the data on the varia-
bility of the walrus are poor; only general information on the number
of teeth in the upper and lower jaws is available (Table 36).

Table 36. Variability of number of teeth in adult male walrus from
the Chukchi Sea (From data of Bel’kovich and Yablokov, 1960)

Pais б 6.0.5 п
Lower jaw, right ee 4.10.32 0.47 11.5-- 0.63 165
Lower jaw, left oe 4.21 0.33 0.53 12.6--0.69 165
Upper jaw, right BA 6.2-40.43 1.08 17.4£0.35 212
Upper jaw, left AY 6.2 10.42 1.00 16.10.77 216

This trait (number of teeth in the walrus) is very specific and for the
time being cannot be compared with any other single trait among those
already studied.

It will be interesting in the future to compare the variability in the
number of teeth in the walrus with that for certain other species of pinni-
peds in the following order: walrus, hooded seal, crab-eater seal, bearded
seal, ringed seal. It can be assumed that the variability of number of teeth
will be highest in the walrus among all seals (as determined by its special
trophic relationships). The usual mode of feeding by the walrus leads
to the fact that it has completely reduced eye vibrissae, unlike the rest
of the pinnipeds (Figure 40), but the lip vibrissae are the best developed.
Yablokov and Klevezal’ (1964) found for this species a somewhat unusual
function. Direct observations in nature and in captivity showed that
walruses always contact the substrate with their vibrissae, perhaps detect-
ing small vibrations of sediment which are caused by travelling snails. The
lip vibrissae form a perfect net at the same time (Figure 41).

We see that the comparison of skull traits carried out among the families
of the order Pinnipedia show that among the families studied there are
common features in the variability of traits. This may mean that such
features are controlled by some environmental factors common to the
whole order. In the absence of large field investigations, it is difficult
to say what these factors may be.

Variability of Populations 127

Figure 40. А young walrus resting on Ruder Beach (Anadir Вау).
Notice the absence of vibrissae near the eyes. Author’s photograph.

But, it seems, a wider comparison of the variability of traits of pinni-
peds with the variability of the same traits for other mammals could answer
this question, particulary a comparison of pinnipeds with Carnivora (near
relatives of pinnipeds) having representatives accustomed to an aquatic
(Kamchatka beaver) as well as a semi-aquatic (otter and polar bear) mode
of life. In such a comparison, traits specially conditioned to an aquatic
life would come to light.

Figure 41. Muzzle of an adult male walrus; the clear disposition of different
rows of vibrissae and an evenly assembled surface of the “‘brush” is seen.
Ruder Beach, 1960. Author’s photograph.

128 Variability of Mammals

A determination of such general regularities in the development of
traits, or of the nature of the variability of those traits (increased or decreased.
as compared to the expectations of the elementary regularities) would give
an opportunity to analyze the mode of variability of a species and of indi-
vidual populations. This would be by comparing the observed deviations
in variability, and by knowing the magnitude and direction of deviation,
depending on the conditions of existence for the whole group. No doubt
during this analysis deviations from the general trends of the whole group
would be found in concrete populations, and these very differences should
be the object of further thorough studies of population morphology, which
is the study of the micro-evolutionary processes going on in such “hot
points.”

Certain Characteristics of Population Variability
on Different Taxonomical Levels

The experiment of making a systematic study of population morpho-
logy for different species within one order at various taxonomical levels
(for the first time carried out in this work) gives an opportunity for esta-
blishing certain general relationships.

The comparison of variability among populations within one species,
among closely related genera within a family, and among the representatives
of closely related families can be carried out in three main directions:
by comparing the means of quantitative characters for structures in different
samples, by comparing absolute and relative values of the standard
deviation and by comparing the variability as expressed in coefficients of
variation.

The first method—the comparison of the means of quantitative cha- _
racters of the samples—is the first step on the way to an analysis of popula-
tion morphology and gives factual material for conducting all further
investigations. However, this direction has an important independent
significance in lower levels of comparison (the comparisons approximately
up to the limits of a subfamily), indicating the existence of morphological
differences between different groups within a population as well as between
populations of different species. By such an analysis it seems possible
to derive a conclusion even about the degree of such differences (which
is not possible to do with the usual methods of comparative anatomy).

This way leads to conclusions having direct ecological significance
because it seems possible to presume the existence of fine adaptive
peculiarities in the groups under study, on the basis of differences in
morphological characteristics which can be gauged by methods of
quantitative morphology. .

The second method of studying variability—investigating the

Variability of Populations 129

distribution of standard deviation values—allows one to obtain data directly
characterizing the degree of similarity between the populations being
studied. Attention must be given to an unjustifiably forgotten method—
the method of comparing not the absolute but the relative values of the
standard deviations, which was proposed by S. R. Zarapkin (1934-1939).
The foregoing modification of this method allows a comparison of popula-
tions having most heterogeneous traits. This method of studying popula-
tion morphology occupies an intermediate position between the study of
absolute quantitative characters of traits and the study of variability itself.

The most important at all levels—from the comparison of intra-popu-
lation samples to the comparison of populations, genera, families and
maybe even of orders—is the third method of studying population mor-
phology : the study of variability expressed in dimensionless (without
any particular unit of measurement) figures of coefficients of variation.
The significance of this method in the investigation of a population con-
sists mainly in the fact that it characterizes the degree of interdependence
and relation of the trait to the effective evolutionary factors (of genetical
and ecological kinds). The variability, expressed in coefficients of varia-
tion, should be looked at as the expression of the most general relations
and interdependence established between the age-sex groups within a
population as well as among those relations by which the whole popula-
tion is held in the biogeocenose.

In a comparison of populations belonging to species of different genera,
the methods of comparing direct and absolute quantitative values of traits
assume secondary importance, since the differences between the traits
of different genera are so clear-cut that sensitive quantitative methods
of analysis are not required for characterizing and finding out these differ-
ences. The usual methods of comparative and descriptive anatomy quite
satisfactorily solve the problems of similarities and differences between
various samples. ‘The methods of analysis of variability assume the fore-
most place during such analysis. ‘The comparison of different species
in absolute quantitative expression of specific traits is useful only for ascer-
taining the presence of larger or smaller differences, but the comparison
of these species with regard to degree of variability in the traits under
investigation, allows one to evaluate the earliest steps in the appearance
of morphological differentiation of the species, which are the very first steps
in the evolutionary process.

It can be said on the whole that the knowledge of absolute values
of variability in traits during the analysis of different genera and families
is useful only for determining the limits of variability for the groups in
the present stage of their development. ‘The variability itself, expressed
in terms of coefficients of variation, seems capable of answering related
questions on the present micro-evolutionary processes.

130 Variability of Mammals

While studying the characteristics of the variability of different spe-
cies, it seems important to pay attention to comparing as large a number
as possible of both taxonomically close and distant species. Such a wide
comparison provides an opportunity to infer which of the characteristics.
of variability in one trait or another are determined by general environ-
mental conditions for populations of species, and which ones are deter-
mined by specific conditions in the direction of natural selection for a
given biotic community (biogeocenose). The comparison of variability
of species placed even widely apart in the taxonomy, but inhabiting one
biotic community, seems to be important in biogeocenosal relations. With
such species there could be a number of common features in the variability
of structures which again could be interpreted as the effect of general
biogeocenosal factors. Such analysis gives an opportunity for recognizing
a developing species and the formation of its peculiarities, and also a system
of regularities acting at the level of populations.

The data considered in this chapter allow one to draw an important
conclusion having great importance for the problem of studying variability
as a whole: Population variability does not depend on the taxonomic
position of a species. In the populations of very close species and in differ-_
ent populations within one species, the coefficients of variation may differ
more significantly than the coefficients of variation for populations of
taxonomically distant species, genera, and families. This differentiation
in coefficients of variation perhaps reflects differences in the force of evolu-
tionary factors.

CHAPTER [У

CLASSIFICATION OF THE PHENOMENA OF VARIABILITY

A Short Survey of Certain Classifications of Variability

Terminology

Researchers usually attempt to classify the phenomena of variability
as phenotypic and genotypic; of the two, phenotypic variability is studied
to a lesser extent. At the same time, it is clear that any phenotypic varia-
bility has a genetic basis. Though the relationship between phenotypic
and genotypic variability is far from clear, it is evident that a separate
classification of the phenomena of phenotypic variability is necessary
in the development of research in the field of population biology.

One of the first investigators, H. F. Osborn (1915) divided characters
into numerical (number of teeth, number of vertebrae, etc.) and propor-
tional (magnitude, form, intensity of color, etc.) characters, and observed
that there is a major and qualitative difference between these two cate-
gories.

One of the best known attempts at classification of the phenomenon
of variability was done by Robson and Richards (1936), who differentiated
variability according to structure, function, and activity, and separated
individual, group, and species variability. ‘They attributed such pheno-
mena as polymorphism, dimorphism, mimetic, and seasonal forms, pre-
sence of different stages of growth, etc., to species variability. While
analyzing this classification, I concluded that the author did not differen-
tiate very clearly between variability and variation in the wider meaning
of the word (see Introduction).

When considering the causes of variability, they separate the
non-hereditary somatic variation (phenotypic variability), the effect of

132 Variability of Mammals

recombining existing genes, and mutational variability (genotypical). This
classification, although with merits on the whole, must now be seen only
from an historical point of view.

In the work of Mayr, Linsley, and Usinger (1956) on the principles
of zoological taxonomy, another classification of variability is presented,
based mainly on the requirements of taxonomists. The phenotypic varia-
bility of mammals is divided by the authors into individual variability
over time (growth and seasonal), ecological (biotopical, determined by
the host, depending upon the density of the population, climatic, alterna-
tion of generations, and variability of colors), and traumatical (brought
about by parasites, by accidents, and teratologically). ‘The hereditary
variability (genotypical) is divided into variability related and not related
to sex, continuous and discontinuous (genetical polymorphism).

In this classification, there are some disputable questions probably
of concern even to taxonomists. First, there is no clear difference between
variability and variation; when the authors talk about growth variability,
they mean growth variation, and when the discussion is on seasonal varia-
bility, they mean seasonal variation, etc.

Moreover, it seems unjustified to combine so many different biolo-
gical concepts like variability from growth and variability from season
(the first is irreversible and the second is reversible) in one type (variability
over time). The variability of growth cannot be seen only from the point
of view of time variability. This phenomenon is complex and deeper
and is not related directly to the concept of generally understood “‘phy-
sical’’ time but is also related to the concept of “biological” time. ‘

Certain other subtypes of variability included in the ecological type
of variability! do not, in my opinion, deserve a special separation; at the
same time, other variabilities should have been separated into individual
types. The variability “depending upon the density of the population”
(1B3, according to the terminology of Mayr её al., 1956 translation), varia-
bility from alternation of generations (1B5), and variability of color (1B6)
fall into the first category. Biotopical variability falls into the second cate-
gory; it deserves placement in a separate type because each individual
population exists only in a specific climatic and geographical environment.

An attempt to classify the phenomena of variability within a popu-
lation was done by P. V. Terent’ev. He considers it possible to distin-
guish topographical (or geographical) variability, the variability of “form,”
(which consists of variation in different stages of growth and ecological

1 Here it will be proper to cite the opinion of E. I. Lukin, who wrote that the terms
“local” or ‘‘ecological’”’ variability have to be used with caution, because the local or eco-
logical forms are usually not the categories of variability, but the systematic units formed
through natural selection on the basis of hereditary variability, traits of which have been
adapted to the places of life for these forms (Lukin, 1940; p. 14).

Classification of Variability Phenomena 133

conditions), and finally, the ‘“‘undetermined’’ variability, which serves
as a normal curve of variability; to determine its properties it is sufficient
to give its mean value, standard deviation, and their errors of measure-
ment (Terent’ev, 1957; p. 79).

This classification also has some points. It appears that all varia-
bility of any type, and not only the so-called “undetermined”’ variability,
may have a distribution in the form of a normal curve. The variability
of “form” distinguished by P. V. Terent’ev is very wide (encompassing all
ecological and growth variability) which hardly makes this classification
useful. This observation applies also to geographical variability, a con-
cept of very wide scope, in which the concepts of biotopical and climatic
variability should also be included.

S. S. Schwartz (1963), while observing the complexity and extent
of variability within a species and the inevitable one-sidedness in forming
any classification of it, proposes to divide variability into individual, bio-
topical, growth, seasonal (chronological), and geographical categories.
This classification is far more constructive than previous ones and reflects
well the differences in expression of variability with regard to the effect
of evolutionary factors in a population in one area. It does not look proper
to separate only “individual” variability from other variability. Any
variability within a population occurs through an individual and only
through an individual; hence all the variability encountered by a researcher
during a population study is individual variability (which can also be
called intra-population or, to be more exact, population variability).

The other observation which must be made with regard to Schwartz’
classification of variability is concerned with the concept of “chronological”
variability. It is doubtful whether this concept could be confined only to
seasonal variability as has been proposed. It seems that all the forms of
variability of time should be included in the concept of chronological
variability instead of only seasonal variability.

The last and most detailed classification of variability has been done
by E. Mayr (1963), in which he proposes to divide all phenotypic varia-
tion into four large groups : (1) individual over time, (2) social (insects),
(3) ecological, (4) traumatical. Notwithstanding here the further frac-
tional differentiation of these main groups (which have specific points
of interest as, for example, fractional classification of time variations in
growth, seasonal, and generational variation, etc.), the basic shortcoming
of such an approach should be noticed. Here there happens to be a mixing
of different types of variability and uncomparable phenomena are compared.
Any individual variations in time could be at the same time ecological ;
therefore dividing variability into variability of time and ecology is not

correct.
Finally, in the same work of Mayr (1963) there is a very complete

134 Variability of Mammals

attempt to classify the phenomena of morphological variability. All the ~
kinds of morphological variability have been examined by Mayr from two
basic positions. First, the variability could be quantitative (measurable)
and qualitative (present-absent)?. Secondly, continuous (greater, smaller,
brighter, darker) and discontinuous (blue, grey, with white, without white,
and the like) variations can be separated among the morphological varia-
tions. The last type, discontinuous variation, is called the phenomenon
of polymorphism. In this classification a clear-cut distinction of morpho-
logical variability can be made into quantitative and qualitative varia-
bility, though the position of meristic variability remains undecided. ‘The
separation of the phenomenon of polymorphism or discontinuous varia-
tion is fundamentally important in the proposed classification.

While evaluating all the above-mentioned classifications of variability,
it can be noted that they have a common deficiency, i.e., the mixing up
of various approaches in the classification of the phenomenon; it appears
that this has not yet allowed the construction of a logical system of varia-
bility. There is no doubt that one of the reasons for such mixing of diffe-
rent approaches in the study of variability is the “narrowness” of the ques-
tion posed, either from a taxonomic point of view or genetical or ecological
point of view. |

Taking into consideration the previous experience of researchers,
and analyzing the data obtained on mammalian variability, it is considered
necessary and proper to group all the phenomena of individual or popu-
lation variability into three large categories, which should be pursued
independently.

1. Division of all the kinds of variability in nature into types (mor-
phological or structural, biochemical, functional or physiological, psycho-
logical or behavioral) .®

2. Classification of variability according to measurements (manifes-
tation of variability) by weight, volume, meristic, colorimetric, etc.

3. Classification of the phenomena of variability with regard to
factors influencing the development of the population (form of variability:
growth, sex, chronological, etc.).

However, before writing in detail about the identified types, mani-
festation and forms of variability, it is necessary to stop and consider the
peculiarities in the usage of certain terminologies. It was said above

2 E. Mayr observes here that non-metrical characters of morphological variability are
characteristic of mammals, birds and molluscs. This assertion of his indicates perhaps that
he had insufficient material on variability of mammals. As shown in various parts of this
work, the non-metrical variability of mammalian traits can be studied quite broadly.

ЗЕ. Mayr (1963) divides all phenotypic variability into two large groups: variability
of behavior and variability of development. The phenomenon of development of different
phenotypes under different conditions of life falls into the last category.

Classification of Variability Phenomena 135

that all the variability studied in a population appears in a population
only through an individual and hence can be fittingly termed individual
variability.

Understanding two aspects of the problem of variability, i.e., within-
population and between-population, originating from the work of Yr.
A. Filipchenko (1916, 1926) which has spread appreciably of late (Simpson,
1948; and many others) does not, in my opinion, result in a “mixing up
of the phenomena of variability and evolution” as indicated by E. E. Lukin
(1940). In the given case, the use or interpretation of the concept of
“group” or “between-population”’ variability need not be declared wrong,
but the exact place of the phenomenon, which is expressed in this concept
as a general theory of variability, should be found out. Further, it is
difficult to take objection to the claim that the phenomenon of “group”’
variability really exists in nature. This between-population variability
should also be clearly classified and should be widely studied; the forms
and methods of such study are examined in this work in passing, and of
course should form an independent object of research although they are
closely related to the study of intra-population variability and are based
on such study.

It is clear from the above that the term “‘intra-specific’’ or only “‘speci-
fic” variability, which has been widely used in the recent past, seems to
combine phenomena of different types such as within-population (indivi-
dual) variability and the between-population (group) variability; more-
over, the latter category of variability seems to be wider than the concept
of “species” variability and “‘within-species”’ variability. To be exact,
perhaps it would have been better if the concépt of “‘intra-specific”’ varia-
bility had been rejected and more specific terms like “within-population”’
variability and ‘“‘between-population” variability had been used. In
the present work, the term “variability” always and everywhere means
individual within-population variability.

Е. Mayr (1963, р. 139) writes: ‘“Моп-сепейс variation adapts the
individual, while genetic variation adapts the population.”’ (Non-genetic
variation is considered by Mayr as variation arising with a single genotype.)
On the other hand, it may be thought that in the framework of variability
in each trait controlled by natural selection, sometimes a situation arises
resulting in the appearance of variability which is not dependent on fac-
tors of selection acting at that moment. There is no doubt that at the root
of any type and form of appearance of variability, there are genetical factors.
Е. Mayr (1963, р. 150) correctly says that ‘‘much of this genetic variation
contributes to the variations of the phenotype.’ All types of variability
in natural populations are genetically constituted, larger in some situa-
tions (or with regard to one trait) and smaller in other situations (with
regard to other traits).

136 Variability of Mammals

Among many difficulties in evaluating genetic and non-genetic varia-
bility of a population, the most serious is the separation of the evolutionary
factors acting on that variability at the present moment and those that
acted in the past. For example: a short tail is an adaptive and advan-
tageous trait in a cold climate but, at the same time, without any sort
of selection for it in a population developing at colder temperatures, the
tail is shorter than in the control (Sumner, 1909; Mayr, 1963). In the
evolution of a species this feature of adaptive reaction is fixed in the normal
way by selection, but in the given generation and in the given population
there cannot be a straight correlation of this trait with the selection. It
is possible that the study of the interrelations of genotypic and phenotypic
variability in a population will be one of the basic approaches in investi-
gating population morphology, but its relation to phenotypic variability
is too complex to be examined in detail in such a cursory account. It
seems important to show the presence of genetical variability in popula-
tions through specific examples. This class of variability can become
manifest only during experimental studies of different populations.

It has been shown through special methods of study (Wright, 1952)
that genetic factors accounted for 43% of all the observable variability
in one experiment, while the remaining 57% (even under experimental
conditions) are controlled by environmental factors. This dependence
on environmental factors increases considerably in inbreeding and decreases
while maintaining the experimental population under conditions approa-
ching normal (Falconer and Latyszewski, 1952).

The data given by Castle (1941) have been used to calculate varia-
bility (Table 15, Appendix) of four lines of white rats. Since in the present
example we can exclude the effect of variability due to age (only indivi-
duals of similar age were studied), and the conditions of existence of these
four populations were identical, it has to be suggested that the differences
observed in the variability depend on genetic factors. It is interesting
to note that the data on females show a considerably higher constancy
than the data on males. |

The analysis of data from the work of Castle, Gates, Reed, and Law
(1936) on the important independent value of variability as a genotypic
character of a trait, shows that in a number of instances the considerable
differences with regard to the absolute value of a trait among the experi-
mental populations are not related to the presence of differences in the
variability of the trait. The reverse is also true. Differences in the magni-
tude of variability are specific between groups which do not differ in the
absolute values of a trait.

An analysis of variability of wild and inbred populations could be
important for understanding the role and value’of “genetic” and “non-
genetic” factors affecting variability. It must be noted that there are

Classification of Variability Phenomena 137

characteristic differences in different populations within a species with
regard to genotypic variability (Timofeev-Ressovskii, 1927; Zarapkin,
1934, 1937; Demerec, 1937; Olenov and Khrmats, 1939; and many
others). The last work shows that the spontaneous mutability in different
populations in nature differs more than ten times in insects. This has a
great significance in creating a heterogeneity among populations in the
magnitude of variability for many morphological and other traits.

On the whole, it is clear that the part of phenotypic variability which
is dependent on genetical factors, makes the analysis of variability observed
in nature somewhat difficult (Romashov and ПУ та, 1943; Mayr, 1963),
but. this situaticn shows how important is the study and calculation of all
phenotypic and also non-genetic variability (Schwartz, 1963, 1965).

~ Types of Variability

Structural variability

Structural variability is the variability of parameters in systems of
organs, individual organs, and their parts. 7

This type of variability is usually investigated by morphologists in
the class Mammalia. Such attention to structural (morphological) varia-
bility is not accidental, because the care in detecting it and performing
exact calculations of structural variations make the study of mammalian
populations considerably easier by giving a chance to carry out the analysis
at a high level with sufficiently specific and exact data.

Information on variability of this type comes from works of quite
different kinds—from some works on ecology to some on physiology, mor-
phology, and zoogeography. Data on structural variability of various
orders of mammals (mainly, works in the last decade) are given below.*

Order Marsupialia. The data are as yet scanty (Tribe and Peel, 1963;
Hughes, 1964; Sharman её al., 1964).

Order Insectivora. A number of researchers in different countries
have studied the morphological variability of small Insectivora and this
information is often published in the current literature (Markov, 1957;
Fateev, 1962; Schwartz, 1962; Dolgov, 1961, 1963; Kogteva, 1963; Mezh-
zherin, 1964a, 1964b, 1965; Lavrova and Zazhigin, 1965; Bazan, 1953;
Borowski and Dehnel, 1953; Pucek, 1956-1963; Stein, 1963; Osborn,
1964; and others).

Order Chiroptera. Special works on variability of Chiroptera are
very few, but among those available there are some interesting data
comparing fossil and recent Chiroptera (Bader and Hall, 1960; Lawrence,

4 In addition to the references cited here, many references to variability of mammals
are given in Chapters П and III and in subsequent sections of the present chapter.

138 Variability of Mammals

1960; Jegla and Hall, 1962; Jegla 1963; Sigmund, 1964; Troyer, 1965).

Order Primates. Comparatively few works are found on Primates in
the zoological literature, but some quite interesting data are found on
variability of the most unexpected traits and structures in some special
anthropological and medical literature (Roginskii, 1954, 1959, 1960;
Derums, 1964; Zhdanov and Nikityuk, 1964; Swindler её al., 1963; Hromada
and Strnad, 1962; Freedman, 1963; Schultz, 1917, 1926; James, 1960;
Lowrance and Latimer, 1957; Basu and Hasary, 1960; Banerjee, 1963;
Decaban and Lieberman, 1964; Cave and Steel, 1964; Nissen and Riesen,
1964; Matsui, Takayasu, 1964; Psenicka and Jurin, 1964; Ashton and
Oxnard, 1964, 1965; Broadhurst and Jinks, 1965; and others).

Order Lagomorpha. Some works have appeared in the recent past
(mainly of Polish researchers) on the structural variability of European
hare. They form a good basis along with the works previously done, for
obtaining a full picture of variability for the representatives of this order
(Filipchenko, 1916; Fateev, 1961; Labedkina, 1957; Bronson, 1958;
Latimer and Sawin, 1955; Sawin, 1937, 1945; Bujalska, 1964; Cabon-
Raczynska, 1964a; Lyne её al., 1964; and others).

Order Rodentia. The largest number of researchers have studied
variability for representatives of this order. It is not possible to give a
complete list of works, but the references made here create a common
understanding about the basic approaches in the study of structural
variability for these mammals (Argiropula, 1946; Artem’ev, 1946; Astanin,
1960, 1962; Bashenina, 1962; Bol’shakov, 1962-1965; Bol’shakov and
Schwartz, 1962, 1965; Vorontsov, 1961, 1963; Gershenzon, 1946; Gladkina
et al., 1963; Zubchaninova, 1962; Kogteva, 1963; Korotova, 1962; Rossolimo,
1962; Krivosheev and Rossolimo, 1964; Kryl’tsov’a, 1957; Larina, 1958,
1962, 1964; Meier, 1957; Ovchinikova, 1962, 1964; Petrov, 1961;
Pokrovskii, 1962; Popov, 1960; Razorenova, 1952; Straka, 1963; Fadeev,
1958; Schwartz, 1959-1964; Sumner, 1909-1932; Clark, 1940; Hoffmeister,
1951; Hoffmeister and Lee, 1963; Newson and Chitty, 1962; Packard,
1960, 1964; Ursin, 1956; Spitz, 1963; Hinze, 1950; Pelt von and Bree,
1962; Pucek, 1956; Kubik, 1953a; Wasilewski, 1952-1956; Haitlinger,
‘1962; Zejda, 1960, 1965; Guthrie, 1965; Schumacher её al. 1965,
Nord, 1963; Klemnt, 1960; Jackson, 1913; Gentile, 1952; and others).

Order Cetacea.® Because of the mass of material obtained from the
whale industry, data on structural variability of cetaceans is abundant
(Tomilin, 1957; Kleinenberg, 1956; Barabash-Nikiforov, 1940; Tryuber,
1937; Ivanova, 196la, 1961Ь, 1961с; Slentsov, 1961; Kleinenberg,
Yablokov, Bel’kovich and Tarasevich, 1964; Yablokov, 1958-1964;

5 The order Cetacea perhaps consists of two independently originating orders, the toothed
and baleen whales (Kleinenberg, 1958; Yablokov, 1964).

Classification of Variability Phenomena 139

Ichihara, 1963; Omura, 1964; Brownell, 1964; Nishiwaki её al., 1965;
Kasuya and Ichihara, 1965; Paulus, 1964; Brown, 1965; and others).

Order Carnivora. ‘This order stands in the third place in the number
of works devoted to structural variability (after rodents and primates).
As in other orders, it is not possible to quote all the works related to
the study of morphological variability for the representatives of this order,
but the list given introduces one to the scale and various approaches of
such studies (Kuznetsov, 194]а, 1941b; Popov, 1959, 1960; Fateev, 1961;
Belyaev, 1962; Bogolyubskii, 1941; Gerasimova, 1958; Krushinskiu, 1946;
Ognev, 1935; Pavlinin, 1962-65; Romashov and Ипа, 1943; Smirnov,
1962; Kurtén and Rausch, 1959; Kurtén, 1964, 1965a, 1965b, Rausch,
1963; Harlow, 1962; Hysing-Dahl, 1954, 1959; Schmid, 1940; Bader,
1955; Crusafont-Pairo and’ Truyols-Santonja, 1956; Latimer, 1936, 1944;
1944a, 1961; Sauer and Ruble, 1946; Voipio, 1962; Wood, McCowan
and Daniel, 1965; and others).

Order Pinnipedia. Works on the structural variability of pinnipeds
have begun to appear mainly within recent years (Popov, 1959, 1961;
Bel’kovich and Yablokov, 1960; Yablokov and Serzhent, 1963; Fedoseev
and Yablokov, 1965; Fedoseev, 1965; Khuzin, 1963, 1964; Bychkov, 1965;
Yablokov, 1963a, 1963b, 1964a; Scheffer, 1962, 1964). Detailed references
to the works on variability of pinnipeds are given in Chapter III.

Order Perissodactyla, Artiodactyla (even- and odd-toed ungulates). Data
on the structural variability of ungulates in comparatively small (Tsalkin,
1956, 1960; Sokolov, 1960; Demidova, 1941; Paaver, 1964; Tumurzhav,
1964; Segal’, 1962; Mezhzherin, 1965; Bader, 1955; Tribe and Peel, 1963;
Bergerud, 1964; Anderson её al., 1965; McMeekan, 1940; Shaw, 1929;
Klein, 1964; and others).

Data on the structural variability of other mammalian orders are
very scanty, or I do not know them (Newman, 1913 on Edentata; Bothma,
1966 on Hyracoidea and none for Monotremata, Dermoptera, Pholidota,
Tubulidentata, Proboscidea, and Sirenia).

While concluding this short survey of works on the structural varia-
bility of mammals, mention must be made of the ever-increasing stream
of publications concerned with discontinuous variability of the type
called epigenetic polymorphism, for quite diverse kinds of mammals
(Berry, 1963; Berry and Searle, 1963; Dunn, её al., 1937; Dunn ef al., 1960;
Stein, 1963; Pucek, 1962; Brownell, 1964; and many others). This approach
is examined in greater detail below in a special section of this chapter.

Many examples and many factual data on structural variability of
mammals are presented further in this chapter as well as in other chapters
of this work and also in the Appendix. On the whole, it may be concluded
that structural variability is the type which provides the basic material
in the current study of population morphology of mammals.

140 Varwability of Mammals

Physiological and biochemical variability

Functional variability. Functional variability is the variability of the
characteristic functions of different systems of organs, individual organs,
and the organism as a whole.

There is an appreciable literature on the physiology of mammals in
which the question of variation in the physiological functions of organisms
in different conditions has been studied. But in the majority of these
works, the variability has not been examined by the method we want.
This can be explained not only, and not very much, by the complex and
troublesome work this approach requires, but rather by the neglect of
physiologists to study the population as a whole. In this regard, the
work of B. P. Ushakov (1959a, 1959b) is very interesting; it shows what
important theoretical conclusions can be drawn from functional researches
in population variability.

It is natural that the works of functional morphology and physiology
be very~close to each other. Likewise, all the works on variability in
weight of organs are definitely very close to the phenomenon of functional
variability because they offer conclusions directly related to the study
of physiological features and the intensity of metabolism in animals. That
is why, these traits are called “morpho-physiological’’ traits in current
works (Schwartz, 1959, 1962; a summary of data on physiological varia-
bility of mammals is also given there).

An example of functional variability in mammals is a work which
studies the prolongation of periods of apnea in the seal Lepionichotes
wedelli under natural conditions (Littlepage, 1963). A. V. Pokrovskii
(1962) analyzed variability of the rate of attaining sexual maturity in
females of Lagurus lagurus during the major part of a year. The experi-
ments of Ponomarev (1944) and Kalkowski (1966) in determining tempera-
tures proffered by martens are interesting and exact from a methodological
point of view. Analogical work on mice has been conducted in
experimental ecology (partial summary by N. V. Bashenina, 1962).

The data of Korotova (1962) on ascorbic acid and glycogen content
in the liver and sugar in the blood, and of Sidorkin and Mikhaley (1964),
Denisova (1966), House её al. (1961), and Groulage (1961) on protein
fractions of blood plasma, should also be grouped under functional or
physiological variability, though these works are more concerned with
biochemical variability, which undoubtedly has an independent importance
in the life of a population. The data of Newson and Chitty (1962) on
variability of the quantity of blood hemoglobin is in the same
category.

Recently, it has become known that it is necessary and possible
to investigate biochemical aspects of purely chemical variability (Vino-
gradov, 1964). ‘These investigations show that different populations of

Classification of Variability Phenomena 141

organisms have different chemical compositions and give the variability
of these indices.

An analysis of the data presented above allows one to conclude that
for ascertaining the finer details of each population, the study of functional
(physiological) and biochemical variability is quite possible and useful.

Ethological variability. This is the variability of characteristic be-
havior. Very little study of this type of variability has been carried out
by zoologists. However, it is necessary to draw attention to the fact
that broadly speaking variability of behavior is shown by the immediate
reaction of an organism to environmental changes. It is very clear that
before physiological or biochemical changes take place, or still further,
before the appearance of morphological changes in the structure of
a trait takes place, an active change in the behavior of the animal occurs
under the effect of the changed conditions (Bei-Bienko, 1959). In this
way, the study of variability of behavior takes the researcher very close to
those first steps of micro-evolution which are the first reactions of all
populations of higher animals to changed conditions in their lives.

Because of the complexities imvolved in studying this type of
variability and similarly, because of an insufficient appreciation of the
importance of obtaining information on the character of a population,
our knowledge in this field is very meager.

The work of M. P. Kol’tsova-Sadovnikova (1931) on the psychological
peculiarities of rats of different genetical lines may be cited as an example
of a detailed work in this line known to me. The animals were let out
in a maze with closed doors; the problem was to find food by the shortest
route and then to remember this route (which was calculated in terms
of time taken for passing through the maze). Our analysis of the factual
data in this work showed that there were no differences in the value of
variability of time taken by laboratory-bred versus wild rats (Table 36).

Table 36. Variability of behavior of wild and laboratory rats (Absolute values
have been taken from the data of Kol’tsova-Sadovnikova, 1931)

Log of number of
minutes taken to
traverse the maze 1.675-+-0.016 20.3-L0.66 2.642+0.043 22.0--1.15 —2.1 —1.2

Mean errors for 10
experiments 58.33411.79 135.4+14.3 55.00-+2.91 54.5+3.74 +0.3 -+5.4

142 Variability of Mammals

At the same time these populations vary significantly (1=2.1) in
the absolute value of the logarithm of the number of minutes taken to
traverse the maze. While the number of errors committed when conduc-
ting the experiment between the wild and the laboratory lines are non-
significant ({=1.2), for variability of this trait, the differences are highly
significant (t=5.4). In this way the study of variability showed that
wild rats are psychologically homogeneous. This can be understood
from an ecological point of view. In the laboratory, selection on “sharp-
ness’ or “quick wits’ was absent for many generations, whereas among
the wild rats the inferior animals died soon. A number of works on
genetics devoted to analyzing behavioral peculiarities of different genetically
pure lines have appeared recently. In these works (for example, R.
Anthony, 1966; P. L. Broadhurst and J. L. Jinks, 1966), interesting data are
given for mice and rats on activity in open space, crossing obstacles,
frequency of defecation, running away from water, and other such |
parameters of behavior.

In the work of St. Skoczen (1962) there are data which show that the
composition of samples changes, depending on the method of catching
and the time of day. (The mole Тафа europaea was studied.) In the
morning, considerably more males are caught, while more females are
caught in the evening. The data of T. V. Koshkina (1964) show a
significant difference between the frequency with which males and females
fell into various types of traps—spiral or grooved. Bank voles (Cleth-
rionomys glareolus) were investigated.

There are data on the variability of parameters (like measurements,
volume, weight) of tree nests built by golden mice (Peromyscus nuttallt)
in. ‘Texas (Packard and Garner, 1964), which again can be taken as a typical
variability of behavior.

All these studies point to the fact that there is a specific sexual
dimorphism in the behavior of animals (see also Krushinskii, 1946), the
variability of which should be investigated in different populations. At
present such fine ecological investigations are not a great necessity, but
it may be remembered that only 20 to 25 years ago investigations were
carried out on samples of animals obtained at different times, at different
places, and of different sexes; now taxonomy requires samples of each
population having one sex and the same stage of growth. There is no
doubt that ethologists will also have to pass to the next step, i.e., the
within-population level of investigation, by using morphological and other
methods of study.

Manifestation of Variability

We have now data for differentiating the following manifestations

Classification of Variability Phenomena 143

of variability: time, weight, linear, and volume measurements; colori-
metry; variability of surface агза, of angles, and of temperature. АП
these manifestations of variability are examined below in detail because
such a classification is being done for the first time.

Variability of time. Varirbility of time is the variability in duration
of some life process.

A good example of such variability is the data presented by Hall and
Schaller (1964) on the variability of time spent by Enhydra lutris while
obtaining different types of food: molluscs, sea urchins, and fish. Another
example of time variability is the data on the time spent out of water and
in water by the Antarctic seal Leptonichotes wedelli (Littlepage, 1963).
Finally, the above-mentioned data by M. P. Kol’tsova-Sadovnikova
(1931) on the variability of time taken by rats of different populations to
traverse a maze, can be cited as an example.

There are further examples of variability of time. Data on the
variability of lifetime in rats maintained on food with or without vitamins
are presented by John Gowen (1936), and A. V. Pokrovskii (1962) have
studied the variability of time taken to attain maturity in females of
Lagurus lagurus depending upon the season of birth (Table 37).

Table 37. Variability of time taken to attain sexual
Е maturity in Lagurus lagurus (Absolute values, according
to the data of Pokrovskii, 1962)

XE Sz, Go Е a
January hy 84.55.05 28.6--4.22
Мау a 21.6-=0.75 14.42.47
June Pa 55.9-+ 10.55 65.5-+13.1
October .. 140.9+14.9 35.71-57.77

It is seen from this table that there is no direct relationship between
variability and the time taken for sexual maturity. This means that
in this case the coefficient of variation may be an important independent
character revealing specific biological facets of the animal under study.

Interesting data on the variability of time taken for passing different
types of food through the intestinal tract and the time taken for its excre-
tion in various types of mice are presented in the work of A. Kostelecka-
Myrcha and Myrcha (1964).

As seen above, the study of time variability is usually associated
with the study of physiological (functional) and psychological (ethological)
variability.

144 Variability of Mammals

Variability of weight. ‘This is the variability of the weight of an
organism, a system of organs, an organ, or their parts.

Variability of weight is one of the most usual manifestations of
variability with which researchers have to deal during the analysis of
data concerning population morphology.

The dependence of weight variability on seasonal factors, sex, growth,
and the general condition of the animal makes it an interesting index— -
one which should be given foremost ecological and physiological impor-
tance (Schwartz, 1960, 1964). This fact is at once illustrated by the
comparative data on variability of weight and variability of other traits
obtained from an analysis of the data of W. Е. Castle её al. (1938). In
this work the effect of reciprocal crosses in various combinations was
studied where the separate phenotypes of male and female mice were:
compared (black, brown, dark or light colors, etc.). It was observed
that the variability of weight was correlated with the linear variability.

In another series of experiments (Castle, 1938), the increase and
decrease in variability of body weight exactly coincides with the trends
of variability of body and tail length. Moreover, it is important to note
that in this case the variability in each individual phenotype seemed to
be higher than the variability of all the males and females in the popula-
tion. Finally, in a third series of experiments with different phenotypes
of one population, ‘an average situation is met when the coefficient of
variation for weight (and also for other traits) for the whole population
appears about average as compared to its values for the seven individual
phenotypes. These features were not noticed by the authors, either
in the first or the second case, because Castle and his colleagues failed
to calculate the coefficients of variation.

The study of variability of weight as a method of genetical research
is promising not only in experimental populations. There are a number
of examples where variability of weight has been used for the study of
ecological aspects of a population in nature. Data from the work of
S. S. Schwartz (1960) on variability of the relative’ weight of organs in
various populations of hamsters can be presented (Table 38).

Unfortunately, the errors of the coefficients of variation have not
been given in that work, which makes it difficult to carry out an exact
comparison of the magnitude of variability of different populations; but
even from the data presented, it can be concluded that the populations inves-
tigated have, perhaps, rather different relations with environmental factors.

In the case cited, the differences between populations were clearly
manifested only between the groups of animals having similar age and

6 The question of variability of relative and absolute indices has been specially discussed
in Chapter V.

Classification of Variability Phenomena 145

Table 38. Variability of weights of certain traits of Citellus
pygmaeus from different populations (Schwartz, 1960)

Group Вора а ааа
Heart Liver Pancreas Kidneys
Adult males Ni I 6.7 21.4 46.8 —
II 19.7 9.0 21.0 14.9
Immature females = I idee 16.4 21.6 —
II 15.4 12.0 113)7/ —

sex (adult males in one case and immature females in the other). It is
clear that all these differences would have been nullified if the variability
of the population had been calculated without forming any groups. For
an exact comparison of populations in such a case, it is necessary to
compare them not only for one season but during the same year. This
rule is especially important in the study of the variability of mammalian
populations having short life spans.

The situation given in the foregoing paragraph is well illustrated
from the data calculated on the basis of variability of body weight of the
common shrew (Sorex araneus) caught in consecutive years in a single
population; the data are presented in Table 39.

Table 39. Variability of body weight of the common shrew (Sorex
araneus) in different months and years in the same population
(Calculated from data of Borowski and Dehnel, 1953)

1949 1950
Month Se SSS tag—50
6. VES. y, С. U.ESe.y
June oe 14.8-+ 1.46 23.73.84 — 2.19
July .. 10.040.94 — 28.04.54 —3.89
August os 11.70.95 10.01.17 +1.15
September Bs 12.2+1.10 13.11.97 —0.39

As seen from the data presented in Table 40, the variability of the
body weight of shrews of the same population in different years and in
different months of the year differs significantly, and these differences
are comparable to the differences in variability for different populations
within a species.

There are many more works on the variability of weight. Some of
them, with data calculated by myself, have been presented in other parts
of this book. Different research workers have investigated this variability

146 Variability of Mammals

Table 40. Variability of body weight of the common shrew (Sorex araneus)
in different populations (с. 9. Е 5с.,.) (All calculations are ours)

Animal reserve of Ok. Belovezh-1 Belovezh-2
(Dolgov, 1963) (Pucek, 1956) (Borowski and Dehnel, 1953)
June-August June-September June-August
9.7-0.68 (males) 8.7+1.14—13.9+ 1.79 10.0+1.17—28.0+4.45

9.7-+ 1.82. (females)

of weight in unexpected ways, in an attempt to ferret out some exact and
significant traits. For example, data are presented on variability of hair
weight in relation to the square of body length in the work of F. B. Sumner
(1903). The variability of hide weight of wild animals of different ages
is presented in the investigations of V. Scheffer (1962) on fur seals. The
weight of the skull and its individual parts, the total musculature and sepa-
rate functional muscle groups, weight of intestines and stomach, weight
of eyeball and eye lens, in fact the weight of practically all the internal
organs has been studied (see Chapter II and the Appendix).

On the whole, it may be said that all those research workers are right
who discern in the weights of organisms of a population an important tool
for knowing the characteristics of populations. One can only add to
this that the study of variability of weights deepens the investigations of
populations and permits conclusions to be drawn which otherwise would
have slipped from the observation of researchers had they calculated only
the absolute parameters.

Linear variability. This is the variability of linear parameters for an
organism, a system of organs, individual organs and their parts.

The study of variability of linear measurements appears to be ie
kind most widely used in mammals. Body measurements for individual
parts of the skeleton, intestinal tract, and hair, and linear measurements
of organs and individual cells are studied. It was linear variability which
until now served as a foundation in works on the taxonomy of vertebrates.
Linear variability has great significance in the study of population mor-
phology because it presents methodological advantages such as simplicity
and accuracy of measurements.

Linear variability, like variability of weight, is related to the sex and
age composition of a sample and also to the seasonal availability in obtain-
ing samples. This fact regarding seasonal (and in that way, chronological)
variability is still used very little today in investigations on morphology
of animals. Because of this, data from many ‘investigations cannot be
utilized for an exact comparison.

Classification of Variability Phenomena 147

This situation compels us to examine in detail certain examples of
the relationship of variability of linear values to the influence of chrono-
logical factors (Table 41).

Table 41. Variability (с. v.--s,.,,) of a number of traits in small mammals
by months (Calculated from data on variability presented in the
works of Adamczewska, 1959 and Borowski and Dehnel, 1953)

ee eee ЕВЕ

Apodemus flavicollis Sorex araneus
Month -—— — — — -— — —— sss ss
Foot length Ear length Tail length Body length
June a 7.10.60 10.50.87 — 4.9 0.79
July Be 9.2+0.23 8.00.36 —— 8.4+ 1.37
August oe 5.2 +0.36 7.6+0.52 31.5+2.25 4.8+0.82
September sis 4.10.17 5.0+0.21 22.1+0.97 4.4--0.66
October Sif 5.8 0.62 5.7-- 0.60 12.31.45 2.00.38
November we 4.0--0.31 5.00.38 11.40.91 —

It can be seen from the data in Table 41 how significantly (in some
instances by more than a factor of two) the variability of a single trait
changes during halfan year. | - р

The difference in the variability of linear traits for a single population
of mammals from year to year is not less significant. From the results
obtained by me as a result of calculations made on data of body length
of the common shrew in Belovezh (Tables 42 and 43), one specific conclu-
sion can be drawn: The variability of the same trait in a population
can sharply deviate from year to year. It is interesting to note that though
the absolute values of these deviations are not always significant, the trend
toward increase or decrease of variability can be detected quite clearly.
For example, during a comparison of data on variability of shrews born
in the 1946-47 season, with similar data for shrews born in the seasons of
1947-48 and 1949-50, for the majority of those months in which the data
were collected, the variability in the season 1946-47 was low (in seven
instances out of nine). Further, during the comparison of variability in
the seasons of 1947-48 and 1949-50, the variability in the latter season
appeared to be higher in all eight cases. The variability of the traits
studied was also higher in the 1949-50 season when compared with the
1950-51 season. The same trend was likewise observed during the com-
parison of mean yearly values of variability.

Another example is taken from the work of V. N. Pavlinin (1962)
(Table 44). From the data presented in this table it is seen that the varia-
bility of linear traits of the skull varies absolutely and characteristically
in different years. Of the four traits investigated in both males and females,

148 Variability of Mammals

Table 42. Chronological variability (c.v.--s, ,) of body length of the common
shrew (Sorex araneus) (Calculated on the basis of data on variability
presented in the work of Borowski and Dehnel, 1953)

Month 1946/47 1947/48 1949/50 1950/51
June be — 3.8 0.30 4.0+0.21 3.60.38
July ее — 3.8 +0.18 4.6+0.15 4.5-£0.28
August di — 4.1 +0.21 4.70.14 _ 4.1-Е0.38
Зее 3 = 3.6 +£0.29 4.9--0.33 4.10.42
October ig — 4.5 0.65 5.70.70 —
April ae 4.30.43 6.8 0.53 — —
Мау sie 3.60.63 5.10+0.60 — —
June ae 4,9+0,65 3.8 +£0.28 8.44 1.37 —
July $0 3.6-Е0.48 4.4 +0.41 4.9--0.79 —
August ate 4.30.54 4.6 0.36 4.80.82 —
September ор 5.00.52 — 4.4--0.66 —
October by — — 2.0+0.38 —

Average 4.3 4.5 4..8 4.1

— Ce ee, Ce, C_ . _ _. — — _ e

Table 43. Comparison of difference (1) in the variability of body length in the
common shrew born in different years (From the data of Table 42)

a i a sa i se ee Se Se ee ремня

NEE 1946/47— 1946/47— 1947/48— 1947/48 — 1949/50—
1947/48 1949/50 1949/50 1950/51 1950/51
June — — —0.58 +0.40 +1.15
July о — — — 3.38 —2.06 +0.03
August i — — —2.36 0.0 +1.46
September We — — —2.89 — 1.00 +1.49
October bic — — —1.25 — —
April a — 3.59 — — — —
Мау о Мб —1.97 — 1.65 — —
June BN +1.59 — — == =
July fe — 1.29 —1.39 —0.56 — —
August 5:0 —0.49 —0.54 —0.33 — —

September a — -- 0.69 Е в wae

—_ дд

Classification of Variability Phenomena 149

Table 44. Chronological variability of skull traits of a population
of American mink (Mustela vison) in Bashkir (Calculated on the
basis of data presented in the work of Pavlinin, 1962)

1957/58 1958/59
Е Е ЕЬ typ
(iy OSES pg. 5 рае
Condylobasal length :
Males ae 4,410.6 2.8--0.4 --2.22
Females ie 2.30.3 2.9+0.4 — 1.20
Width of zygomatic arch:
Males Ne, 12.0-+.1.7 5.6+0.8 -- 3.35
Females a 6.4+0.9 5.0-0.7 1.23
Width between orbits :
Males rae 6.6-L0.9 7.2+1.0 — 0.44
Females Е. 7.9 1.1 6.5--0.9 -- 0.42
Width of snout :
Males Bie 8.5-+-1.2 5.10.7 -- 2.43
Females aes 6.3+0.9 4.5--0.7 + 1.64

—_ ee

in the season of 1958-59, the variability was higher for three traits.

The linear variability is clearly related to the variability of age and
sex. The relation of linear variability to sex variability in morphology
and ecology of animals has been examined also in Table 44. The fact
that the degree of difference among females is somewhat lower than among
males, immediately attracts attention.

On the whole it can be said that the phenomenon of linear variability,
perhaps better called the variability of linear traits, which appears to~
be very simple and clear at first sight, seems to be related to a whole com-
plex of other factors in natural populations. This shows the desirability
of studying linear variability in relation to other phenomena.

Variability of volume. ‘This refers to the variability of volume of ап
organism, a system of organs, individual organs and their parts. ~~

Theoretically, it could be imagined that the variability of volume
is not less important or significant than weight and linear variabilities.
However, the works on variability of volume are not many. These few
references confirm the possibility of using volume traits for characterizing
a population as a whole and also individual groups within a population.
The data on variability of volume of stomach and intestines for minks show
that the males are quite different from the females in this respect
(Table 46).

Specific differences in the variability of the volume of skull have been
observed in the work of J. Hromada and L. Strnad (1962), who studied
the craniology of Macaca (Table 46).

150 Variability of Mammals

It is seen from the data in this table that the variability of volume
appears to differ among age groups : for example, between infants and
adults of М. rhesus and between Young I and Young II in М. cynomolgus ;
and between animals of different sexes (between males and females of
Young I and Young II groups in M. cynomolgus).

Table 45. Variability of volume of stomach and intestines in
(Mustela vison) and fox (Vulpes vulpes) (Calculated from
the data of Sokolov, 1941)

Males Females
6. USey 6. VS. y Зо

Minks

Stomach ss 6.1+0.82 12.9+ 1.86 —3.79

Intestine ae 9.7--1.27 10.51.55 — 0.38
: Foxes

Stomach Be 9.1 +1.37 16.6+2.04 — 2.20

Intestine Types 12.5+1.97 12.6+1.71 —0.01

Table 46. ' Variability (с. ии) of volume of skull in two species of Macaca
(According to Hromada and Strnad, 1962, with addition of calculated
error of coefficient of variation)

М. rhesus M. cynomolgus
ENG TOW) i ee
Males Females Males Females
Infants aid 8.8 +0.72 8,0-+0.75 — Е
Young I te 10.1 40.83 9.3-+0.69 7.30.96. — 13.3-+2.35
Young II Г: 13.0 1.96 9.3 1.60 13.7-Е 1.37 8.9--0.95
Adult 35 13.0+2.11 11.5-Е 1.46 11.80.77

Variability of volume is very closely related to variability of weight
and possibly should be taken as a part of the latter. However, the beha-
vior of the coefficient of variation’ of volume traits does not always coincide
with the behavior of weight traits (see Chapter V).

As can be concluded from the above examples of variability of volume,
this manifestation of variability can be compared with any other

7 Some expressions usually used in mathematics but not in biological works, are found
herein, for example : “‘behavior of coefficient of variation,” “behavior of curve,” etc. These
expressions represent exactly the phenomenon under consideration and so I think they
may be used in some cases.

Classification of Variability Phenomena 151

manifestation with regard to the coefficients of variation, and so there
is no reason why the variability of volume should not be studied during
investigations of a population.

Meristic variability. Meristic or countable variability is the variability
in number of discrete elements in an individual system of organs or in
life processes.

The study of the variability of meristic traits is quite tempting from
several viewpoints: First, the arresting exactness and objectivity of the
data obtained; and second, as will be shown below, the meristic traits
are among the most stable because they are little affected by chronological
and age variabilities. Nevertheless, the last statement is not always correct
and traits like number of hair on a unit area of body surface are affected
by age and seasonal variability. Similarly, the number of small glands
of different types in the digestive tract of animals is affected by the sharp
differences in the food of various seasons. The number of such examples
can be increased. They show that somewhere here lies the line of Пат
cation between qualitative and quantitative variability.

Some of the known meristic traits being used in morphological in-
vestigations of a population, should undoubtedly be attributed to qualita-
tive variability (number of vertebrae or ribs; number of individual elements
in hands and feet; number of embryos, etc.) ; but another part of the meristic
traits, with the same specificity, should be attributed to quantitative varia-
bility (number of hair; number of sensory endings in the skin and other
organs; number of glands in the surface of the digestive tract; number of
cells in histological structures, etc.).

Because of the need for a precise demarcation of the phenomenon of
quantitative and qualitative variability of meristic traits, an approach
proposed by N. A. Bernshtein (1962) appears to be interesting : “That
subclass of the arguments which we considered as ‘insignificant’ (in the
sense of Gel’fand-Tsetlin), mainly determines the continuous traits (mea-
surements, angles, amplitudes of movements, etc.) and forms the corres-
ponding variable populations. In other cases, where the changes are
discrete and related to phenomena of comparatively large numbers like,
for example, numbers of hair on the head, cells or the synaptic contacts
in the cerebral cortex, etc., instead of a discontinuous variability, there
appears a variability having all the characters of randomization. ‘The
determination of one or another large number obtained in each individual
case (number of cones of the fovea centralis of the retina, Purkinje cells,
cilia of chorion cells, etc.) has all the characters of accidental processes
which do not in any manner separate it from the laws of causality. How-
ever, this situation places an important biological question before us:
Where lies the line of demarcation between the area of the numbers which
can be exactly determined and which are not changed by heredity, and the

152 Variability of Mammals

area where the accidental and random processes and numbers dominate
the species ?” (pp. 62-63).

While answering the question, М. A. Bernshtein concludes: ‘“...The
numerical investigations carried out on those frontiers of study where
deterministic exactness yields to variation and chance, are interesting
from the point of view of answering this question. In morphogenesis,
the determined quantities do not exceed two-figure numbers (number of
vertebrae or teeth in animals of one species; number of legs, fingers, eyes,

_ cyto-architectural fields in the brain; etc.). After this, in my opinion, in
a parallel and corresponding action, comes the jump from the determinable
to the accidental....”’ (р. 63).

In the light of the data presented in this work, it is seen that the demar-
cation between random aud determined quantities carried out by Bern-
shtein (1962) is not exact. Many examples of wide variations in the num-
ber of teeth in different mammals are known. Moreover, it is important
to note that only in the toothed whales are there all the gradations from
truly randomized distribution of number of teeth in some species of dol-
phins to an exactly determined number of teeth in others (more than in
Carnivora, Rodentia, and Ungulata) (Yablokov, 1958). The common
dolphin (Delphinus delphis) can be cited as an example of the first type;
it has the highest number of teeth among all mammals (reaching 240 in
number), The beaked whale (<7iphius) and narwhale (Monodon monoceros),
which have only one to four teeth, can be cited as examples of the second
type. It is important to note that there are all types of transitions between
these extreme situations. Unfortunately, data on exact quantitative
variability of organs of this system in cetaceans are very scanty. But those
few that have been obtained are of interest (Table 47).

_ То analyze these data in the desired manner, it has to be noted that the
white whale has the fewest teeth of the three species being compared (seven

Table 47. Variability (с. о. 150...) of number of teeth in certain toothed
whales (According to the data of Kleinenberg, 1956, and the author)

Bottlenosed dolphin Common dolphin White whale
( Turstops truncatus) (Delphinus delphis) (Delphinapterus leucas)
Females Males Females Males Females Males
Lower Jaw
те ey) of 0162 6.7-- 1.44 5.2+1.05 5.10.82 13.8+3.69 9.1+1.01
Right .. 4.8 0.74 7.341.72 5.2+1.06 5.6--0.98
Upper Jaw
Left ,. 4.1+0.59 3.9-++0.77 5.2+1.05 6.20.95

11.4£3.05 8.0+0.88
Right .. 5.5+0.79 6.5-Е 1.23 4.5-+0.95 5.7-Е0.88

Classification of Variability Phenomena 153

to nine teeth on each side of the jaw) ; the bottlenosed dolphin has 19 to 21
teeth in each half of the upper and lower jaws; and finally, the common
dolphin has 41 to 44 teeth in each half of the upper and lower jaws. We
then see that the variability in number of teeth is inversely proportional
to their numbers, i.e., the more teeth, the lower the variability. Such an
exact corresponding relationship (Chapter V), to the elementary regulari-
ties of variability, at once makes investigations difficult “where determi-
nistic exactness yields to variation and chance” (Bernshtein, 1962) and makes
the solution of the question more interesting.

On the basis of the available data on population morphology, it may
be said that the jump—from deterministic to random, which N. A. Bernsh-
tein writes about—lies somewhere within the limits of such traits as number
of extremities, eyes, vertebrae, fingers, and teeth. All these traits are sus-
ceptible to a rather noticeable variability in mammals, even within an order
of mammals, i.e., cetaceans. ‘The number of eyes in toothed whales varies
from two to none (eye being understood here as an organ functioning for
sight, and in this sense it is absent in certain river dolphins (Platanistidae) ;
the number of extremities varies from two to four. Well-developed hind
extremities have been recorded in Physeter catodon (Zemskii and Berzin,
1961). The number of fingers in cetaceans varies from five to three. The
variability in number of teeth has already been discussed above. ‘The data
reveal the variability of vertebrae in practically all regions, even the neck
(Beddard, 1900).

Significant data on variability of number of caudal vertebrae in other
aquatic mammals, certain species of Pinnipedia, have been obtained by
us (Table 48).

Table 48. Variability of number of caudal vertebrae
in certain pinnipeds (Adult males)

Hooded seal (Gystophora cristata) 14.00.17 6.6+0.84 Yan-Maien region, 1962

Greenland seal (Pagophilus gro-
enlandicus) 13.4-0.12 7.8+0.62 White Sea region, 1961

12.6+0.11 4.5+0.61 Newfoundland region, 1963
Common seal (Phoca vitulina) 12.8--0.14 6.5+0.81 Okhotsk Sea, 1963
Bearded seal (Frignathus bar-

batus) 11.9+-0.25 8.0+1.51 Okhotsk Sea, 1963
Baikal seal (Pusa sibirica) 13.00.09 4.40.53
Caspian seal (Риза caspica) 12,140.28 9.30.58
Ringed seal (Pusa hispida) 13.4--0.29 8.11.91 Okhotsk Sea, Тали, 1963

12,9+-0.03 1.5+0.14 Pechora Sea, 1962

154 Variability of Mammals

Even these limited data show that the number of vertebrae varies
considerably not only among different species but within each species;
coefficients of variation reach up to 8 or 9%. It may be remembered
here that pinnipeds have a very short tail and in animals with long tails,
for example in whales, the number of vertebrae may vary more significantly.
The difference in the variability of such an insignificant trait as the number
of caudal vertebrae among different species of pinnipeds would appear
to be significant. In the Pechora seal the coefficient of variation is only
1.5% and in the Caspian seal it is six times more! Such large differ-
ences could hardly be explained on the basis of randomness. As shown
in our investigations on the ice of the White Sea, the tail of the Greenland
seal plays an important role in coital behavior in the breeding season, indi-
cating the degree to which the female is ready to copulate with the given
male, and the degree to which the male is ready to fight for her and follow
her. Thus it would appear that a less significant trait takes on an impor-
tant biological significance.

Coming back to the problem of classification of variability in mamma-
lian populations, one must observe what phenomena the meristic variabi-
lity of traits is associated with and to what extent this variability can charac-
terize the population as a whole and also the groups within a population.
From the data already presented, it can be concluded that the variability
of the same trait may be different in different species. Interesting data
on the variability in number of phalanges for the second and third fingers
of cetaceans have been obtained from the work of I. I. Barabash-Nikiforov
(1940) (Table 48a).

Table 48a. Variability in number of phalanges for 2nd and 3rd fingers in
different populations of the common dolphin (Delphinus delphis) (Absolute
indices have been taken from the data of Barabash-Nikiforov, 1940)

Region Xt Se. с. 9. Е 5у.е. Region * sa 6. U+ESe.y
Yalta .. 7.8340.05 4.70.48 Batumy .. 7.60-+0.08 6.7-L0.71
5,94--0.05 6.2+0.63 5.78+0.06 7.10.79
ly—m ly-p
Novorossiisk .. 7.82-Е0.05 5.0+0.42 2nd finger .. —0.47 —2.32
5.96--0.06 8.6-+.0.71 3rd finger .. —2.51 —0.90

The next example of the use of meristic variability is also concerned
with the variability of certain structures of cetaceans, this time on. baleen
whales (Mystacoceti) (Table 49).

It is seen that the variability of the number of vibrissae on the upper
jaw is greater than for the lower jaw; moreovér, the difference in varia-
bility in this case is not governed by elementary regularities (Chapter V),

Classification of Variability Phenomena 155

Table 49. Variability of number of vibrissae on different parts of
head of the finwhale (Balaenoptera physalus) (Kuril Island, 1959)

Portion of head ХЕ 5х О еб n
Lower jaw, anterior 34.8-+ 1.46 23.72.97 32
The same, sides | 8.50.38 ЕР 8
The same, total 52.8+2.01 20.6+2.70 29
‘Total on upper jaw 21.1-+-1.45 32.4--4.88 22
Total on the head 73.42.80 16.42.70 19

so there is a basis for postulating the presence of specific functional and
ecological reasons which create the given situation. Such reasons have
been worked out in previous studies (Bel’kovich and Yablokov, 1963; Yablo-
kov, 1963b; Yablokov and Klevezal’, 1964), and they are discussed in detail
in Chapter VI of the present work. Here it should be emphasized that the
meristic variability in this case appears to be strongly related to the func-
tional characteristic of the organ.

In this connection, other data on the variability of another trait—the
number of individual fringes in 1 cm of whalebone sheets—in baleen whales
are interesting (Table 50).

Table 50. Variability of number of fringes on 1 cm of the edge of
a plate in certain baleen whales* (Adult animals)

Species 5 6. 0. Sey, n Remarks

Blue whale (Balaenoptera ''

musculus) 30.1-4:1.77 26.24.15 21
Finwhale (B. physalus) 24.6--1.18 21.4+3.39 20 Male of 18.9 meter length
Lesser rorqual' (В. acutoro-

strata) 20.0--1.05. 18.13.69 12
Humpback whale

(Megaptera nodosa) 18.4+0.82 19.8+3.14 20

* The calculations of different baleen plates, taken from the same animal, were carried
out by a Diploma student of the Vertebrate Zoology section of Moscow State
University, T. V. Andreeva.

It is known that the sheets of whalebone with fringed edges represent
a portion of the filtering apparatus for baleen whales. Details are in-
sufficient about the function of this structure. The data presented on
variability in number of fringes show that in Balaenoptera musculus, which
eats plankton and crustaceans, the variability of fringes is greatest, and
in the humpback whale and lesser rorqual, which feed more on fishes than

156 Variability of Mammals

the others, the variability is least. Such an interrelationship cannot be
accidental and needs further careful ecological investigation.

From the scanty data in the literature on the meristic variability of
mammals, interesting results have been obtained by us from an analysis
of data of F. Sumner (1909) on the number of hair per unit of body surface
in mice depending upon their living conditions (с. 0.5. „. and й:

Е и С т

Males Females
Hot Cold Hot Cold
11.72.49 12.9--2,52 15.23.80 17.74.43
t Be hot-cold 100 hot-cold t a о cold i Зо hot
— 0.34 — 0.43 —0.76 —1.13

A comparison of the coefficients of variation shows that, although
the differences are not significant statistically in even a single case, some
specific trends exhibit themselves uniformly :

1. Variability in the cold room is greater than in the hot room;

2. Variability of this trait in males is less than in females, both in the
cold and hot rooms.

The material presented is sufficient to show not only the possibility
but even the desirability for wide use of meristic traits in the study of mamma-
lian populations.

Colorimetric variability. This is the color intensity of an organism and
of different systems of organs.

Variability of colorimetric traits and the nature of mammalian struc-
ture is many-sided in expression. First, it is possible to determine varia-
bility of intensity in some color with the help of a densitometer or photo-
meter; secondly, it is possible to measure variability of area occupied by
the colored field. The last method actually belongs to another type of
variability—variability of field area—but in addition to changes in the
field area, in many instances the variability of color can be expressed simply
by linear measurements of the length of white patches on the tail or indivi-
dual hair, number of hair of a particular color, by number of homoge-
neous patches on the surface, etc.

But the variability of colorimetric traits must be taken as most typical
(more or less intense color with regard to some shade).

Let us examine some examples of colorimetric variability. Нето-
globin content of blood measured with the help of a hemoglobinometer
is widely used in the studies of mammals. In the work of John Newson
and D. Chitty (1962) it is shown that the hemoglobin content of the blood
of field voles (Microtus agrestis) is not directly correlated with the size of

Classification of Variability Phenomena 157

the spleen. An examination of the coefficients of variation for this trait
(Table 51) permits one to draw a more exact and _ specific conclusion:
The greater the weight of the spleen, the lesser the variability in quantity
of hemoglobin.

A determination of the variability of hemoglobin content in the blood
of white-footed mice (Peromyscus) depending on whether they are infected
with botflies allows one to notice characters of comparable groups that
escaped the attention of the author of the paper (Table 52).

Only some higher values of hemoglobin and hematocrit in non-infected
animals are given in the paper, but on the basis of data presented in
Table 52 we can see a greater stability of these traits in healthy animals
and a greater variability in infected animals.

Table 51. Variability of hemoglobin quantity in the blood of field
voles (Absolute data taken from Newson and Chitty, 1962)

ee ee eee_*"

Weight of spleen Е
(mg) ХЕ 6. 0. ESe.y, Ех С. 0. Ес.
Less than 100 111.6+2.85 8.8+1.80 — =
100—250 101.5+2.53 8.6-- 1.76 — и
251—630 102.7+2.19 6.7-+1.51 108.6+ 1.65 5.9-+ 1.07

Table 52. Variability of blood indices in infected and non-infected
Peromyscus (Absolute data taken from Sealander, 1961)

Infected Non-infected
Quantity of hemoglobin, in g/100 ml = 12.2+0.2 14.9-+0.1 14.9+0.1 Xs,
7.5 + 1.16 5.2-Е0.47 4.50.47 uv. 1-5¢.y.
Hematocrit, in percentages 39.8 +0.8 46.3+0.4 46.3 0.3 X45,

OOP COG 292025 ое

a es rm eC C=

Good examples of the study of colorimetric variability are the works
of A. У. Pokrovskii, У. 5. Smirnov, and 5. 5. Schwartz (1962) and У. №.
Bol’shakov (1962), which have been devoted to the study of narrow-skulled
(Microtus gregalis), red-backed and bank voles (Clethrionomys тии; and
Clethrionomys glareolus). ‘Two indices were studied in these works : ‘“‘white-
ness” (coefficient of reflection in white light), and “shade index’ (reflec-
tion of red light in percent of whiteness); stable values in the variability
for one or the other trait have been obtained for each of the species. Data
presented in the work of R. A. Boolootian (1954) on variability of percent of
red coloration in the color of three subspecies of kangaroo rats (Dipodomys
nitratoides) also allow one to draw the same conclusion:

158 Variability of Mammals

р. п. nitratoides О.п. exilis О. п. brevinasus В 3 =
я 5 15.8+ 1.37 16.2 1.87 17.7-Е 1.69 —1.25 --6.33 —4.28
By НЕ. 8.7-=0.82 11.5 1.10 9.60.91 —2.03 —0.73 — 1.28

Data on colorimetric variability in different species of mammals can
be found in many other works (Dice, 1938; Sumner, 1923, 1924; Huestis,
1925; Hayne, 1950).

From the limited data presented, it is seen that the study of variability
of coloration should be widely used during the study of the population
morphology of mammals.

Variability of temperature. ‘This refers to the variability of traits of
mammals expressed in units of temperature; for example, preferred tem-
perature. Calculations of data on temperatures preferred by laboratory
mice show comparatively high coefficients of variability—42 to 112%
(Kalkowski, 1966).

Variability of surface area. ‘This is the variability of parameters of area
characterizing an individual system or portions of organs.

In the review of colorimetric variability it was noted that in many
instances color variability is taken as the variability characterizing the
size of the area of an organ or structure.

JOO
KIO
ROOK

Figure 42. Genetic types of mice differing in color (Charles, 1938).

Classification of Variability Phenomena 159

It can easily be shown that mice of different genetic lines (Figure 42;
Charles, 1938) can be clearly defined in quantitative expressions of light
and dark areas on the hide. A second way is to divide all the variants
into more or less homogeneous groups with regard to size of white area
and then calculate the statistical parameters for each homogeneous group
for sex and age.

An example of such an approach to the analysis of data on variability
of colored area can be found in the wo-k of Dice (1938) in which conven-
tional parameters of color of area around the ears of Peromyscus maniculatus
from nine populations in Arizona were studied (Table 53). As seen from
these data, the variability of size of area of colored spots corresponds well
with the absolute area of colored spots.

Table 53.. Variability in the size of patches in the area around the
ears in Peromyscus maniculatus (Absolute data taken from Dice, 1938) *

Population X55, С. VS. Population X55, 6. 0. +Se,y
I 1.36-.0.06 46.3-43.10 VI 0.75-40.12 72.0--11.4
II 1.52 0.08 37.53.75 УП 1.82 --0.06 28.62.35
Ш 1.36--0.06 43.43.13 VIII 2.08-+0.09 14,9. 3.04
1V 1.66-L0.09 46.4-- 3.80 IX 1.86 =0.09 18.33.45
У 1.95+0.05 20.0-+ 1.81

* The animals were divided into four classes with regard to color area—from 0 to 3.
Data on males and females, aged one year.

It should not be thought that the variability of area is concerned only
with the manifestation of variability of color. Size of glandular surfaces
in the intestinal tube, the size of respiratory surfaces in the lungs, the size
of the body surface as a whole or of its parts, are a few of the many traits
which can be expressed in variability of surface area. М. I. Larina (1966)
gives data on the variability of nuclear surface in the cells of the mesentry
in two closely related species of forest mice, which ranges in two popula-
tions of Apodemus tauricus from 24.3 to 25.3% and in A. sylvaticus from 20.8
to 24.1%.

Angular variability. The variability of angles is the variability of the
angle of attachment or detachment of individual elements of organs for
different systems.

In view of the importance of obtaining the greatest number of different
characters for the organs and structures under study, it is not proper to
leave unattended the variability of a number of traits in angular values,
even though it is not yet widely used.

In the work of E. Ashton and C. Oxnard (1964) an attempt was made
to obtain exact statistical parameters for the angles of disposition of

160 Varability of Mammals

muscles and bones in 26 families of higher and lower primates. ‘The primates
were divided into two large functional groups—quadrupeds and _ those
standing erect. Some characters of angular variability calculated from
these data are given in Table 54.

Table 54. Variability of angular traits of the shoulder girdle in primates*
(Absolute data taken from Ashton and Oxnard, 1964)

Trait Quadrupedal Standing erect P
(n=39) (51)

Angle of attachment ас 8.1-1.66 22.70.71 <0.01

т. trapezius .. 128.64 16.7 17.542.2 <0.01
Angle of attachment of caudal part oe 66.9--3.09 62.0 1.07 0.2—0.1 _

т. serratus magnus a 28.6+3.2 96.6+12.1 <0.01

Angle of orientation .. 80.20.90 57.92.32 <0.01

m. serratus, m. trapezius 56 69.7-+7.9 22.4+ 1.80 <0.01

Orientation of glenoid cavity .. 125.5+0.93 115.7-+0.66 <0.01

4.6-£0.52 3.2-+0.40 <0.01

Bend of clavicle ae 3.4+1.99 34.1--2.2 < 0.01

39.4--4.4 36.2+4.0 0.43

i

* For each pair of lines, the upper represents the mean and the lower represents the
coefficient of variation.

Strictly speaking, the data in Table 54 are from a study of inter-popu-
lation variability and not for an individual population (intra-population).
But it can still be noted from the data how angular variability can be used
in morphological investigations. It is interesting to note that with regard
to the absolute value of the angle of attachment of the caudal part of the
serratus magnus, there are no appreciable differences between the erect pri-
mates and the quadrupedal ones; but with regard to the variability of these
traits, the differences are specific.

The variability of the angle of deviation of the upper edge of the ischial
tuberosity from the axis of the ilium was investigated by V. A. Dolgov
(1961) on two species of the genus Sorex, and by M. Pawlik (1967) on other
species of the same genus.

These few examples of the use of variability of angular characters in
the study of mammals, confirm the correctness of separating this pheno-
menon of variability as an independent group. It is possible that with the
development of exact and simple methods of investigation, the study of
variability of angles will attain a great significance in population morpho-
logy of mammals. :

Polymorphic variability. Epigenetic polymorphism is the presence of

Classification of Variability Phenomena 161

certain quite different, discontinuous characters of an organ or its parts
(of any measurements) within a population.

This category of variability first attracted the attention of researchers
many years ago. It is enough to remember the work by Charles Darwin
(1868, 1877) and an interesting and unforgettable monograph by W.
Bateson (1894), which contained impressive factual data on epigenetic
polymorphism of mammals, birds, reptiles, amphibians, and many inverte-
brates. Since this variability has been studied for the last ten to fifteen
years intensively, at present the number of works is increasing tremendously.
This is a part of the study of population variability where long stagnating
facts, viewpoints, and different approaches suddenly broke through the
barrier of obscurity and are now capturing still more fields of investigation
in a wide and forceful current. The success of the new approach perhaps
depends on the fact that this is the place where a profound intermixing
of two sciences—population genetics and morphology—takes lace; pre-
viously these fields were considered separate.

The study of populations by morphological methods allows the re-
searcher to see that differences in the form of skull bones and in other parts
of the skeleton, differences in the variation of growth of individual bones,
etc., are met with in all the species, populations, lines, and groups studied.
Works on polymorphism of other organs, for example the circulatory system,
internal organs, and skin, are close to the works on epigenetic polymor-
phism of the skeleton. We will look into the basic manifestations of epi-
genetic polymorphism in the same order below. It is considered that both
genetic and environmental factors are at the root of epigenetic polymorphism
and that this phenomenon is brought about by “threshold mechanisms
acting on the background of continuous variation’ (Berry, 1963; Berry
and Searle, 1963; Berry, 1964; etc.). In this way, epigenetic polymor-
phism appears to be a type of continuous variability (first type—genetic
polymorphism) and its study is directly related to the study of mechanisms
in micro-evolution (Timofeev-Ressovskii and Svirezhev, 1966).

With the help of enzymatic maceration, R. Berry (1963-1964) obtained
preparations of the skeleton Юг many series of the grey squirrel (Sczurus
carolinensis), guinea pig (Cavia porcellus), lemming (Lemmus lemmus), field vole
(Microtus agrestis), house mice (Mus musculus), black rats (Rattus rattus),
Malayan rats (Rattus exulans), etc. The skeleton (including skull) showed
55 different deviations from the normal’; moreover, not a single skeleton
was found without any deviations. (Some tens of thousands of animals
have been examined in this respect by different authors.) The number
of such deviating traits varies in each individual case from 12 to 39. Some

8 In the present case, the meaning of “normal”’ loses its meaning because there are по
“normals” as such in nature. There is only a more or less usual group of different deviations
from some mean values.

162 Variability of Mammals

of the traits are as follows : curved nasal cavity; presence of interfrontal
bones; additional foramen in the frontal bones; fusion of certain bones of
the skull (nasal, frontal, and others); presence of a number of additional
foramina in the skull, scapula, and limbs; presence of pterygoid processes
and palatal processes on certain skull bones; absence of certain molar teeth;
fusion of specific vertebrae; presence or absence of individual protuberances
and processes on the body of vertebrae; fusion of pelvic bones and poste-
rior extremities; and so forth. |

It has been determined (Cave, 1930; Deol, 1955; Searle, 1954,
1954a; Griineberg, 1950, 1955, 1963; Berry, 1963, 1964; Manville, 1961; .
Stecher, 1961; Freye, 1964) that epigenetic polymorphism in the skeleton
normally is uot affected by age variations and that clear differences in
the frequency of occurrence of different variants are observed between
individual populations and lines. All this makes the study of the pheno-
menon of epigenetic polymorphism a useful method of differentiating
populations (see also Chapter IV).

Some examples of epigenetic polymorphism are given below. J.
Hromada and L. Strnad (1962) observed different frequencies of occurrence
of unfused parietal bones and the presence of bregmatic bones and fora-
mina in the parietal bone in certain species of Macaca. In the work of
R. Manville (1959), the phenomenon of epigenetic polymorphism with
regard to an individual structure (bregmatic bone) in different populations
of lynx (Lynx lynx) has been shown. On the basis of a study of 1,790
skulls of lynx it was possible to characterize the occurrence of the breg-
matic bone of different populations in the following manner (expressed
in percentages) :

Alabama 0 Texas a 7.0
Georgia 0 Nevada ne 14.6
South Dakota 0 Oregon 16.8
New Brunswick 0 West Virginia = 37.5
Nova Scota 0 Missouri nes 44.0
British Columbia 0 Mexico mA 100.0

Puéek (1962) drew attention to the same phenomenon for a number of
other mammals. R. Haitlinger (1962) records a rare instance of an
occurrence of these bones in a population of field voles (Microtus agrestis).

E. L. Green (1941) showed that variations in the vertebral column,
including additional ribs on the border of the thorax and abdomen, and
similarly in the lumbo-sacral region, are met with in “so many species of
mammals that it seems probable that no species of mammals is free of
such variability” (p. 221).

During a morphological study of the white whale (Delphinapterus
leucas), the structure of its manus appeared to be variable. The constant

Classification of Variability Phenomena 163

Figure 43. Basic types of hand structures in the white whale (Delphinapterus leucas).
Identification of individual elements is given in the text.

164 Variability of Mammals

elements found were as follows : ulnare, radiale, intermedia, pisiforme (first
line) 1 to 4, carpalia distalia (second line), and finally, metacarpalia 1 to 5
serving as a basis for the phalanges (Figure 43). This general structure
of the hand is changed, as a rule, either by an addition of elements between
the first and second row, or the second finger and the additional five
elements—carpalia distalia. Still, it appears in a number of cases that
individual elements like the centralia 1 to 3, pisiforme, and certain carpalia
distalia can be absent in the hand. All this leads to the appearance of quite
different hand structures in animals of even one population. Some of these
have been depicted schematically in Figure 43. Calculations carried
out showed different frequencies of occurrence for different numbers of
connections between limbs and carpal elements in the hands of white
whale populations in the Far East :

Number of connections и! 2 3 4. 5
In % to occurrence 2 16 31.1575 116 mee

and similarly, different frequencies of occurrence of elements in the distal
row of carpals in the white whale :

Number of elements in the line He 1 2 2 4
In % of occurrence a 3.3 27.8 57.4 11.5

We were able to show (Yablokov, 19615; Bel’kovich and Yablokov,
1965), that the structural type of the hand appears to be very close
between the mother and her calf, i.e., the above-mentioned variability
is of a genetic character.

Another peculiarity of hand structure in the white whale was even
known to researchers in the last century—the splitting of the fourth digit
(Kiikenthal, 1890, 1909; Kunze, 1912). One substantial peculiarity
in this splitting of the finger in the white whale came to light: In the
seas of northern Europe and the Kara Sea, the splitting of only the fourth
finger was observed, whereas in the Okhotsk Sea, where the phenomenon
was observed by us, the splitting occurred only in the fifth finger (Figure
44). This served as one of the arguments for proving genetical indepen-
dence between these large populations.

These same investigations showed considerable variability in the
form _and structure of the thorax of the white whale, which consisted of a
variable number of elements and a variable number of attached ribs
(Figure 45). It is interesting that an analogous variability (epigenetic
polymorphism) has been observed by A. Schultz (1917) for a group of
orangutans (Figure 46). Variability in the degree of fusion in the pro-
cesses of the sacral vertebrae in two species of shrews has been observed
by V. A. Dolgov (1961). The number of such examples could be
considerably increased, but I will cite only three which I think show
satisfactorily that the study of variability of the epigenetic polymorphism

Classification of Variability Phenomena 165

Figure 44. Different instances of splitting in fingers of hands found in the white
whale (Delphinapterus), North (lower line) and Far East (upper line).

WU VY
МАИИ

Figure 45. Epigenetic polymorphism in the structure of the sternum
of the white whale (Delphinapterus).

Figure 46. Epigenetic polymorphism in the structure of
the orangutan sternum (A. Schultz, 1917).

166 Variability of Mammals

type reveals the specific biological relations which could be lost sight
of if other approaches are used.

In the work of S. Wright (1926) it is shown that the number of
abnormal individuals in guinea pigs with additional fingers on the extre-
mities (polydactyls), appears to be related to the age of parents (Table
54a). It is interesting that the decrease in the number of polydactyl
individuals in the offspring of older animals is observed not only in the
whole population, but also in the four different inbred lines (A, B, C, D)
having different frequencies of polydactyls.

Table 54a. Number of polydactyl individuals (in %) in four lines of
guinea pigs depending on the age of parents (Wright, 1926)

Parent of polydactyl individuals in offspring Whole
Age’of;parents) Saas Te population
(in months) Line A Line B Line C Line D C.0.-ES, 9,
3 29.3 34.6 68.1 81.0 52.741.8
Gry 7.4 28.2 54.4 69.5 40.0 1.7
9 9.6 29 28.9 50.0 29.2 1.7
12 — — = — 26.7+1.7
15 6.1 т 22.0 30.2 18.5: 1.4
21—46 — = — — 14.2 -- 1.4
Average 11.9 Ра. 38.4 55.8 oiled

Another example (Sawin, 1945) also shows how three different lines
of rabbits differ in relative development of additional ribs and vertebrae
(number of individuals are in %) :

With 27 presacral

Line With 13 ribs
vertebrae
III 95.4 95.3
ТУ 0 13.0
У 34.8 39.3

Finally, the last example concerning skeletal structures is taken from
the work of J. Zejda (1960, 1965) on the occurrence of uncommon or
“simple” form of upper third-molar teeth (M3) in 14 populations of Cle-
thrionomys glareolus at different heights above sea level (Table 55; Figure 6).

A specific interdependence is observed: There is an increase in
the number of individuals with the “заре” №3 when there is a decrease in
the height of the place above sea level, though in certain instances

Classification of Variability Phenomena 167

Table 55. Occurrence of “Simple” form of М3 in a population of
Clethrionomys glareolus (Zejda, 1960)

eee ee ens Oe eee eee ee eee ee

Population Te Utne УТУ АЛЕ 126 SOD OTT ИТ ТУ

eee ee TC

Height from sea level 750—1800 400—600 Less than 400
Number of individuals

with “simple” 3

(in %) 413 1419] eS) 1313 24 26 37 47 44 62

eee eee ee ee | ee I eC ee

(populations V and VIII), this dependence is vitiated. In this and the
preceding examples, the relationship of variability of the epigenetic poly-
morphism type, with environmental factors is apparent. This relation-
ship was experimentally proved by Н. Griineberg (1955), who showed diet
dependence for 12 osteological traits in different lines of laboratory mice.

Body color of mammals is one of the most complex traits; it did not
yield to objective quantitative evaluation for a long time. Because of
an increasing interest in the recent past to study the morphology of
populations, and to search for more efficient and stable traits which could
serve as indicators of a population, some successful attempts in this
direction have been made. V. N. Bol’shakov (1962) successfully used the
colorimetric method for studying the color of mice in a genetical study
of populations; [this method was previously used by F. Sumner (1909-
1924), $. М. Gershenzon (1946), and others]. Voipio (1962) used the
chi-square method to study color variations in martens in Scandinavia.
However, the methods of colorimetry have not been used in investigations
on coloration of those mammals like the Greenland seal which have large
color patches on their bodies.

After examining a large number of animals obtained from all three
regions of their habitat in the northern Atlantic, it was possible to
separate six basic types of positions for dark “wings’’ on the side of the
back. These types have been shown schematically in Figure 47.

It is known that the distribution of chi-square depends on the
number of degrees of freedom. Hence, for a reasonably exact comparison,
it is necessary to determine this number accurately as far as possible, for
the character under investigation. In the present case, as a matter of
a working hypothesis, it would be possible to postulate the number of
degrees of freedom (at the most, say 15) keeping in mind that, in addition
to the six types of coloration observed in the populations, some other
variants could be expected (Figure 47, 7-14).

The comparison was carried out only on the males of the population
to obtain as clean a sample as possible, because a preliminary comparison
of coloration in males and females from the Newfoundland showed the
presence of clear sexual dimorphism in this trait. All the calculations

168 Variability of Mammals

and a fuller comparison have been given in the work of A. V. Yablokov
and У. Ya. Etin (1965). In analyzing the material, the frequency dis-
tributions showed that statistically significant differences were found among
all populations:

85.9 (P<0.01)

| ЖЕ rm
“ia — 1519 (6)
| хз — 2926 0.
a) 0 9 The comparison shows
, - that the largest differences in

distribution of coloration types

are found between seals from

the Yan-Maien region and the

White Sea; the least differences

occur between seals from the

Newfoundland region and the
4 5 6

Yan-Maien region.

|

Unlike the chi-square

method, an aggregate com-
parison using the criterion
“lambda” of A. М. Kolmo-
gorov and N. A. Smirnov,
7 8 9 10

does not require a determi-

| nation of degrees of freedom
(Plochinskii, 1960).
6 F A comparison of the same
material on coloration of
seals, carried out by the
11 12 13 14

“lambda”? method, confirms

Figure 47. Scheme of different types of hide color the results obtained by ПЕ
in adult males of the Greenland seal (Pagophilus): the chi-square method. The
1-6—found in nature; 7-14—some possibilities. differences between the distri-

bution of color types in seals
of the White Sea and Yan-Maien regions appear to be greatest (Ay,3=
1.313), and least between seals from the Newfoundland and Yan-Maien
regions (Aj,3;=0.524). It is interesting to note that the magnitude of
differences in the distribution of color estimated by the “lambda” method
does not in a single case reach the level of significance.

The methodical importance of such a difference in the significance
of the results lies, perhaps, in the reduced number of degrees of freedom
when using the chi-square method. The number of different variants of
coloration for which the differences between the populations of the New-
foundland and Yan-Maien regions could have been taken as non-significant

)\

Classification of Variability Phenomena 169

with the given distribution, by using the chi-square method, should be
close to 50 in number. We found only six variants in the natural popula-
tion. It would be interesting to confirm through special observation the
assumption that many more different types of coloration exist in the
Greenland seal in nature than are presently known.®

It is interesting to note that a recently conducted study of polymor-
phism in blood proteins for three populations of the Greenland seal
(Naevdal and Gunnar, 1966), completely confirms the conclusions about
the degree of independence of individual groups drawn previously on the
basis of quantitative (Chapter III) and qualitative (color) variability.

It is considered that the use of methods of the epigenetic polymorphism
type for quantitative evaluations of coloration in mammals should find a
larger usage in zoological investigations because these are incomparably
clearer, throwing light on the exact similarities and differences between
the populations under study.

In the work of A. I. Argiropula, data on the variability of chest spots
in two populations of yellow-necked mice (Apodemus flavicollis) in Armenia
are given. While comparing several variabilities, the author could only
make the observation that “those traits variable for mice of one popula-
tion are similarly variable for mice of another” (р. 206), i.e., the author
indicated only the presence of variability. The analysis of the same
data carried out by us using the chi-square method (the “lambda” method
appeared inappropriate here because of the small number of samples),
shows highly significant differences between the populations investigated
with regard to the given trait; this allows one to draw more specific
conclusions about its taxonomic and ecological importance. У. М.
Pavlinin (1963) could have drawn much more significant conclusions had
he resorted to analogical analysis while characterizing the coloration of
sables and wild martens in the Sverdlovsk and Tyumen regions. Similarly,
G. Markov (1957) who studied characteristics of tooth pigments while
characterizing different populations of the common shrew, and G. Corbet
(1963) who studied features of coloration at the tip of the tail while
characterizing small mammals of England, could have drawn more
significant conclusions had they resorted to similar analysis. Further,
the data on different coloration of cetaceans could have been much more
significant, had it been collected according to the above plan. There is
no doubt that the coloration of cetaceans is a very favorable object for
investigations of similar species. Many variations in the coloration of
species like Orcinus orca (Figure 48), or the common dolphin (Delphinus
delphis, Figure 49), expressed quantitatively and thus available for

® This proposal is well supported in recent years through new color-type findings
(R. Sh. Khuzin, personal communication).

170 Variability of Mammals

comparison, could answer many yet unanswered questions in ecology on the
inter-relationships of different populations.

The structure of the circulatory
system in mammals is another large
section of variability where epigene-
tic polymorphism is widely mani-
fested. Inthe case of the skeleton,
anatomists had to deviate from the
understanding of the term “‘norms,”’
and similarly in the case of the
circulatory system, because these
“norms” do not exist in nature.
There are many variations which
can be grouped into a given classi-
fication and calculated and com-
pared with the help of simple
mathematical methods. Certain
examples from the structure of the
circulatory system are shown in
Table 56. Two other examples,
taken from the works of Tu Chin
(1963) and P. Psenicka and J. Jurin
(1964), are shown respectively in
Figures 50 and 51.

An interesting example of epige-
netic polymorphism in the structure
of the circulatory system of foxes
(Vulpes vulpes) has been given in
the work of W. Cezariusz (1967).
It was observed that in _ foxes
Figure 48. Different variations of coloration brought up ae Сару (platinum-
in the body of Orcinus orca (Yablokov, 1963). colored and silver-black), there is a

great asymmetry in the development
of the circulatory system of the yeah The possibility of a genetical analysis’
of variation in the structure of the circulatory system has been shown in
this example.

Examples of epigenetic polymorphism in the numbers of different
mammalian species are increasing day by day. It has to be admitted now
that: our understanding of the exceptional structural stability and
sturdiness of chromosomal apparatus is erroneous. ‘The intra-populational
differences in the chromosomal apparatus relate to different numbers of
acro- and meta-centric chromosomes [for example, in three populations of
Peromyscus (Sparkes and Arakaki, 1967)], with regard to form of individual

Classification of Variability Phenomena 171

Figure 49. Epigenetic polymorphism in the coloration of the common dolphin (Delphinus
delphis ponticus) according to Barabash-Nikiforov; from Kleinenberg (1956).

Table 56. Variability of blood supply for different parts of the human heart
(From right or left coronary artery; п=82) (James, 1960) (in %)

Right Left I И Ill IV У УТ
100 0 27 9 5 5 — =
90 10 39 18 37 4 — =
80 20 19 38 34 2 = =
70 30 4 17 17 5 — —
60 40 4 14 4 1] — ==
50 50 4 4. 2 4 — 6
40 60 1 — 1 20 — 9
30 70 1 — — 13 — 17
20 80 1 — — 12 1 30
10 90 — — —- 11 4 23
0 100 — — — 13 95 15

Е ри ри Oe en OO р ee ee

I—Posterior right ventricle; [I—Anterior right ventricle; I[1I—Free wall of right ventricle;
IV—Posterior left ventricle; \У—Егее wall of left ventricle; VI—Anterior left ventricle.

172 Variability of Mammals

Figure 50. Different types of branching in the femoral
artery of Macaca (Tu Chin, 1963):
а. femoralis; P—a. poplitea; S—a. saphena; g—a.
articularis genu suprema; mc—a. musculocutaneous distalis ;
р— а. pergorans distalis.

eYg

Figure 51. Epigenetic polymor- 7.2% 6% 795 4%
phism in the structure ‘of the
circle of Willis (circulus arteriosus)

in Macaca mulatta (PSenicka and
Jurin, 1964),

4% 4% 2% 1%

Classification of Variability Phenomena 173

chromosomes [data on laboratory mice (Bianchi and Molina, 1966)], and
so forth. The study of variability of karyotypes is especially interesting
because it can correlate studies on ecology and epigenetic polymorphism.

Some additional examples of epigenetic polymorphism, showing what
a great scope for studying variability exists in the approach, are yet to
be reviewed by us. In the commercial whale (Delphinapterus leucas) in
the northern seas, I studied the structure of the шегих masculinus. It was
observed that the degree of development in this organ can vary over a
wide range (Figure 52).

E

Figure 52. Different varieties of development of the male uterus
(uterus masculinus) of the white whale (Delphinapterus leucas).

Another example is the frequency of the so-called “погиза?” con-
figuration and possible deviations in the structure of the musculature in
the human hand (Basu and Hasary, 1960) (Table 57).

It is interesting that in this example symmetrical body structures are
investigated and there was more or less direct evidence of clear differences
in the frequencies of different morphs on different sides of the body.

Broadly speaking, the phenomenon of epigenetic polymorphism appears
to be more closely related to other manifestations of variability than it is
supposed to be. As seen from the work of E. Ford and H. Bull (1926),
who studied nearly 7,000 herring, deviation from the typical structure

174 Variability of Mammals

Table 57. Frequency of occurrence of different variations in the configuration
of human hand muscles (n=72) (Basu and Hasary, 1960)

Type of Right Left Type of Right Left
configuration hand hand configuration hand hand
Norm 15 15 У 2 0
I 7 Э УТ 1 0
II 1 1 УП 4 4
ш 10 5 УШ 2 a
IV 3 6

of vertebrae (1.е., the phenomenon of epigenetic polymorphism according
to modern classifications), occurs at greatly elevated frequencies in
individuals with extreme numbers of vertebrae (Table 58).

Table 58. Epigenetic polymorphism in the construction of vertebrae
in herring (After Ford and Bull, 1926; from Fisher, 1930)

— i,

a —_ж_—

53 54 55 56 57 58
Frequency of morph (%) es 0.08 1.06 28.36 61.3 8.91 0.29
Frequency of deviations ‘
from “normal” (%) 5 4.1 1.2 1.1 2.6 10.0

—_ eC CC —

In the detailed work of G. H. W. Stein (1963) оп epigenetic poly-
morphism in the structure of the dental system in different populations
of European moles, certain data appear which allow one to assume that
the regularity observed by Fisher occurs in mammals also. It is necessary
to collect additional accurate data because the solution to this problem
will advance our knowledge about the peculiar phenomenon of epigenetic
polymorphism.

A deeper study of epigenetic polymorphism is important in population
genetics because genetical criteria, which are usually inadequate in mor-
phology, can be used from this point of view. Epigenetic polymorphism
permits the study not only of quantitative but of qualitative differences.
This is one of the most important aspects of this type of variability.

Forms of Variability
Variability may manifest itself in different forms in relation to evo-

lutionary factors. The following forms of variability have been examined
here : age, sex, chronological, biotopic, traumatic and teratological.

Classification of Variability Phenomena 175

Variability of age

Within population age variability is considered to consist of: (1)
the variability (of any dimension) of individuals in a similar age group;
and (2) the differences in variability between different age groups.

The object of embryology is to study age changes in ontogenesis.
However, attention has only very recently been paid to the existence of
age variability in mammals, and I do not know if there are published
works on this subject. This situation deepens interest in studying age
variability as a separate form which undoubtedly plays a great role in
the life of a developing population. I will cite certain factual data in
order to make possible a review of this question (Table 59).

Table 59. Variability (с. v.-:5;¢,),) of body length of the Greenland seal
Pagophilus groenlandicus (White Sea, 1961)

Sex Sexually mature New-born t P
Males ae 4.34--0.31 6.31-=0.80 —2.29 0.05
Females ate 4,44 --0.22 7.34-+0.83 — 3.02 0.01

There is a significant difference between the variability of the trait
in question for newly-born and mature animals; moreover, the significance
increases because of quite similar behavior in the coefficients of variation
for the males and females.

Interesting data on the body length of fur seal embryos (Callorhinus
ursinus) of different ages is given in the work of V. B. Scheffer (1962).
The above-mentioned tendency toward decrease in variability with age, is
clearly seen in Scheffer’s data (see Figure 1).

The variability of tail length in female mice living in a hot room
is not large, but it changes from the age of 1.5 months to 7 months (Sumner,
1909). The length of tail in females become a less variable trait with
age in both the hot and the cold rooms (Figure 53). The differences
between the final values of the coefficients of variation are quite significant.

There is no elimination of deviation by mortality in these experiments,
and when the variability decreases it is clear that we observe the natural
picture existing in the population. In other words, because of the irreg-
ular rhythm of tail growth in the young of the populations, a considerable
difference is observed in animals aged 1.5 months relative to the adult
stage, when a stability of the trait is reached in the population. All these
changes take place on the background of tail-size development in both
the hot and cold rooms. This situation—the reversible dependence of the
coefficient of variation on the size of the trait—is governed by primary
regularities (Chapter V).

176 Variability of Mammals

The data obtained as a result of analyzing the factual material of
E. T. Popova (1941) on weight of sheep, show the importance of calculat-
ing a specific age-variability.
C.V.% It was observed that in Ба
mm bred flocks of sheep (both males
and females) there was a regular
decrease in variability with age
(from 17.2 to 10.8%), whereas
in pure-bred flocks such a dec-
rease was not observed. The
differences’ in individual popu-
lations (perhaps inbred to a
great extent) in age-specific
variability are significant. It is
clear that any practical or theo-
retical investigation of these
flocks will not be complete if
these aspects of variability are
not taken into account.
It can be seen from the

Figure 53. Fluctuations in the magnitude of qatg already presented (see
variability of tail length for mice of different ages Table 46) that the variability

(the same population) in hot and cold rooms: )
a—mean value of absolute length of tail (in mm) of the volume of the skull in
(F. Sumner, 1909). one species of Macaca increases

significantly with age, that the
number of polydactyl offspring of guinea pigs increases with an increase
-in age of parents (Table 54), and that the variability in a number of
traits of yellow-necked mice (Apodemus flavicollis) and common shrews
(Sorex araneus) in one population decreases toward autumn—the season
of general growth in the animal (see Table 41). According to Schwartz
(1962), foot length and tail length in hibernating shrews are less variable
than in the young; variability of body weight in four lines of rabbits
decreased with age for both males and females (Castle, 1931); these data
are confirmed by material obtained from nature by А. I. Kryl’tsov’a (1957)
on lemmings (Lagurus lagurus) (Table 37).

But there are other data which show that decrease in variability
of traits with age is not a uniform occurrence. According to Schwartz
(1962), mentioned above, only two traits out of five in the common shrew
show decrease in variability with age; Sumner ((1909-1924) did not observe
a clear decrease in the variability of mice in experimental populations;
and according to the data published by Roos and Shackelford (1955), the
variability of weight in chinchillas (Chinchilla laniger) sharply decreases
with age in males but remains at the previous level in females.

Tail lengt

Days

Classification of Variability Phenomena 177

The work of McKeever and Quentin (1963) is interesting; here data
show that the width of the zona glomerulosa in adrenals of mongoose
(Herpestes sp.) in mature animals is half that in the young, the width of

Weight of glands
Body weight

С.\.%

Ш IV Vv

Age groups

Figure 54. Coefficients of variation for relative weights of parotid glands
in the common shrew (Sorex araneus L.) in different age groups:
a—Coefficient of variation; b—relative weight of the gland (Buchalczyk, 1961).

the reticular zone is less variable in the young, and the zona fasciculata
takes an intermediate place in variability. In Figure 54, data showing
the presence of age variability in the weight of the parotid glands are

(c.v)?

0 28 60
Age in days

Figure 55. Change with age in the

coefficient of variation of body weight

in two inbred lines of mice and
F, hybrids (Chai, 1957).

presented (Buchalczyk, 1961), and in
Figure 8 data on the age variability for
number of glial and nerve cells in the
brain cortex of rats are given (Вешк,
1964). It is interesting to note that the
relationship of variability to age appears
to be very complex in the experiments of
King (1923), who studied a population
of rats for more than two years, and
also in the recently published work of
Gebczynska (1964) on age variability
for five different traits in field voles
(Microtus agrestis). A fluctuation in the
variability of body measurements (on
the background of an absolute reduc-
tion in body length and an increase in
tail length in very old males was
observed by N. A. Ovchinikova (1966)
for Microtus oeconomus.

А well-known ‘‘model example”
of variability with age, which 15

178 Vartability of Mammals

known to geneticists and zoologists, should be given here. I refer to the
results of the investigations of Chai (1957) on the variability of body weight
in two inbred lines of mice and their hybrids. Variability was highest in
all populations at birth; at four weeks the variability began to decline,
until at 60 days of age it approached a single value for all three populations
(Figure 55). According to К. С. Lewontin (1957) the dependence of
magnitude of variability on age was “elegantly demonstrated” in this
experiment. As can be seen from the examples given above, however,
this dependence does not seem to be unquestionably constant.

In conclusion, it may be said that not only are all types of quantita-
tive variability subject to age variation, but there are also some indirect
relations of qualitative variability (for example, of epigenetic polymor-
phism) with the age of animals!®, From this another important con-
clusion follows: While studying а _ population by morphological
methods, it is always necessary to separate age variability so as not to
disturb the actual situation of the structure under study with regard to
evolutionary factors.

Sex variability

Sex variability considers: (1) the variability (for any dimension)
of individuals of the same sex within a population; and (2) the presence
of differences in variability between the sexes within a population. .

The existence of sexual dimorphism in mammals is well-known.
However, as in the case of age variability, for a long time researchers
paid no attention to the presence of sexual dimorphism in variability
itself or, in other words, to the existence of sex variability. This type
of variability comes to notice rarely, but it occurs more or less frequently
in nature. Although there are many cases where the variability of traits
in males and females does not differ, there are a number of instances
where sex differences in the magnitude of variability are manifested clearly.

The following examples of sex variability (Table 59a) have been
selected from a number of comparable examples obtained during a study
of pinnipeds.

The differences in the variability between males and females are
clear-cut and indicate the force of different controlling factors in different
sex groups. In fact, biologically, the degree of variability of organs like
weight of lungs and weight of liver (Table 59a) cannot be the same.
Direct observation of seasonal changes in the weights of these organs has
shown a sharp development of the liver during pregnancy, which is caused,

10 The determination of components of biological time, which generally do not coincide
with the concept of physical time, is a promising approach ‘for investigations in the field of
age variability (Sikhlyan, 1960).

Classification of Variability Phenomena 179

Table 59a. Sex variability of traits for certain species of pinnipeds*

Males Females
GUAT tS if Е И ПН ВОЕН Зо р
6. Ut Se.y, ЕЕК
Hooded seal
( Cystophora cristata)

Sternal ribs 3.60.34 4,.9+0.55 —2.08 >0.05
Body length 9.2--0.55 7.2-Е0.53 +2.74 >0.01
Ringed seal (Okhotsk Sea)

(Pusa hispida)

Body length of animals,

12 years old 5.00.72 1.40.20 -- 4.84 ‚>0.01
Number of lip vibrissae 6.2 +0.59 4.4+0.42 +2.39 >0.05

Greenland seal
(Pagophilus groenlandicus )

Weight of lungs 17.22.14 35.7--7.61 — 2,34 >0.05
Weight of liver 41.9+5.32 15.64:3.47 +4,14 >0.01
Number of nasal vibrissae . 31.4+2.94 22.82.03 --2.41 >> 0.05

* The variability of only adult males and females obtained during the period of one
season (2 to 3 months) has been compared, in order to avoid the effects of age, bio-
topical, and chronological variabilities.

perhaps, by general physiological changes in the body of females. It
is clear that the control of the weight of the liver among females is rather
stricter than among males. It is also clear that for normal functioning
of the female organism, the intensity of liver function (to the extent that
this can be judged from the weight of the organ) has greater role to play
than in the same season for males. At the same time, the development
of lungs (again, to the extent that this can be judged from the weight of
the organ) is under greater control in males, who are compelled, perhaps,
to hold their breath deeply for a long time when they go into deep waters
after food. This statement could be confirmed by carrying out a com-
parative ecological analysis of feeding and other behavior in females during
the period of caring for the young. I am sure that during this life period
of the Greenland seal, there will be a specific difference in the feeding
spectrum for males and females.

Considering the presence of different degrees of variability for nasal
vibrissae, a trait hardly affected by significant seasonal variations, some
more stable (and not merely seasonal) differences in the feeding and
behavior of male and female Greenland seals can be assumed.

Because of the methods proposed for determining exactly the age
of large mammals with long lives (Klevezal’ and Kleinenberg, 1967),

180 Variability of Mammals

population studies of such animals have become relatively easy. Now,
by obtaining a large sample of a population at one time and by determin-
ing the age of each animal to within half an year, the sample can be accurately
divided by age. This means that features of development in a number of
year-classes can be studied from the material collected in one season.
This method was used оу us (Potelov and Yablokov, 1966) during the
analysis of data on the hooded seal (Cystophora cristata). ‘The data were
collected mainly during the two years of 1962 and 1963, but calculation of
age allowed us to compare variability for animals of several ages (Table 60).

Table 60. Sex variability of body length of the hooded seal ( Cystophora
cristata) (Data of Potelov and Yablokov, 1966)

Males Females
INGO ee ee ee SS

; 6. VAS, y п 6. VSS y п ‘Зо fe
New-borns ти 9.50.45 DD 7.10.35 207 +4.21 50.001
5 years ae 2.1+0.30 20 4.4--0.43 53 —4.60 >0.001
6 years ve 6.1--0.80 27 8.50.80 41 —2.12 >0.05
7 years a 13.82.50 15 6.5--0.80 34 42.78 >0.05
8-9 years aes 8.5+0.40 26 6.0-L0.80 28 +2.77 + >0.01
10-12 years AY, 7.00.80 37 5.7+0.60 4] +1.30 >0.2
13-16 years oe 3.30.70 12 5.2 0.80 23 —1.79 >0.08
17 years and older .. 7.2 +£0.70 48 5.60.80 24 1.15 >0.14

Sexual dimorphism in the trait studied was observed in all measure-
ments; moreover, statistically significant results were obtained for the
majority of the age groups.

Data on sexual dimorphism in variability can be found in many
works. A few of the data are presented in Table 61.

Attention has not been given to sex variability in any of the works
from which data have been taken on the different magnitudes of variability
for males and females. This is the case, perhaps, with all zoological
works and is a serious drawback to using data already obtained as well as
planning experiments.

Data are presented on the weight of the adrenals in male and female
palm squirrels (Funambulus) during different periods of sexual activity
by K. Purohit, 1965. The author concluded that there is a considerable
decrease in the weight of the adrenals in inactive animals. ‘The result
can be strengthened by the nature of the variability. The variability of
the adrenals for both males and females is quite low (a range of 16.3 to
17.7 for males, and 10.2 to 24.4 for females), and is also low in inactive
animals of both sexes.

f
Classification of Variability Phenomena 181

Table 61. Sex variability in certain traits of mammals*

Males Females
Е НЕ ЗОВИ ВЫ РЕ I et ‘Зо р
As НЕО 6. UES, y
Forest mice Tail length 31.54+2.25 22.94+1.66 +3.05 0.001
(Apodemus Height of skull 3.70.37 14+0.17 +5.46 >0,001
flavicollis) = Пуачета 4.60.43 7.840.75 —3.82 >0.001
Sable Width ofskullat zygoma 5.4+0.74 3.4+0.53 42.26 >0.050
(Martes zibellina)
Mink Volume of stomach 6.10.82 12.94+1.86 —3.33 >0.001
(Mustela vison)
Macaque ‘ Volume of skull 15.741.96 31,642.42 ' —5.11 >0.001

(Macaca rhesus)

* All calculations have been carried out from original data presented in the works of
К. | Adamczewska (1959), В. A. Kuznetsov (1944), Е. A. Sokolov (1941), and J.
Hromada and L. Strnad (1962).

Calculations carried out on the data published in the work of I.
Bazan (1956) show that there is sex variability in body weight in different
months in Neomys fodiens even though there is a complete absence of sexual
dimorphism in the absolute measurements of the body. Looking at the
absence of sexual dimorphism in the absolute measurements of the traits,
A. Dehnel (1949) rejected the presence of this type of variability in this
species. Dehnel carried out an analysis of his extensive and valuable
data on population morphology for certain species of insectivores, but
without taking into consideration their sexes. Now, in the light of the
data obtained here, the mistake in such an approach is clear. Moreover,
it can be definitely said that there is now a possibility of revealing some
fine points in the biology of insectivores on the basis of the data of St.
Borowski and A. Dehnel (1953). Skoczen (1962) presents data on a
peculiar sexual dimorphism in the variability of behavior, manifested
in the fact that the sex composition of groups of dead moles changed
according to the time of day.

On the whole, it is important to note that sex variability occurs in
quite different traits, for example in body measurements, number of ribs,
vibrissae, skull measurements, weight of internal organs and behavior.
All this shows how widespread this form of variability is in mammals and
points to the importance of detailed investigations.

Chronological variability
Chronological variability is: (1) the variability (of any dimension)
of individuals of a group within a population in a given period of time

182 Variability of Mammals

(day, month, date, season, year, number of years, etc.); and (2) the
differences in the variability of characteristics of a group in one рорша-.
tion for different days, months, seasons, years, etc.

The concept of chronological variability was named first by S. S.
Schwartz (1963), who included in it only the variability of seasons which
he compared with generational variability. Such a narrow concept of
chronological variability is not proper. Seasonal variability is without
doubt only one of the forms of variability in traits of a population during
a particular time, and generational variability cannot be completely
chronological.

While studying mammalian populations, many researchers have
examined seasonal changes. The attention of zoologists was drawn to-
ward seasonal variation in the morphology of a population after the
_appearance of a series of works by Polish investigators (Dehnel, Petrusevich,
Pucek, and others) showing that apparently stable traits like form and
size of the braincase and weight of the brain are also affected by seasonal
variation. But even after these investigations, particular attention of
research workers was not attracted toward the phenomenon of seasonal
variability. The existence of groups in a population differing in varia-
bility for a single trait at different periods of time, can play an important
role in the life of the population.

As the first example, we will present some data on variability of body
: parameters for forest mice (Apfodemus flavicollis) (Table 62).

Table 62. Variability of certain traits of forest mice (Apfodemus
flavicollis) (Calculated from the data of Adamczewska, 1959)

Tail length

Males ie 31.5-Е2.25 11.40.91 +8.27 0.001
Females a 23.0-+ 1.66 10.90.87 +5.98 0.001
Body weight
Males ae 24.0+1.65 15.9-4 1.24 --3.92 0.001
Females Ne 24.6-- 1.71 15.1-- 1.20 +453 0.001

The high degree of differences in the variability of the same traits
between August and November attracts attention in the data presented.
It is also interesting that the variability of all traits appears to be consi-
derably less during autumn. This conclusion is confirmed, in general,
by other data obtained from the work mentioned in Figure 56 and from
the data of W. Serafinski (1965).

Classification of Variability Phenomena 183

C.v.%

=

vi Vit УИ 1х x Xl
Months

Figure 56. Seasonal variability of measurements of forest mice
(Apodemus flavicollis) (Calculated from the data of Adamczewska, 1959):
1—Tail length; 2—Body weight; 3—Body length; 4—Ear length; 5—Foot length.

An analysis of data on variability of body weight for field voles
(Microtus agrestis) in different seasons (Newson and Chitty, 1962) shows
significant specific differences in the variability of this trait between different
years (Table 63).

' Table 63. Variability of body weight for adult males of
the field vole (Microtus agrestis) in different years
(Analyzed from the data of Newson and Chitty, 1962)

OC

September, 1956 September, 1957

6.2 40.95 9.2+1.01 —2.14 >0.05
December, 1956 December, 1957
5.3-40.63 8.4-40.90 —2.79 >0.01

While studying any type of variability, it is important to obtain a
homogeneous sample so that the effect of unaccounted for factors is
reduced to a minimum. This common requirement fully applies also to
the study of the phenomenon of chronological variability. A good example
in this respect is of the data on time of sexual maturity in Lagurus lagurus
under laboratory conditions (Pokrovskii, 1962).

184 Variability of Mammals

6. 0-Е 5 t le
January .. 28.644.22
2.90 >0.01
May .. 14.42.47
2.61 > 0.01
October о SONS TT.

A number of examples of chronological variability are given in the
sections that follow. The most significant among them are those obtained
from the data of E. Z. Kogteva (1963) concerning the length and thick-
ness of hair of such ecologically different species as the white hare (Lepus
tumidus) and the European mole (Talpa europaea). Different living condi-
tions and different approaches to sampling, led to very different behavior
in the variability of parameters of hair according to seasons.

There are large quantities of data on seasonal variability but data
on yearly variability are much scarce; they are still scarcer for variability
over a cycle of years or for a century. Of these few works, the one by
K. Ya. Fateev (1962) devoted to studying the same population of moles
for three continuous years should be presented (Table 64).

Table 64. Year-to-year variability (c.v.15,,,) for certain
traits of adult males of moles ( Talpa europaea)

—

Weight 1958 1959 1960 ов
Body 16112.07 > 1331193 8511.23 1116 1316 eg
Heart ‚. 3955.28 2081200 17312490 30 3 а
Liver .. 28.44-3.79 18.71.80 13.71198 1231 13955

It is seen from the parameters presented that there is clear yearly
variability. АШ the traits in-the year 1958 are more variable than in
the year 1959 or in the year 1960; furthermore, these differences are quite
large.

While studying chronological variability in mammalian populations,
the continuity of life for individuals of the species under study has to be
taken into consideration. ‘The mean natural lifetime of the small rodents
and insectivores is a few months to, at the most, a year, so when we
investigate a population at half-yearly or yearly intervals, we investigate
entirely different animals belonging mainly to different generations.
Each such generation may have individual characteristics in structure and
reflect the specific influences under which it developed and which continue
to exert influence at the time of investigation.

Classification of Variability Phenomena 185

The situation changes when studying long-lived mammals, such as
large ungulates, pinnipeds, cetaceans or large carnivores. Sexually and
physically mature members of a population may comprise individuals of
several generations (in cetaceans, up to ten generations). Such a situation
significantly changes the reaction of the total population to the influences
of environmental factors (Yablokov, 1961с). The stock of individual
ethological information which has accumulated for such a complex
population seems to be quite large, for “foreseeing’’ a number of influences
and accordingly, adapting behavior in case of a change in some external
factor, without altering the morpho-physiological structure, as happens
in mammalian populations of “‘ephemera”’.

It is known that an organism is most affected by various influences
in its early development. Because of the fact that, depending upon age,
there are several cohorts in a complex population (each developing in
different years and accordingly in different conditions), a great pheno-
typic differentiation among individuals is but natural in such a many-
aged population. In order to test the statement, it is necessary on the
one hand to compare variability of the same trait in different species of
long-lived mammals with ephemera, and on the other hand to compare
variability of different cohorts of large animals. The first type of com-
parison (Table 65) shows that the variability of homogeneous traits is
quite comparable for sables (Martes zibellina), beavers (Castor fiber), seals,
white whales (Delphinapterus leucas), and small rodents (Mus, Sicesta, Apodemus).

The second type of comparison gives interesting results, due possibly
to the exact determination of age in large mammals (Laws, 1956; Chapsku,
1952; Klevezal’ and Kleinenberg, 1967) :

Variability of Variability of
Age body length of Age body length of
ringed seal* ringed seal
(с. о. SE Cnn) (c. о. Е си.)
10 years = 12 years
(year of birth, 1952) 3.4+0.46 (year of birth, 1950) 1.40.20
11 years All ages
(year of birth, 1951) 3.8+0.53 (7 years and older) 6.2 +0.77

*From data of Fedoseev, Okhotsk Sea. 1964, females.

It can be seen that the variability of seals born in the same year is
considerably less than for animals of all ages. Besides this, the varia-
bility of animals sharply fluctuates from year to year; the cohorts born
in 1950 and 1951 are close to each other in variability of body length
(3.4 and 3.8%), but the 1949 cohort is considerably less variable in this trait.

186 Variability of Mammals

Table 65. Comparison of variability (с. v.+5,,) of certain
mammals with various longevities

—_ дж

Condylo- Body Body
Species basal length weight Author
length

ымЫМ—

House mouse (Mus musculus) 2.60.64 9.5 11.86 23.3--5.22 Dinovskii, 1963

Birch mice (55а betulina) 2.30.38 6.6 0.38 — Kubik, 1953
Forest mice (Apodemus Adamcezewska, 1959;
flavicollis) 4,140.63 5.7 +0.43 15.9+-1.24 Ursin, 1956
Sable (Martes zibellina) 2.3+0.34 2.81--0.63 — Kuznetsov, 1941
Greenland seal (Pagophilus
groenlandicus) 3.9+0.39 4.3 +0.31 9.0-+-2.01 Own observations
Ringed seal (Pusa hispida) 2.3+-0:49 6.2 0.7 — Own observations
White whale (Delphinapterus
leucas) — 3.99--0.60 22.4+2.45 Own observations

And so, two different approaches to solving the same problem give
quite different results. The comparison of variability for long-lived
animals and ephemera shows a great similarity in the variability, and the
analysis of individual cohorts of population for long-lived species shows
that each generation is considerably less variable than the population
as a whole. However, in studying the populations of ephemera, only
one cohort was compared which effectively constituted the population.
Could it be said on the basis of these data that small animals are more
variable than large ones? Of course, the data presented are insufficient
to answer this question, but it can be assumed that large animals are
more “protected’”’ during individual development, and are less affected
by the direct influence of those environmental factors which easily change
the structure of the individual in the populations of smaller animals (the
factor of obtaining food, or the stability of the ecological niche in a broad
sense). It is also possible that the comparison made was not sufficiently
“clear” that way; there are always certain individuals born in the begin-
ning, in the middle and at the end of one breeding period in a population
of ephemera.

Returning to the analysis given by the example of the ringed seal
(Pusa hispida) as № the fact of different variability among individual
cohorts of large animals, it can be concluded that the data on the varia-
bility of different cohorts of the hooded seal, like those presented previously
(see Table 60), confirm the opinion of I. I. Sokolov and V. L. Rashek
(1961) that variability in animals of the same age, born in the same year,
and consequently developing in identical conditions, is less than in ani-
mals of the same age brought up in different years. There are almost

Classification of Variability Phenomena 187

no examples of chronological variability in which the variability of the
same populations, or group of populations, has been studied for long
periods of time.

Among the few available examples’! are the work of Elga and Hall
(1962) on Pleistocene and recent populations of bats of the genus Tadarida;
the data of Simpson (1948) on the variability of certain traits of fossil
horses; the data of Hysing-Dahl (1959) on the variability of recent and
Pleistocene populations of otters (Lutra lutra) in Scandinavia; and the
works of Derums (1964) on the variability of human skull traits in south
Russia one to three thousand years ago and in the 15th to 18th centuries.
In summarizing such data, the opinion of Simpson is that the variability
of fossil forms is very close in HEEL to that of present-day forms; this
can be fully confirmed.

The relatively stable variability m the evolutionary development of
groups for quite a considerable period (Simpson’s example considers a
period of millions of years) allows one to pose a still broader question
about the meaning of chronological variability. Perhaps a number of
phenomena concerned with this concept exist which we cannot examine
separately. With our knowledge deepening in the field of ecology and
the genetical structure of a population, modern concepts of chronological
variability should be separated into some individual components. It is
possible that the recently discovered properties of endogenous periodicity
in the functioning of organisms (the presence of a “biological clock’’)
(Byunning, 1964), will play an important role in this separation. It is
usually thought that the phenomenon of “biological clocks” should be con-
sidered only in ecological, physiological, and biochemical investigations.
However, it would also be completely incorrect not to consider this
phenomenon during the study of morphological variability. It is very
possible that we would be able to carry out exact comparisons between
two samples by considering this ‘biological’? time and remove conflicts
between data obtained in apparently identical conditions. This observa-
tion holds good particularly for morphological material obtained during
the study of experimental populations “‘under similar living conditions.”’
It is possible that these other “‘similar living conditions” are not, in fact,
“similar.” Shnoll’ (1964) well observed in this regard: “А doctor should
be accustomed to the thought that the same therapeutic preparations
give different effects depending upon the time of the day they are ad-
ministered....He should always keep in mind that the results of clinical
analysis also depend upon this time factor. Agronomists, veterinarians,

11 In the work of Paaver (1965), there is an excellent analysis of epochal changes and
variability of many south-Russian mammals. Unfortunately this monograph was pub-
lished after the final touches were given to the present book and I was deprived of the
opportunity of using the interesting factual data presented in it.

188 Variability of Mammals

bee-keepers, physiologists and biochemists should also decide their own
approach toward the ‘constant’ conditions—continuous lighting and
constant temperature are also not normal conditions. In conformity
with the concept of endogenous periodicity, during a study on the effect
of some factor on an organism the approach to the principle of ‘other
similar conditions’ must be very cautious. The conflict in the ‘other
conditions may put the ‘control’ and ‘experimental’ into a quite unrealistic
comparison”’ (р. 9).

Hence, while studying mammalian populations, we should not forget
about a number of phenomena based on reversible and irreversible pro-
cesses differing in different periods, going on in each organism, and in a
population as a whole, which are characterized not only by changes in
morphological structures but also by changes in the magnitude of
variability.

Birth-order variability

Birth-order variability is the variability of different traits among
individuals in a population by the order of birth and not related to age
nor to time. One example is of the data on variability of number of
new-borns in a litter in Clethrionomys depending upon the ordinal number
of pregnancies (Olavi Kalela, 1957). An analysis of these data shows a
tendency toward greater variability in number of new-borns in the first
pregnancy. ‘This trait has been shown to be quite invariable in the second
pregnancy. It is possible that this example is not very promising, but
there should be no doubt of the existence of such a type of variability in
nature, related specifically to the order of offspring produced by animals
and plants.

Biotopic variability

The term “biotopic variability’ was proposed by Schwartz (1963).
This concept should include the variability manifested within one popula-
tion or between micro-populations and caused by somewhat different
sets of conditions. It is clear from the above that this type of variability
could be equally classified as micro-population or within-population
variability. But here there is a serious contradiction. By the definition
of population used in this work (see Introduction), all the so-called micro-
populations within which there is panmixis and which, during a given
period of time, are isolated from other such groups, should be taken
as populations having an independent significance in evolution. If these
requisites—panmixis and isolation—are not observed, then the group
cannot be separated as a population and it will not be possible to talk
about the independent significance of variability: for the structure of such
a group.

Classification of Variability Phenomena 189

It is well known that isolation can be temporary and partial. This
situation makes more difficult any single solution to the problem of
classifying the variability of individual micro-populations within a popula-
tion. ‘This phenomenon logically seems to be related to that of geogra-
phical variability. Geographical variability is the same variability of
individuals of different groups, situated in a definite set of conditions—
like biotopic variability, only at a little higher level. Perhaps, it would
be more proper to consider the phenomenon of geographical variability
as between-population variability (down to the level of micro-population
variability); and to take biotopic variability as only within-population
variability (or between-micro-population variability). Ап _ insufficient
resolution of the general theory of population (Naumov, 1958; Beklemishev,
1960; and others), makes it difficult to solve the problem.

Both geographical and biotopic variability represent a phenomenon
of the same type caused by a specific adjustment of a population to its
living conditions. Factors like climate and biotic relations, as well as
the nature of the genetical structure of the populations, will fall under
such “specific living conditions” in both these cases.

Considering that the classification of type of variability proposed
here is preliminary and does not claim any final solution to the problem,
it is better to show the presence and the role of such variability in
mammalian populations without going into further detailed differences
between biotic and geographical variabilities (Table 66).

Table 66. Biotopic variability (с. о. Е 5с. о.) of anumber of traits in the common
hamster (Citellus citellus) in Bulgaria (Analyzed from the data of Straka, 1963)

Length of
МАТА СЕ ТОО ре т О С SS eee
Body Tail Foot Ear
Height above sea (meters) 1*
50 19.7 6.0+0.46 12.1+0.93 4.3--0.33 9.30.71
650 15 5.00.37 9.70.71 4.60.34 9.60.34
1900 7.9 6.00.66 8.20.90 5.2+0.57 9.71.06

*The mean monthly temperature in the active period of life of the population.

While analyzing absolute indices of the traits studied, F. Straka
(1963) confirmed Allen’s rule on the material presented by him. A
comparison of variability for these populations (micro-populations) allows
one to conclude that there is a considerable stability in range of variability
for all the groups for all the traits except tail length. The variability
of tail length is considerably less in hill hamsters than for sea-level ones.
It is not possible to relate these differences specifically to the height of the

190 Variability of Mammals

habitat or to temperature because there are really a large number of
factors which can determine the character of the trait.

A. К. Lee (1963) studied the utilization of oxygen and water in
desert and river-bank populations of Neotoma lipida; his data are concerned
equally with the phenomena of biotopic and geographical variabilities.
The range of variability for these traits (water and oxygen utilization)
is considerably higher in the desert type than in the river-bank type of
Neotoma lipida.

However, the data of Clark (1940) give the most detailed example
of this type of variability; Clark studied absolute indices and indices of
variability for nine body traits (body and skull measurements) in 30 (!)
populations of Peromyscus maniculatus, 8 populations of P. leucopus, and 4
populations of P. eremicus. Analysis of these data shows that there are
significant differences in variability for at least some of the populations
studied. This is one of the few works (if not the only one) where the
phenomenon of population variability has been given independent
“importance in a zoological investigation.

An experimental study of geographical variability was carried out
by Sumner (1924), who studied the changes resulting from the transfer of
two different populations of P. maniculatus to other ecological conditions
in the habitat zone of a third population (subspecies). The results are
quite intriguing. Only the absolute indices appeared to be of interest
to Sumner in this experiment and the variability was left unstudied. This
lacuna can be filled in however, because the factual data presented by
Sumner suffice for the calculation of a detailed characterization of varia-
bility in the original and the transferred groups. The findings are very
significant. Firstly, there is a reduction in the variability of almost all
traits studied—in some to a lesser degree and in others to a greater degree
(of the total number of 16 traits from both groups, only three traits re-
mained unchanged in variability with the transferred group, and there
was a significant increase in variability for one trait). Secondly, this
reduction in variability was rather sharply expressed in the population
of the subspecies P. m. rubidus, where a high degree of difference exists
for five of the eight traits studied. What factors brought about such a
behavior of variability is totally unclear. It may be observed, as con-
cluded by Sumner, that the changes did not go in the direction of the
third subspecies living on this territory. Unfortunately, data are lacking
for an exact comparison of these three subspecies with regard to their
original variability.

That many factors control variability is well seen from sins example.
The genetic mechanisms which, perhaps; determined these changes
in the given case, may occupy an important place among _ these
factors. |

Classification of Variability Phenomena 191

The number of examples of biotopic variability could be considerably
increased; several are given in other places of this work.

Traumatic variability

Traumatic variability is the variability of any dimension manifested
as a result of some trauma sustained by an organism. ‘The manifestation
of traumatic variability can be linear, of weight, of volume, meristic,
colorimetric, or in the form of epigenetic polymorphism, for any organ or
structure. But the importance of an independent treatment of traumatic
variability lies in the fact that first, it allows one to investigate more
exactly the above-mentioned manifestations of variability; and second,
traumatic variability covers a specific number of natural phenomena
and can serve as a particular character of a population. A change in
the structure of the host caused by parasites is known. Such changes
could be and should be taken into account while describing the general
variability of a population, because they represent specific correlations
existing in the “‘environment-population”’ system.
| It is possible that the concept of traumatic variability should be
broadened and should be transformed into a concept of variability caused
by the so-called mechanical reasons. Then variability like that from
‚ grinding down of teeth, which occurs for all mammals but in various
intensities depending upon different ecological reasons, could be classi-
fied with this variability. One. of the complex types of such grinding
down of teeth forms a number of typical “‘scissors’? from the initial non-
differentiated dental system in cetaceans (Yablokov, 1958). The number
of examples of such types could be increased.

Teratological variability

Teratological variability is the variability of any dimension arising
as a result of congenital abnormalities. Usually this type of variability
comes under epigenetic polymorphism (undeveloped individual organs,
like the instances of kidney and adrenal reductions in the Indian gerbil,
Meriones hurrianae described by Kaul, 1966). Examples of such varia-
bility are numerous in man also. Recently, special attention has been
given to this kind of variability because of widespread mutagenic factors
in natural populations.

The proposed classification of variability ends with this account
of teratological variability. Such a classification will inevitably broaden
according to the modes and methods of study of the structure of mammals,
and according to the selection of traits for which variability may be
calculated in exact units.

192 Variability of Mammals

Conclusion

The basic classes, types, appearances and forms of variability which
should be taken into consideration while studying mammalian popula-
tions, have been examined in this chapter. On the basis of the data
presented and from an analysis of the schemes of classification suggested
before, I propose the following classification for the phenomena of
population variability (Table 67).

Table 67. Classification of within-population variability

Manifestations ( kind of measurement )
Variability of time
Variability of weight
Linear variability
Variability of volume
Variability of surface area
Angular variability
Colorimetric variability
Meristic variability
Variability of temperature

Category
Bile ts SNe RNA i AER SR al on pe о
Continuous ог Discontinuous or
quantitative qualitative
Е
T ype

Structural (morphological)
Functional (physiological)
Biochemical
Ethological (variability) of behavior
Form
Age
Sex
Chronological
Birth-order
Biotopic
Traumatic
Teratological

Numerous examples of different types, forms, and manifestations
of variability presented in this chapter show how varied and elaborate is
the phenomenon of variability. This complexity of the phenomenon
shows the importance of exactly observing the class, type, form, and
manifestation of variability and considering the possible relationship of
each to the sample under study. An unexpected cause could be from

Classification of Variability Phenomena 193

unintentional mistakes in the sampling of*material. It has been shown
in the work of Adamczewska (1959) that there are clear differences in
variability (and in other traits) between animals caught in cylindrical
traps and in other kinds of traps. It is clear that some sampling methods
are better than others. Let us imagine that the variability of animals
caught in a cylindrical trap in the virgin forest of Belovezh is being com-
pared with the variability of animals caught in some way in the Rostov
region. It is clear that, although we may observe all rules (same age,
sex, selection at the same time), such a comparison may not give good
results because the variability with regard to the behavior of the animal:
was not studied.

At times, very good data on population morphology are spoiled by
insufficiently clean sampling. V.S. Smirnov (1962) analyzed changes due
‘to age in the skulls of males and females. Coefficients of variation
calculated from Smirnov’s data on the differences due to age in the skulls
of male and female Arctic foxes in Yamal, show no clear-cut trends other
than some sexual dimorphism. However, it appeared that data from
two different years were combined; in each of the age groups, which had
been separated with great care, animals of different ages and not just one
age were included. The development of these animals might have oc-
curred under different conditions. It is very possible that in this case
the conflicting results in the variability obtained could be explained by
the fact that yearly (chronological) and generational variabilities were
not considered.

I clearly understand that my proposed classification for the pheno-
mena of variability cannot be final. But such a temporary “working”
classification is more or less necessary for further developments in the
study of the problem of variability.

CHAPTER V

SOME REGULARITIES OF VARIABILITY

Elementary Regularities in the Variability of Traits

The interrelations of individuals within a population and the relations
of the population as a whole to the biogeocenose are so complex that at
our present level of knowledge we are unable to comprehend these pro-
cesses. It follows from this situation that the long-term need for forma-
lization of biological interrelations among populations and the biogeocenose
has not yet yielded fully to calculations.

However, in all types of biological investigations which are carried
out with the help of formalized methods (during statistical analyses of
biological material,-a known formalization or abstraction—large in some
cases and small in others—always takes place), it is important to know
fully the results of formalization so as not to attribute this or that property
to natural processes when it has arisen only as a result of using some
particular method. In the present case, while studying variability by
the method of coefficients of variation, it is necessary as far as possible
to consider fully the properties of this expression that arise out of mathema-
tical and other simple regularities. It can be assumed that these regulari-
ties are not now directly correlated with the effect of natural selection,
and hence they can be termed “elementary.”

It is evident that these elementary trends or regularities of a popula-
tion, in relation to environmental and within-population interactions,
cannot reflect all the complexities of these interactions and should only be
considered as constantly present parts of these complicated interactions.

1 Analogous to the term ‘‘elementary correlation”’ proposed by A. A. Malinovskii, 1948.

Some Regularities of Variability 195

The available, data allow one to examine the dependence of the
coefficient of variation of a trait on the absolute value of the trait and
on the relative and absolute values of the trait.

Some of these questions have attracted the attention of researchers
(Schmalhausen, 1935; review by Roginskii, 1959), but these questions
were not considered of primary importance for the study of variability
given here.

Relationship of Variability to Dimensionality of Traits

To understand the relation of variability to the dimensionality of
the trait, it is necessary to compare data on the variability of the same
organ or structure, expressed in different dimensions.

Proceeding from the existence of different manifestations of varia-
bility, it is at first desirable to analyze at least two traits which are
biologically close but of different dimensions (which will also serve as
a basis for further general conclusions about the relation of variability
to the dimensionality of traits in more complex series). Logically, the
first step will be to compare different manifestations of variability :2

Variability of weight .. Variability of time
do .. Linear variability
do .. Variability of volume
do .. Meristic variability
do .. Colorimetric variability
do .. Variability of area
do .. Variability of angle
Variability of time .. Linear variability
do .. Variability of volume
do .. Meristic variability
Чо. .. Colorimetric variability
do .. Variability of area
do .. Variability of angle
Linear variability .. Variability of volume
do .. Meristic variability
do .. Colorimetric variability
do .. Variability of area
do .. Variability of angle
Variability of volume .. Meristic variability
do .. Colorimetric variability

do .. Variability of area

? See the classification of variability phenomena in Chapter IV.

196 Variability of Mammals

Variability of volume .. Variability of angle
Meristic variability -.. Colorimetric variability

do .. Variability of area

do .. Variability of angle
Colorimetric variability .. Variability of area

do .. Wariability of angle
Variability of area .. Variability of angle

There are no factual data as yet to examine all these variations.
There is an absence of such comparable data even in the usual investiga-
tions in variability. All this makes it difficult at present to prepare a
full arrangement of the relations of the various manifestations of variability.

More than 30 years ago, while examining methodological problem,
I. I. Schmalhausen (1935) drew attention to the incorrectness of a direct
comparison of variability values of weight, volume, and linear traits.
These observations (made in the basic text of this subject) assume impor-
tance in the context of studying variability problems; I will fully quote
the opinion of I. I. Schmalhausen :

“In the most recent literature, especially American literature
(Schneider and Dunn, 1924; Asmundson, 1932; Angulo and Gonsalez,
1932), it is indicated that coefficients of variation of the same material
appear to be greater for weight measurements than for linear ones, and
on this basis a statement is made about how much more exact linear
measurements are in comparison with weight measurements. ‘This con-
clusion is based on a misunderstanding. The values of coefficients of
variation for linear, surface, and volume traits should not be compared
among each other in the same way, as for one-, two- or three-dimensional
correlations and values in general. It can be proved with the help of
some comparatively easy mathematical calculations that for the same
measurements of the same objects (when the surface and volume expres-
sions are obtained through an analysis of linear measurements of a
geometrically proportional body), there is a specific relationship between
the coefficients of variation for linear, surface, and volume values. For
example, where there is not a very large individual deviation (on the
average less than 10%), the relation is 1: 2:3. In this way, with rela-
tively identical accuracy of measurement, the coefficient of variation of
weight values should be three times larger than for linear ones... .Hence,
to compare the degree of accuracy or variability of data on linear and
weight measurements, it is better to transform all individual weight data
to conventional linear values by extracting the cube root” (Schmalhausen,
1935, рр. 11:12).

There is а known interest in carrying out comparisons of various
manifestations of weight, linear, and volume variability. It may be

Some Regularities of Variability 197

noted here that the relation of variability of weight, linear, and volume
values in the work of Schmalhausen (1935) is concerned with a compa-
rison of parameters of variability of geometrically proportional bodies and
cannot exactly reflect the situation arising during studies of morphological
traits in animals. |

А comparison of the variability of weight and body length for the
common house mouse (Table 68) shows that the variability of body weight
‘is two to three times greater than the variability of body length in all
the sexually mature groups. A similar situation is observed in almost
all cases studied among the most different groups of mammals (Table 1
in the Appendix). The variability of body length appeared to be less
variable than the variability of body weight in all cases.

Of all the data presented in this work, the variability of weight
appeared to be equal to or greater than the variability of body length in
just two cases : in the population of moles studied by K. Ya. Fateev in
1960 in the Kostrom region, and in hybrid populations of white mice
(Law, 1938). It is important to note in the first case that during two
other years (1958 and 1959), the variability of weight in the same popula-
tion of moles was, as usual, a few times greater than the variability of
body length. This shows, perhaps, that in 1960 some unusual changes
took place in the “environment-population” system.? The last proposi-
tion holds good because of the fact that the year 1960 was unusual also
for other traits in this population. The variability of almost all traits
sharply declined (see Table 64).

Table 68. Variability (с. v.+5,,),) of length and body weight of
house mice (Calculated from data supplied by Dynowski, 1963)

Variability of Variability of | Age group Variability of Variability of

secercup weight length weight length
Males Females
II 25.0+1.70 9.9-=0.66 II 30.9+2.28 10.0+0.71
III 15:95 1-26 7.7 +£0.53 III 22,242.34 9.5-+0.93
IV 18.5+3.28 6.9-+1.19 IV 23.35 D222 9.5-+ 1.86

The second instance of greater variability of body length compared
to variability of body weight, in inbred populations of laboratory mice,
is in my opinion one of those exceptions which confirm the general rule.
In fact, the above-mentioned deviation in the behavior of coefficients
of variation takes place in a complex hybrid population maintained under

3 This proposition is true in this instance only if the effect of accidental errors or change
in method of collection and analysis of the material is totally excluded.

198 Variability of Mammals

unusual conditions. The mixing of genetically different populations
vitiates any general relationships in the development of traits. And the
fact that there is such a relationship between weight and length in
genetically pure lines is shown by many genetical works.

But could it be that the variability of length is less than the varia-
bility of weight only when comparing length of body and body weight?
The comparison of weight and linear characters of different organs shows
that this is not so. Thus, the variability of weight of the intestines
appears to be a few times more than the variability of length of the
intestines :

6. U.+S¢,y, Length of intestine 4.2+1.11 8.41.1 25.744

6. 0. 55.0. Weight of intestine 19.7+3.87 45.3460 46.0+15.3

Species and author (Delphinap- (Тара (Macropus sp.)
terus leucas) europaea) ‘Tribe and

Own data Fateev, 1962 Peel, 1963

Data on the variability of linear and weight characters for the skull
of the fur seal are interesting (Table 69).

These data show clearly that variability of weight is always greater
than the linear variability of the same organs.

Table 69. Variability of weight and linear characters for
certain skull traits of the fur seal (Callorhinus ursinus)
(According to data obtained from V. F. Muzhzhinkin)

—_ ee

Weight Length tweight-length Sex Age

С. VUtSe.y с. 9. ESe.y
Skull
17.3+0.84 4.7--0.23 +14.4 Males 2=3 years
16.8-+3.30 3.0+0.76 + 3.8 Fs 4—5 years
10.3-2.31 4.2 4.0.98 -- 2.5 Females 2=3 years
11.5+2.60 3.2+0.71 + 3.1 >» 4-35 years
9.71.82 2.50.47 + 7.6 a 10+ years
Lower jaw

20.8-+ 1.02 5,.3+0.26 +15.0 Males 2=3 years
20.03.93 4,440.87 + 3.9 He 4=5 years
11.7+2.63 3.6+0.80 + 2.9 Females 2=3 years
13.0+2.91 3.7 50.83 а ‚ 455 years
13.5+2.56 2.7+0.51 + 4,1 a 10+ years

Some Regularities of Variability 199

To summarize all the data comparing linear and weight variability pre-
sented here, the following conclusion can be drawn : Weight variability is
significantly higher than linear variability. It should be emphasized
once again that this conclusion is applicable only to the variability of
the same organs or structures; it is reasonable that the variability of body
length may appear to be significantly higher than the variability of weight
for the crystalline lens of the eyes. Instances showing deviations from
this rule should indicate shortcomings in methodology or unusual relation-
ships in the population-environment system.

Certain data permit a comparison of linear and volume variabilities
for the same organs and structures (Table 70).

Table 70. Variability in volume and linear characters of intestines
of foxes and minks (Calculated from data of Sokolov, 1941)

Sex eae aa fyolume-length
Foxes
Males 8.6+1.26 12.5+1.97 —1.71
Females 5.60.66 12.6+1.71 — 3,93
Minks
Males 9.7--1.27 13.6+1.92 — 1.69
Females 10.5+1.55 13.6--1.92 —1.25

It can be seen from these data that the variability of volume was
less than the variability of length in all cases without exception. Un-
fortunately, this conclusion is not sufficiently reliable, as other such
examples are lacking.

A comparison of weight and volume variabilities can be carried out
by analyzing the data of J. Hromada and L. Strnad (1962). From their
data, in 12 out of 13 comparable groups of different sex and age in two
species of Macaca (M. rhesus and M. cynomolgus), the variability of weight
was greater than the variability of volume for the skull; moreover, in
ten cases the differences were statistically significant.

According to the hypothesis developed by Schmalhausen (1935),
the weight and volume variabilities should coincide for the same trait.
It is seen from the example of variability of parameters for arboreal nests
of golden mice (Peromyscus nuttalli) in the data of Packard and Garner
(1964), that this assumption is true in some Cases :

Trait (> Db

Length, width, depth - го 10.3-14.3%
Volume oe 27.0%

Weight 30 38.1%

200. Variability of Mammals

On the other hand, a difference in the variability of linear, volume,
and other traits of the organ under study, indirectly denotes the accuracy
of the methodology used. It is seen from the above data of Hromada

and Strnad that the precision of measurement of the volume traits of
skull appears to be higher than the precision of the weight traits.

Data comparing meristic and linear variabilities for tail length and
number of caudal vertebrae in mice are given in Table 71. In the example
studied, the meristic variability appeared to be two to four times less
than the linear variability in 16 different samples. However, data on
the variability of number of teeth on each side of the jaw, and the length
of jaw, in Black Sea dolphins do not particularly confirm these con-
clusions. Nevertheless, it is quite possible that in the last case it would
be better to compare variability of the number of teeth in a quadrant

with the variability of length of only the dental lamina, and not the whole
jaw.

Table 71. Variability (с. о. -Е5с.,.) of tail length and number of caudal
vertebrae in different phenotypes within one back-crossed population
of mice (Calculated from data of Law, 1938)

Number of Length of

Phenotype Sex Tail length caudal 15th
vertebrae vertebrae
Black I | 4“ 16.52.38 2.90.42 7.0-Е 1.01
ФФ 10.61.67 3.60.57 5.60.88
Blackish-spotted I { 36 23.6+3.40 3.3+0.48 8.6+1.24
ФФ 21.53.24 6.2 0.93 12.01.80
Black, waltzing | ee 18.02.58 4.30.62 6.3-+0.91
ФФ 12.9 -- 1.94 4.90.74 11.51.73
Blackish-spotted, waltzing | 66 21.6+3.04 4.4+0.63 11.2-+1.58
ele) 18.0+-2.77 4.0+0.60 7.8+1.21
Brown | jd 15.7+1.45 3.10.28 7.60.70
9 17.4 1.59 11.7 1.06 11.11.01
Brownish-spotted { && 16.41.70 7.0-0.73 13.7+1.42
ele) 15.8-+ 1.68 7.8 50.83 7.8 +0.83
Black IT | eve 16.1+1.40 4.4+0.37 7.00.60
ФФ 19.2 1.58 8.60.70 7.60.62
Blackish-spotted II | &< 16.6-=1.73 2.90.30 9.3-Е0.97
ФФ 17.81.70 6.5-=0.62 7.60.73

The data of Hall and Schaller (1964) on Enhydra lutris make possible
a comparison of time and meristic variabilities. A sea otter spends diffe-
rent amounts of time breaking up the armored shells of sea urchins,
etc., depending upon what catch the individual surfaces with. Naturally,

Some Regularities of Variability 201

the time taken on the surface Юг eating food can be compared with a
high degree of significance as a homogeneous character of its behavior;
also comparable are the total number of strokes and the individual series
of those strokes inflicted to convert food material (crabs, sea urchins, and
shells) into an edible condition. The variability of time taken on the
surface with these objects ranges from 52.2 to 59.4%, and the variability
of number of strokes and series of strokes is respectively 55.1+3.53%
and 56.3+1.24%. Аз can be seen, the meristic and time variabilities
completely coincide with each other.

A comparison of meristic and colorimetric variabilities is possible
if it is presumed that the quantity of hemoglobin (estimated with the
help of a colorimeter) and the number of erythrocytes can measure the
same trait—the hemoglobin content of the blood in the given species. It
appears that the variability (с. v.=ts,,,.) for number of erythrocytes is
considerably higher than the variability for quantity of hemoglobin :*

ee i a i a eS

Hemoglobin Number of
Breed of sheep’ quantity accord- erythrocytes ‘colorimetric-meristic
ing to Sali in millions
Romni-Marshi 1.76+0.23 3.10+0.40 —2.91
Tsigeiskie 1.59+0.25 2.26--0.36 — 1.52
Е, hybrids 2.28 0.36 3.06 0.48 —1.15

As seen from the data presented, there is no possibility of comparing
the majority of manifestations of variability, but the data already obtained
allow one to draw conclusions in principle for an analysis of variability
of any form and type. ‘Though the coefficient of variation is a dimension-
less number and hence can formally allow a comparison of traits with
any type of measurements, it should be remembered that the nature of
those measurements has an effect on the magnitude of variability. It
follows from this that the variability of a trait under study is, perhaps,
accurately comparable with the variability of another trait, measured in
different measurements, only after considering additional corrections.

Relationship of Variability to Absolute Size of Trait

The first problem confronting us in relating variability to the absolute
size of a trait is the need for comparing measures of body size in different
species. It should be clear in the light of recent studies on seasonal, age,
sex, and biotopic variabilities (Chapter IV) that the assertion of Ya. Ya.

4 Calculated from data presented in the work of Kh. F. Kushner, 1941.

202 Variability of Mammals

Roginskii (1959) is doubtful, that there should be relatively close values of
coefficients of variation for the same trait in different species. In fact, if
the coefficient of variation varies considerably in the same population in
different seasons and years, and depends upon many specific factors, then
how can it coincide with indices of other species ?

For a clean comparison, it would have been better to compare co-
efficients of variation from data which are collected specifically in the same
years, seasons, and months, and belonging to absolutely identical age groups.
But, as a rule, there are no such data available for comparison. Anyway,
let us try to compare the variability of one trait in species belonging to
different orders (Table 72).

It is seen from this table that the statement of Roginskii about the
relative constancy of coefficients of variation for one and the same trait
in different species is wrong. In Sorex minutus (body length, about six cm),
the coefficient of variation of body length appears to be the same as in
sables (body length, more than 40 cm), but the coefficient of variation of
length of the skull is closer to that of beavers and moles; in the hedgehog
(body length, 26 cm) the variability of body length appears to be the same
as that of hamsters, forest mice, bank voles, and certain seals, but the
hedgehog excels all the other mammals presented in the table in variability

Table 72. Comparison of variability of body length and basal
length of skull in different animals (Adult males)

Condylo-
Body length, mm basal
Species length Data from
XESS 5 РАЕН, @ Небо
Sorex minutus 59.641.60 2.8+0.42 1.10.09 Dolgov, 1963
Sicista betulina 65.4--0.996 7.841.08 2.3+0.38 Kubik, 1953
Clethrionomys glareolus 101.0-Е0.98 6.70.69 3.60.37 Bol’shakov and
Schwartz, 1962.
Apodemus flavicollis 105.1 -L0.64 5.7+0.43 4.1--0.63 Adamczewska, 1959
Talpa europaea 140.5 -+ 1.68 4.90.85 1.8+0.31 Markov, 1957
Citellus citellus 202.5+1.31 6.0+0.46 — Straka, 1963
Erinaceus europeus 260.5 -- 4.24 6.5+1.15 13.1--2.34 Markov, 1957
Martes zibellina 412.7+2.50 2.8+0.42 2.3+0.34 Kuznetsov, 1941
Castor fiber == — 1.70.32 Hinze, 1950
Pusa hispida ochotensis 1169.0-+- 12.60 7.2+0.76 2.3--0.49 Own data
Pusa hispida hispida 1185.0-+-14.40 9.1+0.86 3.1-L0.32 Own data
Cystophora cristata | 2208.0-- 17.0 9.2+0.55 4.3 +0.51° Own data

Delphinapterus leucas 4028.0+32.30 4.90.57 — Own data

ee ee ——_——_—

Some Regularities of Variability 203

of skull length. Thus, the variability of the same trait may be different
in different species.

Going to another aspect of the problem, let us compare the variability
of different traits within one population. In the opinion of Roginskii (1959)
there should be a negative relationship between the absolute size or
magnitude of a trait and its variability value.

It is seen from Table 73 that if all the traits of organs of different systems
are combined and examined, no negative relationship between the magni-
tude of the trait and its variability is observed either in the meristic or the
linear traits. Thus, the variability of body length is close to the variability
of width of snout (the absolute values of these measurements differ more
than 40 times), but the variability of skull width between the orbits appears
to be inseparable from the variability of the “‘anus-navel’? measurement
(the differences in the absolute values are more than 50 times). Many
examples of this type could be cited.

Table 73. Relation of variability to absolute size of
trait (Ringed seal of the Okhotsk Sea, Pusa hispida;
adult males from Summer collections of 1963)

6 : Coefficient of
Trait Absolute size о

variation

Meristic (in numbers)
Eye vibrissae 3.0+0.15 28.0+3.7
Sternal ribs 10.0-L0.02 1.4+0.1
Caudal vertebrae 13.4+0.29 8.1-L1.9
All lip vibrissae 50.6+0.42 6.2 -40.6
Tracheal rings 91.0-+1.03 5.7+0.8

Linear (in mm)
Skull width between orbits 5.9+0.17 15.4--2.0

Width of snout 26.9 0.30 5.7+0.7
Length of lower jaw 108.0 +0.55 2.9+0.4
Length of skull 166.0-+-0.74 2.4+0.3
Anus-navel measurement 319.0--7.80 13.6+1.7
Body length 1169.0-+ 12.60 7.2 -40.8

Therefore, in general, the statement about the negative relationship
between the magnitude or size of a trait and its variability, when comparing
different traits of one population, appears to be incorrect. However,
this statement would appear to be correct for organs of one system.

If the variability for number of all vibrissae in the rows of the nasal
group in the Greenland seal is compared with the absolute number of
vibrissae in each row (Figure 57), it appears that a smaller number of

204 Vanability of Mammals

vibrissae per row completely coincides with a greater coefficient of variation.
This observation is confirmed by all other data on the variability of
vibrissae in pinnipeds of all ages, male and female, for all species studied
(Yablokov, 1963a, 1963b; Yablokov and Klevezal’, 1964; see also Table 3
in the Appendix). The data of Huestis (1925) confirm these observations
with regard to meristic traits.

But this rule is true not only for meristic traits. It appears that the
negative relationship between the magnitude of a trait and its coefficient
of variation occurs also in linear traits of one organ or a system. Many
examples of this type of variability for different animals can be found in
the data already presented.

But it must be said that for some traits this regularity rarely occurs
in a typical way. It can be seen from Figures 18, 21, and 22 that there
is no clear-cut relationship between the magnitude of the traits and their
coefficients of variation according to the rule indicated. However, a
specific trend is observed in the behavior of the majority of traits which,
briefly, is that the traits of greater size have generally smaller coefficients
of variation. |

C.V.%

100

Number of vibrissae in a row

2 3 4 1 5 6 7 8 9 Rows

Figure 57. Relationship of coefficients of variation to the absolute
number of vibrissae in the lip group in the Greenland seal:

a—Number of vibrissae in a row; b—Coefficient of variation.

The data of Lowrance and Latimer (1957) show not only the dependence
of coefficients of variation of length on the absolute size of the trait, but
also the correlation of traits of weight with their absolute values at the
same time. As seen from Figure 58, the hyoid bone, with the lowest
weight, appears to be the most variable bone in the human skeleton, and
the skeleton as a whole has the lowest variability. However, it is seen

Some Regularities of Variability 205

from the same figure that a number of measurements do not fall in the
framework of this rule. Thus, for example, a comparatively low variability
occurs for the weights of the lower jaw, the pelvic bones, and the vertebral
column, while the weights of the ribs and scapulae have a comparatively
high variability. But the part having the least length, the manubrium
of the sternum, has the greatest variability of length, and the part having
the greatest length, the hip, has the least variability.

C.V.% д

40 то
1234567 8 9 1011 12 13 1415 16 17 18 19

Figure 58. Relationship of absolute magnitude of weight Юг parts of the human
skeleton with the magnitude of the respective coefficients of variation
(according to the data of Lowrance and Latimer, 1957):

a—Weight; b—Coefficient of variation. Individual bones, in order of increase
in weight, are shown on the abscissa.

Data exist on the relation between the magnitude of colorimetric
variability with the absolute value of the trait (Table 74), and on the
relation between time variability and the magnitude of the trait (Littlepage,
1963). These data generally confirm the conclusion of the presence of a
negative relationship between the size of the trait and the magnitude of
its variability. Therefore it appears that during a comparison of functionally
similar traits or the characters of organs of one system, there is a negative
relationship between the size of the trait and the magnitude of its variability; the
higher the absolute value of the trait, the lower its variability. ‘This regularity or
dependence occurs in nature only in a general way, and in large series of
measurements quite a few deviations are observed.

206 Variability of Mammals

Such deviations from these regularities are not accidental. It is
already clear from the general situation explained in the beginning of this
chapter, that there must exist some more complex regularities of the
appearance of variability of traits in a population, in addition to the primary
trends or regularities. And it would be difficult to assume that these
primary regularities could be brought to light in pure form in natural
It becomes necessary to examine them through the complex
mesh of other relationships. ‘There could be cases where primary regularities
absolutely do not appear. But such cases cannot prove the absence of such
primary regularities; they only show the resultant of the total situation

material.

in nature.

Table 74. Variability of back color in Peromyscus maniculatus of
Florida and Alabama (Absolute data from Klein, 1964)

Population Red Green Bluish-violet
т { 18.9+0.74 15.30.82 13.90.60 Rss
17.99.76 2481381 198.305 соо
т { 14,140.40 10.50.20 — 8.90.30 я
16252203) ПТО 2 бо.
Ш { 22.440.48 9540.31 7.60.29 KLSz
17.5 1258 4s 7c 2033) Пр ро
= { 11,140.28 8140.31 674039 #45,
73:51 1178 162-123 2035 во
у | 13.70.33 9.80.36 7.20.36 R55
30.442.23 36.342.77 49.443.63 c.v.45,.
VI | 10.70.22 8140.31 6.70.39 я
17.91.45 33,312.54 ` 50:64:4.11 со
т { 8.80.51 6940.31 5.5+0.24 R45;
57.9+4.08 44.83.16 43.743.07 c.u.+5,.9
at { 7.7106 601019 52016 Я
26-239 231224 м Бо
| 7440.86 5.50.68 5.2+0.52 Хх
66.2+8.70 65.38.16 67.98.49 c.v.+5¢.y

The question of relating the absolute value of a trait to the magnitude
of its coefficient of variation appears to be far from simple and specific,
as it seemed from a superficial examination. It is seen from the formula for

the coefficient of variation,
с. 100

С. 0. == -

x

that in fact the coefficient of variation must decrease with an increase
in the absolute value of the trait (x) in the denominator (except when

Some Regularities of Variability 207

the standard deviation itself increases proportionately with the mean
value).

Because this rule is not observed in a natural sample, a speculation
arises that some powerful forces capable of changing and regulating
variability of structures, act beyond simple mathematical regularities.
The generally understood ecological and genetical factors must be the
first of such forces.

After examining more carefully the graphs of the relationship of the
absolute value of a trait with the magnitude of variability in a sufficiently
large number of traits, and by restricting the higher and lower peaks, it is
possible to obtain the general nature of this range of real variability for the
group of traits in the organ under study, a “‘variability drift”? in which the
most variable traits form the lower border of the drift, and a certain number
of traits of intermediary significance appear. It is better from a practical
point of view to introduce into this “drift”? a certain hypothetical “mean
drift,” in relation to which it would be possible to calculate more exactly
the deviations both on the side of increased and that of decreased variability,
and which to a certain extent can show the magnitude of deviations from
the influence of the elementary regularity between the magnitude of the
trait and the magnitude of variability.

Such “drifts” of variability in the skull measurements for a number
of species of pinnipeds have been given in Chapter III (see Tables 15,
О ЭР ес).

It is necessary to compare data belonging exclusively to one organ,
one system of organs, or with a functional complex of similar measurements,
for a practical analysis of this drift. Sufficiently large amounts of data
(not less than 15 to 20 traits) are a necessary prerequisite so that the upper
and lower boundaries of the variability drift are brought out; and finally,
it is necessary to arrange at least a large majority of the traits in an ascending
or a descending order of their absolute values (and not group them around
one or a few traits close to them in absolute value). This method of finding
out the “variability drift’? attains a special significance for analyzing struc-
tures with a greater or lesser variability, and for a morphological-ecological
study of population.

Relationship of Absolute and Relative Variability Values (Indices)

The study of indices (relative values) sometimes makes it possible
to eliminate the effect of the largeness or smallness of an animal (Filipchenko,
1916) during comparisons, and brings to light more vividly the specific
properties of the organ under study, or of the animal as a whole. This
method of indices is widely used in modern taxonomy (Mayr, Linsley
and Usinger, 1956) and in studies of the ecology and physiology of mammals

208 Variability of Mammals

(summary of data by Schwartz, 1960). However, as far as I know, the
variability of relative values has not especially attracted the attention of
researchers. Because of the widespread use of indices in zoological
investigations, the question of comparing the magnitude of variability for
relative and absolute indices should be taken into account during the study
of morphological variability as a whole.

The variability of indices is in abstract numbers without dimensions,
and Ya. Ya. Roginskii (1959) observed that instead of calculating the co-
efficient of variation, the standard deviation may be calculated directly.
However, the formula is

2 2 2
49 | 9 Oy 27бабь
oz — (= eed ps aI

2 B2
a bo аобо

where с” is the variance of the index a ag and by are the means of the
measurements a and р ; and г is the coefficient of correlation of a and 6.
We see that the magnitude of the index is influenced by the standard
deviation itself, and the coefficient of variation has to be used. “Ву dividing
the standard deviation by the mean index, we rid ourselves of the influence
of the latter’? (КослазКи, 1959; р. 86).

To develop the viewpoint of Roginskii, it would appear proper to
propose that for a complete study of the variability of an index, it is
necessary to know these two parameters: (1) the coefficient of variation
for each separate trait; and (2) the coefficient of correlation between the
traits under study. And then (in case the indices under study are similar
in magnitude and there is a considerable correlation between the traits), `
the coefficient of variation of the index will be small even if the value of
variability for each trait is large or small or however diverse. With a
low correlation between the two traits under investigation, it is possible
that the variability of each trait will be comparatively low while the co-
efficient of variation for the index appears to be very large. But then what
does calculating the coefficient of variation for the index reveal if we already
know the separate coefficients of variation for each trait and the correlation
coefficient which expresses the relation between the traits ?.

Ya. Ya. Roginskiu concludes that “the coefficients of variation give an
idea of the structural mode of variability of indices” (1959; р. 86), and in
this the role of coefficients of variation differs considerably from the one they
play in the study of any individual trait.

After these general observations, let us compare some appropriate
data on the magnitude of variability for absolute and relative values of
a trait; then let us make an attempt to determine the relation between
them by looking at specific examples. :

Data on the variability of body length and length of intestines for

Some Regularities of Variability 209

the Greenland seal (Table 75) show that the variability of body length has
comparatively low values (3 to 4%), that the variability of intestinal length
is three to four times higher (11 to 12%), and that the variability of intestinal
length in relation to the body length is about 11 to 12%, i.e., is retained on
the level of variability of the more variable member of the comparison.
The variability for length of large and small intestines has values quite
similar (15 to 17 and 11 to 12% respectively); the variability of the ratio
of length of large intestines to that of small intestines is very different in
males and females (11 and 24%); moreover, in this case, the variability
of the index in females is retained on the level of the less variable member of
the comparison, while in males it somewhat exceeds the level) of the more
variable member of the comparison. ‘The sexual dimorphism in variability
of body length did not influence the variability of the index during the
determinations of variability of the relative length of the intestine, but the
sexual dimorphism, which was not reflected in the variability of absolute
traits, clearly appeared during the determination of the variability value
for the small intestines relative to that of the large intestines.

Table 75. Variability (с. 0.5%...) of absolute and relative
values of length of intestines of adult Greenland seals
(Pagophilus groenlandicus), Yan-Maien region, 1962

Males Females
Body 4.0+0.28 3.10.24
Intestines 12.042.35 11.7+1.89
Small intestines 12.242.40 11.1+2.03
Large intestines 17.22.30 15.3+2.54
Intestines/body, length 11.6+2.46 10.6+1.81
Small intestines/body length 12.6+2.80 11.01.88
Е Small intestines/length оЁ]агое intestines 24.05.37 11.31.93

Data on the variability of absolute and relative weights of musculature
for six different groups of rabbits are available (Latimer and Sawin, 1955).
The variability of absolute muscle weight appeared to be practically the
same as the variability of body weight (11.7 to 21.5 and 12.5 to 21.0%
respectively), and the variability of relative muscle weight was three to
four times less than any of these parameters (3.9 to 5.5%).

As seen from the data given below, the variability of the index of
intestinal length (in relation to the body length) in white whales (adult
males, Novaya Zemlya, 1957) appeared to be twice the variability of either
of the components in the comparison :

210 Variability of Mammals

ee ee, eee ee

Length of 6. 0: 5. #5;
Body 4.00.60 397.3 4.22
Intestine 4.2 Е 1.11 29.6--0,47
Intestine/Body 8.6 +2.30 7.7+0.25

In this case, there is a reverse situation from the one observed for the
variability of height index for muscles in rabbits.

The data on variability of relative length of the feet of yellow-necked
mice (Table 76) also show the complex behavior of the index of variability
(in relation to the variability value of each of the traits in the given index).

Table 76. Variability of yellow-necked mice
(Calculated from data of Ursin, 1956)

Population Body length Foot length = —————————

© 552.0 6. 9-Е. Co 0-Е 56.0. Go, ВНЕ Чо» 5х
Latvian 4.8+0.28 10.5+0.61 1.50.34 20.7-+0.10
Danish—1 4,640.49 8.8+0.96 3.5+0.27 20.8+0.08
Danish—2 4.4--0.35 8.6 0.88 — 6.10.88 — 20.80.26

oe a eee ss SS

It is seen from the data presented in Table 76 that the coefficients
of variation. for body length and foot length are practically the same in
the three populations, whereas the variability of indices of these measure-
ments differ. It is interesting that the indices of absolute values are
unequivocally the same in all populations.

The distribution of the variability values for indices of intestinal length
in four sexually mature groups of hooded seals (Cystophora cristata), may depict
a still more complex picture (Yablokov, 1963). The variability of the
index for intestines (length of intestines/body length) in mature females
is greater than that for absolute length of intestines in new-born males
and females, but in mature males the situation is the reverse. The
variability of the index of intestines (length of small intestines/length of
large intestines) appears to be greater than the variability of absolute
measurements for large intestines in both mature males and new-born
females. Since it is true that all the differences mentioned here are insigni-
ficant, it can be said that the variability value of the index in these cases
approaches the variability of absolute values of the trait.

It is important to examine large numbers of measurements in their
absolute and relative expressions so that the relationship between the
variabilities of the absolute and relative values becomes clear. An example

Some Regularities of Variability 211

of this is found in the work of Lowrance and Latimer (1957), where the
variability of weights of bones of the human skeleton is given. Out of
16 traits, the coefficients of variation for 13 indices appear to be lower
than the corresponding absolute values of the traits; moreover, in the
majority of cases, the differences are considerable. However, the disposition
of the coefficients of variation of the indices in order of ascending variability
remains éxactly the same for the coefficients of variation of the absolute
weight of the skeleton and its parts.

A comparison of the variability of absolute and relative values for a
number of traits in a laboratory population of mice (data of Sumner,
1909) shows a different tendency. The variability of indices throughout
appeared to be higher than the variability of absolute measurements. The
length and weight measurements are exceptions to this (weight of hair/
square of body length). But even this behavior of coefficients of variation
of indices is not constant; thus, for example, the variability of the weight
of the eyeball is significantly less than the variability of the index of the
weight of the eyeball (according to Latimer, 1951).

The last example in this series is from the data of V. I. Tsalkin (1960)
on the variability of metapodials in horned cattle (Bos taurus). It appeared
that the variability of all indices of the metacarpus and metatarsus of bulls,
bullocks, and cows of Kalmitskii cattle was, in all cases, lowest in cows and
greatest in bullocks. A similar distribution is observed also for the variability
of height at withers in the groups under study.

The examples presented here and in other chapters allow one to draw
‘certain conclusions. ;

First, the variability of indices is not always governed by those general
tendencies which are characteristic of the variability of the absolute values
of traits. Second, specific trends in the behavior of variability of indices,
characteristic of all the appearances of a variability of this type, are not
found. Both of these conclusions confirm the assumption that the variability
of indices reflect relations and regularities different from those which express
variability in absolute values.

Attempts are often made to use different traits as independent variables
in a determination of relative characteristics in order to discover general
regularities in the variability of indices. ‘Wood её al. (1965) showed that
indices obtained on the basis of dividing the weight of an organ by the
body weight were not sufficiently accurate, and they proposed that in such
calculations it is better to use heart weight than body weight. Talbot,
Lee, and McCullich (1965) think that weight of the heart is not the most
useful independent variable but rather perimeter of the heart, whereas
in many other works the relative weights of organs are calculated in weight
units per 100 kg (or 100 g) body weight (for example, Kramer, 1964). All
these attempts reflect a non-specificity and insufficiency of determination

212 Variability of Mammals

of the relative parameters of organs. It is not possible to say how closely
the variability of indices is related to the micro-evolutionary processes
in populations, unless some special critical investigations are carried out.
And to carry out such an analysis, large quantities of factual data on all
types, forms, and appearances of variability are necessary.

It is quite likely that the data on variability only seem heterogeneous
and can be explained by the complexity of the whole phenomenon, which
does not yield to interpretation from the viewpoint of the study of absolute
values. In any case, the variability of relative values (indices) is not
comparable in magnitude to the variability of absolute values.

Comparative Variability and its Importance
in Population Biology

In the previous chapters, from II to IV, the variability values of traits
characterizing one system of organs or an individual organ were examined
separately without reference to the variability of other traits in the popula-
tion. This problem—the correlation of variability of different traits in the
same population—has also not been examined by other research workers.
This has happened possibly because data for even posing the question were
not then available.

Data on two basic aspects are needed to obtain a unified picture of
comparative variability in a population : Data from the same population
at different times (after months, seasons, years, a number of generations,
etc.),° and data on variability of a number of the same traits in different
populations of close species in the same season and during a number of
seasons.

In the work of K. Adamczewska (1959), data appear which characterize
a population of yellow-necked mice (Apodemus flavicollis) in different months
of one year. Absolute measurements and coefficients of variation calculated
on the basis of factual data are presented in Figure 56.

Let us try to distribute the coefficients of variation in each monthly
series in declining or increasing order of variability, and then compare
the data obtained for different months. It appears in all cases (not only
for males as shown in Figure 56, but also for females) that with regard to
the lowest coefficient of variation, “length of feet” (trait 5) comes first,
“length of ear” (trait 4) second, “Бо4у length” (trait 3) third, and’ “tail
length” (trait 1) in fourth place. It seems that, according to the magnitude

5 Strictly speaking, by investigating the animals at the same place after sufficiently large
intervals of time, exceeding the lifetime of one generation, we investigate not one but two
different populations. It is quite logical to restrict a population by time, and in future
developments of population morphology such an approach’of demarcating the Ве of
population by the life of one generation will perhaps be widely used.

Some Regularities of Variability 213

of variability, each of the traits maintains its place in the series. It is
important to note that the “‘maintenance (of place) in the series” is observed
in spite of the considerable overlapping of absolute values of coefficients
of variation from month to month. Thus, the coefficient of variation for
ear length in males varies more than twice—from 5% in September to
10.5% in June; the coefficient of variation for length of feet varies almost
as much—from 4% in November to 7.1 % in June; the coefficient of variation
for tail length varies almost three times (Table 77).

Table 77. Seasonal variability of body measurements for male

and female forest mice (Apodemus flavicollis) (Collected 1955;
calculated from the data of Adamczewska, 1959)

Month Sex KE Se Co 6.9. Е 5.0 n
Body length
August 33 100.0 -+0.88 8.98 — 8.98--0.68 104
ФФ 97.00.73 TS) Па NO?
September 3¢ 102.0-0.48 6.81 6.67+40.27 248
Фо 99.00.40 69 679029 29
October BS 103.6 0.95 6.47 6.25--0.65 46
Фо ‚ 96.61.14 6.54 6.77-0.83 33
November 4 105.1--0.64 5.96 5.67+0.43 86
Фо `98.2 0.70 6.29 6.41+0.51 80
: Body weight .
August Be 28.70.67 6.88 23.97+1.65 105
Фо - 25.60.62 6.29 24.5741.71 103
September 3 29.8 +0.37 6.34 21.28+0.87 301
ФФ 26.30.31 5.14 19.54-40.83 278
October BE 30.8 +0.89 5.60 18.18-2.03 40
eo 24.0+0.79 4.47 . 18,634+2.33 32
November 44 32.40.57 5.15 15.90+1.24 82
Фо 26.2 0.45 396 1511120. 79
Tail length
August 3S 102.043.24 32.12 31.49+2.25 98
ФФ 101.0+2.37 23.18 22.95--1.66 96
September < 107.9+1.48 23.08 22.06-40.97 259
ele) LOAM 2717 76 1011.20 1235
October Ae 105.94+2.18 13.05 12.3241.45 36
®®) 102.3 2.01 — 10.05 9.824 1.39 24
November 44 107.9 1.39 12.29 11.39-+0.91 78
ФФ 103,141.28 11.14 10.81-50.87 76

(Contd.)

214 Variability of Mammals

Month Sex K+ Sy б 6.9. 50.0 п
Ear length
June && 17.60.22 18.50 10.51-+0.87 73
ele) 17.2+0.18 1.58 9.19+0.74 77
July 3d 17.40.09 1.39 7.99 +-0.36 245
9 17.20.09 1.44 8.37 0.38 240
August 33 17.5-Е0.13 1,33 7.60 +0.52 105
ФФ 17.40.11 1.14 6.55-+0.46 101
September gd 17.80.05 0.89 5.00-+0.21 292
Фо 17.7+0.05 0.91 5.14-+0.22 282
October 33d 18.10.15 1.03 5.69-=0.60 45
ФФ 17.80.17 1.01 5.67 +0.69 34
November 3¢ 18.10.10 0.90 4.97-40.38 85
Foot length
June 3d 24.49 +0.21 1.74 7.10-0.60 70
ФФ 23.60--0.16 а 5.97 0.48 78
July 3d 24.80 +0.008 1.77 5.12 0.23 250
ФФ 24.20-0.09 1.41 5.83 -Е0.27 241
August 33d 25.40+0.13 1.31 5.16--0.36 104
ole) 24.60-+0.11 1.09 4.43+0.31 101
September Gd 25.30+0.06 1.04 4.11+0.17 295
ФФ 24.70--0.07 1.11 4.49 0.19 286
October eves 25.20-+0.22 ПЕ 5.83 +0.62 44
9 24.90 +-0.17 1.00 4.02 --0.49 34
November Go 25.30+0.11 1.02 4,03+0.31 86
oe) 24.60+0.11 0.97 3.94+0.31 80

And thus, in each moment of the life span of a population (in the
given example, in each month), the coefficients of variation for the same
traits are distributed in exactly the same order : 5-4-3-1.

In the work of Sumner (1924), the same four traits were studied in
four populations of Peromyscus maniculatus : in two original populations of
different subspecies (Р. т. sonoriensis and P. т. rubidus), and in transplanted
populations of these subspecies. In the first case, the experimental popula-
tion was studied after seven generations, and in the second case, after 12
generations of development in captivity.

The absolute values of coefficients of variation for a number of traits
considerably overlapped, as in the above-mentioned case of forest mice.
But if we distribute the traits according tothe coefficients of variation at one
time, it appears that the disposition of all traits relative to each other is
fixed absolutely specifically in all cases :

P. m. sonoriensis Р. т. rubidus
Wild population 3212 In nature Sra 12
After 7 generations 3412 After 12 generations 3412

Some Regularities of Variability 215

As in the case of forest mice, in all the contemporary samples, tail
length (trait 2) appears to be the most variable. Then in the order of
reduced variability follow body length (1), ear length (4), and length of
feet (3).

It can be seen from the two comparisons cited above that the same
traits in taxonomically remote species are exactly similar with regard to
variability. As will be shown below, such a situation is not characteristic
in general of the given traits. This is shown by the analysis of data obtained
by us on house mice sent by Dynowski. The traits arranged themselves
in the following manner with regard to variability in one population:

Age group Males Females
п 9210: 32410
Ill 3—210 34210
IV 314) 2 150 34120

* The nomenclature of traits is the same; 0—Body weight.

In the majority of cases (and without exception in males), the disposition
of the traits in the series is absolutely specific, but it differs from the disposi-
tion of the same traits in yellow-necked mice and deermice in that the
variability of body length appears to be less than the variability of tail
length. It is important to emphasize again that, on the average, all the
variabilities for all the traits considerably overlap (6.9 to 10%, body
length, 6.1 to 10.4%, tail length; 5.7 to 7.3%, ear length). This means
that it appears to be impossible for traits to maintain their disposition in
the samples with regard to variability because of specificity of coefficients
of variation.

There are data in the above-mentioned work of Adamczewska which
allow one to compare variability not only of body measurements, but also
of six different measurements of the skull in one population in different
months. ‘The results obtained are not as uniform as those secured in the
comparison of body measurements, but the presence of a general trend
toward a similar disposition of traits is undoubtedly there:

Males Females
October 165 4 3 2* 615432
November 561342 156432

* 1—condylobasal length; 2—diastema; 3—length of palate; 4—width between
orbits; 5—skull height; 6—height of braincase.

In these samples it is seen that traits 1, 6, 5 are always positioned
in the left half of the series and traits 4, 3, 2 are always in the right half.

216 Variability of Mammals

It should be remembered here that the coefficients of variation for a number
of traits are very similar and an accidental change of place in traits is possible
(see below).

In the work of K. Ya. Fateev (1962) on the morphology of moles from
the Kostrom region, data on weight of different organs of males and females
caught by the same method in different seasons during 1958, 1959, and 1960
are presented. In Table 78 the calculated error (+5s,.,.) of coefficients
of variation is added in each case to the coefficients of variation presented
in this work (a necessity when comparing coefficients of variation of samples
which differ considerably in their values). |

Table 78. Variability (с. v.+s5;.,.) of weights of certain organs
in one population of moles in different years (According to Fateev, 1962,
with calculated error s,_,_)

1958 1959 1960
Weight of ©

organ Males Females Males Females Males Females
1—Body 16.1+2.1 23.843.1. 13.341.2 13.8+1.3 8.5 1.2 5.70.8
2—Heart 39.5-5.3 44.345.9 20.8+2.0 26642.6 17.342.5 16.7+2.4
3-Liver 28.443.8 32.544.3 18.741.7 23.642.3 . 13.742.0 19.642.8
4—Lungs 23.6-3.2 17.842.4 19.541.9 17.242.9 21.243.1 15.742.3
5--Kidneys 30.54+4.1 34.044.5 17.641.7 13.341.3 19.742.9 21.1+3.1
6-Spleen — — 43.2--5.1 29.94+3.5 37.745.2 18.2+2.6
7-Stomach 51.8+6.9 27.843.7 26.942.6 47.24-4.5 38.9+5.6 18.2+2.6

Let us try to compare the variation of weights of different organs in
different years.

If the traits are arranged in the order of increasing value characterizing
their variability, the following picture emerges :

Males Females
1958 о ил)
1959 1534276 5143967
1960 1325467 1427356

* Character 6, spleen weight, was not recorded in the year 1958.

Without disturbing the traits in the series, these series could be divided
as follows for analytical convenience:

Males ~ Females -

27 ea О 9 fo
а 26 и 6

и, О Weak oe) BiG

Some Regularities of Variability 217

On examining these series, the following conclusions may be drawn:

1. In the left half of all the series the males have traits 3 and 1; the
females have traits 1 and 4;

2. In the right half of all the series, the males have traits 6 and 7;
for females, only trait 6 appears in the right half of all the series;

3. The series of the traits arranged according to absolute values of
coefficients of variation are more similar within each sex group than between
the sexes.

To analyze these samples further it is necessary to determine somehow
the “‘reliability” of disposition for each of the traits in a particular place
in one or another series, because many coefficients of variation differ from
each other. Even with considerable differences in the mean values of
coefficients of variation, the differences could seem insufficiently reliable
because of large errors caused by the size of the samples examined. For
example, traits 3 and 4 seem to differ from each other appreciably in the
year 1958, 28.4 and 23.6% respectively, but their larger errors (3.8 and 3. ap)
make these differences less clear-cut and in fact insignificant.

Let us say that the differences could be considered “‘significant”’
If С. 0.3—C. V.g>S,41 + ор 16., if the conditions were deliberately
made to give more significant differences than do the ones usually used in
statistics.

The following picture emerges if the series are rewritten after considering
the significant differences between variability values of individual traits:

Males ` Females
1958 45 2 41 (3,5, 7) 2 —
1959 В: 52) 76 (54) 3267
1960 #3 (24. 5) 67) TED (Dias Seiad Weve 0)

Now, taking into consideration the possibility of a chance in place
of the traits put into parentheses (because of the absence of detectable
differences in the values of their coefficients of variation), let us do the
final possible rearrangement :

Males Females
1958 134527— 415372 —
1959 1345276 1453267
1960 1345276 1453267

These lines make it possible to’ draw more specific conclusions, in
principle, coinciding with those which have been drawn above for the
linear measurements of forest mice and deermice. The disposition of
traits in variability during the period of three years is practically the same

218 Variability of Mammals

in males; such a conclusion is not completely correct in the case of females
but it cannot be overlooked that, even in this group, the sequence of
variability of traits is specific and quite similar from year to year. _

A question arises of whether such a uniform distribution of variability
of a trait is merely a reflection of a constant coefficient of variation for every
trait. However, a simple comparison of the coefficients of variation for
traits in different years shows that this hypothesis is not correct. Thus
for males, the variability value of body weight ranges in three years from
8.1 to 16.1%; of the heart, from 17.3 to 39.5%; of the liver, from 13.7 to
28.7, 2/ ЕЕС:

And the large variations in variability notwithstanding, the causes
ordering the majority of the traits in the series have a tendency to keep
constant from year to year. Hence, though the absolute value of coefficients
of variation for the same trait in the same population can vary considerably
in different years, the relative value of coefficients of variation (the position
of a given trait in the sequence of other traits with regard to variability
value) has a tendency to remain constant.

On the basis of this conclusion, I now propose that the phenomenon
of comparative variability, or the variability of disposition of traits, be
singled out from the usual study of variability of organs and structures
in animals.

Analysis of Comparative Variability for Some Mammals

The parallels in the comparative variability of the same body measure-
ments of mammals like forest mice and Peromyscus, have already been seen.
In these species, the comparative values of coefficients of variation for four
body measurements completely coincided. The results of other researchers
(Clark, 1940) are in accord with this:

P. maniculatus (30 populations)

I subspecies в II subspecies 6953142
Ш subspecies 653412 IV subspecies 653412
V subspecies 6539412 VI subspecies 635412
VII subspecies 653412 VIII subspecies 635412

P. leucopus (8 populations)
ча

Р. eremicus (4 populations)
653312

(Length of : 1—Боду; 2— а; 3—feet; 4—thigh; 5—lower jaw; 6—skull).

Some Regularities of Variability 219

In the few instances of some other disposition of coefficients of variation
(1, 4 in one case, and 3, 5 in another two cases), it seems that the values of
coefficients of variation are so close to each other that there are no detectable
differences between them (4.3+0.26% and 4.2+0.30% in one case,
2.88+0.21% and 2.92--0.22% in another case). Only in one case is
there a formal difference in the values from the general disposition of co-
efficients of variation, and that is for the VIII subspecies of P. maniculatus;
the coefficients of variation for the third and fifth traits are 2.75--0.07 and
3.05-40.4% respectively.

A comparison of these data with those presented above (Sumner,
1924) shows that the disposition of the same traits is similar in the sequence
of variability. However, the data presented do not agree with the data of
D. Hoffmeister (1951) on another species, P. truei, in which the arrangement
of the traits in order of variability is : 6, 2, 1, 3 (traits 4 and 5 were not
investigated). The data of Hoffmeister and Lee (1963) shows that in one
more species of this same genus, P. merriami, the arrangement of the traits
corresponds well with the data of Clark and Sumner : 3, 4, 1, 2 (traits 5
and 6 were not investigated).

Many times the variability of incomparable traits has been investigated
in different species of mammals. Such a situation restricts the comparison
of comparative values of variability to only closely related forms as a rule.
But even such a comparison appears to be interesting.

From the data sent to me for analysis by J. Dynowski, it is possible
to compare variability of traits in six age and sex groups for a series of
measurements of the body and skull of house mice (Table 79). First of all,
we shall examine those groups of traits concerned with body measurements
(2-body length, 3—length of tail, 4-length of feet, 5—length of ear):

Age group Males Females
а в Е BS) > By as
Ш 2.3 24 ИЕ |
ТУ ЗЧ 5 4

It is seen that the arrangement of coefficients of variation coincides well
in the first two age groups in both males and females. In the third age
group, traits 2 and 3 change places for females but, as has happened in
examples presented previously, the differences between the values of
coefficients of variation are so insignificant compared to the error (6.9+1.2
and 6.1-+1.0% in females) that such an arrangement of the coefficients of
variation could even be accidental. =

The analysis of distribution of coefficients of variation for skull traits
shows a generally similar picture (Table 80).

220 Variability of Mammals

Table 79. Variability of some traits of male and female house mice
(Calculated from data obtained from Dynowski)

Body weight

ii) AK 11.90.29 2.98 25.0-Е 1.70 108
Фо 12.60.41 3.89 30.9 --2.28 92
ПО on a 14.4+0.26 eal 15.9+1,26 80
ФФ 15.7 --0.52 3.48 22-2 |-2.34 45
му. ge 15.80.73 2.93 18.5-- 3.28 16
Фо 15.01.11 3.50 28.35.22 10

Body length

Е 74.9--0.70 7.38 9.9 + 0.66 110
ele) 75.30.75 7.50 10.00.71 99
To а 80.6 + 0.61 5.42 7.7--0.53 80
ele) 83.6+1.10 7.90 9.5 + 0.93 52
IM о О REP 6.9+1.19 17
99 84.82.22 8.02 9.5+ 1.86 8.

Tail length

i WR 67.0+0.58 6.07 9.140.61 110

fee) 68.0 +0.56 5.63 8.30.59 100
Die as 72.2+0.56 5.04 7.00.55 80
22 73.4--0.83 5.97 8.10.81 50
У. oo 73.91.09 — 4.50 6.11.04 И
оо 76.52.21 7.96 10.4 12.04 13

Foot length

SEES a 16.8-+0.06 0.65 3.9+0.26 108
ФФ 16.9--0.07 0.66 4.00.29 94

Ui es Cavey 17.0+-0.07 0.62 3.7+0.29 80
; ele) 16.8 --0.09 0.66 3.90.39 52
ТУ 52 17.10.12 0.50 2.90.50 17
oo 17.0+0.16 0.58 3.4+0.67 13

Ear length

Less, 12.5--0.08 0.83 6.60.45 110
ФФ 12.50.09 0.91 7.30.55 97

Gul Ge) 13.3+0.11 0.76 5.7+0.57 50.
и. да 13.1+0.19 0.75 5.7+1.01 16

ФФ 13.40.27 0.98 ‚ 7.31.43 13

Some Regularities of Variability 221

It is seen from the final data obtained as a result of rearranging
coefficients (after taking into account the possible errors), that even the
skull measurements may well be ordered quite similarly in the first two age
groups of males and females and in a very close but somewhat changed order
in the third age group. The values of coefficients of variation of comparable
skull measurements overlap considerably ; this means that it is impossible
to explain a regular arrangement of coefficients of variation in a sequence
by an absolute variability inherent in each trait.

Together with such easily calculable data, more complex cases have
been met which show that the concept of a comparable variability is far
_ from complete in spite of the general regularities already noticed and some
other reasons yet unknown which influence the regularities previously
observed.

In the work of Saint Girons and van Bree (1962), factual data are
presented on variability for a number of traits in three populations of
forest mice (Apodemus sylvaticus) of North Africa.

Table 80. Arrangement of skull measurements in different
age groups of a population of house mice in decreasing
order of coefficients of variation (Calculated according
to the data of Dynowski)

° Distribution according to absolute value, с. и.

Е: (uncorrected data)
Males Females ^
II 7 610 9118 6710 911 8
III i YQ 10) ie Й ОО ©
ТУ О. О И, B®) 7910 6 811
Distribution after taking error into
consideration, с. и.
II 7 (6; 10) (9. Ш) 3 (67) NWO (9 by 8
III (6, 7) 10 (11, 9) 8 (6, 7, 9, 10) 11 8
IV (G 10) Gh 8» GD) GG, в sy)
Final possible distribution
II 7610 Os: 7610 9118
Ill 7) 0) И © Фе
ТУ 6 10 711 89 79 61011 8

The coefficients of variation for body measurements calculated on
the basis of these data (Table 81) are arranged in two populations in exactly
the same way as they were in the Polish population of this genus : 3, 4, 1, 2.
The arrangement of the coefficients of variation in the third population
is different : 1, 2, 4, 3. Even after a possible rearrangement of traits
1, 2, and 4, the comparative variability of this population will differ from

222 Variability of Mammals

the one which is characteristic for the rest of the populations of the species.
The comparative variability of skull traits also seems irregular (Table 82).

It can be said only in a most general form that there are certain common
characters between the variability of the populations (traits 7 and 5 are
always on the left and traits 10, 11, 8 are on the right). However, the
coincidence of arrangement for coefficients of variation is far from complete.

Table 81. Variability of body and skull measurements of three
populations of forest mice (Apodemus sylvaticus) of North Africa =
(Calculated from data in the work of Saint Girons and van Bree, 1962)

a nS SS

Trait 6. 01-Е 5с. у. 6. 9.2 tSe.y.2 6. 0-3 Е 5..3
Length of
Body 6.7-£ 1.93 8.9+1.06 — 5.5+1.03
Tail 6.9--2.17 9.2 + 1.09 6.14 1,19
Feet 3.71.05 5.90.68 9.5 -- 1.79
Ear 4,341.24 6.7-£0.85 бе =
Condylobasal length 2.70.85 4.9--0.63 3.8-£0.77
Width of skull
Zygomatic 3.6-+ 1.44 5.2 -+0.67 3.6-0.77
Maximum 1.0--0.32 2.7--0.34 2.7+0.65
Minimum 3.71.18 4,3+0.49 4.2 40.83
Diastema 2.9+0.83 6.30.73 3.40.66
Length of lower jaw — 3.8+ 1.08 5.6+0.65 3.8-L0.75
Upper dental row 4.8 -- 1.37 5.0-=0.57 5.00.98
Lower dental row 3.3+40.94 3.90.44 4.6 0.90

Table 82. Comparative variability of skull
measurements for three populations of
forest mice of North Africa

According to absolute values, с. о.

I 7 5912 68 10 11

II 7128 511610 9
III 7 96.510 8 12 11

With calculation of error, 5¢-y,
I 7 (5, 9, 12) (6, 8, 10) 11
II 7 12 (5, 6, 8, 11) 109 >
ш й (5.6, 9. WO) 8 GBs Wb)
Possible final

I 7 591210 6 811
II Й 26 809
Ill 7 59 610 8 12 11

м Some Regularities of Variability 223

Data on variability of some traits in 11 populations of bank voles
(Clethrionomys glareolus) have been presented in the work of V. N. Bol’shakov
and 5. 5. Schwartz (1962). Because of the small size of some samples,
I have analyzed data for only seven populations (Table 83).

Table 83. Geographical variability (с. v.4-5,.,.) of adult bank voles
(Clethrionomys glareolus) (Calculated from the data of Bol’shakov and Schwartz, 1962)

Population Region body a oF bar
I Kuvandik 8.7+1.07 11.0--1.40 7.1+0.87
II Bashkiria 3.7+0.55 6.2+0.92 8.11.19
Il Raskuikha* 6.7--0.69 12.6 -Е 1.30 9.5 -+-0.98
IV. Pishminik . 5.60.74 8.0+1.05 9.9+-1.28
У Verkotur’e** 6.30.85 8.51.16 5.6+0.76
VI Deneshkin Kamen 6.0-+0.83 11.8+ 1.63 7.3+1.01
VII Komi ASSR 5.00.67 5.40.72 8.3-+1.10

* Perhaps, males and females together.
* Sverdlovsk region.

A comparison of coefficients of variation for body measurements shows
that the comparative variability is similar in five of the seven populations:

According to After rearrange- According to After rearrange-
observed ment with calcu- observed ment with calcu-
values, с. о. lation of error values, с. и. lation of error
5%.с 5.5
I ти Shee eas 2: Мох Иа 3142
II ia nati: 3 12 4 м Belted 2 Silos 42
Ill За Ве VAT 34 Srl 2-4
IV 11342 Зо

a a ee ll

Let us note, as an aside, that the sequence of variability of body
measurements in bank voles differs from that in forest mice and deermice.
The differences in the absolute values of coefficients of variation are so great,
as is the case also in all the instances examined above, that it seems impos-
sible to relate their arrangement to a general variability of individual traits.

A general analysis of variability for eight measurements of the skull
in the same populations of bank voles (Clethrionomys glareolus) confirms the
presence of a specific similarity in the disposition of coefficients of variation
for these traits between different populations which as before, could not be
related to a constancy of the variability values of different traits :

224 Variability of Mammals

Population After correction with standard error, 5с.у =
I 12 11 5 8 10 7 9 6
II 12 11 5 8 10 6 7 9
Ш 8 12 5 11 10 6 7 9
ТУ 11 3 5 10 6 12 7 9
У 12 5 8 11 10 6 a 9
VI 12 11 5 8 7 6 9 10
VII 12 5 8 10 7 9 12 6

It should be noted that the populations studied up to now include
males as well as females. This undoubtedly makes it difficult to under-
stand the natural situation, because in the data presented for the house
mouse and mole it was always observed that the comparative variability in
males is not similar to that in females. In addition, the authors do not
give the date of collection of material for each of the populations. Judging
from the geographical distances between the populations—South Urals to
Sverdlovsk region and Komi, USSR—it is very probable that the data were
collected in different years; if this is the case, the natural picture could.
have been distorted by the effect of varied differences in elements of geo-
graphical, biotopic, and chronological variabilities. These same observa-
tions are similarly applicable to the following data on red-backed voles
(Clethrionomys rutilus) by the same authors (Table 84).

Table 84. Variability of body and skull measurements (с. v.=-5;.,,)
of mature red-backed voles (Clethrionomys rutilus)

Kurgan Shipitsino, Denezhkin Yamal
Trait region Sverdlovsk Kamen В. Polui В. Khadita
n=69 region n=40 n=94 п= 105 n=93

Length of

Body 7.0+0.60 5.9+0.56 9.9+0.72 6.3+0.43 9.5+0.92

Tail 13.64+1.16 7.7--0.86 9.14+0.66 14.6+1.00 13.8+1.34

Feet 4740.40 3.440.388 11.4+0.83 4.8+0.33 6.0+0.58

Ear 11.1+0.94 7.9+0.38 — 10.50.77 — —
Condylobasal length 3.30.28 7.30.82 5.80.42 3.10.21 4.50.44
Diastema 5.9+0.50 5.2-+-0.58 6.6+0.48 4.4+0.30 7.3+0.71
Width between orbits 4,440.37 3.2 +0.36 4.90.36 5.40.37 3.70.36
Zygomatic width 5.2+0.44 4.8-+0.54 5.10.37 6.3-40.43 6.2 0.60
Height of zygoma ‘3.80.32 2.8+40.31 3.2+0.23 3.50.24 2.50.24
Length of dental row 5.40.46 2.7+0.30 4.1+0.30 4.4+-0.30 3.2-0231

Some Regularities of Variability 225

The pattern of coefficients of variation in body measurements (traits
1 to 4) differs sharply only in population III from that which characterizes
the other three populations. ‘The skull measurements show, аз before,
a more mixed picture of comparative variability for the red-backed vole.
Traits 11, 5, and 12 are little variable in all populations and traits 6 and 8
are more variable. ‘The places for the rest of the measurements are not
exactly fixed in the sequence:

Population Corrected data with calculated error, 5¢.y,
I 3 42 11 Beg DVO Ne ОВ
II Зо ИИ КО о)
Ш РЭ И WD бо о 0 7) 9
ТУ Е Ее 11 Se ae Ome Sars 2 Oa
У И Л 12 9 §& й © 10 8

Viewed from a general analysis of the comparative variability of
skull measurements, an assumption can be made to some extent explaining
the irregular arrangement of coefficients of variation in a series. When
the body measurements are compared, traits having different relations to
the controlling factors are undoubtedly compared (length of ear helix,
length of feet, length of tail—each of these measurements has a specific
ecological importance). When one compares many measurements of the
skull, it is possible that traits having “‘equal relation’”’ to natural selection
are compared, the variability of which is controlled to a great degree by,
say, genetical factors. It is natural that in such a case the relative variability,
controlled by natural selection, will not be manifested so regularly.

To continue the analysis of comparative variability of traits in
mammalian populations, I shall consider the_results obtained during a
comparison of variability for three populations of common hamsters (Citellus
citellus), calculated from data by Straka (1963) (Table 66). It appears
that the variability for length of body (1), of tail (2), of feet (3), and of ear
(4), form identical sequences of comparative variability (3, 1, 4, 2) for the
populations living at 50, 650, and 1,900 meters above sea level. The
positions of the coefficients of variation in the sequence are maintained
in all cases despite appreciable differences of absolute values of organs
and despite a scattering of values of the coefficients of variation.

All the above cited examples of comparative variability belonged to
various species of rodents. The material presented in the work of V. A.
Dolgov (1963) makes possible an analysis of comparative variability for
body and skull measurements in populations of three species of insectivores

(Tables 85, 86, 87). -

226 Variability of Mammals

Identical sequences of variability are observed in each of the three
species of shrews from the data on comparative variability. There are
also identical sequences of comparative variability between the common
shrew ($. araneus) and 5. isodon while the lesser shrew (5. minutus) differs
from the first two species in the sequence of traits in variability (the position
of coefficients of variation for body measurements and the position of

Table 85. Variability (с. v.+5;,y,) of some traits of the common shrew
(Sorex araneus) (Calculated from data of Dolgov, 1963)

дж

Males Females
Trait. 2) 52305 SS SSS SS Ss a a

Immature Mature Immature Mature

1. Body weight 8.2 +£0.37 9.7 +£0.68 9.2+0.41 9.7 +£1.08
2. Body length 2.8-40.12 3.1+0.21 2.9+0.13 3.5+0.31
3. Tail length 5,440.24 6.0-=0.41 5.70.24 6.4-Е0.68
4. Foot length 3.440.15 3.4+0.23 3.9-+0.15 3.5+0.31
5. Condylobasal length 1.5+0.07 1.6-L0.11 1.7-L0.08 1.7-- 0.15
6. Maximum width 1.9--0.09 1.90.13 1.90.08 1.80.17
7. Height 2.4+0.11 2.40.17 2.60.12 2.50.22
8. Inter-orbital width 3.0+0.14 3.8 +0.26 3.6+0.15 3.80.33
9. Pre-orbital width 3.9-40.17 3.50.23 3.60.16 3.5+0.31
10. Upper dental row 1.9-40.08 1.9-L0.14 1.9-=0.08 1.9-=0.14

Table 86. Variability (с. 0.5...) of some traits of the lesser shrew
(Sorex minutus) (Calculated from data by Dolgov, 1963)

ey eee eC, —

Males Females
а ee аи
Immature Mature Immature Mature

1. Body weight 5.80.80 6.50.56 7.5 1.14 —

2. Body length 3.20.47 2.7+0.23 2.40.43 2.3-40.66
3. Tail length 3.8-+0.53 5.8+0.50 4,2 +0.69 4.5+1.20
4. Foot length 2.5+0.35 2.4+0.20 2.7-40.45 1.7-0.45
5. Condylobasal length 1.6+0.25 1.3+0.11 1.30.25 1.6--0.42
6. Skull width 2.4--0.43 1.6+0.14 2.0+0.40 2.6+0.69
7. Skull height 2.10.36 2.9+0.25 3.30.68 2.7+0.72
8. Inter-orbital width 2.30.34 2.6-+0.22 3.0-=0.51 1.50.40
9. Pre-orbital width 3.80.57 3.60.31 3.70.40 3.70.98

10. Upper dental row 1.6+0.24 1.30.12 ° 1.70.30 1.9-=0.51

Some Regularities of Variability 227

Table 87. Magnitude of variability (с. v.+-5;,),) im a shrew (Sorex isodon)*

Trait Male Female Trait Male Female
1. Body weight 6.7--0.97 6.7-1.19 7. Skull height 2.0+0.30 2.0-0.37
2. Body length 3.10.43 1.3+0.22 8. Inter-orbital
3. Tail length 4.3--0.60 3.1-+0.55 width of skull 3.4+0.48 3.7+0.65
4. Foot length 3.4+0.47 1.9-+0.34 9. Pre-orbital width
5. Condylobasal of skull 4.20.59 2.7-+0.47
length 1.7+0.27 1.9+0.33 10. Upper dentalrow 1.7+0.25 2.0+0.35
6. Skull width 1.6+0.24 1.9-+0.36

* Only immature.

coefficients of variation for some measurements of skull). Characters 1
through 4 of Tables 85 through 87 differ from those in the sequence

diagram.

Sorex minutus Sorex araneus
Sees +2) 4: 5 0 © 8 7 3:2 a 5 OO 10 7.8 9
ЗИ 4 ) IO 6 8 7 п йе 5 & IO 7 99
Sd 24+ DS 10 © 8 7 9g №524 бое 9
ЗИ 2-4 ВОО 9 ПС 5 6 10 7 8 9

Sorex isodon

По Or ПИ

PB Dod BG Ore 9 8

* After corrections.

Data characterizing the skull measurements of four types of shrews
from Yamala are given in the work of Schwartz (1962). Parameters of
variability calculated on the basis of these absolute measurements are
presented in Table 88.

The samples under investigation are not of similar sex and age and
are not very large, all of which makes one approach the results rather

cautiously :

Without correction | After correction
5. minutus Oe Ses ЕЯ 9 Ой
5. arcticus Sie OMG mae ОЭ BO) 6:8 97
5. daphaenodon 5 — 10 9 — 8 в 9
5. araneus 9108 й ® 5 10 9

298 Variability of Mammals

Table 88. Variability (с. v.+:5¢.y.) of traits for shrews from Yamala
(Calculated from data by Schwartz, 1962)* |

Sorex araneus ~S. daphaenodon,
Trait Segoletki females breeding — 5. arcticus 5. minutus
in Segoletki

5. Condylobasal length 1.90.30 2.10.48 2.9-+.0.46 1.40.26

6. Skull width 3.20.50 = 3.4+0.55 3.2+0.59

7. Skull height 4.6+0.71 — 6.30.99 5.2 0.94
8.-Inter-orbital width 811045 = 6514 4.1-- 0.64 4440.80 -
9. Pre-orbital width 5.00.72 5.51.16 5.50.86 1.10.19
10. Upper dental row 2.7+0.41 2.80.63 5.10.81 3.60.66

О en

* The time of collection of the data has not been mentioned exactly; for the third graph,
the males and females are combined.

As can be seen, after correction, all the traits correspond well in all
species except trait 9, which retains its position at the head of the line for
Sorex minutus (which itself has to head the line). The non-coincidence of
this trait is so significant that it appears that an accidental mistake may have
(unwittingly) been committed while calculating the material.

A comparison of these data with the data obtained by Dolgov from
the Oksk region shows that in the Yamala samples, the comparative
variability of the same traits appears to be similar in all the four species
investigated, and is similar to only Sorex minutus of the Oksk collection.
Whether this is caused by specific natural conditions of Yamala—an extreme
region for all Sorex—or can be explained by an insufficient study of the
within-species taxonomy, is not fully clear. But it looks as though the
study of the phenomenon of comparative variability permits one to throw
light on relations which have slipped from the attention of ecologists and
morphologists. j

All the series of comparative variability presented above were con-
cerned with body measurements, skull measurements, and weights of
individual organs. It appears that variability of color in different popula-
tions within a species polymorphic for color, has a similar tendency to main-
tain the sequence of traits. In Chapter IV, data on variability of photo-
metrical indices for the posterior and lateral sides of Peromyscus maniculatus
are presented. After a small number of rearrangements, the individual
color indices for eight different populations arrange themselves (in order of
increase in variability) in the following manner:

Some Regularities of Variability 229

I ВТ У GON By ВУ У RE), У С BY BY
II ВГУ Со BOMB: VI ВС В в
III Rel Ge baw bv, VII RES Yu Gy. в ву
IV RT У В С BV VIII Во У С АВВ

(RT—Reddish tinge; Y—Yellow; G—Green; B—Blue; BV—Bluish-violet).

For all populations together, the values of the coefficients of variation
for all the shades overlap each other considerably : RT—11.0 to 16.6;
Y—11.3 to 16.3; G—13.0 to 18.3; B—13.8 to 21.3; BV—14.7 to 26.1%.
This means again that such maintenance of trait positions cannot be ex-
plained through constantly very low or very high values of variability for
a given trait. It is interesting to note here that the comparative variability
of color shades of the back does not show as much uniformity as has been
observed for the variability of color shades on the sides of the body.

Interesting data on variability of color are presented in the work of
Hayne (1950), who studied photometrical indices of color for nine different
populations of P. maniculatus from Florida and Alabama. А comparison of
absolute photometrical indices shows an identity inithe position of bluish-
violet, green, and red shades in all the populations (red has the greatest
value, and bluish-violet has the least).

A comparison of variability values shows practically full overlapping
of coefficients of variation for all three shades in different populations.
But in comparing positions in order of increased coefficients of variation
for shades within each population, a more complex situation appears:

I RT В G У RT G B
II отв УТ ВТ с В
Ill G Bo Ra УП В @ RT

IV G B RT VIII B G RT
IX G RT B

(RT—Reddish tinge; G—Green; B—Blue).

It can be seen that populations Ш and ТХ, III and IV, У and VI,
and VII and VIII appear to be the same in the sequence of traits in order
of variability. There is reason for thinking that such similarity is not
accidental but is determined by common features of the environment,
and here the analysis of comparative variability values points to phenomena
which are overlooked in other approaches.

An examination of variability for a number of traits within one popula-
tion, as well as the comparison of relative variability of traits between
different populations, is part of the widespread subject of “comparative
variability.” The phenomenon observed is not universal and it is not

230 Variability of Mammals

completely specific in some cases. The available facts of course are
insufficient to describe comprehensively the character of this phenomenon,
but based on the general conclusions it is possible to advance an hypothesis
of ‘‘comparative variability,” the basic aspects of which can be formulated
in the following manner :

1. In mammalian populations, variability of all traits is in a mutually
coordinated condition during each moment of evolution;® and

2. The variability value of each trait changes for different specific
reasons, but these changes are related to changes of variability of other
traits also, as a result of which the position of this trait has a tendency to
keep constant in the series of traits.

The corresponding positions of coefficients of variation of the same
trait in the sequence of variability of all traits in some populations, and their
different positions in other populations and species, allow one to assume
that there are specific reasons for the existence of such a phenomenon.
It is possible that these reasons play a major role in the formation of the
whole morphological appearance of different populations.

It is obvious that a parallel detailed ecological analysis of the group
under study is necessary for analyzing these reasons. ‘The experimental-
ecological approach could throw light on the relationship of variability
to structures having some importance to the life of separate individuals
as well as to the whole population. The study of micro-evolution could
not even be thought of without such an analysis.

But on the other hand, even the most exact experimental-ecological
analysis appears to be incapable of solving evolutionary problems without
subsequent extrapolation of ecological data obtained in the field of popula-
tion morphology, and without testing by genetical methods the results
obtained. Hence the essential aspects of the study of micro-evolutionary
processes are population morphology, genetics, and ecology.

The hypothesis of comparative variability advanced above is an example
of such a situation, when the regularity discovered by morphological data
cannot be understood without an ecological understanding of the morpho-
logical material, and without testing the results thus obtained by exact
genetical experimentation. The importance of a morphological approach
to population biology is thus confirmed.

6 This situation has something in common with the phenomena observed by P. V.
Terent’ev (1959) under the name of “correlation pleiades’* because it is possible that the
correlation pleiades of the traits underlie their mutual variability.

CHAPTER VI

VARIABILITY AND THE PROBLEM OF VESTIGIAL ORGANS

After an examination of the problems of “selection and variability,”
a chapter devoted to the manner of phylogenetic formation of organs and
structures—namely, vestigiality and vestigial organs—follows naturally.
An attempt must be made to show that population morphology can throw
light on a number of confused problems of evolutionary morphology.

A Short Historical Account of Viewpoints on the
Problem of Vestigial Organs

The concept of “vestigial organs”’ is often encountered in the morpho-
logical literature. Modern understanding of this concept was mainly
formulated by Charles Darwin, and in order to study the problem objectively
it is logical first to draw attention to his work. ‘Useful organs, however
little they may be developed, unless we have reason to suppose that they
were formerly more highly developed, ought not to be considered as
rudimentary’ (Charles Darwin, Sobor. soch., Volume II; р. 637; ed. 6,
р. 440).1 A few lines later, Darwin adds: ‘“‘Rudimentary organs, on the
other hand, are either quite useless, such as teeth which never cut through
the gums. . . . They have been partially retained by the power of in-
heritance, and relate to a former state of things” (p. 638; ed. 6, p. 440).

Darwin also lists the wings of penguins, milk glands of male mammals,
the reduced portion of lungs in serpents, teeth in the embryos of baleen
whales and ungulates, as further examples of vestigial organs. He comes to
this general conclusion: “Г would be impossible to name one of the higher

1 The 1939 edition has been referred to here and hereafter,

232 Variability of Mammals

animals in which some part or other is not in a rudimentary condition”
(ae C802 Gelb ©, jo.) 429))).

There are some contradictions in the understanding of the concept
of vestigial organ even in the above-cited statements. In the first citation
which determines an organ as “‘rudimentary,”’ stress is given to the peculiari-
ties of its structure, and whether it was well developed before—its usefulness,
its functional importance, is not the deciding factor. In the second citation,
complete emphasis is given to another criterion, the functional utility of the
organ. ‘This concept of a primacy of form in the definition of vestigial organs
becomes more difficult: “Ап organ, serving for two purposes, may become
rudimentary or utterly aborted for one, even the more important purpose,
and remain perfectly efficient for the other. . . . An organ may become
rudimentary for its proper purpose, and be used for a distinct one: in certain
fishes the swim-bladder seems to be rudimentary for its proper function of
giving buoyancy, but has become converted into a nascent breathing organ
or lung”’ (р. 637 ; ed. 6, р. 440). :

Such mixed functional and morphological criteria in the understanding
of the concept of vestigial organs has been retained even to this day in
many works. А. М. Severtsov observes that “‘Vestigial organs are absolutely
useless, more or less dwarfed and secondarily simplified’? (1939, р. 554).
I. I. Schmalhausen gives essentially the same specifications: ‘Vestigial
organs are those organs which have lost their significance, and are re-
presented in the form of insignificant remnants; their position, correlation,
development, and sometimes even their structure indicate their origin from
the more developed organs of their direct ancestors, in which they played
a more significant role” (рр. 13-14). A few pages later, Schmalhausen
again turns to this problem and writes: “Strictly speaking, оЁ course, it
could not be said that vestigial organs have no function at all. They remain
to a less extent as the active point of some aspect or other of morphogenetical
regulations . . . .” Further on he states: ‘“‘Vestiges preserved in the
adult state usually have a specific function, even though it may be insigni-
ficant and generally has nothing really in common with the previous
function”’ (р. 30). .

There is no point in pondering over many such definitions occurring
in various works. The above-given definitions suffice to reflect clearly
the existing conditions. And the reader is no wiser for knowing that there
are about 100 vestigial organs in man because here again there is a very
imprecise definition of the concept: “. . . remnants of the fully deve-
loped organs of ancestors” (Villey, 1959; p. 588). Such vague definitions
of the concept. of vestigial organs are characteristic of works in the field .
of evolutionary morphology and general biology. А. A. Malinovskii (1939)
writes: “. . . However, besides the traits vital for life, there should
also exist such traits which do not have any significance for the organism”

Problem of Vestigial Organs 233

(р. 609). Later he adds: “. . . Actually, each small organ, if it has
no advantage for the animal, is in itself harmful” (p. 610).

It is clearly seen in all these examples that the vague or imprecise
understanding of the concept of vestigial organ present in the works of
Darwin, persists to the present day.

Recently, comparatively speaking, there has been a renewed interest
in the problem of vestigial organs in the works of 5. G. Kryzhanovski
(1950), G. V. Nikol’skii (1955), G. V. Nikol’skii and V. A. Pikuleva (1958).
G. У. Nikol’skii writes: “АЦ the organs considered to be useless (vermiform
appendix, wing of ostrich, small feathers in flying birds, visceral vessels
in the embryos of all vertebrates) appear to have an adaptive significance
after detailed investigations” (р. 725).2 С. У. Nikol’skii nowhere directly
rejects the presence of vestigial organs, but from his discussions (especially
in the work of 1955, p. 978), this conclusion is almost evident.

It is understandable that the problem of vestigial organs can only be
examined confidently by taking some concrete examples of the structures
of organs belonging to the category “vestigial.” A number of structures
of marine mammals will serve as such examples.

Structure of the Pelvic Girdle in Toothed Whales

From the time of Charles Darwin to the present, the two small bones
in whales in the place of the well-developed posterior extremities of terrestrial
mammals, have been considered to be a fine example of vestigial organs
(Prout, 1964). What are these pelvic bones and should they really be
considered vestigial ? To answer these questions, let us examine the
structure of pelvic bones in the common small whale of our northern and
southern seas—the white whale, Delphinapterus leucas.

The pelvic girdle so characteristic of all terrestrial mammals is
practically absent in the white whale as well as in all other cetaceans,
because there are no posterior extremities and the small bones are not in
direct contact with the vertebral column.

The pelvic bones in the male differ in position and structure from
those in the female. In the male they are larger (in similar-sized animals),
more curved, and their surface is rather raised (Figure 59). The pelvic
bones of the female are rather smooth, without sharp crests, and have a
generally rather round form. The difference between the absolute measure-
ments of these bones for males and females is not large proportionately:
the maximum length is 13 to 15 cm in males and 8 to 10 cm in females (when
the length of the animal is 3.5 to 4 meters).

2 It is best to understand that this viewpoint is rather close to the one held by I. I.

Schmalhausen (1947), who also considers that the vestiges always have a function however
insignificant.

234 Variability of Mammals

Figure 59. Construction and position of pelvic bones in male and
female white whales (Delphinapterus leucas).

The pelvic bones in the male are situated on the ventrolateral surface
of the bulb of the penis, at the base of the corpus cavernosum (Figure 59).
In the female, the pelvic bones lie on the posterior part of the abdomen
between the wall of the vagina and the mammary gland, as in other cetaceans.
The pelvic bones in the male are connected to the bulbocavernosus muscles
at the base of the penis; in the female they are connected to the muscles of
the vagina.

The presence of the pelvic bones makes penis erection possible in the
male; they help in the effective contraction of the vagina in the female.
These muscles make it possible to close the anal aperture in both sexes.
For these reasons, they are important functional structures.

In the evolution of cetaceans, as the aquatic mode of life became
prominent the tail became the main organ for locomotion. Initially the
pelvic bones served as a support for the posterior extremities. Many
muscles related to other systems of organs were attached to these bones.
Gradually, the function of this support had to dwindle and the morpho-
logical readjustment of the bones progressed, under the new conditions,
in the direction of more successfully carrying out the functions concerned
only with the muscle activity of the urogenital system and the anal aperture.
In present-day toothed whales (both male and female), the pelvic bones
have lost all contact with the function of body movement and appear to be
related only to the urogenital and digestive systems.

It seems that we have a typical example of principal function change
in an organ as understood by A. Dorn (1936). On the whole, the pelvic
bones as an organ do not lose their significance fer the organism but simply
change this significance. As a result, it would appear that the significance

Problem of Vestigial Organs 235

of the pelvic bones under the new conditions of existence is also vital for
the whole organism. The existence of these structures in cetaceans is perhaps
not less important than the presence of hind limbs for other mammals,
because the process of copulation and the normal function of the digestive
tract is only made possible by these “‘vestiges.”’

From the viewpoints of Charles Darwin, A. N. Severtsov, and I. I.
Schmalhausen, the pelvic bones of cetaceans should, no doubt, be labeled
vestigial organs because they were more developed in the ancestors of these
animals. But, as seen from the brief analysis of their structural and
functional significance, these structures not only retained their significance
for the organism, but their presence has been proved essential for the
normal functioning of life. It follows logically therefore that these
organs should not be labeled incipient, embryonic, or “rudimentary”’
(“‘rudimentum’’—unformed) on the one hand, or vestigial on the other.

Structure of the Vibrissal Apparatus in Cetaceans

It is possible to arrive at the same conclusions (i.e., those given in the
preceding paragraph), after examining the so-called “‘remnants of hair
cover” in baleen cetaceans. It is known that there are individual “‘hairs”’
on the head of whales and these are usually considered a very clear example
of vestigial organs (Kiikenthal, 1909; Beddard, 1900). However, the work
of A. Jafha (1910) shows that these ‘‘vestiges” have a very complex structure
and well-developed nerve supply for each “hairlet.”

According to Jafha’s calculations (1910), about ten thousand nerve
fibers, combined in big bundles, approach each of the vibrissae of the
blue whale (Balaenoptera musculus). Based on this fact, and also the
observations of A. Malm (1866) regarding the behavior of shoaling blue
whales on touching their vibrissae, Jafha concludes that W. Kikenthal’s
opinion (1909)—that the vibrissae are only “remnants of a previous hair
соуег”’—маз wrong. In Jafha’s opinion, the vibrissae can play a large role
in the life of the animal as tactile organs. A still more specific statement
about this has been made by Nakai and Shida (1948): “. . . the vibrissae
have a very important significance in the life of baleen whales because they
use them as instruments for feeling out food” (р. 47).

Recent investigations (Yablokov, 1963b; Yablokov and Klevezal’,
1964; Solntseva, т litt.) have confirmed the observations of Jafha and
Nakai and Shida. Results of histological investigations of vibrissae for
five species of baleen whales show tha it is a complex organ.

The general structure of this organ is shown in Figure 60. The
hair follicle extends deep into the dermal layer of the skin and is well
embedded in the connective tissue fibers. These fibers differ considerably
in size and disposition from similar ones in the connective tissue of the

236 Variability of Mammals

Figure 60. Diagram of structure of vibrissae and hair follicle (longitudinal section) of
(A) Fin whale (Balaenoptera physalus) ; embyro of common dolphin (В) (Delphinus delphis);
Greenland seal (С) (Pagophilus groenlandicus) :

1—Epidermis ; 2—Core of vibrissae ; 3—Hair follicle ; 4—Bundle of connective tissue fibers ; 5—Blood
sinuses ; 6—Cross-sectioned blood vessels ; 7—Bundle of nerve fibers ( Yablokov and Klevezal’, 1964).

dermis. While protecting all sides of the hair follicle, they connect the
follicle with the surrounding dermis and fix it in the hide to make it immov-
able. The thickness of the wall of the hair follicle varies somewhat in the
different species of cetaceans investigated; it is greatest in the hooded
whale and least in the sei whale.

Between the core of the vibrissae and the inner edge of the hair follicle
there are wide spaces separated by partitions. ‘These wide spaces are
located all along the portion of the vibrissae embedded in the dermis in the
fin whale and the blue whale, but are considerably less developed in the
hooded whale and the grey whale. In the sei whale, these spaces are wide,
asymmetrical, and found only in the upper half of the hair follicle imme-
diately beneath the epidermis. Many small blood vessels pass through the
walls of the hair follicle from the sides and from underneath the follicle.

In all the species examined, there are well-marked nerve tracts in the
base of the hair follicle, the endings of which are lost in the thickness of the
wall (Figure 61). At the bottom of the hair follicle there are epithelial
layers in the form of glands, which cause continuous growth in the vibrissae.

In baleen whales the vibrissae are found only on the head, where they
form separate groups. ‘The first and the most significant group in its numbers
is situated at the end of the lower jaw (see Figure 4). In the blue whale,
about 42°% of all vibrissae are found here; in the sei whale and the fin whale,
about 46 to 47%. In rorquals, the vibrissae in this region are situated in
two vertical rows. ; ; oe

The second group of vibrissae is arranged along the external edge

Problem of Vestigial Organs

237

ir follicle of a fin whale

he walls of the ha

ing into t

Bundles of nerve fibers pass
(Balaenoptera physalus) vibrissae. Photo by author from sample prepared by G. N. Solntseva.

Figure 61.

238 Variability of Mammals

of the lower jaw beginning about 0.5 to 1.0 meter aborally from the first
group. of vibrissae. Here the vibrissae are usually arranged in one row
though there are instances (fin whale) in which the vibrissae are arranged
in two rows; a second lower row of vibrissae always has fewer hair than
the upper row. Usually, vibrissae of this group extend to the end of the
lower jaw. A relatively larger number of vibrissae in this group are found
in the blue whale (38% of the total); in the fin whale, 31%; and in the
sei whale, only 24%.

The third group of vibrissae is situated on the dorsal surface of the head
and usually comprise four more or less well-defined parallel rows located on
both sides of the central crest formed from the intermaxillary bones, and
in certain rare cases, a number of vibrissae situated in the region of the
external nares. ‘The greatest number of vibrissae in this group, both rela-
tive and absolute, are in the blue whale (up to 50 in number; on an average,
38% of the total); the fin whale comes next (36 in number; average,
31%); and finally, the sei whale (18 in number; average, 24.6%) (Table 89).

Table 89. Vibrissa distribution on the head of
whales (Balaenopiera) (In % of total)
(From Yablokov and Klevezal’, 1964)

End of Sides of | Upper

НЕС mandible mandible jaw
Blue whale (B. musculus) 29.2 18.5 52.0
Fin whale (B. physalus) 47.7 23.3 29.0
Sei whale (B. borealis) 45.2 31.6 25.9

It is interesting that in the sei whale, vibrissae were not even once
observed between the external nares, while in the blue whale and the fin
whale, vibrissae were found there quite frequently.

The data presented above on the anatomical structure of vibrissae
and the manner of their disposition on the body of animals correspond well
with the already known ecological characteristics of different species of
baleen whales. Of the species investigated, the blue whale requires plankton
and crustaceans for food (Nemoto, 1959; Betesheva, 1961); hence for it,
the development of a tactile apparatus which permits an accurate orienta-
tion to the surrounding food field is most significant. Also, the blue whale
vibrissae are arranged in a very uniform manner on the surface of the head;
this fulfills the requirement of obtaining uniform information about the
disposition of an object which can be accidentally met in front, at the sides,
and under or above the head.

Plankton and crustaceans are found in the food of fin whales regularly,
but apparently not in as large a quantity for sei whales, whose stomach

Problem of Vestigial Organs 239

content in the region of the Kuril’skii Islands is mostly small fish. Cor-
respondingly, the number of vibrissae sharply increases at the end of the
lower jaw, on the very front receptor portion.? Ofcourse, the concentration
of a number of vibrissae on the end of the jaw in the sei whale (and to a
lesser degree in the fin whale), should be considered as an adaptation for
collecting more accurate and quick information about the quicker and more
elusive fish and squid shoals.

Thus, the differences in the distribution of vibrissae in ecologically
different types of whales are such that the characteristics of their distribution
can be viewed as adaptations to feeding habits.

Specific differences in the mode
of distribution of vibrissae are ob-
served between the sexes in the
same species. As seen from the
data presented in Figure 62, the
limiting line of the scatter diagram
of male fin whales completely
includes the scatter diagram of
females, showing considerably less
variability for number of vibrissae
in females. In the ecology of large
cetaceans, it is difficult with our
present knowledge to point to a
specific biological significance for
these traits, though some basis for
this is available in the work of
T. Nemoto (1959), who observed
И 900 food differences between

Mandible male and female baleen whales.
In principle, sexual dimorphism

Side of the mandible

10 20 30 49 50 60 70
End of mandible B

Upper jaw

Figure 62. Scatter diagram of distribution - ,
of whiskers on the side and at the end of the the development of Organs that

lower jaw (A) and in the upper and are considered to be “‘vestigial”’ is
lower jaws (B) of the fin whale: of great significance.

1— Males ; 2—Females. It is difficult to imagine that an

organ which has no significance for

an organism develops to a different degree in the different sexes; on the

contrary, it seems more probable that the presence of sexual dimorphism

in the development of an organism points to an important functional

significance for that organ.
In view of the above arguments, it is absolutely incorrect to consider

8 It should be noted that sight is excluded during the catching of a fish; the whale sees
nothing in his direct vicinity (for details, see Slijper, 1962; Yablokov, 1964).

240 Variability of Mammals

the “remnants of hair cover” growing on the body of whales as vestigial
organs. ‘They comprise a complex and specialized receptor which helps
in a special way in the finding of food.

Specialization and Vestigiality

The two examples of the so-called “‘vestigial organs” presented above
show that not only do these vestigial organs have a function in the organism
but they are highly specialized structures, perfected for carrying out com-
plex and delicate functions as in the case of the pelvic bones in the present
toothed whales, or vibrissae in baleen whales. The structure of these
organs was modified by a significant change in function at some time in
their evolution.

The basic criterion for defining vestigial organs, based оп their
relatively poor development as compared to similar organs of their ancestral
forms, does not serve the purpose adequately, because it does not distinguish
between the specialized organ and the vestigial organ.

If the pelvic bones of cetaceans are to be considered ‘‘vestigial.”’ then
why not also consider as vestigial the front limbs of these animals, which
have a simpler construction (in the number of muscles retained and the
absence of movement by different parts of the limbs, the structure of
skeletal, circulatory, and nervous systems, etc.) when compared to the
limbs of the present and past terrestrial mammals which are the direct
ancestors of cetaceans ?

Proceeding from these principles of determining the “vestigial” nature
of an organ from morphological criteria, why not consider as vestigial
the few remaining digits so characteristic for mammals but mostly lost by
the majority of ungulate mammals ?

Such examples can be increased considerably, and moreover, as shown
by A. N. Severtsov (1939), the reduction of an organ is caused by those
very processes (substitution of organs and functions, change of functions,
and decrease in the number of functions) which cause the specialization
ofanorgan. This situation serves as one more argument for the impossibility
of an exact demarcation between specialized organs and vestigial organs.
as they have been understood in the past.

One example from the already presented data on the structure of the
vibrissae in cetaceans, illustrates this conclusion. The embryo of Delphinus
delphis has some vibrissae on its upper jaw arranged in two erect rows.
After birth, or possibly before it, these vibrissae fall off.

The study of the characteristics of these vibrissae in the dolphin embryo
(Yablokov and Klevezal’, 1964) showed that the depth of the vibrissae in
the skin of the embryo is relatively less than that for all the baleen whales.
The hair follicles of the vibrissae of the dolphin have only a weak attachment

Problem of Vestigial Organs 241

to the connective tissue (see Figure 54); the core of the hair is completely
free in the follicle, does not have any contact with the wall of the follicle,
and looks like a foreign body. ‘There are no blood sinuses in the hair follicle
characteristic for the vibrissae not only in cetaceans but in all mammals,
and there are no traces of blood vessels passing through the hair follicle.
Unlike baleen whales and pinnipeds, there is no epithelial papella at the
base of the hair pouch in dolphins, which indicates that the processes of
growth have stopped. Nerve fibers were not observed in either the hair
follicle or the surrounding parts of the dermis.

But there are other facts too. It appears that there are many vibrissae
on the body surface of mature river dolphins of the Amazon (Inia geoffrensis)
and the Ganges (Platanista gangetica). ‘The functional and morphological
analysis of the vibrissae of these species has not yet been done, but coupled
with the factors connected with their habitat in very turbid waters of
tropical rivers, and full or partial blindness in this type of dolphin, the
vibrissae of these species should be taken as important functional devices.

The question arises as to when (depending upon the nature of their
construction) vibrissae should be considered specialized organs, and when
vestigial organs ? ‘The possibility is not excluded that the developing
vibrissae in the embryo dolphin play a specific functional role in the early
stages of its development. This hypothesis cannot be ruled out completely.

On the whole, it is very clear that vestigial organs cannot be dis-
tinguished from specialized ones with the help of morphological criteria.
The use of a mixed morpho-functional criteria is not correct in principle;
first, because it is not possible to calculate exactly the “part taken’ by
one criterion or the other in the determination of an organ as “vestigial,”
and second, because it is not possible to find a functional criterion of a
phenomenon without knowing the significance of that phenomenon. Having
examined the data presented here and elsewhere, a question arises as a
logical consequence of this material: Do vestigial organs exist at all ?

In all the usually cited examples of vestigial organs, the organ is
present in all the individuals of a given species. Further, it is observed
in all cases that such organs or structures, inherited by the whole population
have a functional significance and logically cannot be named as vestigial.
But should not those structures be taken as vestigial which develop in some
individuals but are not characteristic of the whole population ? It has
been known for a long time that such organs exist in animals. ‘The following
can be cited as examples: the appearance of additional mammary glands;
bregmatic bones in the mammalian skull (Manville, 1959; Pucek, 1962);
the development of the palatopharyngeus muscle only in some individuals
of such species as horses and dogs (Gimmel’reikh, 1962); and the develop-
ment of posterior extremities and a male uterus in toothed whales. I will
discuss the last two in detail as I have been involved in their study.

242 Variability of Mammals

It is known that in some individual cetaceans underdeveloped posterior
extremities—like small fins or protuberances—are visible on the underside
of the body. Such instances have usually been described as proofs that
once the posterior extremities in the ancestors of whales were more signi-
ficantly developed.

Such findings among toothed whales have been made in dolphins
(Slentsov, 1939) and in sperm whales (Ogawa and Kamia, 1957). ‘To-date,
only six instances of remnants of posterior extremities have been found
in the sperm whale (Physeter catodon).

In June, 1962, V. I. Borisov observed a sperm whale with well-developed
protuberances on the ventral region of the body,* while working in the
whale factory at “Skalistii’? (Central Kuril Islands). One of these pro-
tuberances could even be X-rayed. Aided by already known facts, this
helped us to separate some types of construction in sperm whales (Figure
63). The first type is represented by a single pelvic (?) bone found without

Figure 63. Known variants of the structure of pelvic girdle (“vestiges of posterior
extremities”) in the sperm whale, Physeter catodon:
I—Usual variant ; II—From author’s data; III—Data of Ogawa and Kamia (1957); Ша—
Data of Г. I. Borisov; IV-IVa—Data of A. A. Berzin and Г. A. Zemskii (1960).

4V. 1. Borisov kindly passed all the material of this find on to me.

Problem of Vestigial Organs 243

exception in all the animals of this species and playing an important role
in the function of the urogenital and digestive systems (see above). The
second type is the pelvic bone with additional bony elements (perhaps,
remnants of a femur), observed by me in June, 1959, in a male sperm
whale in the whale factory at “‘Podgornyi,” North Kuril Islands. The
third type of posterior extremity in sperm whales was found and described
by T. Ogawa and T. Kamia (1957); a variant of this type was found by
V. I. Borisov which differed only in being more complex in structure.
A fourth type of extremity, also from the North Atlantic, has been described
by A. A. Berzin and V. A. Zemskii (1960) ; in this instance, the metatarsus
and remnants of phalanges had developed.

The presence of so many types of extremities in the sperm whale
can be taken as a clear proof of the fact that the evolution of this organ
in cetaceans is far from complete and we observe only a particular stage
in this process. Much material passing through the hands of investigators
(almost tens of thousands) has made it possible to examine extremely rare
variants in such mammals that would have otherwise been left undetected.
Such findings point to the fact that in the mobilized reserve of variability
of sperm whales, the possibilities of evolution in other than modern directions
have been preserved in the evolutionary ““memory”’ of the species. Given
some condition or other, these evolutionary directions, characteristic of
the distant ancestral forms, which have apparently been discarded because
of adaptive evolution, could be realized once more.

Another example of the same kind is the development of a male uterus
(uterus masculinus) in certain toothed whales. Many authors have found
significant variations in the development of a male uterus, including its
complete absence, while studying the male reproductive system of different
species of cetaceans (review by Yablokov, 1961b). The degree of develop-
ment of a male uterus has been advanced as a taxonomic criterion not only
for the species, but also for larger taxonomic units even up to the level of
family and sub-order.

As a result of works on the morphology of the white whale (Delphinapterus
leucas), it was possible to investigate the structure of the male uterus for
tens of thousands of this type of whale (Yablokoy, 1961а, 19616; Kleinen-
berg, Yablokov, and others, 1964). In some animals, a well-developed
male uterus was found which opened into the seminal vesicle, ventral to the
opening of the vas deferens. A tube two to three mm wide and up to 100 to
150 mm long (depending upon the size of the animal) ran up to the bifurca-
tion (Figure 52). After bifurcation, the horns of the uterus were symmetri-
cally arranged on the common mesentry of the vas deferens. Such a

5 Because I had no data on the variations in the construction of the pelvic girdle at that
time, this find did not attract much attention and was recorded in the diary as “‘normal.”

244 Variability of Mammals

structure of the male uterus is conventionally taken as the normal and is
found in some individuals of the species under study.

In the majority of individuals of the species, the male uterus develops
with different deviations from this “average” type. One of the commonly
occurring deviations is the absence of horns. In some cases, the horns are
completely absent; in others, they are considerably reduced (see Figure 46).
The most usual variant is the division of the internal lumen of the organ
into various isolated ones; usually, the division is observed up to bifurcation,
but it has been observed farther than this also. In a few animals there
were absolutely no traces of the uterus. None of these deviations were
related to the age of the animal in any way.

Most of the various stages of reduction of the male uterus were observed
in the white whale. The functional importance of this organ is not clear
at present. It can only be assumed that the lymphatic fluid contained in
the cavity of the male uterus points to some relation of this organ to the
function of the lymphatic system of the sex apparatus. Whatever the
case may be, the development of this organ in some animals and its complete
absence in others, shows that this organ is certainly not needed by some mem-
bers of the population during their normal life span.

The organs described above—posterior extremities of the sperm whale,
and the male uterus of the white whale; similarly the bregmatic bones,
additional milk glands, and the palatopharyngeus muscle—could perhaps
fall into the category of atavism, 1.е., organs appearing in the development
of present forms and indicating the condition of their ancestors (“‘atavus’—-
ancestor in Latin).

I feel that the solution to the problem of vestigial organs is closely
related to the phenomenon of atavism. It is impossible to imagine that an
organ developing in an organism in conjunction with other organs, is not
related to it in various other ways. In the case of so-called “atavisms”’
appearing for reasons not yet known, the morphogenetical processes followed
the path characteristic for ancestral forms. But because the development
occurred by such an old path, the corresponding structures would necessarily
appear to be essential and important for the given direction of morphogene-
tical processes.

It is difficult to imagine that such a disturbance in development would
take place uniformly in many systems of organs at once. We should
not forget the system of control that acts by selective elimination, influencing
all the developmental stages of individuals and destroying greatly deviating
forms (Schmalhausen, 1943), but individual organs and parts of organs
could change (as is observed in nature) in a similar way. If during this
the individual as an object of selection successfully passes the test of natural
selection, then such “‘ancestral”’ types will also be observed in nature.

In any case, each such organ, developing in an uncommon way,

Problem of Vestigial Organs 245

happens to be an active participant of specific morphogenetical processes
and carries out a specific function, significant only for that organism.
Such functions are important for that very organism in which such a
structure exists, but are not significant for the population.

It is easy to imagine that genetic mechanisms play an important role
in the appearance and distribution of such structures. It is possible that,
after some given time, it would be significant to have such a useless organ
subject to selection, and then those individuals with, let us say, a male
uterus, will either be eliminated or, on the other hand, will have distinct
advantages; and according to the rule of genetics, this trait will either
vanish or spread widely in the population.

Here we approach a most interesting problem, i.e., the usefulness of
an organ retained in a state of considerable variability, which is likely
to be compared with a stage of typical pre-adaptation. An organ which has
become useless at some stage of evolution does not vanish immediately and
completely, but is transformed into a potentially hidden stage with the
help of genetical mechanisms and, after a lapse of quite a considerable
period, may again be “picked up” by natural selection if the evolutionary
direction of the group changes.® In this way, the considerable variability
inherent in the vanishing organ and structure becomes understandable
(Simpson, 1948; and many others), and finds expression in a well-known
conclusion: “It is so commonly true that degenerating structures are
highly variable that this may be advanced as an empirical evolutionary
generalization” (Simpson, 1948; р. 74; 1944, р. 39). The proposed
explanation of the phenomenon may perhaps help to transform this empirical
generalization into a theoretical one.

The phenomenon of increase in the variability of vanishing structures
may be a consequence of the slowing down of the elimination of alleles
from a population as a result of weakening the selection on particular
traits. Such a decreased elimination of alleles theoretically may lead
to an increase in phenotypic variability, especially that part of it which
is controlled by genetic mechanisms.

And so, it is proposed to examine the functional significance of an organ
or a structure not “generally” but relatively, either to the population as
a whole (as an object of evolution) or to the organism (as an object of
selection).? Thus a discrepancy may arise regarding the interpretation

6 This aspect of the relation of natural selection to variability is examined in greater
detail in the next chapter.

7 I wrote about the possibility and the necessity of such a division of an organism and a
population in an evolutionary study incorporated in my review of M. M. Kamshilov’s book
(1959; see oologicheskit Zhurnal, 1962; vol. 41: 1438-1441). This article concerns the
process of micro-evolution; species and genera could become objects of selection during
the process of micro-evolution.

246 Variability of Mammals

of the concept of “‘vestigial organ’? proposed by previous investigators :
Vestigial organs do not exist and cannot exist in relation to an organism;
such organs exist only in relation to a population and are carried by some
members of that population. In such cases, their relative insignificance
for the life of the population at a given moment of evolution serves as a
criterion for describing them as vestigial.

Proceeding from the fact that the function (in its wider sense) deter-
mines the structure of an organ, would it not be more correct to look first
at the process of the evolution of vestigiality primary changes (the
changes in functions), and not the secondary changes (the changes in
organs)? In such an approach it is more logical to talk about the vesti-
giality of functions rather than the vestigiality of organs. In the first
place, to talk about the vestigiality of an organ it would be necessary
to know clearly with regard to which function a degeneration of an organ
has taken place. It is natural that all the general conditions which were
put forth for determining the concept of vestigial organs and the
criteria for defining such an organ, should accordingly be fully applicable
to the concept of the vestigial function. Such an approach is justified
because it is not correct to relate the changes of a whole organ exclusively
to the changes in one function of this organ (which happens when the organ
is said to be a vestigial one).

In conclusion, it is necessary to note the close relationship of the
problem of vestigiality to the problem of morphological variability (epigenetic
polymorphism type). The majority of the examples presented in this chapter
belong to that type of variability—epigenetic polymorphism. This shows
once again the outstanding importance of the phenomenon of epigenetic
polymorphism for analyzing the problems of evolution.

With such widely differing ideas about populaticns, it becomes
necessary to revise many of the basic hypotheses of evolutionary biology.
This need is recognized by leading present-day investigators: Beklemishev,
1960, 1964; Livanov, 1960; Berg, 1957-1964; Schmalhausen, 1958-1965;
Timofeev-Ressovskii, 1965a, 1965b; Simpson, 1953; Russell, 1958; Collin,
1960; Lewontin, 1961; Mayr, 1962; and many others. Such a revision is
going on incessantly through the impact of revolutionary ideas from conti-
guous biological fields, 1.е., molecular biology, genetics, cybernetics, physio-
logy, and others.

СнаАРТЕВ VII

VARIABILITY AS AN ADAPTATION

“Study is more hopefully directed to variability for its own sake, directly observable
and a primary factor in evolutionary rates and modes.”

(Simpson, 1948; p. 79; 1944, p. 44)

The specific material presented in the previous chapters allows one
(in my opinion) to define precisely the well-known position that population
variability is an independent evolutionary factor closely related to the
effect of natural selection.?

This statement, in such a general form, can hardly be objected to.
But the interaction of the processes of natural selection and variability in
populations are areas which have had almost no investigation and offer
great possibilities for new ideas and new methods of approach.

The study of populations by morphological methods inevitably leads
the researcher to pose general questions regarding the place and role of
the study of variability in evolutionary biology. What is variability from
the viewpoint of evolution ? What is the nature of the relationship of
variability as a leading evolutionary factor to natural selection? The
answer to these fundamental questions does not lie, of course, only in the
field of study of within-population variability to which the present work

1 “Selection” is, by definition, the result of specific relations (see Introduction). How-
ever, the process of differential elimination by the reproduction of organisms comes under
selection. In a third approach, selection can be defined as an important factor of evolu-
tionary processes. All three approaches are useful and show once more that the categories
of formal logic (“‘either—or’’) cannot be used in analyzing the phenomena of biological
order. In this chapter, the third approach to “selection” has been used—that it is an im-
portant factor of evolutionary processes.

248 Variability of Mammals

has been devoted. But the important problem in all such investigations of
this type, is to discuss at least some such questions, out of so many, on the
basis of available factual material.

The problem, unfortunately, is not only inevitable but also ignoble,
as its solution lies somewhere in the future and cannot yet be realized.
Some hypotheses and speculations have to be advanced which will be re-
jected sooner or later. However, Simpson is very correct in saying that
factual data will never be sufficiently complete and hence it is cowardly
to withdraw efforts to determine solutions to the problems of today simply
because tomorrow those efforts may be evaluated as insufficient.

There can hardly be any opposition to the point of view that variability
is an important evolutionary factor, standing in close relationship to the
evolution of specific populations. An important conclusion from many
works devoted to the study of variability of different organisms is a hypothesis
regarding the relationship of variability to the degree cf its influence on
evolutionary factors.

Darwin wrote: “‘Heretofore, I said that variations—so common and
so manifold in organisms under domestication and rare under natural
conditions—were more or less accidental. Such a statement is, of course,
quite incorrect, but it clearly shows our lack of understanding of each
individual variation”? (Sbor. Soch, Volume IV; 1951, р. 641). Thus, by
taking variations in an organism and afterwards even the variability in
a population as his basis for reasoning, Darwin took a major step toward
the solution of this problem.

The idea, based on Darwin’s theory, that all the variations are actually
adaptations, was theoretically developed in the 1920’s Бу В. A. Fisher. It
is noteworthy that this conclusion not only offered a possibility for conducting
a study of population from qualitatively new positions, but even happened
to be (as has been admitted by Ashby) an original source for the develop-
ment of cybernetics (Russell, 1958).

By the 1940’s development of this situation had taken place in all
the fields of biology concerned with populations. And, more important,
experimental evidence for this idea is accumulating. The presence of a
relationship between the intensity of selection and the frequency of appear-
ance of a particular mutation can be shown by genetical investigation.
“The rate of mutation in each species is an adaptive trait determined by
natural selection” (Dubinin, 1940). Simpson (1948, p. 77; 1944, p. 42), while
summarizing data on paleontology, comparative anatomy, and genetics,
observes that “the extent and nature of variability are themselves important
group characters subject to natural selection and other evolutionary
factors.”’ These points of view express a general trend in progressive evolu-
tionary ideas. On this В. Г. Berg said well: ‘““The-concept of group adapta-
tion is more and more- occupying the attention of scientists” (1957, р. 133).

Variability as an Adaptation 249

It is generally accepted by researchers that the magnitude of variability
is definitely related to the intensity of the evolutionary processes going on
in nature. But a simple comparison of different points of view shows that
the nature of this relation is understood quite differently by different
researchers. To compare different viewpoints, it is considered reasonable
in the first stage te simply cite excerpts from various works. Of course, it
is impossible to include here all the scientific works concerning this problem,
but a sample of 25-30 individuals is usually considered sufficient to under-
stand the morphological features of a population, and similarly, to attempt
to show the contradictory opinions of biologists, zoologists, paleontologists,
morphologists, ecologists, and geneticists.

An examination of the material presented below shows that all the
possible combinations can be stated in terms of the following four variants:

Weak selection leads to strong variability,
~ Strong selection leads to strong variability,
Weak selection leads to weak variability,
Strong selection leads to weak variability.

According to one point of view, the variability in “favorable” conditions
(under which comes a longer life span due to free access to food resources,
i.e., a general weakening of the pressure of natural selection) increases
considerably. In “unfavorable” conditions (correspondingly, during
intense selection), the variability decreases.

According to another viewpoint, variability decreases in “favorable”’
conditions. During intensive selection, variability increases—which 13
to say that during unfavorable conditions an increase in variability takes
place.

But before considering the different viewpoints presented in the next
few pages, it is necessary to mention that Simpson pointed out in 1944
that one of the reasons for obtaining exceptionally low or exceptionally
high values of variability may be an insufficient genetical homogeneity of
the sample. If we are working on a sample genetically more homogeneous
than the whole population, then the variability may be considerably less
than that really existing in nature.

2 Here and henceforth, while determining the relation of variability to selection pressure,
I do not propose to examine the genetical control of any variability, accepting an important
role for such control in the formation of a mobilized reserve of variability in mammals.
The study of the regulation of the formation of such reserves and variability is in the province
of genetics and cannot be carried out from the position of population morphology. But
the ways in which this reserve of variability manifests itself in the process of evolution, should
be simultaneously studied through the genetics, morphology, and ecology of populations.
The present work, as has been mentioned many times before, is supposed to be the morpholo-
gical part of such a common investigation.

250 Variability of Mammals

On the other hand if we are working with a sample where the genetical
heterogeneity of the population has been artificially increased (for example,
by rearranging variants within classes and reducing the number of common
variants), the magnitude of variability could be increased. But to make
consideration of the problem easier, let us assume that the samples with
which we are working are sufficiently accurate to reflect the real situation
of a population.

Relation of Variability to Intensity of Selection
(According to Different Authors)

Weak selection (increase in numbers)—high variability

1. “On the other hand, vestigial and non-functional characters—
those with low selective value—have higher average variability” (Simpson,
19435 р. 235; 19 р. 93: italics mame):

2. “It is clearly seen from this, that the more favorable the living
conditions, the greater appears the within-population polymorphism”
(Sinskaya, 1948; p. 71).

3. “The first phase—increase in numbers in favorable conditions
when the influence of natural selection is weak—is correlated with an
accumulation and recombination of mutations (with an increase in individual
variability)’ (Schmalhausen, 1946; р. 190; 1949, р. 116).

4. “Under favorable conditions, with a weakened struggle for
existence and a decrease in the intensity of elimination of deviations,
mutations accumulate rapidly and reach rather high frequencies. Indivi-
dual differences accumulate great diversity and to a rather considerable
extentan (Ша. 195).

5. “. . . The degree of within-species diversity to a considerable
extent depends on the total intensity of selection; to be exact, of its stabilizing
form. Moreover, even in the most homogeneous conditions, in a micro-
population with weak stabilizing selection, the genetic heterogeneity sharply
increases. Because of this, the number of aberrant individuals increases”

(Olenov, 1961; p. 107).

Strong selection—high variability

1. ‘Intense selection for deviations (1.е., against the previous norms)
may lead to an increase in the variability’ (Schmalhausen, 1943; р. 196).

2. “Increased variability in populations living under unfavorable
conditions’... Ваз been recorded? (@bid®, р: 207. 192 ира о

3. “The amount of variability increases when the food supply dimi-
nishes” (Naumov and Nikol’skii, 1962; р. 1137).

Variability as an Adaptation 251

Weak selection (reduction in numbers)—low variability

1. “In a very small population an almost stable insignificant variability
and weak selection is: observed. Therefore a static situation, disturbed
from time to time by accidental occurrences of rare mutations inevitably
leading to degeneration and death, is likewise seen”? (Wright, 1931, cited
by Klein, 1964; Italics mine).

2. ‘Variability decreases in favorable conditions’ (Naumov and
Nikol’skii, 1962; pp. 63-64).

Strong selection (reduction in numbers)—low variability

1. “. . . functional, integrated structures of adaptive significance—
i.e., those that are under strong selective pressure tend to have low, although
still appreciable, variability. . . .” (Simpson, 1948; р. 135; Italics mine).

2. “АЦ these facts demonstrate that the more intense the selection
under certain conditions, the smaller is the number of visible variations”
(Schmalhausen, 1946; p. 190; 1949, p. 117).

3. ““The second phase—a relative stability associated with the condi-
tions of competition and also a type of direct struggle of a species for
existence—is related to an effective selection of more favorable combinations
with a decreased variability. The third phase—a more or less sharp decrease
in numbers under the influence of powerful eliminating factors (usually
not having any selective significance)—-may be associated with a further
Geetease imivaniabilitys ” (bids, р. 190: 1947, р. 116; Italics mine).

4. “. . . a rigid individual elimination leads to a decrease in variabi-
lity’ (Schmalhausen, 1939; р. 213; Italics mine).

In noting the possible reasons for an increase in variability in natural
populations, we have to consider fluctuations in numbers (analysis by
I. I. Schmalhausen, 1946; summary by М. P. Naumov, 1945). During a
period of increase in numbers, the variability of all populations inevitably
increases. The mechanism of such an increase in variability is adequately
understood: In the period of increasing numbers, there appear in great
quantities diverse extreme variants which get eliminated under other
conditions. The genetical heterogeneity of the population increases with
an increase in numbers and increases the chance of accidental influences
affecting the development of a considerable number of individuals. After
a sharp decrease in numbers, brought about by non-selective elimination
as shown by Schmalhausen (1946), the genetical constitution of the popu-
lation is likely to change qualitatively, and in the new growth period with
an increase in numbers there appears a population with a different gene
pool. This complex cycle of phenomena, now yielding to only a most
general analysis, shows that there are considerable possibilities in a close
collaborative study of the problem of micro-evolution from the ecological,
genetical, and morphological points of view.

252 Variability of Mammals

Some contradictions in the statements of different authors. can be
eliminated if we consider that during unfavorable conditions, and under
great pressure of selection, these conditions would not affect all organs
and traits of the animal to the same extent. Elimination, or the final
removal of the individual, in other words, of the whole ontogeny, will always
have individual traits as focal points. This means that during the most
intense selection, despite pressure on one trait or on a group of traits, the
variability of other traits could not even be affected. And it is known that
the majority of authors, although realizing the inaccuracy of biological
terminology, use the concept of variability in conjunction with the whole
organ; strictly speaking, this is not possible and cannot be done without
qualifications.

But despite all this, many viewpoints still remain contradictory. ‘This
contradiction shows that the generalized concept of ‘‘natural selection,”
“pressure of natural selection,” “improvement” or “deterioration” of living
conditions, etc., seems to be inadequate for the analysis of the phenomenon
observed. -

One of the most important achievements in the development of Darwin’s
theory was separation of a unique process of natural selection into different
forms (Wright, 1931; Schmalhausen, 1939). And the explanation of the
above-mentioned controversial viewpoints is obscured without observing
the influence of different forms of selection on variability.

Modes of Selection and Variability

Arising out of elimination, natural selection manifests itself in two
basic modes: directional and stabilizing (Schmalhausen, 1939, 1946).

The directional mode of selection is caused by the occurrence of a
selective advantage in changed environmental conditions with specific
deviations from the ‘‘normal”’ (that characteristic of the previous living
conditions). This form of selection is also named “centrifugal” or “linear”
(Wright, 1931; Simpson, 1948). Selection in this mode (with intermediate
types also considered) is graphically shown in Figure 64, taken from the
works of I. I. Schmalhausen and G. G. Simpson.

The stabilizing mode of selection appears when there is a selective
advantage for the ‘‘normal”’ organization above all the deviations. ‘This
type of selection, also called ‘centripetal’? (Wright, 1931), counteracts
any displacement in the character of a population. ‘This type of selection
is likewise graphically shown in Figure 64. An examination of these graphic
representations of different modes of selection permits one to assume that,
with regard to a specific trait, the directional mode of selection can be taken
as basic. Thus directional selection, if directed uniformly in different
directions away from the central point, gives centrifugal selection and

Variability as an Adaptation 253

Figure 64. Schematic representation of centripetal (A,_,) and centrifugal (B,_,) selections:

A,, В. and C—‘landscape’’ of selection ; C—variant of an asymmetrical centripetal selection.
Arrows—pressure and direction of selection (Simpson, 1948).

selective vectors directed from various points toward the center give rise
to centripetal selection. A linear selection in different directions gives
rise to an asymmetrical type of selection (which has been known before).
This subsuming of the graphical representations of different types of selec-
tion under the linear mode makes the study of further theories on the structure
of variability rather easier to understand.

From the viewpoints of different investigators given above, it can
be seen that variability may be similar at different selection pressures and,
contrarily, it may be different during similar selection pressures. 5. Wright
(1931) and Schmalhausen (1939), who have described different types of
selection, showed that variability decreased only with an increase in the
centripetal selection, or ‘“‘stabilizing’’ selection, to use the terminology of
I. I. Schmalhausen. When directional (centrifugal) selection increases,
the variability increases. These well-established facts are unfortunately
not always remembered in zoological research. ‘The authors of these works
continue to use the word “‘selection”’ from only one point of view, and it is
imperative that they should indicate which form of selection is being dealt
with.

To summarize, it can be said that the magnitude of variability depends
not only on the intensity of selection, but also on its direction. Different

254 Variability of Mammals

modes of selection, similar in magnitude but different in direction, can give
rise to variability of different amounts.

After such general comments, let us now analyze the manifestation
of variability under the influence of different forms of natural selection.

Stabilizing selection and variability

The relation between the stabilizing mode of selection and variability
is expressed by the same normal Gaussian curve which serves as a base
for the overwhelming majority of statistical calculations on variability.

Stabilizing selection expresses the relations which have been formed
over a long evolutionary period between structure and environmental
requirements. If the direction and pressure of stabilizing selection with
regard to variability of any trait is shown by vectors, then it is seen that
the vectors with minimum value will be arranged on the curve against the
modal group of variants (Figure 65), and the ones with maximum value
will be arranged against the extreme classes of variants.

А JA

Increase of stabilizing selection Influence of stabilizing selection

/
LAs ~ : : № SS VN >
Weakening of non-selective Original distribution Increased non-selective
elimination elimination

Figure 65. Some variants of transformations of the original distribution
of the variability of a trait.

It is easy to show that for the variability of the whole population,
as determined by the sum of the variability of (n+1) traits, the resultant of
its interaction with stabilizing selection appears аз а ‘“‘hill,”’ the transverse
section of which will give the variability of a practically unlimited number
of traits (Figure 66).

During an increase in intensity of stabilizing selection, there is first
a proportional and uniform reduction of variants in each of the classes,
as a result of which the general variability is unchanged, but as the number

Variability as an Adaptation 255

AA A;D;

В, Bz B3 Az Ag Ay
B A

Figure 66. Variant of the graphical representation of the “hill” of variability
‘ and the effect of type of selection:

AB, A,B,, A,B,, A;B;—some of the possible cross sections of the “hill” of variability
without pressure from the directional type of selection. CD, C,D,—some possible
cross sections with the influence of the directional type of selection.

of individuals in the population decreases (Figure 65), then, through the
continuous action of selection, the frequency of variables begins to decrease
until they reach an equilibrium; with even stronger selection they can be
completely eliminated.

In a weakening of intensity of stabilizing selection, a proportional
increase in the frequency of variants takes place in the beginning in the classes
already available; then the appearance of extreme classes is possible, which
leads to a general increase in variability restricted perhaps only by genetical
limitations. During the change in intensity of stabilizing selection, the
shape and disposition of the variability curve may remain constant. This
means that the mean and the standard deviation and coefficient of variation
will remain constant. It is possible to have a situation where the distri-
bution would become changed in shape by changed intensity of stabilizing
selection, and then the standard deviation and coefficient of variation
would also change. But, in all cases, the arithmetic mean of the population
would remain unchanged.

Directional Selection and Variability
The relations of this type of selection to variability are considerably

more complex. Let us look at some examples. In the first, as a result of
a sharp selective elimination by truncation, a large or small proportion of

256 Variability of Mammals

the variants are destroyed from some part or other of the curve. This
example is presented in a more or less general form in Figure 66, where some
of the variants of the hypothetical “hill” of the general variability of the
population, have been eliminated.? During the continuous influence of
this form of selection in the given case, a mixing up of the whole curve of
variability for specific traits takes place in the direction opposite to the one
where elimination occurs. After a time, the curve should take up the normal
shape, but the arithmetic mean will have moved. ‘The same can be said
with regard to the whole “hill” of variability. At various stages of this
prolonged process different asymmetrical forms of the curve can be observ-
ed, always with a more elongate part away from the region of elimination.

The second example concerns the selective elimination of modal
variants and has been worked out in detail by G. D. Polyakov (1961, 1962)
on fish populations. During a sharp deterioration of feeding conditions
in the water, the modal groups were in a worse state than those capable of
feeding on several atypical types of food. The extreme variants, which
were accustomed to food atypical for the whole population, appeared to
be in a better position; they had a selective advantage in which was hidden
the possibility of forming polymorphic groups or forms. This example,
which to some extent can be taken as a model for some organisms, could
hardly be true as such for mammals. But it should be remembered once
again that not a single type of natural selection mentioned above and,
correspondingly, not a single relation of variability with different types of
selection, occurs in nature in pure form.

In observing these types of hypothetical and model examples, we are
only making an attempt to penetrate the sort of phenomena for which the
on-going processes in nature can be examined; but not one of these pheno-
mena can occur individually in nature.

It would seem that other more complex cases of relations between
directional selection and variability could exist, but all of them would be
based on the two instances which we have seen.

Prolonged Variability and its importance
in the Evolutionary Process

In nature the variability of a population is under constant control
of natural selection in its different forms. However, in specific periods
of population development (because of the environmental changes taking
place), such control will be different in intensity and in direction. Simpson

3 It can be seen in Figure 66 that the selection will always act only on some specific
traits and not on all. Even those traits which it affects in a given situation, are not uniformly
affected: Three cross sections (AB, CD, and EF) show different degrees of influence by
eliminating effect of the given direction of selection.

Variability as an Adaptation 257

(1948) observed that during periods of weak selection particularly, the
role of other factors increases and becomes decisive in the evolution of a
population. Here this proposition as applied to variability is examined
in a general form.

During a weakening of selection on some trait, variability itself should
act as an important evolutionary factor in determining those trends which,
on the whole, will ultimately control. the future of a population. The
weakening of selection on a particular trait or group of traits (which very
often happens in the evolution of a group) should be a main turning point
capable of determining the future of an organ, or perhaps the future of an
entire population for a long time to come.

When the selection weakens with regard to a given trait, genetic drift
begins to play a greater role in the control of variability. With the lifting
of all types of selection pressure, through the slowing down of the elimination
of alleles, the hidden mobilized reserve of variability appears on a greater
scale. The continued existence of the organ or trait (which has had an
appreciable period of existence in the gene pool with its phenotypic expres-
sion) will be determined mainly by morphogenetic relations with other
important structures by a selection. If the organism, as a part of a popula-
tion, does not reconfirm the importance of the existence of the organ by means
of an interaction with the biogeocenose (Schmalhausen, 1959, 1965) (because
even the mere presence of a useless organ in an individual may act as an
adverse trait from a selective viewpoint), these relations will start to be
replaced by those which better conform to the present conditions. The
structures and traits which have thereby become useless will morphogene-
tically drift more and more to earlier (and more general) developmental
stages in the organism.

The disintegration of such relations, leading to the vanishing of a
structure from the phenotype, is not a sudden but a very prolonged process.
The phenotypic retention of a trait remains in an incomplete but variable
stage during a long evolutionary period (the selection pressure has been
reduced, the reserve variability has appeared); through this increased
variability a breach of the morphogenetic relationship takes place, trans-
‘forming its position, fixed by previous selection, into another. This
process has been depicted in Figure 67, where the graphical expression
of variability transformed on to a time scale should be helpful.

I would like to propose that the prolongation of a trait in a condition
of elevated variability be named ‘prolonged’ variability, taking into
consideration the protracted nature of this stage in the evolution of an organ
and structure. Some of the above-mentioned examples would surely
permit one to examine the scope of such prolonged variability.

Certainly, the great variability of posterior limbs in cetaceans has
been maintained for some tens of millions of years.

258 Variability of Mammals

ААА,

iit

Figure 67. Diagram of variability of a trait during a period of time:
Arrows—selection pressure; horizontal scale—time; vertical lines—
amplitude of variability. See text.

Prolonged variability is of outstanding importance in the evolution of
morphological aspects of animals. ‘The prolonged continuation of a vanish-
ing organ in a condition of considerable variability may be taken as an
important evolutionary property of a population. For a period of many
generations, a population seems to “‘test’’ or check whether a structure
which became useless some time back, will again become useful and attain
selective importance. It is possible that the origin of structures in descend-
ants which differ considerably from those of their ancestors, can be understood
in this way.

Perhaps the proposed hypothesis of prolonged variability will
considerably simplify our understanding of the processes going on in
nature. These processes are undoubtedly very complex because, due to
the overall multi-functionality of an organism, an independent variability
of structures occurs quite rarely before an appreciable reduction in size;
and also, because environmental changes may occur to restore the selective
value of a structure from time to time and frequently change the direction
of its evolution. An analysis of some developmental features of baleen
and toothed whales shows the possibility of such complex organ development
(Yablokov, 1964).

Here it would seem important to give only a general picture of the
possible process to show the importance of the study of population variability
for morphological structures from a purely ‘phylogenetic aspect. The
collaborative work of paleontologists and morphologists on morphological

Variability as an Adaptation 259

variability will show how useful the proposed hypothesis is. In concluding
this section, it would seem desirable to emphasize the fact that the mobi-
lized reserve of variability is formed completely in higher animals (as
shown by Schmalhausen, 1946), and hence the whole hypothesis of prolonged
variability is applicable primarily to those animals.

Adaptive Importance of Variability

Because of the many examples of the relation of population variability
to the processes of natural selection, the question naturally arises as to
the adaptive significance of population variability and the fundamental
nature of variability in general or, to express itin another way, the usefulness
or uselessness of variability as observed in nature.

In looking at population variability as an important adaptive property
of a population, an appreciable influence of it on evolutionary factors is
implied. Now the question is posed from another point of view: Is the
variability useful initially ? Is the observed end result—the coordination
and adjustment of variability for all traits in a population in exact con-
formity to environmental requirements—true at the beginning ?

These views on the possible usefulness of specific changes in organisms
have been advanced many times since Lamarck, and subjected to criticism
by zoologists. However, recently these viewpoints have often been
advocated (Raicu, 1960), which makes it necessary to examine this
problem.

Recently, G. V. Nikol’skii (1963) re-emphasized that a population’s
reaction to any influence should be primarily adaptive and rapid. Such
a о. in his opinion, is usual in all situations wie arise during the
group’s development.

Let us examine a hypothetical example. Several changes occurred
in a population of mice as a result of increased doses of radiation: deviations
in the growth rate of young animals, disturbances of embryonic development,
and disturbances of the characteristic indices of variability of other traits
(number of vertebrae, number of hair per unit area of body, etc.). The
reaction is clearly of an inadaptive nature, but the population continues
to exist in the biogeocenose (as a result of changing some individuals to
different food, others becoming less susceptible to predators, or a weakening
of the different forms of the struggle for existence). There probably appear
at this level of change in the population some individuals which, in the
newly-created conditions, have a specific selective advantage. After
some time, the population will come out of this difficult phase of development
with some changed characteristics (because of the greater reproduction of
some of these individuals).

Looking at this process more generally, it can be seen that there is a

260 Variability of Mammals

direct relationship between the environmental influence and the adaptive
change. But this seems to be only the external aspect of the process. The
imaginary usefulness of accidental and regular relations, and as a result of
the constantly unbalanced relationship between the “‘micro-structure of the
environment” (Verheyen, 1969) and the variability in a population, this
forms a basis for self-development for the ““‘population-biogeocenose”’ and is
the main cause of the impossibility of direct adaptation.

If we study this example rather closely it can be seen that a typical
process of “Darwinian” selection takes place through the elimination of
non-adapted animals. This very elimination confirms the presence of a
typical picture of population variability.

In the majority of cases, the variability of a trait observed in nature
is expressed by the curve of the normal distribution. While accepting the
statement about the initial usefulness of variability, it should be said that
any environmental factors influencing the innumerable individual traits
always give a normal distribution of values of the affected individuals, and
the non-adaptive variants do not even appear in the world.

In the examination of population variability from the viewpoint of
present controlling factors, it can be said that in fact the normal curve acts
as the “stamp” which “separates” (leaves alive) only those variants out of
the innumerable ones appearing in nature which appear to be capable of ~
existing under the given conditions. When these conditions continue, a
preferential reproduction of the more efficient variants inevitably takes
place over long periods (Berg, 1964; p. 5), which inevitably leads to the fact
that fewer and fewer individuals in succeeding generations fall out beyond
the border of the “stamp” and get lost. In this way, a hereditarily. fixed
character for the norm of reaction takes shape in the nature of the curve,
reflecting the variability for each trait.

A problem arises as to whether these are population traits whose
variability would be the same for the whole population; this is directly
related to the problem of the usefulness of variability. Are there some
individuals in a population who are carriers of traits which make no
difference to the population ?

A sufficiently specific answer to these questions has been given in the
previous chapter on vestigiality of organs. ‘There are such traits and,
accordingly individuals in a population with such traits. But if there are
individuals in a population with traits which make no difference at a given
moment of evolution to the life of the whole population, then we must admit
that the variability of such traits should also make no difference to the
population at that moment of its development. Because of a progressive
reduction in the pressure of controlling factors under the changed conditions
of their existence, these traits cannot make any difference for the population.
When selection pressure is reduced below a certain level, the variability |

Variability as an Adaptation 261

ceases to be controlled by external factors, and its range can be determined
only by hereditary mechanisms and genetic drift (Prout, 1964).

The Possibility of Studying Natural Selection
through the Understanding of Variability

There is no doubt as to the importance of studying the properties
of natural selection in a population, for the cognition of general evolutionary
mechanisms (Gilyarov, 1959; Sokal, 1962; and others). By the study of
variability, a specific contribution to the study of natural selection is possible.
It is necessary to obtain data on the variability of the same organs and
structures under different living conditions in order to understand the
importance of natural selection, so that a comparison of the differences in
variability under different living conditions will yield a key to determining
the nature of the influence of natural selection. This is the first stage in
studying the relationship between natural selection and variability. Then
the second stage is inevitable, 1.е., the functional analysis of different
structures for determining the importance of the selective value of a given
structure for the individual and the population as a whole. It is clear that
in any case the answer obtained may only be indirect.

The study will initially yield then a good knowledge of those environ-
mental conditions which are capable of influencing the variability of an
organ or a trait. One of the main difficulties in solving this problem is
obtaining a solid understanding of these conditions. We do not even know
as yet how the variability of traits or the manifestation of an individual
trait is determined in individual ontogeny and the successive changes in
ontogeny (phylogeny). Molecular biology has just made an initial step
toward an understanding of this process.

Strictly speaking, it is not possible at present to compare the environ-
mental conditions of one population with those of another with any
satisfactory degree of accuracy. ‘That is why it is interesting to compare
the features of variability in genetically similar populations placed in
different environments. During such a comparison, at least there is one
common denominator known—the identical initial genetical structure of the
population and its general potentialities. As a result, the observed
phenotypic variability can throw light on the differences in the influences
of evolutionary factors.

One of the kinds of such an investigation is the comparison of the
variability of wild and inbred populations. By completely taking off the
pressure of natural selection in an inbred population, we can observe a
somewhat “basic” variability, which has been deformed by the effects of
selection (possibly, through reduction) in natural populations. But on
the other hand, the inevitable genetical impoverishment of the inbred

262 Variability of Mammals

population decreases its potential variability as compared to natural
populations.

Attempts to compare the variability of wild and inbred populations
of mammals with regard to different traits have been made by different
investigators. R. Bader (1956) made a serious attempt to compare the
variability of a number of wild and inbred populations of rats, house
mice, and guinea pigs through tens of thousands of individuals maintained
in captivity. It was observed that on an average the variability of wild
populations differed insignificantly from the variability of inbred populations,
and the observable differences were more or less directed toward an increased
variability in the wild populations. Looking at these data, it can be said
that the removal of selection pressure, which was supposed to increase
the variability of all traits, appeared to be less effective than the specific
processes in the natural populations which increased the variability of the
same traits.

V.N. Pavlinin (1962) presents data on the variability of skull measure-
ments for wild Bashkir minks and minks from the Mramorsk ranch. ‘The
initial genetical material is the same: All the animals were transported
from North America, apparently from a quite homogeneous group. Some
were released into the new environment, where they had to fight for life in
conditions differing greatly from those which existed in their motherland;
others were placed in artificially-created favorable ‘conditions where the
eliminating factors were sharply reduced and all the animals having a
normal quality of fur had a chance to live fully and reproduce.

An analysis of the variability values obtained (Table 90) leads to
the conclusion that the variability of almost all traits in the females appeared

Table 90. Variability (с. v.+5¢,y,) of skull measurements of wild and farmed
minks (Calculated from the data of V. N. Pavlinin, 1962)

ee

AT ее a
Males Females Males Females
Skull length 2.7+0.38 2.2+0.32 3.4+0.43 1.7+0.22
Condylobasal length 4.60.36 2.10.30 3.20.41 1.70.22
Zygomatic width 5.6+0.78 4.3+0.62 3.8+0.49 2.1+0.27
Height of skull 2.9+0.41 6.2+0.89 20.2+2.60 24.9+0.92
Mastoid width 4.3+0.60 3.0-Е0.43 3.80.49 1.90.24.
Width of snout -6.2+0.88 4.04+0.58 41.145.30 33.744.34
Length of upper canines 3.5+0.49 3.5+0.46 24.53.50 33.64.33
Length of tympanic bone 5.0--0.70 5.3-+0.77 4.2+0.54 3.4+0.44

CC

Variability as an Adaptation 263

to be somewhat increased in nature over the variability of animals in
captivity; in the males, the situation was rather reversed. In both males
and females in captivity, there were traits which showed a greater variability,
such as height of skull and width of snout. Some problems arise, the
answers to which are very problematic as yet. Why did the variability
in males and females for skull length go in different directions ? Why did
the variability value of some traits increase so much whereas the variability
of other traits was maintained at a very low level ?

Е. Sumner (1924) studied changes in the morphological characteristics
of two subspecies of Peromyscus maniculatus (Р. т. sonoriensis and P. т. rubidus )
under the influence of the climatic environment of a third subspecies,
Р. т. gambelli. The animals of the first two subspecies, taken from natural
populations, were transferred to the habitat (hills) of the third subspecies
and raised in large open-air cages for several generations. The conclusions
drawn by the author are very interesting when viewed with regard to the
problems posed above. All the transferred groups considerably changed
characteristics for a number of traits, but these changes were not closer to
the characteristics of the subspecies which inhabited this region; instead,
they deviated so much that the differences among the three natural sub-
species, in the traits of color and body measurements, appeared to be less
than those between the transferred populations and the local subspecies.

An analysis of the data presented by Sumner (1924) shows that in the
transferred populations, the variability decreased in 12 out of 16 measure-
ments, it remained practically unchanged in three measurements, and
increased slightly in only one trait. It is still difficult to explain the pheno-
mena observed. The possibility of the presence of some regularities,
important in principle, and which could be brought to light by conducting
a number of planned analogous experiments, cannot be excluded.

An interesting way to study the relationship of selection to the magnitude
of variability is to examine the variability in animals of different ages.
In Chapters III and IV, some examples of chronological variability were
analyzed, bringing to light some specific trends in the changes of variability
parameters for traits of animals of different ages which had been subject
to abiotic influences that were inevitably different from year to year in the
early stages of their development. The promise of such an approach in an
analysis of selection under natural conditions was shown by the investigations
of K. I. Kopein (1964) conducted on ermines in the Yamal-Nenetskii region.

Another way to study the relationship of variability and selection,
with a view to understanding the regularities of selection, is to compare
the variability of natural populations which have lived for a long time
in similar and sufficiently dissimilar conditions. This last approach opens
up the possibility of bringing to light the influence of specific environmental
factors on the development of variability.

964 Variability of Mammals

Here, the investigations must proceed in the direction of finding
out “natural experiments,”’ which in the wider perspective is the study of
the influence of specific factors in nature. К. М. Lyutikov (1931) noted
the importance of the detection and study of such “involuntary experi-
тет.» In recent years a large number of ecologists and evolutionists
- have been working in this field.

To set up an experiment in nature does not necessarily mean disturbing
the flow of natural processes, but it is essential to separate. the influencing
factors, the object of the influences, and the result of such influences.

The transport of muskrats in the year 1929 to the Solovetskii Islands
in the White Sea resulted in a strong and permanent population of these
animals there. After some years, a number of them were transferred
from these islands to the Arkhangelsk region, where they successfully
reproduced. Thus a genuine opportunity arises for comparing the
morphological characteristics of the initial and the transferred populations—
both the characteristics of the environment and the characteristic action
of natural selection. If data on the population of muskrats obtained from
central and southern Europe, western Siberia, and Kazakhstan are included
in such a comparison, then a real possibility exists for discussing not only
the significance of the size of the population, the roles of temperature and
ice cover, the features of water resources, the effect of particular food, but
even more, the common regularities regarding the development of a popula-
tion under the control of natural selection would come to light and thus,
indirectly, provide an opportunity for understanding the process of selection
itself, and the process of micro-evolution as a whole.

“The number of such “natural experiments’’ representing a ready-made
laboratory for a population-morphology study of mammals can be multiplied
in our country: the squirrel in the hills of Crimea and the Caucasus
compared with the original Altai type; the characteristics of the population
morphology of the coypu in its natural habitat in the Trans-Caucasus
compared with the same in central Asia; and of course, the island forms of
mammals, beginning with the slightly isolated Ol’Khon Islands in Lake
Baikal, to the more isolated Shantar Islands, the completely isolated Kuril
Islands, etc. Here the researcher has an opportunity to select practically
any degree of isolation, during quite a varied time period. It is significant
that more and more investigators—both ecologists and morphologists—are
attracted toward isolated natural populations (Blair, 1964; Berry, 1964;
and many others), a tradition having its roots even in the work of Darwin
and Wallace.

In the same context, it would be useful to study the distribution of
variability of forms characterized by ring-like, looped, or ribbon-shaped
areas. The regularities of changes of variability: for traits in such regions
are, perhaps, directly related to the general regularities of micro-evolution

Variability as an Adaptation 265

(Timofeev-Ressovskii, personal communication).

It would be unjust to underestimate the difficulties involved in the
interpretation of data obtained from such investigations, but nevertheless,
only such studies of natural populations are capable of bringing something
new to the solution of the general problems of micro-evolution.

The functional analysis of the organs and structures under investigation,
which forms a necessary stage in the general study of the interrelation of
selection and variability, also presents a number of major difficulties.

A thorough study of an organism allows only to understand the inherent
functions of the organism in a unique way. — Retrospectively, any functional
analysis of any organ carried out 25 to 50 years ago must be outdated.
And, 50 years hence, even the present functional analysis will similarly
appear insufficient and outdated.

We can analyze with a high degree of accuracy the mechanical function
of the digits of the hand in bats with respect to their function in flight,
but recently it was found that the wings of bats are used not only for flight but
also as a sort of net for catching insects. What fraction of variability, if
that can be said, is controlled by selection for wings as an organ of flight,
and what fraction for wings as an instrument for catching prey ? At the
moment, this is entirely unclear, and only a critical ecological analysis of
individual species and perhaps of individual populations, together with an
analysis of the ecology of their food which is caught in this way, may solve
the problem. And after conducting such a very difficult analysis, there
arises the problem of determining the degree of importance and the role of
the size of the digits and accordingly the size of the membrane in the pro-
duction of vitamin D, or in the mechanism of thermoregulation, etc.

An ecological study of an animal will always bring about a closer
understanding of its properties and its relation to its environment. A
serious difficulty develops in such a study, namely, the necessity and
inevitability of searching for newer methods and approaches for finding
solutions to the problems posed by the investigation. It is generally
apparent that these newer methods and other approaches lie hidden in the
armory of the research techniques used by related natural sciences. One
such method, perhaps, could be the ‘“‘black box” which has been successfully
developed by cybernetics. We know only the information entering and
leaving the system, but it is not clear what processes take part in the analysis
of the information which comes out. Models are being prepared, as if to
strengthen the “black box,” based on the knowledge of those aspects of
objects which have been well-studied and have achieved, as far as possible,
a full correspondence between the effect of the model and the effect of
the modeled system. On the whole, understanding of structure appears
to be possible through understanding of the model. ‘Ша other words,
the impossibility of knowing the processes and properties of the objects

266 Vartability of Mammals

in the chain of related processes, is not a great obstacle to the study of these
processes at this time. It is very possible that experiments of ecologists
with model populations (Odum, 1959; Nikol’skii, 1963a) will give a new
impetus to the solution of the problems of micro-evolution.

The general aspects of the theory of variability mentioned in this
chapter naturally cannot settle the innumerable vexing problems which
arise during the study of variability in nature. The majority of these
difficulties have been examined in the past and hence their mention here
is justified only because I think it necessary to draw attention to those
general problems of population morphology which have not yet been seriously
considered.

CHAPTER VIII

CONCLUSION

The aim of this investigation has been an attempt to systematize,
summarize, and theoretically understand. the data available on the variability
of mammals, and to form specific working hypotheses for additional planned
intensive studies of the phenomenon of population variability.

The data obtained in a morphological investigation of a population
should only be analyzed against the background of ecology; the regularities
of the appearance and disappearance of characteristics in a population
cannot be made without the help of genetics. However, the relationship
of population morphology to population ecology is not one-sided. Popula-
tion morphology helps in planning ecological investigations of a specific
group and in demarcating ecological questions which are not yet clear, and
hence are awaiting solutions. The purpose of this analysis is as follows:
to obtain information on the quantitative morphological characters of an
organ, to bring to light particular trends in the variability of traits, to
compare the trends which have appeared with the requirements of the
elementary regularities of variability, and to bring out all the deviating
cases which exhibit a typical difference from the usual relationships, and
which accumulate in the evolution of groups with regard to these traits.

An animal population acts in different ways in different seasons and
years. This means that the investigator cannot twice investigate absolutely
identical populations; the response of a population to the changes in external
conditions will be different at different times.

Finally, all the categories of within-population variability taken
together have a direct relationship to the dynamics of the numbers of
individuals in populations. For example, without calculating the variability
of growth, our idea about the dynamics of a population seems to be fairly

268 Variability of Mammals

incomplete. The role of growth variability appears to be different at
different times, depending on when a new generation enters a population,
characterizing each of the developmental stages of the variability itself, etc.

The study of population morphology may serve as one means of
intensive ecological investigations of a group. By using ап ecologic-
morphological approach, a unified functional understanding of the organ
and of a system of organs is obtained. In a study of population morphology
where an ecologic-morphological approach has been added in a unified
way, a group emerges as a developing evolutionary unit.

In exploiting mammalian species which are important from a com-
mercial point of view, it becomes essential to determine the optimum
population catch from the group. During systematic, epidemiological,
and geographic studies of mammals, questions about the real dimensions
and the size of individual populations confront the investigators. It is
necessary, as far as possible, to determine accurately the degree of in-
dependence (isolation) of individual populations, to decide these and many
other questions.

Data on population variability are capable of specifically and con-
vincingly contributing to the solution of this question. The values
characterizing the variability of traits have a great importance in determining
the independence of a population. Comparing populations in many
traits of any dimensionality is obvious, and can be done by comparing
_ standard deviations of relative values.

The study of a population by morphological methods should necessarily
be carried out in conjunction with genetical and ecological methods. An
elaboration of the best methods of mathematical analyses of biological
material should be the important approach during these researches. But
even the combination of population morphology with the genetics and
ecology of a population, and the elaboration of the best mathematical
methods of analyzing the material, are complex problems that are far from
solved to this day. However, this should not stop investigations in the field
of population morphology proper. ‘These studies—and among them, the
study of population morphology is the foremost—are capable of making a
substantial contribution to the study of the mechanism of the evolutionary
process.

SUMMARY*

1. The work is devoted to a study of the characteristics of variability
of mammalian organs, structures, and functions within a population
(‘variability is meant as the presence of differences between individuals
within a population). Original data of the author concern the morphological
variability of Pinnipedia, Mystacoceti, and Odontoceti; apart from them,
data on the variability of many characters of several hundred populations
belonging to 10 other mammalian orders are used in the work.

с.100
x
tion (с), despite some disadvantages, were chosen as the main indices
of quantitative variability; qualitative variability of the type of epigenetic

polymorphism is estimated by means of the methods of x? and A.

2. Whether a given genotype is preserved during evolution depends
upon the fitness of the phenotype. That is why a study of any phenotypic
variability in natural populations contributes to the understanding of
the nature of micro-evolution. Unfortunately, the study of the population
as an elementary unit of evolution receives but little attention from morpho-
logists and comparative anatomists.

3. The “importance” of any character is relative, as all organs are
multifunctional; they function at various degrees in various periods of
ontogeny, being correlated, so that we can never say we know all the
functions of the given organ. On the other hand, general regularities of
variability must be the same for all characters (irrespective of whether
this concerns the length of tail or the number of cells in the cerebral cortex).
Because of this, general regularities of variability can be studied on any
characters.

4. Ву the value of variability (с. v.), mammalian characters in samples

The coefficient of variation (« д. = ) and standard devia-

* Presented in English in the 1966 edition of the Russian work.

270 Vartabiltty of Mammals

of the same sex, the same age, and the same time (i.e., in “pure samples’)
can be conditionally divided into three large groups (Chapter III):

a. variability is less than or near 10%, e.g., that of linear dimensions
of the postcranial skeleton and the skull, the number of tracheal rings, the
absolute weight of the brain, linear dimensions of hair and of the body,
the number of vibrissae;

b. variability of 10 to 15%, e.g., linear dimensions of the intestine,
the amount of hemoglobin, the number of elements in the postcranial
skeleton and of the nervous system; and

с. variability over 15% is typical of the weight of the body, skeleton,
skin, dimensions of the urogenital system, the number of hair and erythro-
cytes, the weight of the skull, musculature, heart, lungs, stomach, and of the
majority of the glands.

These figures characterize the mean variability values of characters
that may be higher or lower in specific cases. Such deviations from the mean
values may show the trend of the action of the controlling evolutionary factors.

5. A comparison of variability of populations belonging to different
species, genera, families, and orders (Chapter II), shows that the amount
of variability does not depend on the taxonomic position of the populations
compared: Variability of the same characters in populations within a
species can differ more than that of populations of remote species.

Such a diversity in variability seems to reflect the difference in the trend
of controlling factors, both ecological and genetic ones.

- 6. In previous classifications of the phenomena of phenotypic
variability, different categories of variability were mixed up. It is suggested
that variability be regarded in at least three independent ways (Chapter IV).
By type, variability can be morphological, physiological, biochemical,
and ecological; by dimensionality, we may distinguish linear variability,
that of volume, weight, area, color, angle, temperature, time, meristic
variability, and epigenetic polymorphism; by form, variability may concern
age, sex, generation, chronography, biotope; it may also be geographical
and teratological.

Experiment shows that only the most “pure”? samples (1.е., when
taking into consideration the variability categories distinguished) provide
a reliable comparison of populations as to their variability.

7. When studying the variability of many series of functionally related
and similar characters, variability shows a trend to decrease with an increase
in the absolute value of the character. On the background of such ele-
mentary relationships, some characters can be found nearly always that are
distinguished by a higher or lower variability in the general “Ном.” These
characters, deviating in the value of their variability, are individually related
to the controlling genetic and ecological factors:and may be the objects of
special investigation.

Summary 271

Other elementary regularities determine the relationship of variability
to the dimensionality of the character, and the relationship of the variability
of absolute and relative indices.

8. If several different characters are arranged by their increasing
or decreasing value of variability, it turns out that in many instances,
despite considerable fluctuations of variability from season to season and
year to year, the position of each character among others remains the same.
Such a “comparative” variability shows that within a population, variability
of all the characters at any moment of evolution is correlated.

9. The fact that an organ or a structure disappearing in phylogenesis
undergoes a considerable variability (Figure 67), is an important adaptive
property of the population. During many generations the population
“checks” whether a structure that has become useless may again acquire
selective importance. Such a “prolonged” variability permits adaptation
to periodical environmental changes; it restores the selective value of the
structures and permits a change in the function of organs and the direction
of their evolution.

10. А study of variability may serve аз an effective tool for insight
into the regularities of micro-evolution when:

a. comparing variability of natural and experimental populations
kept under strictly controlled conditions ;

b. comparing variability of populations under the conditions of
“natural experiments,”’ i.e., different types of band-shaped ranges, island
forms, and cases when it is possible to clearly distinguish the effect of environ-
mental factors;

c. establishing relations of the variability of structures to their
selective value.

11. A great amount of work still has to be done aimed toward the
development of a general theory of variability and the understanding of
the characteristics of variability in the main groups of animals and plants.
A study of variability within a population gives a key to a deeper ecological
investigation of the group. ‘The ideas and methods of population morpho-
logy contribute to a solution of the problems of micro-evolution.

APPENDIX

THE VARIABILITY OF BODY MEASUREMENTS AND
INTERNAL ORGANS OF MAMMALS*

Table 1 .. Coefficients of Variation for Some Body Measurements

Table 2 .. Coefficients of Variation for Parameters of Hair Cover

Table 3 .. Coefficients of Variation for Number of Vibrissae

Table 4 .. Coefficients of Variation for Skull Measurements

Table 5 .. Variability of Linear Measurements of the Mammalian Skull

Table 6 .. Coefficients of Variation for Linear and Meristic Traits of
the Postcranial Skeleton

Table 7 .. Coefficients of Variation for Absolute Weight of Some Glands

Table 8 .. Coefficients of Variation for Length and Weight of Gut

Table 9 .. Variability of Some Characters of the Respiratory System

Table 10 .. Coefficients of Variation for Absolute Weight of Heart,

Kidneys, and Skin

* The number of samples investigated are indicated in parantheses throughout.

Appendix 273

(‘рзиво)

LOGI ‘лозхе — 9°6 0:8 ЕП с'9 “8 8 — snotupUns snasvuLary
C96] “SuoT a — — — (5)175-1'6 °° * ppnvnnasg эта
C961 ‘SuoT = == = — с*6 227200
C96] “SuoT =. == —= — Lv а. РОЛ
С96Т ‘24лемЧос = = 8's 9:9 — о
6961 *“Z1I@MyoG О ТТ — 0’? 7'6 OSG © v bs 2: uopouanydop AS!
C961 ‘2лемЧ2$ — —= 75 £9 LG
SG6I ТэиЧэ( ‘tysmoiog (51)5-с1-Р"8 = a —= (98)9:9-2°5
8961 ‘Aosjoq (6)с-/-8°6 == ($) 5-11 ($)3°6-8:6 сео °° °° о ми
$961 “Aosjog (<)/`9 = F'S-6'1 SPs >
896Т “OxUI I] Ol = = — ==
9661 “99nd ~—s (F) FF I-Z'9 — == = (D)FL-1'S
6961 “O96I “Zizemyog — (Z) 9° 1 Z-G 6 = (ео (Z)6'9-8'0 S11
866Т ‘Teuyeq ‘“Язмолоя (6 Z)Z'87Z-1'6 os = — (97)2'8-0°5
$961 ‘ло8То(Т ($) 16-61 = ($)с*6-Р'6 ($)с:9-9`6 ($)5'6-8'2 ° * * * = * о 502042 406
BVIOAIZIISUT
1 9 © P 6 с I
леч 39эЯ пет, Apog
eyed О areata ios ee dine ae Sas satoadg
jo ysuo'T

s[eulwieu эито$ JO 53заэшэлизезит Арод эихо$ лоу UOTVIIVA Jo S}UIIOWF209 uo езер рэгтлеолонис *] э] чет,

Variability of Mammals

274

QG6I “UIST

G96] ‘SuOT

8961 “SuOT

OZ6I “72 12 UMOIg

1861 “pseO

6661 “GG6T “UIMeEg ‘rou eT
Z96] JUNE ‘цозлэриу
9661 ‘чозчола

E961 ‘ЭЭ7Т ‘1э351эон

С96Т “499787

$66Т 1Ээ9Чэ “Hsmosrog
096Т ‘237е^Ч5$

$661 Ч29Чэ( ‘тЯ$молоЯ
9661 ‘uezeg

Sc6l ‘тай

LOGI ‘лоте

(¢)9°S1-9°SI
(9§)6'81-8'F
(9)0°12-S'ZI

(Z)§ 11-86

(9)8°§2-L°S
HS
(L1)6'F1-6'8

TSS
(9)9°91-9'6

TS1-8°61

(11)8'9-6'S
="циэро

LCl-8 Tl

eydsour0sey]

VG-CP

рапицио)—1 e192 L,

(11) #'F1-2'F

8`61-7'8

0°51-9°/

(11)6°ZI-1'S

(11)8°9-0'§
AG
CrCr

(§)S°F-1'S

(9)0°6-0°4
(8)8°8I-S'F

(¢1)8°9-0'S
тот

5107100 snuapody

* 5022244 ви010120

57]12П9 5192] 01244()
* омотяюцу “т
* 5721440102 51921

qognpnp 580110]

vavgoina #90,

* sninjouUoud "М

suatpof stuoanr
* 211124 10151215,

snavgoina “i

Е a Е Е ee a a er eg НЕЕ

("p1U09)
$961 ‘99 PUL лэззтэацон

1761 ‘тео

Appendix 275

ТУбТ ето
096Т ‘23темЧэ$
096Т ‘23темчЧэ$
0961 ‘22лемчЧэ$

C961 ‘Зиот

$76Т “UIODH

5961 ‘АзЧО рие иозмэм
POG] “eysuAzoqory

S96] ‘лэзапиен

C96] ‘Зиот

СО6Т ‘23лемчЧэс ‘AoyZeys [og

8961 “Ozu TT
С96Т ‘23лемЧоб “aoyeys [og

1961 ‘Aets1095) “aoysog

6661 “eysmozourepy

96] ‘тт

0961 ‘23лемЧ2$

C961 “921g чел ‘зиоле) JUTeESg

(Z)9°91-§'SI
(FI)L°6-1'F

8 sl

(в) 5. ст-б"/

- (8)9°75-Г"61Т

CGT

(¢)8'F-0'S

(/)<"6-9'6
(в) Гт 1-6°/

(8)0°L-I'F
(Z1)S°01-0'S

(g)2°9-E°

($)0°5-6'1
($)5'5-с°<

(3)9"5-с°5

(/)6'9-6°6

(S)P'LI-F'§

(Z)L°01-S°2
(ZI) I°L-6'S
(CI) 91е=Аа

($)G°6-L'S

(¢)2'9-4'F
(+)0°9-8'F

(8)F°L-4'S

(9)F°§-8°0
PC

(L)9°Z1-#'S
(S)9°FI-L'L
(8)¢°1§-8'°6
(Z1)S°91-6'F

(¢)2'°6-1'9

($) 56-16
($)5°5-с°6

(8)9°¢-z'¢

fv

(Z)¢"01-9'8

1S
(/)/'8-1°6
(5)5`/-9°$
(¢)6°6-0'S
(3)0:6-/°6
(Z1)6°T1-9°9

(5)6'8-6"6

TUDIALOUL ‘d

SnUa4a * J

57140212] 51250421

5144594421 2]02104 V

5212040 ИГ

5700230 * AT

* 42150504120 ‘ИГ

5721120 ]45ии24 ИГ

58159480 5710491]

* 2499425 *15

57]024018 17)

snpyns 50и0142] ")

5124100] 5 `У

276 Variability of Mammals

6961 ‘HIsmoudg
3661 “72 12 э13$°0
3561 LSeD
8661 ‘мет

9661 “7 12 913580
IF6I ‘2050

606] ‘лэчитае
Z96I ‘7темЧ2$
6661 ‘РлоДэхэеЧ$ pue $00

с96т ‘Темцеча
6561 ‘ЗАтУ

0961 ‘23темЧэ$
E961 ‘232118

1661 ‘тэз5тэауоН

S96] ‘72 12 чомахе о)

0261 ‘zeuuINS
1P61 ето
Poel ‘JouwINS
E76 ‘touUINS
8661 ‘°1d

(9)6'05-6'61
(8)0°Z1-8'6
(с)5'6-6'8
(25) /'65-9'6

(8)с'8-5'9
(71)8`75-Р"С1
(ето
(5)6`11-0:9
(23)5'85-6'8
(Z)8'12-G' LI

(9F) 181-801

(с)5:/-/'6

«(Z)1'9-6'P
(F) PPL'S

($)L°6-8'6

(91)8°S-L°Z
(¥)1°G-9'§
(11)/'6-8'6
(6)6°S-L°%

(9)0:5-6'5

«(Z)9°9-2°9
(2)7'8-6'5

(5)5'6-5'$
(21)6'6-0:8

(эт)т'8-6°1Т
(0$)0'5-6'5
(8)S°§-6'I
(ZZ) 1-9
(6)6°§-2°%

(9)F'01-1°9
PY AL 2 LAG
61
(¥Z)S°92-9'01
(S1)9°¢-I'F
(8)0:5-8°6
*(Z)G°L-§°S
(Pp) 11-F'§

(§)1°Z1-2°8
(21)5:/-С'$

*(91)8'Р-5'8
(9т)56-6'$
(06)6'3-/ "7

(2)7’/-р'6

«(11)9°9-8'°F

(6)8°8-0'S

‘yysugy] элиеэл Jo AIqeIeA y

(9)0°01-6'9
(3)8°6-6'5
(¢)§°1-2°0

(FZ)0'16-6°L

(8)9°C-L'I

(2)9:9-6'6

[79
(Z)¢°8-6°L
(§)0°9-0'S

(21)9:/-5'8

(05)9'6-1'$
(2)7'6-б'7
(11)L'9-0'F
(6)5:6-Г'$

ропицио)—1 этает,

> snynasnue пит
22112912 виерио

чабзив] 11172и1 4)

5704 SNYDY

5120544 °”)

зиррарэ зтраио)
1214 “qi

snjououog “gq

* SNIDINIIUDU * J

Appendix 277

ВСбТ ‘позлеза
866Т ‘чозхеэ4
Е о
8667 ‘позе
СЪ6Т “AOYSpurTS
IL61 ‘Арго
096Т ‘1эЯ2ТРЕТ
096Т ‘тЭЧЭтртТ
C961 ‘Зчо’Т
961 “Suo'T
8961 ‘Su0T
$961 “OAeTM
P96] ‘зэбтетУ
0961 ‘ртеяова
0961 ‘рхеяэза

LOGI ‘2,A081 [ATS]

(6)0°ZI-F'1

(8)§°§1-0'°01

(Z)0'0

(5)6'6-6'6

(3)с°с-8'8

VE-1e

(<)7'6-5'6
(Z)L°S-F'S

(Z)1'4-0'S

(8)L°L-@'S

(5)'6-6'8
(Z)6°S-L'S

(Z) 9-29

(6)6:4-1-9

OL

5'9-0'6

(<) т0т-7'8
(“аа
(8)8-8-/"Р

9°5
8'6-6'6

OF
(Z)1°S-L°E
(Z)G°b-8'S

lh
(€)9°€-8'E
-(Z)0°S-8°S

(Z)6°S-S'S

шттри Ча

* muyosffjom “Yq
eS во Ча
адамтр уно ИИА

* 57]22409 D10d/)

1pi0 а

SHEE XT

пита sCuopogrd
WUOSPADYIL "6
snjyoulan snpryqousads
saprogyny stumouoy 1,
snouospny sndvy>
SnIDINIINGUN 52ио14 ИГ
140] @7 Я

571251 5иотЯ

574150] пап

278 Variability of Mammals

6561 ‘чозпаен

С96Т ‘1232428

076Т ‘лочечуэа

езер чмо.

вер чмо
eyep имо
езер имо
езер чмо

езер имо
езер чмо

езер имо

eyep чмо
еуер чмо

езер UMC
е]ер UMC

(0Z)0°S11-0°S

yys1om Apog

(Z)8°11-2'SI
(<)5'81-6°1Т
5:95
6'3

G66
9°6

OTT
£91
8°SG

($)971-6'8
(5)9:6-0°/

(sojddiu
u90M}0q)
stuod-snuy

рапицио)—1 эта,

(2)5:91-5°/
(Z)€°Z1-8'8
8'55-9°6
$'/-Р'9
(Z)6'01-F'8
(4)8°Z1-9°9

erpedraurg

[э^леч- паху

(Z)§°6I-1'6

(¢)2'6-9'F

J@9-9SON

(<)6`01-6'8
(05)7'51-5'8
0:8

(21)7°/-5'6
(61)9°L-F'1
(F)E'6-G°L
(1)Z°8I-1'8
(с)28-Р°5
(02)2'6-§'F

(SI)PL-1'S

i ee ee

* 10192] vounape py
* snuisin Snuty40770‘)
° рипио 0204

2171250 0204014494]

7р1451 vsng

72141915 DSN

2214502 vsng

* 01015142 0401901547)

5171рир]иа045 5п]и/4050]

i a a a a a a ee a oo

es es ee oe ee oo oe oe eo oe АА oe oo. a a

Appendix 279

(‘p7u07))

9561 ‘лэчизет-

1861 ‘^э98О

Т86Т ‘^298О

8961 ‘чот

L961 “uredoy
1961 SD 72 4993} Я

IF61 ‘AOsjouzny

OF6I ‘ЧэПУ
F961 “UIST
<96Т ‘.1e82S

F9GI “UALS
096Т “ACTOXOS

IF61 “eaodog

(9)L°S1-0'°S1
(+)<'61-Р'6

(g)F'11-3'8

(96)/'05-0'6

= (6)5'/-7'9
елолталео)
= xx(b)E°C-0'S
a(E)TL-9'S (2) FT 1-8'6
еАзэеротзху
19Э ет,
jo чЧ18иэ7

"sIOYIIM 3е 34819Ч АроЯ x x

"413 9э] элце[эл JO Азтаемелд x

(2)6°9-8'F
6%
Uh
48-1"
(9)P°S-9'E
(9)6°L-€'%
(O1)8°E-L'T

CL
LT
(+)6'6-6"6

#*ж0'7-Р'С
xa(b)S°h-8'S

* 57402 524
* 5141] АРИЯ
* DAN) РАПТ
* 57х01 DapixD J,
DIU 2121577
saqjna 5ад тд

2112912 SAID

имея заодолрОН

* *ф 5п21020РО

* snpuning safiouny

> 574781 sog

“ds 50

a a a we rm ae ae Ry yg a | ee | nO С aL Se eg

рапицио)—1 этаех,

Variability of Mammals

280

9961 “Su0'T Е — — —- 16 2 : к 5722142 DJUOPOXOT

vaprosoqorg

98561 ‘5м^эЧ433= == = == = (Z)Sh-Ch ° ° * 5122404 viadouavjog
28861 ‘SMOYNLI = == — — (с)+'91-0'° ° `` snpvydooossvm ааа
©э2е3э))
L961 “10 42 пзтатя 2:91 pa pe = == "+ + 59020 snaayjidoosar
LOGI “12 42 изытЯ — (7) FL 1-8'F1 = = — — ae 5 6 и,
зэзеитл
G96T ‘Зчот a == = = (Z)VL0°G ° ° `` * Sagronyf snurysajup
eiperdnsavyy
C961 “SuOT =r = = — L9 * * snurywun зтриалоциио
езептэлзоцо{
G96I ‘SuO'T о == = —= CZ отт apuasumo, то
C961 ‘Зиот = rs = = 9'6 "tt 51251} зп5ЯдА
C961 ‘чот те = = = IEG (7220217557494
елэзЧолт4о
L 9 G т 5 G |

рапицио)—1 этает,

281

Appendix

‘(ТРбТ “PAcdog) (8)F'E% 0 9:01 “4 5120 UO (6061 ‘1эиитс) (z)g'9z 01 9:65 ‘snjnosnu snpy UO теч Jo 14SIOM\—Z
“(6061 “t9uUINS) (5) /°/Т 03 "ТТ ‘ии snpy Ul ээвута$ Apoq этап UO меч до лол -—Т : зчязоучуно YIHLO

GZ6I ‘susONyT = — = =

(4)0°L-G°¢ * 521425 Ошо

$961 ‘эээчеЯ = а oa aa (0F)8°61-8°¢ " 5121405 Ошо

ТУбТ ‘ело4оа (8)6'6-9'05 — (8)6'65-0'6Т oe zs = = с

8©6Т “BAousersy) (6)$'01-8'9 (05)6'6-9"5 (€)9°9-1°@ (S)$°GE-1'S1 (0Z)0°OF- 1°/ (€)2°L-8°9 22912 samy

8S6I ‘лээрея = (E)O'ZI-E'Z = (ST) Z01-0'¢ (€)9°Z1-8°S (g)Z°GI-8'¢ — (61)6`01- 671 (S)GI-1'T * 579602 40150207

E961 “PARISON (F) F'01-7'G (8)8°6-6'1 (Z)G°1-€'1 ($)9°11-8°5 (8)¢°b- 8`0 (6)8`0 52610042044 5адпала12\[

$961 “2^э330 (9)Z'9-2'0 (42) F°9-1°% (9)5"/-5°6 (9)6°0I-1'8 — ($6)6'с- 8'0 (9)/-5-9'0 " snpuun тд

E961 ‘2^э780 GEE (F)0°L-1°% ($)5'8-58 . (¥)I'S-9'1 (€)E°I-1S°0 (+)L°1-9°0 vandoina 20 7,

“att, LEB Baer pia ane pram ame
полу рэ поте) Чаи] SSOUROT T, 2

S[PUUUALUE SUTOS UO лэлоэ ATEY JO Stoj}0UIVAET лоу UOTILIALA JO SJUSTOYJa0D °z эает

282 Variability of Mammals

Table 3. Coefficients of variation for

ee Oe Oe OS eS ee ee eS eS нивИЯ

Species 0 ss
Ist 2nd 3rd 4th 5th 6th

Tibia о м
Phoca vitulina . . - 0.0 3.8 5.6 2.0 5.3 6.5
Histriophoca fasciata . ИО = KO) 6.7 7.9 13.7 14.2

Равор из groenlandicus 4.2-10.6 Вт 5.3-7.7 ' 6.5-16.7 8.4-13.2. 19.2267
(17) (17) (17) (17) (17) (17)

Cystophora cristata . . 8.5-9.8 7.7-11.1 5.9-9.7 6.8-9.0 8.3-16.3 12.3-38.6

(6) (6) (6) (6) (6) (6)
Pusa сариа . . . 54-88 4.6-56 — 5.477 5.8-82 4.9-11.3 14.2-21.9
(8) (8) (8) (8) (8) (8)
Рша sibirica . . . 4.2-52 5.2-7.8 5.5-6.8 3.9-64 6.6-11.1 11.0174
(6) (6) (6) (6) (6) (6)
Pusahispida ©. . 63187 56100 5466 63-129 211
(6) (6) (6) (6) (6) (6)

Pinnipedia Е _0-17.1 3.7-11.1 5.3-9.7 2.0-16.7 4.9-16.3 6.5-38.6

number of vibrissae in some mammals

Appendix 283

—S eC CI

Total lip
vibrissae

Eye
vibrissae

Nose
vibrissae

—_ eC OC

7th 8th 9th
11.0 266.7 —
67.5 32.0 —
30.1-47.8 130-594.0 0-980.0
(17) (17) (17)
15.6-136.8 — =
(6)

20.8-28.8 38.5-74.2 214-342
(8) (8) (8)

21.0-29.4 43.1-101.0 150-600

(6) (6) (6)
3.5-46.0 65-1189 —
(6) (6)
3.5-136.8 32-1189 0-980

2:7
10.0

3.8-7.8
(17)

5.4-6.9
(6)

4.4-6.0
(8)

3.2-6.8
(6)

4.1-8.6
(18)

2.7-10.1

30.5

25.8-37.2
(17)

16.6-19.6
(4)

11.3-22.6
(8)

1.6-35.8
(9)

9.3-36.2
(5)

1.6-37.8

13.5-35.3
(15)

13.7-78.7
(4)

26.6-53.0
(8)

30.8(1)

20.8-31.0
(2)

13.5-78.7

ole) ad, Okhotsk Sea
SS + $9 ad, Okhotsk Sea
So ad, Okhotsk Sea

All sexually mature groups,
1961-1963, North Atlantic

Different sexually mature
groups, Greenland Sea, 1962

Different sexually mature
groups, 1962

Different sexually mature
groups, 1962

Pechora, Okhotsk, Bering,
and Chukchi Seas

284 Variability of Mammals

Table 4. Summarized data on coefficients of

Condylobasal

length

Mastoid
width

Inter-orbital Width of

width

Length of

toothrow in

upper jaw

Dipodomys nitratoides

Apodemus sylvaticus .

A. flavicollis

Microtus agrestis

M. pennsylvanicus

M. ochrogaster

Peromyscus maniculatus .

P. leucopus
Р. eremicus .
Р. merriami .

P. truer .

Clethrionomys glareolus .

Cl. rutilus

Cl. gapperi

Mus musculus

2.4-3.1(3)
6.2-15.2(3)
2.5-5.5(10)
2.7-4.9(3)

2.2-5.7(10)

2.0-6.8(8)
2.9-3.4(4)

2.6
0.4-3.6(6)
4.7-5.0(2)

2.6
1.8-2.7(8)
1.3-2.3(12)

2.50(30)

2.3(8)

2.0(4)
1.9-2.6(3)
1.3-2.2(7)

2.2-3.6(7)
2.6-6.2(4)

3.2-7.3(5)

125)

2.2-9.8(2)
3.0

2.6-5.1(6)

7.8-13.1(3)

1.0-2.7(3)

3,7-4.3(3)

3.7-4.2(4)

2.4

5.7-6.1(2)

2.6-4.2(3)
2.5

2.4-7.0(0)

3.2-5.4(5)
2.4-5.3(2)

4..0

2.3-3.7(6)

Rodentia

3.3-5.0(3)
2.7-4.6(10)

1.4-4.0(13)

Tee

4.3-5.8(2)

Do?

2.6-3.1(3)
3.3

1.8-6.1(7)

2.7-5.4(5)
3.0-3.4(2)
3.0-3.2(2)

3.5

Appendix 285

* e e
variation for skull measurements in some mammals

Length of Height Ris и
] Ni F skull matic of the Data
PS ead os width palate
й 8 9 10 11
— — — — Boolootian, 1954
— — — — Pelt, van Bree, 1962
— — 1.8-4.9(10) — Резвеу and Georgiev, 1961
3.8-5.6(3) — 3.6-5.2(3) — Saint Girons, van Bree, 1962
— — — — Ursin, 1956
— — — — Ursin, 1956
— 2.9-3.9(4) — 3.5-5.6(4) | Adamczewska, 1959
— 9.6 2. 10.0 Haitlinger, 1963
= 2.7-3.9(6) 3.2-3.6(2) — Wasilewski, 1956
= — 0.4-1.4(6) — Gebczynska, 1964
— `` 5.6-6.5(2) — — Goin, 1943
— 4.1 — — Long, 1965
— — — — Sumner, 1920
— — — — Sumner, 1923
2.7(30) — 2.4(30) — Clark, 1941
2.6(8) — 2.3(8) — Clark, 1941
2.3(4) — 1.8 (4) — Clark, 1941
— — — — Hoffmeister and Lee, 1963
— — — — Hoffmeister, 1951
— 2.2-4.0(7) 2.0-3.9(7) — Bol’shakov, Schwartz, 1962
— — _ — — Wasilewski, 1956
— 2.5-3.8(5) 4.8-6.3(5) — Bol’shakov, Schwartz, 1962
— 1.3-3.3(2) -- — Т?епКо, 1968
— 3.5-4.0(2) — — Manning, 1956
— 3.4 — — Long, 1965

= — 2.7-4.4(6) — Dynowski, 1963

286 Variability of Mammals

Table 4
во
Caiter fier о ee ae
. 1.7-2.2(2) 2.6-5.3(2) -— — —
Ondatra zibethica . . 4.3 2.8 6.1 — an)

4.5 — 6.8 — 4.0.

1.2-2.1(2) — = one gre
Lagurus lagurus. . . 1.2-5.3(10) — | 6.2-10.6(8) — 4.5-6.2(8)
Baiomys musculus . . 1.8-2.3(2) 2.2-2.7(2) 2.2-2.9(2) — 2.8-3.2(2)
Вю, . 22272) 3.3-3.6(2) 3.2-3.3(2) — 2.6-2.7(2)
Spermophilus armatus . 3.1-3.2(2) — — — 2.2-3.2(2)
S. сратфот . . . 3.0 — = — 4.0
Rattus rattus . . . 5.9 3.6 Ag — 4.9
Кари; hudsonius. . . 27 3.6 5.8 — 3.6
Сала porcellus . . . 3.0-3.2(2) 6.5-7.5(2) — — —
Thomomys talpoides . . 3.1 — 6.5 — 6.3
Dipodomys ordi . . . 2.3-2.5(2) 2.6-3.0(2) 3.4(2) 5.7-6.7(2) 4.6-5.0(2)
Dsvmerriamie а №5172) 1.1-2.3(2) 2.7-3.2(2) — 5.4-6.1(2)
Devinsularisi secs ate 1.8 2.3 2:1 — 1.9
Sicista betulina . . . 2.3-3.0(2) 2.5-3.2(2) — — —
Phyllotis darwint . . 2.9-4.5(3) — 3.8(3) — 3.7-4.6(3)
Phvosilae . . 2) 3.2-4.0(2) — 3.7(2) — 3.6-3.9(2)
Ph. wolffsohni . . . 4.0 — 4.6 — 4.0
Вл апанни (ie Guise oe 123) 2.3-3.2(3) — — 4.2-4.6(3)

Monotremata
Ornithorhynchus anatinus 6.0 — 6.8 — —
Marsupialia

Didelphis marsupialis . 6.3 — "— — 4.0

Appendix 287

—Continued
аи i
р
он mad = — Hinze, 1950
4.1 isan 4.9 10.0 Latimer and Riley, 1934
see = 4.5 — Long, 1965
— — 2.1-2.4(2) — Schwartz, 1962
— 3.3-6.6(8) 2.1-5.4(10) — Schwartz, 1962
— 2.5(2) — — Packard, 1960
— 2.5-3.6(2) — — Packard, 1960
— 2.8-3.3(2) — — Long, 1965
— 39 — — Long, 1965
— 4.9 — — Dhaliwal, 1962
г 4.0 — — Whitaker, 1963
— — 2.3-3.1(2) — Strandskov 1942
Ей 5.9 — — Long, 1968
2.9-3.2(2) 1.9-2.2(2) 3.0-3.1(2) — Desha, 1967

mae fut — Lidicker, 1960
= pas oe = Lidicker, 1960
— 2.2-2.7(2) — — Kubik, 1953

ak a pad — Pearson, 1958
— Pearson, 1958
— 3.8 — — Pearson, 1958

— — — — Pearson, 1958

— — 5.4 — Long, 1968

5.6 — Long, 1968
(Contd.)

288 Variability of Mammals

Antechinus flavipes

Sorex araneus

S. minutus

S. daphaenodon .
5. palustris .

S. cinereus

S. vagrans

5. nanus .

S. raddet

5. arcticus

$. isodon

Crocidura suaveolens .

С. leucodon .

Talpa europaea .

Erinaceus europaeus .

Е. romanicus

Tadarida brasiliensis
Т. sp.
T. fossilis

Myotis sodalis

2.4-4.9(2)

1.5-1.7(4)

2.3-3.0(2)
1.9

1.6-1.3(4)
1.3
1.4
DI

25)

2.9
17180)

2.3

DD)

2.0-5.2(12)
1.5-2.9(3)

1.3-5.9(8)

2.5-13.1(3)

17
0.6-2.5(2)
2.8

1.7

1.8-1.9(4)
3.2

1.6-2.6(4)
3.3
3.2

1.6-1.9(2)
2.5

45.7

3.1-3.9(2)

Shi

2.0

Table 4
4 5 6
2.6-4.5(2) — 1.5-5.1(2)
Insectivora
3.0-3.8(4) = 1.9(4)
0
1.5-3.0(4) — 1.3-1.9(4)
ЗЕ == 2.9
4.4 — 3.6
6.6 = 2.8
5.9 — 3.2
6.0 — 3.3
3.3 — 3.8
4.5 — 5.1
29 = 21
4.1 — 5.1
3.4-3.7(2) — 1.7-2.0(2)
3.9 — 3.8
5.0 — 25

3.1-3.8(2) 3.9-4.1(2) 1.0-2.2(2)
3.6 5.3 2.5

Chiroptera

— т SAO)

= = 2.8

Appendix 289

—Continued
Е ое О
ор ce
= 2.4-2.6(4) = — Dolgov, 1963
— 1.8-4.0(11) — —- Рибек, 1956
—- 4.6 — — Schwartz, 1962
pie eae — -- Dolgov, 1963
— 4.3 — — Lavrova, Zazhigin, 1966
— 5:2 — - — Schwartz, 1962
aut == = — Schwartz, 1962
ae: zen = = Long, 1965
se Е aes — Long, 1965
о aie = = Long, 1965
Е = ws — Long, 1965
— 4.7 = — Lavrova, Zazhigin, 1966
— 6.3 == = Schwartz, 1962
— 2.0(2) — —- Dolgov, 1963
— 3.6 == —— Lavrova, Zazhigin, 1966
— 3.2 — — Lavrova, Zazhigin, 1966
ute za — — Stein, 1963
ae —_— = — Skoczen, 1962
2.6-2.9(2) — — — Markov, 1957
3.4 -= — — Магкоу, 1957
— — — — Long, Jones, 1966
1.5-1.4(2) — — — ]е а, Hall, 1962
0.8 — — — Теа, Hall, 1962

aon = — — Bader, Hall, 1960
( Contd.)

290 Variability of Mammals

Table 4
a ee 20) oe 4
Mitte oe Te
Macrotus waterhousti . 1.5-2.0(4) 2.0-2.9(4) 2.9-4.3(4) — —
Eptesicus fuscus . . . I) AN Dall 9) — —
Plecotus townsendi . . 2.5 3.6 1.8 — --
Perissodactyla
Equus hemionus . . . 4.1 — — — 3.3
5.8 Bel — = pale
ES И ee eae nts 2.4 25) — — —
E.quagga . . . . ВВ а — — —
Е. рггеша ки . . . 3.7 3.8 — — —
И Сара 5 55 10.6 18.4 — — 8.3
Artiodactyla
Rangifer tarandus . . 1.5-5.0(9) — — — 4.0-5.9(9)
Hydropotes inermis . . 0.6-3.8(2) — — _— 3.4-30.5(2)
Muntiacus reevesi . . 1.9-3.4(2) — — — 3.8-7.4(2)
SIS SHOE По 3.2 3.7 — — —
о hemionus . . — — = = =
О. virginianus . . . 4.6 Dall — — 5.4
Merycochoerinae. . . 3.49(5) — 9.68(5) 6.23(5) —
Merychyinae. . . . 4.93(5) — 8.06(5) 7.15(5)- —
Capra саисахка . . . 2.9-3.5(2) — — — —
С. cylindicornis . . . 2.6-3.3(2) — == — а

Cervus elaphus . . . 3.6-4.0(2) — 3.3-5.6(6) 4.2-4.7(2) 3.6-5.3(2)

Appendix 291

—Continued

Be ane — — Bader, Hall, 1960
ee == — — Anderson, Nelson, 1965
a 3.3 — — Long, 1965

ue 3.0 was ut Long, 1965

a sal 4.0 mt Allen, 1940
ue Hay, 1915

ua ane ae a Hay, 1915
am i an ua Hay, 1915
ня a ee Hay, 1915

— Nehring, 1884

my eee as ti Banfield, 1961
гы elas 0.7-6.5(2) = Allen, 1940
zie 4.0-4.8(2) os Allen, 1940
we 3.1 wR = Schultz, 1926
0.2-3.9(10) aa a ae Bergerud, 1964
а tas га Phillips, 1920
6.01(5) ae 6.45(5) 7.28(5) Bader, 1955
6.21(5) = 7.21(5) 9.37(5) Bader, 1955
a 11.7* 8.6** ss Tsalkin, 1956
os 9.3* 6.7** a Tsalkin, 1956
aut = Brna, 1964

(Contd. )

292 Variability of Mammals

Papio sp.

Alouatia palliata

Ateles geoffroyt

Macaca iris .

Tupaia clarensis

Homo sapiens
(Ancient Egypt)

(England)

Cercopithecus aethiops

Lepus europaeus .
L. mandschuricus

L. timidus

L. europaeus .
L. flavigularis

Oryctolagus cuniculus

Sylvilagus auduboni .

Ochotona princeps

Loxodonta africanus .

Physeter macrocephalus

Balaenoptera borealis

2.0-3.5(2)
2.4-3.2(2)

3.2.3.8(3)

2.2

2.7-3.2 (4)
3.3-3.5(2)

5.8
2.9-5.0(5)

1.5-3.2(3)

2

18.8

5.6-7.8(4)

3.0-4(2)
3.4-3.8(2)

3.9-5.2 (2)

4.1-5.3(4)
4.3-4.6(2)

7.0-9.0(2)

5.7-6.8(2)

3.4

4.0
Lagomorpha

4.8
Sl

3.8
4.8

4.5

2.6-4.1(3)

Proboscidea

Cetacea

Appendix 293

—Continued
7 8 9 10 1]
5-6.0(2) — — 5.0(2) Freedman, 1963

wel pas aes = Schultz, 1926
pat. pao == — Schultz, 1926

zo i un 6.2-8.0(2) Kirisu et al., 1967

— — 13.7 — Long, 1968

— — 3.6-4.8 (4) — Pearson, Davin, 1924
— 4.1-4.3(2) — McDonell, 1907

sale pe. a= 4.0 Kirisu ef al., 1967

ane se zi 10.4 Lebedkina, 1957

== о — 6.5 Lebedkina, 1957

a= = — 9.1 Lebedkina, 1957

oes aoe — 6.9 Filipchenko, 1916

a oe mul 10.7 Filipchenko, 1916

wee so ee = Anderson, Gaunt, 1962

-- ee dhe as ‘Long, 1968

= — — — Latimer, Sawin, 1959
= 2.1-3.1(3) — — Hoffmeister, Lee, 1963
= 519 — — Long, 1965

= — 10.4 — Long, 1968

= pest pet = Matthews, 1938a

— = Matthews, 1938b
( Contd.)

294 Variability of Mammals

Table 4

a

— ee

1 2
Delphinus delphis 2.9-3.7(2)
Tursiops truncatus 1.0-4.8(2)
Gulo gulo 0.8-3.6(7)
Lynx lynx 2.3-5.6(6)
L. rufus Е 3.5
Martes zibellina 1.7-2.9(11)
1.5-3.4(10)
M. foina 1.2-3.3(3)
M. martes 2.3-4.4(4)
M. americana 1.4-4.1(10)
Vulpes vulpes 2.7-3.4(3)
2.7-3.1(2)

Canis familiaris . 1.8-3.9(2)

(boxer) 4.2-6.6(3)

(““Rassenhunde

ingesamt’’, N=503) 37.8
С. lupus . 3.6-6.0(6)
Taxidea taxus 2.4-2.6(2)
Mephitis mephitis 3.6-3.8(2)
Enhydra шт . . . 2.5
Felis catus 2.4-5.0(3)
Mists vison 1.7-4.4(8)
М. erminea . 2.3-3.5(6)
Alopex lagopus 1.4-3.7(12)
Ursus americanus 2.9-5.5(4)

2.0-2.9(2)

3.3-6.1(2)

2.8-3.4(8)

3.8-4.5(2)

1.9-4.3(4)

6.0-8.3(4)

3.0-3.8(2) 3.1-4.4(2) 5.6-4.7(2)
4.0-5.8(2) 3.9-7.8(2) 3.2-5.8(2)

Carnivora

4.9 16308)

— — 3.8

3.7-6.0(11) 2.7-4.8(11) 1.8-2.6(2)

3.6-4.4(3) 2.5-5.0(3) 1.5-3.7(3)

2.2-6.5(4) 2.7-6.0(4) 1.9-6.3(4)

é

6.2-8.0(2) о
6.1-6.9(2) — 30410)
12.4-13.6(3), — 6)
29.5 ты 40.5
6.6-11.6(6) — 2.4-7.6(6)
a — 298)

4.5 Е 2.9
3.5(2)

6.5-7.9(4) 4.0-8.6(6)

4.3-5.7(4) 4.1-5.7(4) ae

Appendix 295

—Continued

7 8 9 10 11
4.9-5.1(2) 2.5-3.7(2) aa в Е IHS
3.2-4.9(2) 1.5-5.0(2) = — Kleinenberg, 1956
= — 3.3 3.5 Kurtén and Rausch, 1959

sls —_ == — Rausch, 1963

— — Ao — Long, 1965

= 2.6-4.3(8) 2.3-5.9(11) = Pavlinin, 1962
= — 2.2-6.8(10) 3.3-5.9(10) Kuznetsov, 1941
== 1.0-1.6(3) 2.9-4.4(3) = Kuznetsov, 1941
= 2.7-5.8(4) 2.4-6.5(4) = Kuznetsov, 1941

— — = — Hagmeier, 1961

2.8-3.4(3) — 2.0-3.6(3) — Churcher, 1960
== — 2.9-4.4(2) 3.2-4.6(2) — Rausch, 1953
— — 2.1-7.1(2) 3.4-4.6(2) | Rausch, 1953
— — 4.0-6.3(3) — Stockhaus, 1965
— — 24.2 — Stockhaus, 1965
— —- 5.0-10.2(6) — Stockhaus, 1965
РР == = — Long, 1968
— — 4.1-4.3(2) — Latimer, 1937
— — 2.6 — Ognev, 1931
— — 5.8-8.3 (2) — Latimer, 1936

2.5-6.2(3) 2.1-12.0(8) —- —- Pavlinin, 1962

ты | oh = = Kopein, 1967
— Smirnov, 1962

— 4,1-8.3(4) 5.6-8.4(4) 3.3-5.5(4) Harlow, 1962
( Contd.)

296 Variability of Mammals

Table 4
a ON Ga oe ee
О а. 215226) = — — —
4.4: — 6.5 — ii:
Сани |. 22-262) 3.3-3.5(2) | 5.2(2) 3.8-4.2(2) 2.1-2.7(2)
Entra luna ce 23-535] 3.2-6.2(6) 1.4-7.5(6) 0.9-6.2(4) 3.0-4.7(4)
‘Pinnipedia
и о а с ПО 2.6-5.1(9) 5.1-24.8(9) 5.7-10.1(9) 3.9-10.6(9)
Callorhinus ursinus . . 2.5-4.7(5) 2.5-5.6(5) — — 3.0-6.1(5)
Cystophora cristata . . 3.8-4.9(2) 2.4-5.0(2) 7.7-9.6(2) 9.4-10.6(2) 6.1-9.2(2)
Pagophilus groenlandicus 2.2-3.6(6) 2.7-5.0(6) 12.4-31.3 5.3-8.6(6) 4.6-5.1(6)
Hydrurga leptonyx . . 2.6 — 3.7 — 3.9
Eumetopiasjabata . . 2.4-4.3(2) 5.5-6.6(2) 5.8 — 8.5
Odobenus rosmarus . . 9.3-10.5(2) 11.7 — — —

* Length of 2nd to 5th segments of horns of males.
** Circumference of horns at base.

Appendix 297

—Continued
а eS
— — — — Rausch, 1963
— — 4.1 — Long, 1965
2.3-3.0(2) 2.6-4.0(2) 3.9-4.8 (2) = Chernyavskii, 1969
3.6-9.2(4) — 2.2-6.3(6) 2.9-6.0(5) | Hysing-Dahl, 1959
2.9-5.9(9) — 4.3-7.1(9) == Own data
1.2-3.4(5) 2.3-4.3(5) 5.0-9.6(5) — Muzhzhinkin (zn litt.)
4,2-3.2(2) — 5.4-5.9(2) = Own data
3.0-5.3(4) — 3.5-4.8 (4) a Own data
— — 3.8 — Hamilton, 1939
= = 5.3 = АПеп, 1880

= — 11.0 Ognev, 1931

Variability of Mammals

298

‘g[dures ,,uvojo,, Апиэтотуазит ло ‘suUOTVe[NI[Zd oY} UE холл o[QIssod илизуер onstzajOvIVYOUL.) „,

i i a SS ea ee

эзетеЯ oy}
jo Ч139эТ

(x2°61) SF
PG

yp
эцемозАЙ

mys
JO 398эН

Mel лэмо]
jo Ч18чэт

911-01
(9°91)
5'9-Р"1

9'01-0'8
(45°07)
9'1I-S'I

8'6-6'5
8`7-9°6

BOVE

Mef soddn
: мол 43003
jo Ч38чэ7Т

9'01-6'0

L9-L'S
9'01-°6
(0°14)

[0-60

SLES

ynous 9y}

© РТМ

Чарим
Тезтало-лэза1

Чар!
рто35е |

933991 [е5еа
-0]4риоэ

еее

ециэроЗ
erpodruurg
елолталео)

° 829810)
eydiowose’]
* Baprlosoqorg
* зэуеитаа

eyAjoeponry

e[Ajoepossiiog

eraydoaryy

° BIOATIDOSUT
* erperdnsieyy

еуемэлЗочо

[IMYS UelpeururvUr эцз JO ззиэлэлизезит лезиц JO (U7) Азитеетлел UO езер peziaeWUING ‘с этает,

Appendix 299

Table 6. Coefficients of variation (Lim,.,,) for linear and meristic
traits of the postcranial skeleton of some mammals
(From my data and those of other authors)

WMextebracy i. В в 2.0 — 14.9 (27)
О И а О о и ее 7.4 (15)
а о о. 22—66 (18)
И еасатра ан паева ата ее oe OO 94 (39)
О о О о ОСИ nee eae ат 2.3— 8.3 (68)
се 27-Е (Ag)
оО: 2411228)
Phalanges . Не NSIS eer ake 3000.1 (90)*

—_ — — — — — — — — — — — —— ——— —— — ——- —-

Number of ribs

Sechna Mees ppm eta оси. 00—97 (G0)

А: 00-214 5 (271)
Number of vertebrae

Бока сте ginal hinealopiea Kas ol Gate el Garh 19 92.0)

Caudalie Е Senegal eon caw may НБ У (49)

* Tadarida brasiliensis, 10.8 and 18.5 (Long et al., 1967); Myotis sodalis, 17.5 and 20.3
(Bader and Hall, 1960).

300 = Variability of Mammals

== * 57] [22409 121007)

G6] ‘тэбиет Fae 6°66 O16 = GLI 7 65 = =
1961 ($)

‘AID pue UosMaNy = — — — — — — с'08-6`06 — 51452490 ТОЛИ

GS6I “PIF (5) (5) `` 428]
речь рисков а aon Е PV CTS 66 9197-Е 5'88 GOOF 1['66-/'УС =e т aay PERE)

961 ‘23лемЧэ$ Jo (6) 0 ETS
ERNE} WWE S) [DEM НО) oa Tas = ae = = Е == 875-16 ИНО)

59212]
веры мО a re ee Se та 6'67 PITL 6LTIS) GSTY6o Рина

(5) (5)

ера“) = =: Sa = ars == Е ОУ СИЕ ^^ РН aN)
ALA BING) =e ca Sree = = San ZGOFI9Z LILFOLE 6'9-Е8'65 °° 2214509 вт

(z) @ fee
ззер чмо =e = = =: = =- a 057-6'88 $'756-7'6б 2101401540)

(с) (<) 517 рив]иго45
взер UMO = oe = = = = = 6'8-6'86 6 1P-9'C1 57119055

18] WEEN) = т = a Е == = O°9L 6'LE 21156
22049049 Е

рие[8 stsAyd d
улет АЕ ртолАЧТ, $1151, Алело от зтеиэлр\ svoioueg чээ]а$ JOATT

S[VUIUIVUL ат SpUL]S JUIOS Jo 34 8тэм JIN[OSGL лоу UOTVIALA JO SJUZIONJI0 ‘/ эт,

Appendix 301

(5) (5) (5) (5) (5) (5) (с) snyojnainsun
Ф96Т “OWI 6'61-6М 8'75-8'86 SEOTE6 6STS6C 265-86 6'65-6'8Т 1`65-Р/с G'S9-L'8S $'68-8'85 5201427
9561
‘2 42 UMOIG {6661 (5) (9) (6) (6) (9) (6) (6) "плит
‘UIMeg ри 127 = = ["59-8'96 LBS"HIE I'PE-S'BI S'Eh-9'6S 08-156 = GTS-B'OS = BIS"1°2Z 51010120)
(6) (+) (+)
956Т “72 32 лээзеЯ к 8`0/-8"66 = as = ar SbF 01 == GHC-O'S `` 529та sagjng
Ф9бТ 22а ри замл, CLF bs 91616 — == == — — = = > «ds sndosmypy
(9)
gc6] “Uezeg = 0°1/-6'06 Ба = = cae = = — хиаро{ stuoayy
(+) (9)
PARTE SEMAN SE os oe aE: a ст == SEbSl «GELS 22040 59 1,
(5)
2961 “PA01OIST — — — == = с*65-0'65 — — — 51.44524421 2] 021047

302 Variability of Mammals

(5)

S961 ‘129d PUL эамл, = = 0:97-8'96 Т'1Т-8'96 8906S PHTS'GZ — = = * «ds 51904207
(Z) : (Z)
LG) ACCS) GOI-L'6 — 6’51Т-Г'9 — — = — Q’SI-ZIl ° ия vjaysnpy
1761 ‘A0jOx0S (5) (5) (<) (<) (9)
1Р6Т “79 72 Ase — = QZI-S°ZI 9°9I-1°6 L'6-S'6 L:9-¥'6 = == SLES ° sadjna saging
рпг]
азер UME, == = о = == == 6'с-8°Р5 6'6-/°6Т ТТ-Р 5 мари а

Сбт “UrULIsy Aq
езер итол} рэпе) == = == — — — — = ГЕ-Р9ТТ ° 54 тт

2961 ‘итае3$\У Aq

вер WO рэзиэ[ео == = === = — — — = PEST Snsisaas0u snyoy
9561 “72 12 UMOIg (6) (9) (9) - уйти
C6] “towne T == = 6'56-1'91 = —= = = 6'15-8°ТТ 8°11-0'9 51901012440
(yoeuroys ау TIM (9) (5) (9)
3981 М) $961 ‘^2э3е = == == = = = SIG-Z8l L9F-S'Sh L8I-1'8 22040402 vqjv 7,

эицзэзит эи15эзат
SIP] [jews
yysuo'T, yisus'y

yisust snseydoso oumjoa QUINOA
Ч2еи1015 Ч13щ5Т = OUJsaqUT «= ORION

JYSIOM 108 sJOYM 318 э|оЧм

ВИТО 422101 = UBIO yysuaT

S[VUIUIEUE это$ лоу INS Jo 2YSIOM pur
y3sueq Jo (‘9 "21/27 ‘sajdures axoux ло 0M} JO asva ит? “9s 2 :2) потретхел JO SJUDTIOWJeOr) ‘8 этает,

Appendix 303

FOG] ‘лэбтехУ

эуер чмо

езер UME

еуер имо

Zep “мо

вер чмо

Zep UMO

Sear Gel Ш

(<) (5)

USES ALU

(Z) (Z)

9°51-5"8 Tel al

ВР ^

9'5-2 5'8
SITE'S

(<)

0'GI-L°01

($)

0`9Т-9`0Т

* snypjnainsun

52и04а И

27958 Д
2201401.452 ЕТ

2814514 vsng
221502 vsng
722141915 DSN

572рит]и2048

571144090]

> 2101542
72401401547)

304 Variability of Mammals

Table 9. Variability ([1тс.,.) of some characters of the respiratory system

Number of
tracheal rings

Lung weight

Vulpes vulpes .

Oryctolagus cuniculus
Talpa europaea
Delphinapterus leucas .
Pagophilus groenlandicus
Cystophora cristata

Pusa hispida

Pusa caspica

Pusa sibirica
Histriophoca fasciata
Phoca vitulina .

Meriones unguiculatus .

Table 10.

16.1-34.0(3)

15.5-50.6(4)
15.7-23.6(6)
19. 1-23.3(3)
17.2-35.7(3)
24.4-37.1(2)
62.5-63.7(2)
19.4-27.0(2)
25.0

25.1-27.1(2)

6.7-10.4(15)
9.1-10.1(8)
5.7- 6.3(4)
3.2- 5.8(3)
4.4-10.0(2)
5.0- 5.3(2)
3.8- 6.7(2)

Fateev et al., 1961; Bogolyubskii,

1933

Latimer and Sawin, 1955
Fateev, 1963

Own data
Own data
Own data
Own data
Own data
Own data
Own data
Own data

Kramer, 1964

Coefficients of variation (Lim,.,,) for absolute weight

of heart, kidneys, and skin of some mammals

Histriophoca fasciata
Pagophilus groenlandicus
Cystophora cristata

Pusa caspica

Pusa sibirica

Callorhinus ursinus

Delphinapterus leucas .
Ondatra zibethica .

Canis familiaris
Talpa europaea

Oryctolagus cuniculus .

Meriones unguiculatus .

18.1
10.9-70.9(3)
22.8-25.3(2)
14.6
13.0-16.2(2)

11.1
16.1-17.0(2)

24.9
16.7-44.3(6)
7.8-19.4(9)

15.9-21.7(2)

15.9-26.0(5)
11.9-22.2(5)
25.9-26.3(2)
18.5-19.7(3)

12.7-39.9(6)
24.2-43.1

13.3-34.0
10.9-22.1(9)

14.8-17.3(2)

6.5—13.6
(13)
14.8

Own data
Own data
Own data
Own data
Own data
Scheffer, 1962

Own data

Calculated from data
of Schwartz, 1962

Latimer, 1961
Fateev, 1962

Latimer, Sawin,
1955 ; Brown её al.,
1926

Kramer. 1962

BIBLIOGRAPHY*

ApamczEwskA К. A., 1959. “Untersuchungen @Бег die Variabilitat
der Gelbhausmaus Apodemus flavicollis flavicollis (Melchior, 1834),”
Acta Theriol., 3 : 141-190.

ALEXANDER, М. M., 1960. ‘“‘Shrinkage of Muskrat Skulls in Relation to
Aging,” 7. Wildlife Manag., 24 : 326-329.

Auten, С. М., 1940. The Mammals of China and Mongolia. Pt. П, 1350 pp.
[Cited from Long (1968).]

ALLEN, J. A., 1880. History of North American Pinnipeds . . . U. S. Geol.
& Geogr. Surv. Terr. Misc. Publ. 12 : 1-785. [Cited from Long (1968).]
Aupatoy, V., 1924. “Izmenchivost’ i nizhnie sistematecheskie kategorii’’

(‘Variability and Lower Taxonomic Categories”), Zool. Zh., 4 : 240-244.

Атратох, V., 1956. “Ро povodu за С. У. Nikol’skovo ‘Ob izmenchivosti
Organizmov ” (“The Article of С. У. Nikol’skii “On the Variability
of Organisms’”’), Zool. Zh., 35 : 373-375.

Aupatoy, W. W., and A. М. BoscHKO-STEPANENKO, 1928. “Variation and
Correlation in Serially Situated Organs in Insects, Fishes, and Birds,”
Amer. Naturalist, 62 : 409-424.

ANDERSON, A. E., L. G. Frary, and R. H. Stewart, 1964. “A Comparison
of Three Morphological Attributes of Mule Deer from the Guadalupe
and Sacramento Mountains, New Mexico,” 7. Mammal., 45 : 48-53.

ANDERSON, S., and A. $. Gaunt, 1962. “А Classification of the White-
sided Jack Rabbits of Mexico,” Amer. Mus. Nov., 2088 : 1-16.

ANDERSON, S., and С. Е. Netson, 1965. “А Systematic Revision of
Macrotus (Chiroptera), Amer. Mus. Nov., 2212: 1-39.

AnpreEv, I. F., 1954. “К voprosu О putyakh formoobrazovatel’nogo
protsesa na primere izmenchivosti zheltogorlii myshi у stepyakh
Moldavii” (“ТЬе Problem of the Action of Developmental Processes
from the Example of Variability of Yellow-necked Mice in the Steppes
of Moldavia’’), III Ekol. kon. о Клео. Tez. Dokl. III.

* Some works have been listed in the bibliography that have not been referred to in
the text. This has been done because the factual data presented in these works on variabi-
lity agrees well with that discussed already. Secondly, I came to know these works after
the final draft and after handing over the manuscript for printing. (Some of these works
were published in 1964 and a majority in 1965.) Hence this section of the book may be
seen as of independent importance for reviewing variability of mammals from contemporary
literature—Author.

306 Bibliography

ANDREEVA, T. V., and А. У. УАвтокоу, 1965. ‘“‘Novoe у razrabotke
metodiki opredelemiya voprosa usatykh kitov’’ (“New Method of Age
Determination of Whales”), Zool. zh., 44 : 145-146.

Arata, A. A., 1964. “А Comment on the Utility of the Testicular Blood
Supply in Taxonomic Studies,” 7. Mammal., 45 : 493-494.

ArGIROPULA, A. I., 1946. “К voprosu ob individual’noi 1 geograficheskoi
izmenchivosti а nekotorikh vidov года Apodemus Kaup (Mammalia)”
(“The Problem of Individual and Geographical Variability in Certain
Species of the Genus Apodemus Kaup’’), Trudy Zool. Inst. Akad. Nauk
SSSR, 8 : 195-220.

ARTEM EV Уч. T., 1946. “К sravnitel’noi morphologii suslikov volzhako-
kamskovo kraya”’ (“Comparative Morphology of Susliks of the Volga
and Kamskii Regions”), Tez. Dok. Itogovor Nauch. Aspirant. Konf. 1964,
Kazansk. Univ., рр. 83-85.

AsHou, М. R., J. О. Bicgcers, A. McLaren, and О. Мисни, 1958. “The
Effect of the Environment on Phenotypic Variability,” Proc. Roy. Soc.
London, ser. B, 149 : 192-204.

Азнтох, Е. H., and С. Е. Oxnarp, 1964. “Functional Adaptations in the
Primate Shoulder Girdle,” Proc. Zool. Soc. London, 142 : 49-66.

Asuton, Е. H., С. Е. Oxnarp, and T. Е. Spence, 1965. “Scapular Shape
and Primate Classification,” Proc. Zool. Soc. London, 145 : 125-142.

AsTANIn, Г. Р., 1960. “‘Sootnositel’naya izmenchivost’ cherepa i mozga
nekotorykh mlekopitayushchikh” (“Correlated Variability of Skull and
Brain in Certain Mammals’’), Arkhiv Anatomit, Gistol. 1 Embryol., 39 :
39-45.

Asranin, L. Р., 1962. “‘Materialy К izucheniyu tserebralizatsii blizkikh
vidov mlekopitayushchikh”’ (‘‘Materials for the Study of the Cerebraliza-
tion in Related Species of Mammals’’), Uch. Zap. Stavr. Med. Inst., No. 5.

Bapamsuin, В. Г., 1949. ‘“‘Nekatorye novye dannye о biologii kaspiiskogo
tyulenya vo ’dakh”’ (“Some New Data on the Biology of the Caspian
Seal on Ice”), Rybnoe khoz., 3 : 39-44.

BapamsuHin, В. Г., 1963. ‘‘Biologicheskie 1 tekhnicheskie predposil’ki К
uvelicheniyu dobychi kaspiiskogo tyulenya na ostrovnykh lezhbishchkh”’
(‘Biological and Technical Prerequisites for Increasing the Catch of
Caspian Seals from Island Colonies”), Tez. dokl. Viorogo Vsesoyuz. Soveshch
о Izuch. Morskikh Mlekopitayushchikh.

Bapamsuin, В. Г., 1965. ‘‘Osobennosti lin’ki kaspiiskovo, tyulenya”
(“Characteristics of Moulting of Caspian Seals”), in Sb. Morskie
mlekopitayushchie. Moscow : Izdatelstvo ‘Nauka’.

Baper, В. 5., 1955. ‘Variability and Evolutionary Rate in the Oreodonts,””
Evolution, 9 : 119-140.

Baber, R. 5., 1956. ‘Variability in Wild and Inbred Mammalian Popula-
tion,” Quart. 7. Florida Acad. Sci., 19 : 14-34.

Bibliography 307

Baper, К. Ъ. and J. 5. Hatz, 1960. ‘“‘Osteometric Variation and Function
in Bats,’’ Evolution, 14 : 8-17.

BANERJEE, А. R., 1963. “Variations in the Medullary Structure of Human
Head Hair,”’ Proc. Nat. Inst. Sci. India, Part В, 29: 306-316.

BANFIELD, А. W., 1961. “А Revision of the Reindeer and Caribou, Genus
Rangifer,’ Canada Nat. Mus. Bull., 177 : 1-137.

BaraABASH-Nixiroroyv, I. I., 1940. Fauna kitoobraznykh chernogo morya
(Cetaceans of the Black Sea). Voronezh : Izdatelstvo Voronezhskii
Gos. Univ.

BasHentna, N. V., 1962. Ekologiya obyknovennot polevki i nekotorye cherty ee
geograficheskot izmenchivosti (Ecology of the Common Field Vole and
Certain Aspects of Its Geographical Variability). Moscow : Izdatelstvo
Moskovskii Gos. Univ.

Basu, 5. S., and 5. Hasary, 1960! “Variations of the Lumbrical Muscles
of the Hand,” Anat. Rec., 136 : 501-503.

Bateson, W., 1894. Materials for the Study of Variation, Treated with Special
Regard to Discontinuity in the Origin of Species. London : MacMillan
Company.

Baur, E., 1925. “Пе Bedeutung der Mutationen fir das Evolutions
Problem,” ©. indukt. Abstammungs- und Vererbungslehre, 37 : 107-115.
Bazan, I., 1953. ‘“‘Zmiany morfohistologiczne grasicy u Sorex araneus Г.

w cyklu zycoiwym,” Ann. Univ. М. Curte-Slodowska, Sect. С, 7 : 253-297.

Bazan, Г.,, 1956. “Untersuchungen @Бег die Verdnderlichkeit des
Geschlechtsapparatus und des Thymus der Wasserspitzmaus (Neomys
fodiens fodiens Schreb.),” Ibid., Sect. С, 9 : 213-258.

Bepparp, F. E., 1900. A Book of Whales. London : J. Murray.

Вет-Втемко, С. YaA., 1959. “Ргицмяр smeny statsii i problema nachal’noi
divergentsii vidov”’ (‘Principle of Change of Habitat and the Problem
of the Initial Divergence of Species’), Zhur. Obsch. Biol., 20 : 351-358.

Векил, N., 1964. Statisticheskie metody о biologi (Statistical Methods in Bio-
logy). Moscow : Izdatelstvo ‘Mir’.

BEKLEMIsSHEV, V. N., 1960. ‘“‘Prostranstvennaya 1 funktsional’naya
struktura populyatsii”’ (“Spatial and Functional Structure of a Popula-
поп”), Bull.. Mosk. Obshch.. Ispyt. Prirody (Biol.), 65 : 41-50.

BektemisHEv, V. N., 1964. “Ob obshchikh printsipakh organizatsii zhizn1”’
(“General Principles of Organization of Life’), ВиЙ. Mosk. Obshch.
Ispyt. Prirody (Biol.), 69 : 22-38.

BEL’ Kovicu, У. М. and А. У. Улвгокоу, 1960. ‘“‘Osobennosti stroeniya
zubnoi sistemy morzha”’ (“Characteristics of the Structure of the Dental
System of the Walrus’), IV Molodezhn. Nauchn. Ronf. Inst. Morfol.
XKhivotnykh Akad. Nauk SSSR, Moscow, 3-7.

Вег’коутсн, У. М. and A. У. УАвгокох, 1961. “‘Sredi morzhei’’ (““Among
Walruses’’), Priroda, 1961 (3) : 50-56.

308 Bibliography

BeL’Kovicu, У. М. and А. У. Улвьокоу, 1962. ‘‘Nablyudeniya morzhei
na lezhbishakh у Anadirskom zalive i Chukotskom morye”’ (Observations
on Walruses of the Colonies of Anadir Bay and the Chukchi Sea’’),
Kraevedcheskie zap. Magadansk. obl. kraeved, Muzeya, 4 : 156-174.

Bev’Kovicu, У. М. and A. У. YasLoxov, 1963. “‘Obitateli morya delyatsa
opytom s konstruktorami’’ (“Inhabitants of the Sea Shore: Their
Experience in Construction’’), Nauka i zhizn, 1963 (5) : 61-64.

Вег’ко\тен, У. М. and A. У. УАвгокоу, 1963a. ‘“Molodost’ drevnei
nauki: Morfologia zakluchaet soyuz s tekhnikoi” (“Youth of an Old
Science : Morphology Establishes a Relationship with Technique”),
Priroda, 1963 (8) : 20-30.

Bev’Kovicu, У. М. and А. У. УАвтокоу, 1965. “О strukture stada
zubatykh kitobraznykh” (“Оп the Composition of a Group of Toothed
Whales”), in Sb. Morskie mlekopitayushchie, рр. 25-29. Moscow:
Izdatelstvo ‘Nauka’.

Betyaev, О. K., 1962. “О nekotorykh problemakh korrelyativnoi
izmenchivosti i ikh znachenie dlya teorii evolyutsii i selektsii zhivotnykh”’
(“Certain Problems of Correlated Variability and Their Significance
for the Theory of Evolution and Animal Breeding’’), Izv. Sibirsk. ОШ.
Akad. Nauk SSSR, 10 : 11-124.

Bere, L. $., 1910. ‘‘Fauna Baikala i ее proiskhozhdenie” (“Fauna of the
Baikal and Its Origin”), Biol. zh., 1: 10-45. |

Berc, В. L., 1957. “Tipy polimorfizma”’ (“Types of Polymorphism’),
Vestnik Leningradski Gos. Univ., 1957 (21) : 115-139.

Berc, В. L., 1959. ‘“‘Ekologicheskaya interpretatsiya korrelyatsionnykh
р!еуаа” (‘‘Ecological Interpretation of Correlation Pleiads’’), Vestnek
Leningradskit Gos. Univ., 1959 (9) : 142-152.

Berc, В. L., 1959. “А General Evolutionary Principle Underlying. the
Origin of Developmental Homeostasis,” Amer. Naturalist, 93 : 103-105.

Berc, В. L., 1964а. “Korrelyatsionnye pleyadi i stabiliziruyushchu
otbor’” (“Correlation Pleiads and Stabilizing Selection”), Primenenie
matematicheskikh metodov v biologi, (3) : 23-60.

Berc, В. L., 1964b. Stabiliziruyushchii otbor, у evolyutsii razmerov
tsevtov i semyan travyanistykh rastenii (Stabilizing Selection in the
Evolution of Size of Flowers and Seeds of Grasses). Avtoref. Diss. na
Soisk. Uch. Step. Doktora Biol. Nauk. Leningrad. Biologicheskii Institut,
Akad. Nauk SSSR.

BercEeRup, A. T., 1964. ‘“‘Relationship of Mandible Length to Sex in
Newfoundland Caribou,” 7. Wildlife Manag., 28 : 54-56.

BERNSHTEIN, М. А., 1962. “Рай razvitiya fiziologii 1 svyazannye $ nimi
zadachi kibernetiki’ (‘““The Ways of Development of Physiology and
the Problems of Cybernetics Related to Them’’), in Sb. Buologicheskie
aspekty kibernetiki, Moscow : Izdatelstvo Akad. Nauk SSSR, 52-65.

Bibliography 309

Berry, R. J., 1963. “Epigenetic Polymorphism in Wild Populations of
Mus musculus,’ Genet. Res. 4 : 193-220.

Berry, В. J., 1964. “The Evolution of an Island Population of the House
Mouse,” Evolution, 18 : 468-483.

Berry, К. ]., and A. С. ЗЕАвтЕ. “Epigenetic Polymorphism of the
Rodent Skeleton,” Proc. Zool. Soc. London, 140 : 577-615.

BetesHevA, E. I., 1961. Puitanie promyslovikh kitov Kuril’skogo raiona (The
Feeding of Whales in the Kuril Region). Trudy Soveshch. Ikhtiol. Komis.
Akad. Nauk SSSR, 12.

Buatnacar, О. S., and С. 5. Taneja, 1960. ‘Variation in the Performance
of Hariyans Cattle,” Sez. and Culture, 25 : 595.

Brancul, М. O., and МотлмаА Omar, 1966. “Autosomal Polymorphism
in a Laboratory Strain of Rat,” 7. Heredity, 57 : 231-232.

Biccers, J. D., 1958. ‘‘Biometrical Analysis of Variance,” Proc. Roy.
Soc. London, 148 : 199-202.

Вилл, K., 1959. Biologiya (Biology). Moscow: Izdatelstvo Inostrannoi
Literatury.

Buair, W. F., 1964. ‘The Case for Ecology,” Biol. 56., 14: 17-19.

Восотховзки, S. N., 1941. Sootnoshenie massy razmerov organov и razvodimykh
Mustelidae (Correlation of a Number of Organ Measurements in Farm-
bred Mustelidae). Trudy Morsk. Zootekhn. Inst., 1.

‚ Вот?знАкоу, У. N., 1962. Individual’naya izmenchivost’ okraski doukh vidov
lesnykh polevok (Clethrionomys rutilus Palli Cl. glareolus Schreb) [Individual
Variability of Coloration of Two Species (Clethrionomys rutilus Pall and
Cl. glareolus) of Field Voles], Trudy Inst. Biol. Schreb. Uralsk. Fil. Akad.
Nauk SSSR, 29.

Bow’sHaKov, У. N., 1964а. “О ее geograficheskoi 1zmen-
chivosti” (“The Species Specificity of Geographical Variability’),
in Sb. Voprosy vnutrividovoi izmenchivosti. . . . 1 mikroevolyutsi1, Sverdlovsk,
17-18.

Вот’знлкоу, У. N., 1964Ъ. “О taksonomicheskom polozhenii lemming-
ovidnoi polevki У svyazi s nekotorymi problemami nadvidovoi sistema-
tiki”? (““The Taxonomic Position of the Lemming-like Field Vole with
Reference to Certain Supraspecific Тахопоту”), in Sb. Voprosy vnutrivi-
dovoi izmenchivosti. . . . 1 mikroevolyutsii, Sverdlovsk, 19-21.

Вот’знАкоу, V. N., 1965. ‘‘Materialy ро sravnitel’nomu izucheniyu
geograficheskoi izmenchivosti inter’ernykh priznakov blizkikh vidov
polevok” (“Data on a Comparative Study of Geographical Variability
of Internal Traits of Related Species of Field Voles’’), Trudy Inst. Biol.
Uralsk Fil. Akad. Nauk SSSR, 38 : 53-60.

Вот’знаАкоу, У. М. and 5. $. Scuwartz, 1962. Nekotorye zakonomernosti
geograficheskoi izmenchivosti gryzunov па sploshnom uchaske ikh
areala (na primere roda Clethrionomys) [Some Regularities of Geographical

310 Bibliography

Variability of Rodents in a Continuous Part of Their Range (an Ex-
ample from Clethrionomys) |, Ibid., 29.

Вот’знАкоу, У. N., and S. S. Scuwartz 1965. “Моуот podvid krasnoi
polevki (Clethrionomys rutilus tundrensis)”’ [“А New Subspecies of the Red
Field Vole (Clethrionomys rutilus tundrensis)”’|, Ibid., 38 : 63-65.

Воогооттлм, В. A., 1954. “Ап Analysis of Sub-specific Variations in
Dipodomys nitratoides,’ 7. Mammal., 35 : 570-577.

Borev’, E., 1964. Veroyatnosti’ 1 dostovernost’ (Probability and Significance).
Moscow : Izdatelstvo ‘Nauka’.

Borowski, St., 1964. ‘“‘Moult of Shrews (Sorex L.) Under Laboratory
Conditions,” Acta Theriol., 8 : 125-135.

Borowskl, St., 1964. ‘Studies on the European Hare: 1. Moulting and
Coloration,”’ Acta Theriol., 9 : 217-231.

BorowskI, Эт., and A. DEHNEL, 1953. ‘‘Materialy do biologii Soricidae,”
Ann. Univ. M. Curie-Slodowska, 7: 305-436.

ВотнмА, J. pu P., 1966. ‘Color Variation in Hyracoidea from Southern
Africa,” 7. Mammal., 47 : 687-693.

Bruna, J., 1964. ‘‘Prilog poznavanju beljskog jelena,”’ Zelen : 19-31.
Broapuurst, Р. L., and J. L. Jinks, 1965. “Parity, as a Determinant
of Birth Weight in the Rhesus Monkey,” Folia Primatol., 3 : 201-210.
Вводрновзт, Р. L. and J. L. лмкз, 1966. “Stability and Change in the
Inheritance of Behavior in Rats : A Further Analysis of Statistics from

a Diallel Cross,” Proc. Roy. Soc. London (В) 165 : 450-472.

Bronson, Е. H., 1958. “Моез on Body Size of Blacktailed Jack-rabbits,”’
Trans. Kansas Acad. Sci., 61 : 109.

Brown, S. С., 1965. ‘’ТБе Color of the Baleen Plates in Southern Hemi-
sphere Sei Whales,” Norsk Hvalfangst-tidende, 54 : 131-135.

Brown, W. H., L. Pearce, and С. М. van ALLEN, 1926. “Organ Weights
of Normal Rabbits,” 7. Exper. Med., 43 : 733-741.

BRown_ELL, К. L., ]в., 1964. “Observation of Odontocetes in Central
Californian Waters,” Norsk Hvalfangst-tidende, 53 : 60-62, 64-66.

Bucuatczyk, A., 1961. “Variation of Weight of the Internal Organs
of Sorex araneus Linnaeus, 1758: Г. Salivary Glands,’? Acta Theriol.,
9,2 229-252.

Во]АтзкА, G., 1964. “Studies on the European Hare: IV. Variations
in the Pelvis and Sacrum,” Acta Theriol., 9 : 287-304.

BycuKxov, У. A., 1965. “О polovoi 1 vozrostnoi izmenchivosti cherepa
kuril’skogo morskogo komika”’ (“Sex and Growth Variabilities of the
Skull of the Kuril Fur Seal’’), in Sb. Morskie mlekopitayushchie. Moscow :
Izdatelstvo ‘Nauka,’ 183-188.

Byunninc, E., 1964. Biologicheskie chasy (Biological Clocks). Moscow :
Izdatelstvo ‘Mir’. .

Cason, K., 1958. “Untersuchungen tiber die Schadelvariabilitat des

Bibliography 311

Wildschweines Sus scrofa Г. aus’ Nordostopolen,” Acta Theriol., 11 :
195-266.

Cason-RaczynskA, K., 1964a. “Studies on the European Hare: II.
Variations in the Weight and, Dimensions of the Body and Weight of
Certain Internal Organs,” Acta Theriol., 9 : 233-248.

Cannon, H. G., 1959. Lamarck and Modern Genetics. Manchester : Man-
chester Univ. Press.

Carmon, J. L., Е. В. Gottey, and В. С. У/иллАмз, 1963. “Ап Analysis
of Growth and Variability in Peromyscus polionotus,’’ Growth, 27 : 247-254.

Castie, W. E., 1931. ‘“‘Size Inheritance in Rabbits : the Backcross to the
Large Parent Race,” 7. Exper. Zool., 60 : 325-338.

Caste, W. E., 1938. ‘“The Relation of Albinism to Body Size in Mice,”
Genetics, 23 : 269-274.

Слзтье, W. E., 1941. “Influence of Certain Color Mutations on Body
Size in Mice, Rats and Ка ББ,” Genetics, 26 : 177-191.

Caste, W. E., and 5. С. Reep, 1936. ‘‘Studies of Inheritance in Lop-eared
Rabbits,”’ Genetics, 21 : 297-323.

Cast_e, W. Е., 5. С. Reep, and W. H. Gates, 1936. “Studies of a Size
Cross in Mice,” Genetics, 21 : 66-78.

Castie, W. E., 5. С. Reep, W. H. Gates, and L. W. Law, 1936. “Studies
of a Size Cross in Mice: II,”’ Genetics, 21 : 310-323.

Cave, А. J., 1930. “Оп Fusion of the Atlas and Axis Vertebrae,” 7. Anat.,
641337.

Cave, А. J. and Е. L. О. Этеег, 1964. ‘‘Craniometric Sex Determination
in the Colobus Skull,”’ Proc. Zool. Soc. London, 143 : 503-510.

Crzariusz, WILAND, 1967. “Krag totniczy i tetnica podstawowa mozgu
u lisow,” Prace Komis. Nauk Roln. Komis. Nauk Lesn. Poznan. Towarz.
Przytacio Nauk., 23 : 305-324.

Cuar, С. K., 1957. ‘Developmental Homeostasis of Body Growth in
Mice,” Amer. Naturalist, 91 : 49-56.

Cuat, C. K., 1966. ‘‘Selection for Leukocyte Counts in Mice,” Genet. Res.,
8: 125. 142.

СнАРЗки, К. K., 1938. “‘Morskoi zayats (Erignathus barbatus Pall.) Karskogo
i Barentsova morei” [‘““The Bearded Seal (Erignathus barbatus Pall.) of
the Kara and Barents Seas”], Trudy Arkticheskovo Nauch.-Issled. Inst.,
123 : 7-70.

Cuapsku, К. K., 1952. “Оргедеше vozrosta nekotorykh mlekopitayu-
shchikh po microstrukture kosti” (“Age Determination of Some Mammals
by Microstructure of Bones”), Izv. Istestv. Nauch. Inst. im P. Е. Lesgaft,
Moscow, 25 : 47-66.

Cuapsku, К. K., 1952a. ‘‘Vozrostno-polovaya izmenchivost’ kraniologi-
cheskikh priznakov i ее vliyanie na diagnostiku nekotorykh lastonogikh”
(““Age-sex Variability of Cranial Traits and Its Effect on the Diagnosis

312 Bibliography

of Some Pinnipeds’’), Ibid., 25 : 47-66.

Cuapsku, К. К., 1955. “К voprosu ob istorii formirovaniya kaspiiskogo
1 baikal’skogo tyulenei”’ (“Оп the Problem of the History of the Forma-
tion of the Caspian and Baikal Seals’), Trudy Zool. Inst. Akad. Nauk
SSSR, 17 : 200-215.

Cuapskil, К. K., 1955а. “Opyt peresmotra sistemy i diagnostiki tyulenei
podsemeistva Рйостае” (“Ап Attempt to Revise the Taxonomy and
Diagnosis of Seals-of the Subfamily Phocinae’’), Ibid., 160-199.

Cuapskil, К. K., 1961. ‘‘Nekotorye ekologicheskie obosnovaniya sezonnoi
diagnostiki a reala belomorskoi popyulyatsii grenlandskogo tyulenya
(Pogophoca втоещапйса)” [“Зотае Ecological Bases for Seasonal Dis-
tinctions of the Habitat of the White Sea Population of the Greenland
Seal (Pogophoca groenlandica)”’|, Trudy Vsesoyuzn. Soveshch. po Izucheniyu
Morskikh Mlekopitayushchikh, 150-163.

Cuapskil, К. K., 1963. ГП. Otryad Pinnipedia : Lakstonogie. (Order
Pinnipedia : Seals), Mlekopitayushchie Fauny SSR, 2 : 895-964.

Cuapskil, К. K., 1963a. “‘Resursy zveroboinogo promysla, ikh sostoyanie
ispolzovanie 1 izuchenie”’ (“Resources of the Whale Industry : Their
Condition, Use and Study’), Tezizy Dokladov Vtorogo Vsesoyuznogo
Soveshchamya po Izuchenityu Morskikh Mlekopitayushchikh, рр. 43-47.

CHERNYAVSKU, Е. В., 1969. ‘‘Kraniometricheskaya kharakteristika belogo
medvedya (Ursus maritimus Phipps, 1774) Sovetskoi Arktiki” [‘‘Cranio-
metric Characters of the Polar Bear (Ursus maritimus Phipps, 1774) of |
the Soviet Агсис”], ш Sbornik Belyi Medved’ i Ego Okhrana о Sovetskoi
Arktike, pp. 54-67. Leningrad : Gidrometeoizdat.

CHETVERIKOV, 8S. 5., 1926 (1965). “О nekotorykh momentakh evolyutsion-
nogo protsesa $ tochki zreniya sovremennoi genetiki” (“Оп Certain
Aspects of the Evolutionary Process from the Standpoint of Modern
Genetics”), Bull. Moskov. Obshch. Inst. Рит. (Biol.) 70 : 33-74. (Trans-
lated 1961, Proc. Amer. Philos. Soc., 105: 167-195).

CuurcHER, С. 5., 1960. “Cranial Variation in the North American Red
Fox,” 7. Mammal., 41 : 349-360.

Сглвк, Е. H., 1940. “Correlation and Body Proportions in Mature Mice
of the Genus Peromyscus,’ Genetics, 26 : 283-306.

Сотллм, R., 1960. Evolution : Hypotheses and Problems. London : Burns
and Oates.

Corset, С. В., 1963. “The Frequency of Albinism of the Tail-tip in
British Mammals,” Proc. Zool. Soc. London, 140 : 327-330.

СвозАЕОКТ-РАтво, M., and I. Truyors-Sanronya, 1956. “А Biometric
Study of the Evolution of Fissiped Carnivores,” Evolution, 10 : 314-332.

Damier, R., 1952. “‘Remarque au sujet du polymorphism du mulot en
Belgique,” Bull. Inst. Sci. Natur. Belg., 28 : 1-11.-

Danrortu, С. H., 1930. ‘‘Numerical variation and Homologies in

Bibliography 313

Vertebrae,” Amer. 7. Phys. Anthropol., 14 : 463.

Darwin, C., 1868. The Variation of Animals and Plants under Domestication.
London : J. Murray.

Darwin, C., 1877. The Different Forms of Flowers on Plants of the Same
Species. London : J. Murray.

Darwiy, C., 1939. Sobranie sochinenii (Collected Works). Vol. ПТ, Moscow.

Decasan, A., and J. Е. ГлевЕВМАх, 1964. “Calculation of Cranial Capa-
city from Linear Dimensions,” Anat. Rec., 150 : 215-219.

DEHNEL, A., 1949. ‘“‘Studies on the Genus Sorex L.,’? Ann. Univ. М. Curie-
Slodowska, Sect. C, 4 : 17-102.

Denne, A., 1950. “Studies on the Genus Neomys Каир,” Ibid., 5 : 1-63.

Demerec, M., 1937. “А Mutability Stimulating Factor in the Florida
stock of Drosophila melanogaster,” Genetics, 22: 190-195.

Demipova, Z. A., 1941. Zavisimost’? mezhdu_ dlitel’nostyu embryonal’
nogo razvitia, rostom i nastrigom shersti u ovets (Relation between the
Length of Embryonic Growth, Development, and Wool Clippings in
Sheep). Trudy Mosk. Zootekhn. Inst., vol. 1.

Denisova, I. А., 1966. “Elektroforeticheskoe issledovanie vozrostnykh
izmenenii у otnoshenii belko’vykh fraktsii sivorotki krovi malogo suslika”’
(“Electrophoretic Investigations on Protein Fractions of Blood Plasma
of Pigmy Ground Squirrels of Different Ages’), Trudy Vsesoyuzn.
Soveshchaniya “Vnutrivid. Izmenchivost Nazemnykh Pozvonochnykh Zhivotnykh
$ Mikroevolyutsiya.”’ Sverdlovsk, рр. 237-241. .

ПЕог, М. S., 1955. “‘Genetical Studies on the Skeleton of the Mouse :
XIV. Minor Variation of the Skull,” 7. Genet., 53 : 498-514.

Derums, У. Ya., 1964. ‘“‘Nekotorye paleoantropologicheskie dannye о
zhitelyakh Pribaltiki” (““Some Paleoanthropological Data on Inhabitants
of the East Baltic Area’), Arkhiv Anat. Gistol. Embryol. 16 : 80-90.

ПЕзнА, P. G., 1967. ‘Variation in a Population of Kangaroo Rats,”
Dipodomys ordii medius (Rodentia : Heteromyidae) from the High Plains of
Texas,” Southwestern Nat., 12 : 275-290.

DHALIWAL, S. S., 1962. “Studies on the Body Measurements and Skeletal
Variations of Two Taxa of Rattus rattus in Malaya,” 7. Mammal., 43 :
249-261.

Dice, L. R., 1938. ‘“‘Variation in Nine Stocks of the Deer Mouse, Peromyscus
maniculatus from Arizona,” Univ. Mich. Occasion. Paper, Mus. Zool.,
Эк: 19:

Dice, L. R., and Н. L. Leraas, 1940. “Есо]оглс and Genetic Variability
within Species of Peromyscus,’ Amer. Naturalist, 74 : 212-221.

DoszHaAnsky, Тн., 1937. Genetics and the Origin of Species. New York :
Columbia Univ. Press.

Dopson, Е. O., 1960. Evolution, Process and Product. New York : Reinhold.

Doucov, У. A., 1961. “ОЪ izmenchivosti nekotorykh kostei postkranial’-

314 Bibliography

nogo skeleta zemleroek (Mammalia, Soricidae)” (‘Variability of Some
Postcranial Skeletal Bones of Rodents (Mammalia, Soricidae)”|, Acta
Theriol., 5 : 203-227.

Dotcov, У. A., 1963. “Ob izmenchivosti burozubok poimy г. Oki.
(Mammalia, Soricidae)” [““Variability of Shrews in the Flood Plains of
the Oka River (Mammalia, Soricidae)’’|, Bull. Mosk. Obshch. Ispyt.
Prirody, (Biol.), 68: 135-140.

Dorn, A., 1936. Printsip smeny funktsi (Principles of Change of Function).
Moscow : Biomedgiz.

Dornesco, С. T., and С. Масвост, 1961. “Ftude Comparative du Crane
des Cervidés,”’ Rev. Biol., 6: 261-294.
Dusinin, №. P., 1931. ‘“‘Genetiko-avtomaticheskie protsessy 1 ikh znachenie ^

Фуа mekhanizma organicheskoi evolutsi”’ (““Genetic-automatic Processes
and Their. Significance for the Mechanism of Organic Exelon i
Kh. Eksperim. Biol., 7 : 463-479.

Dusinin, М. P., 1932. “О nekotorykh osnovnykh propia eel ен
(“Some Basic Problems of Genetics”), Bzol. Zh., 1(1-2) : 112-146.

Dusinin, М. P., 1940. ‘‘Genetika i ‘Proiskhozhdenie vidov’ Ch. Darvina”
(‘Genetics and the ‘Origin of Species’ by Charles Darwin’), ch. Obshch.
Biol Nee SVelae

Dusinin, М. P., and О. О. Вомазнох, 1932. ‘‘Geneticheskoe stroenie vida
i ego evolyutsiya’’ (““Genetical Build-up of a Species and Its Evolution’),
Biol. Zh., 1(5=6) : 52-95.

Dusinin, М. P., М. №. бокогоу and С. G. Tinyaxov, 1937. “‘Vnutrivido-
vaya khromozomnaya izmenchivost’ ”
Variability”), Biol. ch., 6: 1007-1054.

Dunn, Г. С., A. В. BeasLtey, and H. Tinker, 1960. “Polymorphism in
Populations of Wild House Mice,” 7. Mammal., 41 : 220-229.

Dunn, Г. C., and О. С. Cuartes, 1937. “Studies of Spotting Patterns :
I. Analysis of Quantitative Variations in the Pied Spotting of the House
Mouse,”’ Genetics, 22 : 14-42.

Dwyer, P. D., 1962. ‘‘Studies on the Two New Zealand Bats,” Zool. Publ.
Victoria Univ., Wellington, 28 : 1-28.

Dynowsk1, J., 1963. ‘‘Morphological Variability in the Vialowieza
Population of Mus musculus Linnaeus, 1758,” Acta Thertol., 7 : 51-67.

Enruicu, Р. В. and P. W. Ногм, 1962. ‘Patterns and Populations,”
Science, 137 : 652-657.

Exton, C., 1927. Animal Ecology. London : Sidgwick and Jackson.

Emerson, A. E., 1961. ‘‘Vestigial Characters of Termites and Processes
of Regressive Evolution,’ Evolution, 15 : 115-131.

ЕАрЕЕХ, Е. V., 1958. ‘‘Vozrostnye izmeneniya mekha nutri” (“Age

‚ Variations of Fur of Nutria”’), Trudy Vsesoyuzn. Nauch.-Issled. Inst. Khivoin.
_ бура Pushniny, 17 : 161-179.

(‘“Intra-species Chromosomal

Bibliography 315

Fatconer, D. S., and М. Latyszewsx1, 1952. “Selection for Size in Mice
on High and Low Planes of Nutrition,” Quantitative Inheritance, London :
Agricultural Research Council, pp. 145-151.

Fatrev, К. Ya., 1961. ‘“‘Morfobiologicheskie izmeneniya krolikov v
raznykh usloviyakh sredy” (““Morpho-biological Changes in Rabbits
under Different Environmental Conditions’) Zh. Obshch. Biol., 22 :
388-392.

Fatrev, К. Ya., 1962. “Izamenchivost’? vnutrennykh orgonov evropeis-
kogo krota (Тара europaea)” [Variability of Internal Organs of
the European Mole (Talpa europaea)”, Zool. КЛ., 41: 1700-1705.

Fateev, К. Ya., M. V. Kuromova and L. 5. Guzova, 1961. “Izmenchivost’
vnutrennykh organoy и serebristo chernykh lisets (Vulpes —fulves
Docm.)”’ [‘‘Variability of Internal Organs of Silver-black Foxes (Vulpes
fulves Оост.)”], Zool. Kh., 40: 1090-1098.

ЕЕроЗЕЕУ, С. A., 1964. “ОБ embrional’nom, post-embrional’nom roste
i polovom sozrevanii okhotskoi kol’chatoi пегру” (“On the Embryonic
and Post-embryonic Growth and Sexual Maturity of the Ringed Seal
of the Okhotsk Sea”), <ool.. gh.; 43 : 1228-1235.

ЕЕРОЗЕЕУ, С. A., 1965. “‘Sravnitel’naya kharakteristika populyatsi
kolchatoi nerpy pribrezhnykh vod Chukotskogo poluostrova” (“Сот-
parative Characters of Populations of the Ringed Seal in the Coastal
Waters of the Chukchi Peninsula”), Zzv. Tikh Nauchno-Issled. Inst.
Куб. Khoz. i Okeanogi (Vladivostok), 59: 194-212.

ЕЕРОЗЕЕУ, С. A., and А. У. Улвтокоу, 1965. ‘“Morfologicheskaya
kharakteristika okhotskogo kol’chatoi nerpy (Pusa hispida, Pinnipedia)
okhotskogo тогуа” [‘‘Morphology of the Ringed Seal (Pusa hispida,
Pinnipedia) of the Okhotsk Sea”’], Zool. Zh., 44 : 759-765.

Fepotov, О. M., 1959. ‘‘Filogeniya zhivotnykh kak kompleksnaya prob-
lema”’ (“Phylogeny of Animals аз a Complex Problem’’), Vestnik Akad.
Nauk SSSR, 1959(9) : 9-12.

Fetten, H., 1952. ‘Untersuchungen zur Okologie und Morphologie
der Waldmaus (Apfodemus sylvaticus L.) und der Gelbhalmaus (Apodemus
flavicollis Melchior) im Rhein-Main-Gebiet,” Bonn. ool. Beitr., 3 :
187-206.

ЕилрснЕМКО, Yr. A., 1916. ‘‘Izmenchivost’ i nassledstvennost’ cherepa
mlekopitayushchikh” (‘Variability and Heredity of the Mammalian
Skull’), Russk. Arkhiv. Anat. Gistol. Embriol., 1 : 311-403.

ЕилрснЕекко, Yr. A., 1926. Izmenchivost’ i metody ee izucheniya (Variability
and Methods of Its Study). Leningrad.

Fisoer, В. A., 1930. The Genetical Theory of Natural Selection. Oxford :
Clarendon Press.

Forp, Е. B., 1955. “Rapid Evolution and the Conditions Which Make
It Possible,’ Cold Spring Harbor Sympos. Quant. Biol., 20 : 230-238.

316 Bibliography

Forp, Е. B., 1958. ‘‘Darwinism and the Study of Evolution in Natural
Populations,” 7. Linnean Soc. London, 44 : 41-48.

Forp, E. B., 1964. Ecological Genetics. London: Methuen.

Forp, Е. В., and H. О. Butz, 1926. ‘Abnormal Vertebrae in Herrings,”’
J. Marine Blol. Assoc. United Kingdom, 14 : 509-517.

FRANKEL, О. H., 1959. ‘‘Variation under Domestication,” Austral. 7.
alan) S 27-32.

FREEDMAN, L., 1963. “А Biometric Study of Papio cynocephalus Skulls from
Northern Rhodesia and Nyasaland,” 7. Mammal., 44 : 24-43.

FREEMAN, У. A., 1939. ‘‘Variation in the Number of Vertebrae in Swine,”
J. Heredity, 30 : 61-64.

FREIMAN, 5. Yu., 1935. “Ваззргедеше lastonogikh у moryakh Dal’nego
vostoka”’ (“Distribution of Pinnipeds in the Far East Seas”), Trudy
Vses. Nauch-Issled. Inst. Ryb. Khoz. 1 Okeanogr., 3 : 195-198.

ЕВЕУЕ, H. A., 1964. “Variabilitat und Fehlbildungen in der Occipital
Region von Ondatra zibethica (L. 1766), ©. Siugetierkunde, 29: 331-336.

Frick, H., 1957. “Quantitative Untersuchungen an 4thiopischen
Saugetieren : I. Absolute und Relative Gewichte von Herz, Leber,
Milz und Nieren,” Anat. Anz., 104 : 305-333.

Слтввлдгтн, О. B., 1964. “The Agouti Pigment Pattern of the Mouse ;
a Quantitative and Experimental Study,” 7. Exper. Zool., 155 : 71-89.

GesczynskA, Z., 1964. ‘‘Morphological Changes Occurring in Laboratory
Maicrotus agrestis with Age,” Acta Theriol., 9 : 67-76.

GEGENBAUR, C., 1898-1901. Vergleichende Anatomie der Wirbeltiere. 2 vols.
Leipzig : W. Engelmann.

GenTILE, J., 1952. “А Craniometric Study of the Norway Rat,” 7.
Mammal., 33 : 190-197.

Gerasimova, М. А., 1958. Kachestvo mekha Barguzinskogo sobolya,
akklimatizirovannogo v Tomskoi oblasti (Quality of Fur of Barguzins
Sable Acclimatized to the Tomsk Region). Trudy Vsesoyuzn. Nauch.-
Issled. Inst. Khivotn. Syrya 1 Pushniny. 17.

GERSHENZON, S. M., 1941. “ ‘Mobilizatsionnyi rezerv’ vnutrividovoi
izmenchivosti’”’ (“ ‘Моше Reserve’ of Intra-species Variability’),
Kh. Obshch. Biol., 2 : 85-108.

GERSHENZON, S. M., 1946. “Ко estestvennogo otbora vrasprostranenil
i dinamike melanisma и khomyachkoy (Cricetus cricetus L.)” [“Вое of
Natural Selection in the Spread and Dynamics of Melanism in Hamsters
(Cricetus cricetus L.)”], Zh. Obshch. Biol., 7: 97-130.

Gityarov, М. 5. (Сншлвоу), 1959. “‘Problemy sovremennoi ekologii 1
teoria estestvennogo otbora”’ (“Problems of Modern Ecology and the
Theory of Natural Selection”), Usp. Sovrem. Biol., 48 : 267-278.

Gityarov, М. 5. (GuitARov), 1960. ‘‘Nekotorie obshchie zadachi evolyut-
sionnoi morfologii bezpozvonochnykh”’ (“Зое General Problems of

| Bibliography 317

Evolutionary Morphology of Invertebrates’), Usp. Sovrem. Biol., 49 :
214-224.

Gityarov, М. 5. (GuILARov), 1964. ‘“‘Sovremennye predstavleniya о
gomologii” (“Modern Concepts of Homology”), Usp. Sovrem. Biol.,
57 : 300-316.

GIMMEL’REIKH, С. A., 1962. “М. palatopharyngeus у svete evolyutsii
bronkhiel’noi muskulatory mlekopitayushchikh” (“М. palatopharyngeus
in the Light of the Evolution of the Bronchial Musculature of Mammals’’),
Arkhiv Anat. Gistol, Embriol., 43: 76-82.

GiapDkina, Т. 5., М. М. Meier and Т. М. Moxeeva, 1963. “О vnutrivido-
voi izmenchivosti melkikh gryzunov” (““Intra-species Variability of
Small Rodents”), Dokl. Akad. Nauk SSSR, 148 : 962-965.

Сом, О. B., 1943. “А Study of Individual Variation in Maicrotus р.
pennsylvanicus,” 7. Mammal., 24: 212-220. [Cited from Long (1968)].

Сотозснмтот, R., 1935. ‘“‘Geographische Variation und Artbildung,”
Naturwissenschaften, 23 : 169-176.

Gowen, J. W., 1936. “Inheritance as It Affects Survival of Rats Fed on
a Diet Deficient in Vitamin D.,”’ Genetics, 21 : 1-23.

Green, Е. L., 1939. ‘““The Inheritance of a Rib Variation in the Rabbit,”
Anat. Rec., 74 : 47-60.

Green, Е. L., 1941. ‘Genetic and Non-genetic Factors which Influence
the Type of the Skeleton in an Inbred Strain of Mice,” Genetics, 26 :
192-222.

Groves, С. P., 1963. “Results of a Multivariate Analysis of the Skulls of
Asiatic Wild Asses; with a Note on the Status of Mucrohoppus hemionus
blanfordi Pocock,” Ann. and Mag. Natur. History, © : 329-336.

GRUNEBERG, H., 1950. ‘“‘Genetical Studies on the Skeleton of the Mouse :
I. Minor Variation of the Vertebral Column,” 7. Genet., 50 : 112-141.

GRUNEBERG, H., 1955. ‘‘Genetical Studies on the Skeleton of the Mouse :
XV. Relations between Major and Minor Variations,” 7. Genet., 53 :
515-535.

GRUNEBERG, H., 1961. ‘“‘Evidence for Genetic Drift in Indian Rats
(Rattus rattus L.),” Evolution, 15 : 335-338.

GrineBerc, Н., 1963. The Pathology of Development. Oxford : Oxford
University Press.

Guitpay, J. E., 1957. “Individual and Geographic Variation in Blarina
-brevicauda from Pennsylvania,” Ann. Carnegie Mus., 35 : 41-68.

Сотные, В. D., 1965. ‘Variability in Characters Undergoing Rapid
Evolution : an Analysis of Microtus Molars,” Evolution, 19 : 214-233.

Harcket, E., 1894-1896. Systematische Phylogenie, Ed. 1-3. Berlin: G.
Reimer.

Hacmeier, Е. M., 1961. ‘Variation and Relationships in North American
Marten,”’ Canadian Field Nat., 75 : 122-138.

318 Bibliography

Harituincer, К., 1962. ‘Morphological Variability in Apodemus agrarius
(РаНаз. 1771)’ Acta. Тео’. 6: 239-256.

Hajek, К., 1962. “‘Vedecka hodnota teoretickeho odkazu Ch. Darwina,”
Biologia (Czechoslovakia), 17 : 546-551.

Harpang, J. В. S., 1949. ‘‘Suggestions as to Quantitative Measurements
of Rates of Evolution,’ Evolution, 3 : 51-56.

Hapane, J. В. 5., 1955. “The Measurement of Variation,” Evolution,
9 : 484.

Hatt, К. В. L., and С. В. ScHatier, 1964. ‘‘Tool-using Behavior of the
California Sea Otter,” 7. Mammal., 45 : 287-298.

НАмАймА, Fusanori, 1964. ““The Life History of the Japanese Mouse,
Mus molossinus Temminck and Schlegel : XI. Measurements of External
Characters, Skull, and Reproductive Organs in Adult Mice,” 56. Bull.
Fac. Agric. Kyushi Univ., Fukuoka, 21 : 73-82 (English Summary).

Hampurcu, М. and Е. Lynn, 1964. “The Influence of Temperature on
Skeletal Maturation of Hypothyroid Rats,” Anat. Rec., 150 : 163-168. |

Hampurcu, M., Е. Lynn, and Е. Werss, 1964. “Analysis of the Influence
of Thyroid Hormone on the Prenatal and Postnatal Maturation of the
Rat? Anat Rec, 150: 147-161.

Hamitton, J. E., 1939. “The Leopard Seal Hydrurga leptonyx (de —
Blainville),”’ Discov. Rep., 18 : 239-264. [Cited from Long (1968)].

Hay, О. P., 1915. ‘‘Contributions to the Knowledge of the Mammals
of the Pleistocene of North America,” Proc. U. 5. Nat. Mus., 48 : 515-575.
[Cited from Long (1968) ].

Hayne, О. W., 1950. ‘‘Reliability of Laboratory-bred Stocks as Samples
of Wild Populations, as Shown in a Study of the Variation of Peromyscus
polionotus in Parts of Florida and Alabama,” Contribs. Lab. Vert. Gen.
Univ. Mich., 46 : 1-56. ;

Harrow, В. F., 1962. ‘“‘Osteometric Data for the Florida Black Bear,”
Quart. 7. Florida Acad. Sct., 25 : 257-274.

Hetsop-Harrison, ]., 1959. “Variability and Environment,
ge Е: é

Hicuton, R., 1960. ‘‘Heritability of Geographic Variation in Trunk
Segmentation in the Red-backed Salamander, Plethodon cinereus,”
Evolution, 14 : 351-360. ;

Hinze, С., 1950. Der Biber. Berlin : Akademie-Verlag.

HorrMeister, О. Е., 1951. “А Taxonomic and Evolutionary Study of
the Pinon Mouse, Peromyscus ие,” Illinois Biol. Monogrs., 21 : 1-104.
Horrmerster, D. F., and М. В. Lez, 1963. ‘“The Status of the Sibling
Species Peromyscus merriami and Peromyscus eremicus,” 7. Mammal., 49:

201-213. Rae

Howe 1, А. B., 1924. ‘‘Individual and Age Variation in Muicrotus montanus

of Yosemite,” 7. Agric. Res., 28 : 974-1016.

29

Evolution,

Bibliography 319

Howe t, J. M., and Р. В. Sprmce, 1966. ‘Morphological Effects of the
Manx-factor in Cats,” 7. Heredity, 57 : 100-104.

‚ Hromapa, J., and L. Srrnap, 1962. “Prispevek ke kraniologii makaku,”
Ceskosl.. Morfol., 10 : 341-351. |

Hromapa, ]., and L. Strnap, 1964. “Beitrag zur Kenntnis der Austritt-
stellan des Nervus trigeminus und des Foramen symphyseos mandibulae
an den Schadeln des Macaca mulatta und Macaca irus (cynomologus),”
К. Morphol. und Anthropol., 55 : 368-379.

Huestis, К. R., 1925. “А Description of Microscopic Hair Characters
and of Their Inheritance in Peromyscus,’ 7. Exper. Xool., 41 : 429-270.

Hucues, В. L., 1964. “Sexual Development and Spermatozoan Morpho-
logy in the Male Macropod Marsupial Potorous tridactylus (Kerr), Austral.
ool 2 42-81.

Нохгеу, J. 5., 1942. Evolution, the Modern Synthesis. London : С. Allen
and Unwin.

Huxtey, J. 5., 1957. ‘Ве Three Types of Evolutionary Process,”’ Nature,
180 : 454-455.

Нохтеу, J. 5. (ed.), 1940. The New Systematics. Oxford : Clarendon Press.

ustnys dil) 185855); On the Wheory of the Vertebrate эк.” Proc.
Roy. Soc. London, 9 : 381-457.

Нузме-РаАнт, Снв., 1954. “Den Norske Greyling, Meles meles (L.),”
Arbok Univ. Breen Naturvit. rekke, 16 : 1-57.

Hysinc-Daut, Снв., 1959. ‘““The Norwegian Otter Lutra lutra (L.) : A
Craniometric Investigation,” Ibid., 5 : 1-44.

ТснтнАВА, T., 1963. “‘Identification of the Pigmy Blue Whale in the
Antarctic,” Norsk. Hvalfangst-tidende, 5: 128-130.

П’вмко, A. Г., 1968. ‘“‘Vliyanie zagryazneniya uchastka mestnosti
strontsiem-90 na izmenchivost’ v populyatsiyakh melkikh mlekopitayush-
chikh” (“The Influence of Pollution by Strontium-90 in Different
Geographical Regions on Variability of Populations of Small
Mammals’’), Zool. <h., 47: 1370-1377.

Ivanova, Е. I., 196la. ‘“‘Morfologicheskaya kharakteristika kashalota
(Physeter catodon L.) raiona kuril’skikh ostrovov”’ (‘‘Morphological
Characteristics of Physeter catodon L. in the Region of the Kuril Island’’),
Trudy Inst. Morf. Zhivot. Akad. Nauk SSSR, 34 : 151-173.

Ivanova, Е. I., 1961b. “Ргорогзи tela i rost finvala (Balaenoptera physalus
L.) dobyvaemogo и kuril’skikh ostrovoy”’ [*““Proportions of the Body and
Growth of the Finwhale (Balaenoptera physalus L.) Found Near the Kuril
Islands’’], Ibid., 34 : 174-194.

Ivanova, Е. I., 1961с. ‘‘Proportsii tela Seivala (Balaenoptera borealis L.)
dobyvaemogo и kuril’skikh obstrovov” [“Во4у Proportions of the Sei
Whale (Balaenoptera borealis L.) Found Near the Kuril Islands”’], Ibid.,
34 : 195-204.

320 Bibliography

Jackson, С. M., 1913-14. “Postnatal Growth and Variability of the Body
and the Various Organs in the Albino Rat,” Amer. 7. Anat., 15 : 1-68.

ТАЕЕЕ, Е. А., 1951. “A Quantitative Study of the Islets of Langerhans
in the Rabbit,” Anat. Rec., 111 : 109-122.

ТАЕНА, A., 1910. “Die Haare der Waltiere,” <ool. Jahrb., Abt. Anat.,
92: 1-92.

James, Tu. N., 1960. ‘The Arteries of the Free Ventricular Walls in
Мар,” Anat Rec., 136 : 371-376.

JEGLA, TH. C., 1963. “А Recent Deposit of Myotis lucifugus in Mammotl
Cave,” 7. Mammal., 44 : 121-122.

ТЕСТА, Tu. C., and J. 5. Harz, 1962. “А Pleistocene Deposit of the Free
Tailed Bat in Mammoth Cave, Kentucky,” 7. Mammal., 43 : 474-481.

JEWELL, P. A., and P. J. Futtacar, 1966. “Body Measurements of Small
Mammals : Sources of Error and Anatomical Changes,” 7. ool.,
150 : 401-409.

КАгЕГА, Отлут, 1957. “Regulation of Reproduction Rate in Subarctic
Population of the Vole Clethrionomys rufocanus (Зала), Suom. Tied. Toimit,
Ser. A, IV, No. 34 : 7-60.

Клгко\зкт, W., 1966. ‘Preferential Temperatures in the White Mouse
under Age Differentiation,’ Folia Biolog., 14 : 273-289.

Kamsuitov, М. M., 1939. ‘‘Otbor kak faktor menyayushchii zavisimost’
priznaka ot izmenenii vneshnykh uslovii’’ (“Selection as a Means of
Changing the Relations of a Trait from the Changes in Environmental
Conditions’’), Dokl. Akad. Nauk SSSR, 23 : 361-364.

Kamsuitov, М. M., 1959. “Vzaimnye otnosheniya mezhdu organismami
i ikh rol’ у evolyutsii”’ (“Mutual Relations between Organisms and Their
Role in Evolution”), <h. Obshch. Biol., 20 : 370-378.

Kasuya, Tosuio, and T. Тен\надва, 1965. ‘‘Some Information on Mink
Whales from the Antarctic,” Sci. Rept. Whales Res. Inst., Tokyo, 19 : 37-43.

Kau, О. K., 1966. “Ап Instance of Congenital Right Renal Agenesia
in the Indian Desert Gerbil, Meriones hurrianae Jerdon,” Proc. Zoolog.
Soc. India, 19 : 121-125.

Kern, A., 1958. Vidz evo evolyutsis (The Species and Its Evolution) Moscow:
Izdatelstvo Inostrannoi Literatury.

Kuuzin, В. Su., 1963. “К voprosu о somostoyatel’nosti trekh stad gren-
landskogo tyulenya’”’ (“Оп the Problem of the Independence of Three
Groups of Greenland Seals”), in Sb. Rabot. Sev. Otd. Polyarnogo Inst. Ryb.
Khoz. 1 Okeanogr. (Arkhangelsk, 1962), pp. 130-136.

Kuuzin, В. 5н., 1964. ‘“‘Morfo-ekologicheskaya kharakteristika gren-
landskogo tyulenya Yan-Maienskogo з{а4а’” (‘““Morpho-ecological
Characters of Greenland Seals of the Yan-Maien Group”), Materzaly
5655 Uch. Soveta Polyarnogo Inst. Куб. Khoz. 1 Okeanogr. po Rezul'tatem
Issledovanii, 1962-1963 ; Murmansk, pp. 226-234.

Bibliography 321

Kuuzin, К. Su., and А. У. УАвгокоу, 1963. “О nekotorykh chertakh
funktsionirovaniya pishchevaritel’nogo trakta khokhlacha (Cystophora
cristata) у period molochnogo рцалиуа” [‘‘“On Certain Aspects of the
Function of the Digestive Tract of the Hooded Seal (Cystophora cristata)
During the Period of Milk Feeding”’] Zool. Zh., 18 : 1273-1275.

Kuuzin, К. Эн., and М. Ya. YAKovenxo, 1964. “О rasspredelenii 1
biologii khokhlacha grenlandskogo morya”’ (“Оп the Distribution and
Biology of the Hooded Seal of the Greenland 5еа”), Trudy Polyarnogo
Inst. Ryb. Khoz. 1 Okeanogr. (Murmansk), No. 4.

Krister, H. H., 1943. “Relation of Certain Endocrine Glands to Body
Weight in Growing and Mature New Zealand White Rabbits,”
Endocrinology, 33 : 250-256.

Kine, H. D., 1923. ‘““The Growth and Variability in the Body Weight
of the Norway Rat,” Anat. Rec., 25 : 79-94.

Kina, J. A., 1965. “Воду, Brain and Lens Weight of Peromyscus,” Kool.
Jahrb., Abt II, 32: 177-188.

Kine, J. A., О. Mass, and В. С. Weismann, 1964. ““Geographic Variation
in Nest Size among Species of Peromyscus,” Evolution, 18: 230-234.

Kirisu, K., 5. Kanna, and T. Kanazawa, 1967. “А Metrical Study
on the Bony Palate of Macaca ати; and Cercopithecus aethiops,” Med. 7.
Osaka Univ., 18 : 11-18.

Kirpicunikoy, А. A., 1964. “О proiskhozhdenii Kaspiiskogo tyulenya”’
(“Origin of the Caspian Seal’’), Bull. Mosk. Obshch. Ispyt. Prir. (Biol.),
69 : 136-139.

KirpicHnikov, У. A., 1935. ‘‘Rol’nasledstvennoi izmenchivosti у protsesse
estestvennogo otbora (gipoteza о kosvennom otbore)”’ [‘“The role of
Hereditary Variability in the Process of Natural Selection (Hypothesis
of Indirect Selection)”’] Biol. Zh., 4 : 775-801.

Kirpicunikov, У. A., 1940. ‘“‘Znachenie prisposobitel’nykh modifikatsii у
evolyutsii”’ (““The Significance of Adaptive Modifications in Evolution”’)
Kh. Obshch. Biol., 1 : 121-151.

Kircnen, H., Е. М. Purnam, and W. J. Taytor, 1964. ‘‘Hemoglobin
Polymorphism : Its Relationship to Sickling of Erythrocytes in White-
tailed Deer,” Sczence, 144 : 1237-1239.

Kiem, О. R., 1964. ‘‘Range-related Differences in Growth of Deer
Reflected in Skeletal Ratios,” 7. Mammal., 45 : 226-235.

KLEINENBERG, S. E., 1956. Mlekopitayushchie Chernogo i Azovskogo moret
(Mammals of the Black and Azov Seas). Moscow : Izdatelstvo Akad.
Nauk SSSR.

KLEINENBERG, S. E., 1958. “К voprosu о proiskhozhdenii kitoobraznykh”’
(“Ор the Question of the Origin of Cetaceans’’), Dokl. Akad. Nauk SSSR,
122 : 950-952.

322 Bibliography

KLEINENBERG, 5. E., 1961. “Izuchenie morskikh mlekopitayushchikh у
SSSR 1 osnovnoe napravlenie semiletnego plana issledovanii’”’ (‘‘Studies
on the Marine Mammals of the USSR and the Basic Direction of the
Seven-Year Plan of Researches’), Trudy Soveshch. Ikhtiol. Komissit,
12 : 5-16.

KLEINENBERG, 9. E., V. М. BEL’KovicH and А. У. YaBLoxov, 1965. “О
vyrobotke edinoi metodiki izucheniya morskikh mlekopitayushchikh”
(“Оп the Development of a Unified Method of Study of Marine
Mammals”), in Sb. Morskie mlekopitayushchie, рр. 242-249. Moscow :
Izdatelstvo ‘Nauka’.

KLEINENBERG, S. Е., А. У. УАвгокоу, У. М. BeEL’Kovicu, and М. М.
TarRASEVICH, 1964. Belukha. Opyt- топовтайсйезКово issledovaniya vida.
Moscow : Izdatelstvo ‘Nauka’. [Translated 1969, Beluga (Delphinap-
terus leucas) : Investigation of the Species. Jerusalem : Israel Program
for Scientific Translations].

KLEINENBERG, S. E., and A. У. Улвгокоу, 1964a. ‘‘Morfologicheskaya
izmenchivost?’ mlekopitayushchikh 1 ee znachenie Фуа ponimaniya
strukturi i dinamiki chislennosti populyatsii’’ (““Morphological Vari-
ability of Mammals and Its Significance for the Understanding of the
Structure and Dynamics of Population Numbers’’), Materialy Soveshch.
po Sovrem. probl. Dinamiki Chislennosti, pp. 50-51. Moscow.

KLEINENBERG, о. E., A. V. YABLOKOov, С. А. KLEVEZAL’, У. М. BEL’KOVICH,
and У. YA. Erin, 1965. “Spravochnye pokazateli po kharakteristike
nekotorykh lastonogikh и kitoobraznykh” (“Handbook of Indices for
Characteristics of Some Pinnipeds and Cetaceans’’), Sbornik Morskie
mlekopitayushchie, pp. 252-257. Moscow : Izdatelstvo ‘Nauka’.

Ктеммт, L., 1960. “Quantitative Untersuchungen an Apodemus sylvaticus
(Linnaeus, 1758),” Zool. Anz., 165 : 249-275.

KLEVEZAL’, С. A. and S. Е. KLEINENBERG, 1967. Opredelenie vozrosta
mlekopitayushchikh ро. sloistym strukturam zubov 1 kosti (Determination of
the Age of Mammals by the Layered Structure of Teeth and Bones).
Moscow : Izdatelstvo ‘Nauka’.

Ктомоу, 5. К. and Е. S. Cuuzuyaxina, 1955. “Akiba: kol’chataya
nerpa (Phoca hispida ochotensis Pallas, Phoca hispida krascheninikovi)”’
[‘“Akiba : the Ringed Seal (Phoca hispida ochotensis Pallas, Phoca hispida
krascheninikovi)”’|, Trudy Inst. Okeanol., 14 : 104-107.

Костеха, Е. Z., 1963. “‘Sezonnaya izmenchivost’ 1 vozrostnie osobennosti
stroeniya kozhii volosyannogo pokrova krota, zaitsabelyaka 1еп. otovidnoi
sobaki”’ (“Seasonal Variability and Age Characteristics of Skin Structure
and Hair of Moles, White Hares, and Raccoon Dogs”’), in $b. Nauchn.
Stater Zap. Ош. Vses. Nauch.-Issled. Inst. Zhiv. бут. Push., 2 : 213-271.

Ko.’ TsovA-SADOVNIKOVA, М. P., 1931. ‘“‘Geneticheskii analiz psikhiches-
kikh sposobnostei Кгуз” (““Genetical Analysis of Psychological Characters

Bibliography 323

of Rats”), Zh. Eksperim. Biol., 7 : 265-283.

Konstantinoy, К. G., 1956. ‘‘Napravlennaya izmenchivost’ ili estestven-
пу! otbor ?” (“Directed Variability or Natural Selection ?”), Zool. Xh.,
Зо 2- 977,

Koren, К. I., 1964. ‘‘Opyt izucheniya estestvennogo otbora у prirodnykh
usloviyakh” (“Ап Experiment to Study Natural Selection under Natural
Conditions”), Sb. Voprosy Vnutrividovor Izmenchivosti Nazemnykh Pozvono-
chnykh Zhivotnykh 1 Mikroevolyutsiya, рр. 61-62, Sverdlovsk.

Корым, К. I., 1967. ‘‘Morfo-fiziologicheskie osobennosti severnykh
populyatsii gornostaya” (“Morpho-physiological Characteristics of
Northern Populations of Weasels”), Trudy Moskouskogo Obshch. Ispyt.
Prir., 25 : 40-48.

Ковотоу, L. G., 1962. ‘‘Izmeneniya nadpochechnikov i uglevodmogo
obmena и vodyanoi polevki (Arvicola terrestris) У vesene-letnyi period”
[‘“‘Changes in Adrenals and Carbon Dioxide Metabolism in Water Voles
(Arvicola terrestris) in the Autumn-summer Period], Trudy Inst. Biol.
Uralsk Fil. Akad. Nauk SSSR, 29 : 129-140.

Korzuuev, Р. A., 1964. Gemoglobin : Sravnitel’naya fiziologiya 1 biokhimiya
(Hemoglobin : Comparative Physiology and Biochemistry). Moscow :
Izdatelstvo ‘Nauka’. ;

Kosuxina, T. V., 1964. ‘‘Mezhvidovaya konkurentsia lesnykh polevok
i regulyatsiya ikh chislennosti” (‘‘Intra-species Competition of Field
Voles and the Regulation of Their Numbers”), Matertaly Soveshch.
Sovremennye Problemy Izucheniya Dinamiki Chislennosti Роршуазй Zhivotmkh.
Inst. Morf. Zhivotnykh Akad. Nauk SSSR, pp. 51-53.

KostTeLecka-Myrcua, A., and A. Myrcna, 1964.. “The Rate of Passage
of Foodstuffs through the Alimentary Tracts of Certain Maicrotidae
under Laboratory Conditions,” Acta Theriol., 9 : 37-53.

КозтегескаА-МувснаА, A., and A. Myrcua, 1964a. ‘“‘Choice of Indicator
in the Investigation of the Passage of Foodstuffs through the Alimentary
Tract of Rodents,” Acta Theriol., 9 : 55-65.

КозтегескА-Мувсна, A., and A. Мувсна, 1964b. “Ваше of Passage of
Foodstuffs through the Alimentary Tract of Neomys Fodiens (Pennant,
1770-1771) under Laboratory Conditions,” Acta Theriol., 9 : 371-373.

Kozaxevicu, У. P., 1967. “‘Soderzhanie askorbinovoi kisloty v nadpoche-
chnikakh i nekotorye ekologo-fisiologicheskie osobennosti malykh
Suslikov” (‘Ascorbic Acid Content in the Adrenals and Some Ecologo-
physiological Characteristics of Small Susliks’’), Bull. Mosk. Obshch.
Ispyt. Prir. (Biol.) 72: 107-113.

Kramer, А. W., ]в., 1964. “Body and Organ Weights and Linear Measure-
ments of the Adult Mongolian Gerbil,”? Anat. Rec., 150 : 343-348.

Квгуознеку, V. С., and О. L. Воззоммо, 1964. “Vnutrividovaya
izmenchivost’ sibirskogo lemminga’’ (‘‘Intra-species Variability of the

324 Bibliography

Siberian Lemming”), Sb. Vopr. Vnutrivid Izmenchivosti Nazemnykh
Pozvonochnykh Zhivotnykh i Mikroevolyutsiya, pp. 65-66. Sverdlovsk.
Квознимзки, L. V., 1946. “Ро]оуое razlichie у povedenii sobak” (“Sex

Differences in the Behavior of the Dog”), Zh. Obshch. Biol., 7 : 131-146.

KryzHanovski, 5. G., 1950. ‘‘Teoroticheskie osnovy embriologii”’
(“Theoretical Basis of Embryology”), Usp. Sovrem. Biol., 30 : 382-413.

Kryw’tsov’a, А. I., 1957. “Izmenenie vesa stepnykh pestrushek (Lagurus
lagurus) У zavisimosti ot ikh pola i vozrosta”’ (“‘Changes in the Weights
of Lagurus lagurus Due to Their Sex and Growth’), Zool. Zh., 36:
1239-1250.

Kusik, J., 1951. ‘Analysis of the Pulawg Population of Sorex araneus and
Sorex minutus minutus L.,” Ann. Univ. М. Curie-Slodowska, Sect. С, 5 :
194-215,

Ковк, J., 1953a. “Badania nad morfologia i biologia smuzki (51615
betulina Pall.) 2 Bialowieskiego Parku Narodowego,”’ Ibid., 7 : 1-64.
Ковк, J., 1953b. ‘‘Micromys minutus Pall. м Bialowieskim Parku

Narodowym,” Jbid., 7 : 449-527.

Ковота, Kinziro, 1967. “Comparative Anatomical and Neurohistological
Observations on the Tongue of Elephants (Elephas indicus and Loxodonta
africana) ,”’ Anat. Res., 157 : 505-515.

KiGKENTHAL, W., 1890. ‘Uber die Hand der Cetacean,’” Anat. Anz.,
5 : 44-52.

KUKENTHAL, W., 1909. ‘“‘Untersuchungen an Walen,” Х. WNaturwiss,
45 : 454-588.

Kunze, А., 1912. “Uber die Brustflosse der Wale,” Kool. Fahr., Abt. Anat.
Ontog., 32 : 577-651.

Kurten, B., 1964. ‘The Evolution of the Polar Bear (Ursus maritimus
Phipps),”? Acta Zool. Fennica, 108 : 1-30.

Kurten, B., 1965. “Оп the Evolution of the European Wild Cat (Felis
silvestris Schreber),” Acta Zool. Fennica, 111 : 1-29.

Ковтем, В., 1965a. ‘The Carnivora of the Palestine Caves,” Acta Zool.
Fennica, 107 : 1-74.

KurtEn, B., 1965b. “The Pleistocene Felidae of Florida,’ Bull. Florida
State Mus., Biol. Sci., 9 : 215-273.

Kurten, BS and В. Rauscu, 1959. ‘Biometric Comparison between
North American and European Mammals,” Acta Artica, 11 : 1-44.

Кознмев, Ku. F., 1941. “О geneticheskoi i fiziologicheskoi prirode
geterozisa i intsukht Чергеззи” (‘““The Genetical and Physiological
Nature of Heterosis and Inbreeding Depression’’), Zh. Obshch. Biol.,
2: 229.271.

Kuznetsov, В. А., 1941а. ‘“‘Geograficheskaya izmenchivost’ sobolei 1.
kunits fauny SSSR” (“Geographical Variability of the Sable and
Marten of the USSR’’), Trudy Moskovskovo Zootekhn. Inst., | : 113-133.

Bibliography 325

Kuznetsov, В. A., 1941b. “‘Aberrativnaya izmenchivost’ okraski pushnykh
гуегег” (“Abnormal Variability of Colors of Fur Animals”), Trudy
Mosk. Zootekhn. Inst., 1 : 134-140.

Kuznetsov, У. V., 1956. ‘‘Nekotorye zakonomernosti vidovoi morfo-
logicheskoi i biologicheskoi raznokachestvennosti na primere morskikh
bezpozvonochnykh” (“Some Regularities of Morphological and Bio-
logical Species Differentiation with Examples from Marine Invertebra-
Gesu), 00) <n 35 5 1118-1130.

Larina, М. Г., 1958. “К voprosu о diagnostike blizkikh vidov lesnoi i
zheltogorloi myshi” (“Оп the Question of the Diagnosis of the Close
Species of Field and Yellow-necked Mice”), Zool. Zh., 37 : 1719-1732.

Larina, М. I., 1962. “Individial’naya vozrostnaya и geograficheskaya
izmenchivost’ shukhovykh kostochek и gryzunov (Na primere blizkikh
vidov rodov Apodemus i Citellus)’? [Individual Age and Geographical
Variability of Ear Bones in Rodents (with Examples from Related
Species of the Genera Apodemus and СиеЦиз)”]. Zool. Zh., 41 : 1536-1547.

Larina, М. I., 1964. ‘‘Vnutrividovaya izmenchivost’ yader mezoteliya
u lesnoi i zheltogorloi myshi” (“‘Intra-species Variability of Nuclei of
Mesothelium in Field and Yellow-necked Mice’’), Sb. Vopr. Vnutrivid
Izmenchivosti i Mikroevolyutsiya, pp. 69-71. Sverdlovsk.

Larina, М. I., 1966. ‘‘Vnutrividovaya izmenchivost? razmerov yader
mezoteliya bryzhzheiki и lesnoi (Apodemus sylvaticus L.) i zheltogorloi
(A. tauricus Pall.) myshi” [‘‘Intra-species Variability of Measurements of
Nuclei of Mesothelium of Mesentery in Wild (Apodemus sylvaticus L.)
and Yellow-necked (A. tauricus Pall.) М1се”], Trudy Vsesoyuzn. Sovesh-
chaniya Vnutrivid. Izmenchivost? Nazemnykh Zhivotnykh 1 Mikroevolyutsiya,
pp. 255-258. Sverdlovsk.

Latimer, Н. B., 1936. “Уаз: and Linear Measurements of the Adult
Cat,” Amer. 7. Anat., 58 : 329-347.

Latimer, H. B., 1937. ‘“‘Weights and Linear Dimensions of the Skull and
Some of the Long Bones of the Skunk (Mephitis mesomelas avia),” 7.
Morphol., 60 : 379-391.

Latimer, H. B., 1944. ‘“‘The Prenatal Growth of the Cat: XIV. The
Weight of the Skeleton in the Fetal and in the Adult Cat,” Growth,
8 : 149-158.

Latimer, Н. В., 1944а. ‘“The Prenatal Growth of the Cat : XV,” Growth,
8 : 205-219.

Latimer, H. B., 1951. ‘“The Weight of the Eyeballs in the Guinea. Pig,”’
Anat. Rec., 110 : 349-358.

Latimer, H. B., 195la. ‘‘Weights, Percentage of Weights, and Correlations
of the Endocrine Glands of the Adult Male Guinea Pig,” Anat. Rec.,
В] - 299-515.

Latimer, Н. B., 1961. ‘‘Weights of the Ventricular Walls of the Heart

326 Bibliography

in the Adult Dog,” Univ. Kansas Sci. Bull., 62 : 3-12.

Latimer, H. B., and В. В. Виху, 1934. “Measurements of the Skull and
of Some of te Long Bones of the Muskrat (Ondatra zibethica GEE)
F. Morphol., 56 : 203-212.

Latimer, H. B., and P. B. Sawin, 1955. ‘Morphological Studies of
Rabbits : хи. Organ Size in Relation to Body Weights in Adults of
Small-sized Race X,”’ Anat. Rec., 143 : 81-97.

Latimer, H. B., and P. B. Sawin, 1955a. “Morphological Studies of

_ Rabbits : XIII. The Influence of the Dwarf Gene upon Organ Size
and Variability in Race X,” Anat. Rec., 143 : 447-466.

Latimer, H. B., and Р. В. Sawin, 1959. “Могрвосепейс Studies of the
Rabbit, XXII : Linear Measurements of Large Race III and Small
Race X,” Anat. Rec., 134 : 69-86. :

Lavrov, М. P., 1953. “Sistematicheskoe polozhenie ondatri akklimatizi-
rovannoi у SSSR, 1 vliyaniya uslovii sredi na izmenenie priznakov”’
(“Taxonomic Position of Ondatra acclimatized in the USSR, and the
Effect of Environmental Conditions on the Changes in Traits’’), <ool.
92 ПТИ.

Lavrova, М. УА., and У. 5. и, 1965. “О sistematike 1 biologi
zemleroik Kracnodarskor kraya s otsenkoi ikh roli v leptospiroznykh
ochagakh” (“Taxonomy and Biology of Shrews of the Krasnodar
Region, with an Evaluation of Their Role in the Leptospira Breeding
Grounds”’), Zool. Zh., 44 : 101-105.

Law, L. W., 1938. “Studies of Size Inheritance in Mice,”’ Genetics, 23 :
399-422.

Lawrence, B., 1960. ‘‘Fossil Tadarida from New Mexico,” 7. Mammal.,
41 : 320-322.

LeBEDKINA, N. S., 1957. ‘‘Morfofunktsional’nyi analiz chelyustnogo
apparata хайзеу” (““Morpho-functional Analysis of the Jaw Apparatus
in Hares) 002 Ai, 30 27 1539-1555:

Ler, A. K., 1963. ‘“‘The Adaptation to Arid Environment in Wood Rats
of the Genus Neotoma,” Univ. Calif. Publ. Zool., 64 : 57-96.

Генмдм, E., 1961 (1962). ‘‘Langschwanzmause der Gattung Apodemus
in Hohe Venn,” Decheniana, 114 : 177-185.

Levine, С. ]., МУ. Mann, H. С. Hones, J. Автет, and О. DuPont, 1941.
“Distribution of Body Weight in the Organs and Tissues of the Rabbit,”
Proc. Soc. Exper. Biol. and Med., 47 : 318-321.

Lewenz, М. A., and М. A. У’ытегекх, 1902. “А Second Study of the
Variability and Correlation of the Hand,” Biometrics, 1 : 345-350.

Lewontin, В. C., 1957. “The Adaptations of Populations to Varying
Environments,” Cold Spring Harbor Sympos. Quart. Biol., 22 : 395-408.

Lewontin, В. C., 1961. «‘‘Evolution and the Theory of Games,” 7. ‘Theoret.
Biol., 1 : 382-403.

Bibliography 327

Lipicker, Jr., 1960. “An Analysis of Intra-specific Variation in the
Kangaroo Rat, Dipodomys merriami,” Univ. Calif. Publ. Zool., 67 : 125-218.

Lirttepace, J. L., 1963. “Diving Behavior of a Weddell Seal Wintering
in McMurdo Sound, Antarctica,” Ecology, 44 : 775-777.

Livanov, М. A., 1960. “О zakonomernostyakh zhivoi prirody” (“Оп
the Regularities of the Living World’’), Uch. Zp. Kazan’ Univ., 120 :
3-32.

Lone, С. A., 1965. “The Mammals of Wyoming,” Univ. Kansas Publ.
Mus. Nat. Hist., 14 : 266 pp.

Lone, С. A., 1968. “Ап Analysis of Patterns of Variation in Some
Representative Mammalia : Pt. I. A Review of Estimates of Variability
in Selected Measurements,” Trans. Kansas Acad. Sci., 71 : 201-227.

Lone, С. A., 1969. “Ап Analysis of Patterns of Variation in Some Re-
presentative Mammalia : Pt. П. Studies on the Nature and Correlation
of Measures of Variation,” in Contr. in Mammalogy : A Volume Honoring
Prof. E. R. Hall. Univ. Kansas Mus. Nat. Hist. Publ., 51 : 289-302.

Lone, С. A., 1970. “An Analysis of Patterns of Variation in Some
Representative Mammalia : Pt. ПТ. Some Equations on the Nature of
Frequency Distributions of Estimated Variabilities,’ Acta Thervol.,
15 : 517-528.

Lone, С. A., and С. J. Jones, 1966. ‘Variation and Frequency of Occur-
rence of the Baculam in a Population of Mexican Free-tailed Bats,”
Southern Nat., 11 : 290-295.

Lone, С. A., С. J. Jones, and P. Kamensxy, 1967. “‘Osteometric Variation
and Function of the High-speed Wing of the Free-tailed Bat,” Amer.
Midland Natur., 77 : 452-461.

Lowrance, E. O., and H. B. Latimer, 1957. “Weights and Linear
Measurements of 105 Human Skeletons from Аза,” Amer. 7. Anat.,
101 : 445-459.

Lukin, Е. E., 1936. “О parallelizme nasledstvennoi i nenasledstvennoi
izmenchivosti’’ (“Оп the Parallelism of Hereditary and Non-Hereditary
Variability”), Uch. Zap. Khar’k. Univ., Vols. 6-7.

Lukin, Е. E., 1940. Darvinizm 1 geograficheskie zakonomernosti о izmeneniu
organismov (Darwinism and Geographical Regularities in the Variation
of Organisms). Moscow, Leningrad : Izdatelstvo Akad. Nauk, SSSR.

Lyne, A. G., С. 5. Могумеох, К. Myxytowyeaz, and P. Е. PARAKKAL,
1964. “The Development, Structure, and Function of the Submandi-
bular Cutaneous (Chin) Glands in the Rabbit,” Austral. 7. Zool., 12:
340-348.

LyusisucHEv, А. A., 1923. “О kriterii izmenchivosti organizmov”’ (“On
the Criteria of Variability of Organisms’’), Izv. Biol. Nauch.-Issled. Inst.
Permsk. Gos. Univ., Vol. 1, Мо. 7-8.

LyupisucHev, А. A., 1959. “О primenenii biometrii у sistematike” (“Оп

328 Bibliography

the Use of Biometry in Тахопоту”), Vestnik Leningradskii Gos. Univ,
1959 (3) : 128-136.

Туоткоу, К. M., 1931. ‘‘Geneticheskii analiz у otnoshenii medlinno
plodyashchikhsya, zhivotnykh” (‘‘Genetical Analysis of Slowly Maturing
Animals’), Trudy ТУ Vses. S’ezda Zoologov. Anatomov i Gistologov., Kiev,
pp. 142-145.

Млднмоузки, A. A., 1939. “Вог geneticheskikh i fenogeneticheskikh
yavlenii у evolyutsii vida : Chast’ I. Pleiotropia”’ (““The Role of Genetical
and Phenogenetical Phenomena in the Evolution of Species : Part I.
Pleiotropy’’), 120651. Akad. Nauk SSSR, Seria Bool., 1939 : 575-614.

Млдикмоузки, А. A., 1948. “Elementarnye korrelyatsii i izmenchivost’
chelovecheskogo organizma”’ (“Elementary Correlations and Variability
of the Human Organism’’), Trudy Inst. Tsitol. Gistol. 1 Embriol., 2: 136-
198.

Maun, А. V., 1866. Ezniges von den Walfischen im Allgemeinen und Balanoptera
carolinae im Besonderen. Berlin.

Mannine, Т. H., 1956. ‘““The Northern Red-backed Mouse, Clethrionomys
rutilus (Pallas) in Canada,” Nat. Mus. Canada Bull., 144 : 1-67. [Cited
from Long (1968) ]. :

Мамупле, В. H., 1959. “Bregmatic Bones in North American Lynx,”

Science, 130 : 1245.

Мамупле, К. H., 1961. “The Entepicondylar Foramen in Ochrotomys,”

JF. Mammal., 42 : 103-104.

Markov, G., 1957. Nasekomyadnite bozainitst о Bulgariya. Fauna Bulgariya,

Vol. 3, Sophia. `

Матзот, TaKayasu, 1964. “The Medio-ventral Muscles of the Abdomen
in Macacus cyclopsis (М. rectus abdominis and М. pyramidalis),” Acta Med.
Nagasakiensia, 8 : 63-80.

Mattruews, L. H., 1938a. ‘““The Sperm Whale, Physeter catodon,’ Discov.
Reports, 42 : 93-168. [Cited from Long (1968) ].

Mattuews, L. Н., 1938b. ‘“The Sei Whale, Balaenoptera borealis,” ‘Discov.
Repts., 42 : 183-290. [Cited from Long (1968)].

МаАттнех, R., 1949. Les Chromosomes des Vertébrés. Lausanne : Е. Rouge.

Marveey, В. S., 1936. “Зоугетеппуе zadachi evolyutsionnoi morfologir’
(“Modern Problems of Evolutionary Morphology’’), Izv. Akad. Nauk
SSSR, Sertya Biol., 1936 : 863-893.

Matveev, В. S., 1954. “Вог morfologii у razreshennii ocherednykh
problem biologii” (““The Role of Morphology in the Solution of Major
Problems in Biology’’), Zool. <h., 33 : 743.

Mayr, E., 1942. Systematics and the Origin-.of Species. New York : Columbia
Univ. Press. |

Mayr, E., 1947. Sistematika i proiskhozhdenie vidoy. (Systematics and the
Origin of Species). Moscow: Izdatelstvo Inostrannoi Literatury.

Bibliography 329

Mayr, E., 1962. Animal Species and Evolution. Cambridge : Harvard
University Press.

Mayr, E., Е. Linsey and К. Оямсев, 1956. Metody i printsipy zoologiches-
kot sistematiki (Methods and Principles of Zoological Taxonomy).
Moscow : Izdatelstvo Inostrannoi Literatury.

McDoneti, W. R., 1907. “А Second Study of the English Skull, with
Special Reference to Moorfields Crania,” Biometrika, 5 : 86-104. [Cited
from Long (1968) ].

McKeever, 9т., 1963. ‘Seasonal Changes in Body Weight, Reproductive
Organs, Pituitary, Adrenal Glands, Thyroid Gland and Spleen of the
Belding Ground Squirrel (Crtellus belding:)”, Amer. У. Anat., 113:
153-173.

McKeever, Эт., 1964. ‘Variation in Weight of the Adrenal, Pituitary
and Thyroid Glands of the White-footed Mouse (Peromyscus maniculatus) ,”
Amer. F. Anat., 114 : 1-16.

McKeever, Эт., and T. P. Quentin, 1963. ‘“‘Observations on the Adrenal
Glands of the Mongoose,” Anat. Rec., 147: 163-169.

McLaren, J. A., 1960. “On the Origin of the Caspian and Baikal Seals,
the Paleoclimatological Implication,” Amer. 7. Sci., 258 : 47-65.

МсМЕЕкАм, С. P., 1940. “Growth and Development in the Pig with
Special Reference to Carcass Quality Characters,” 7. Agric. Sci., 30 :
276-311.

Merer, М. N., 1957. “О vozrostnoi izmenchivosti malogo suslika
(Citellus pigmaeus Ра|.)” [“Оп the Growth and Variability of the Small
Suslik (Citellus pigmaeus Pall.)”], Zool. Kh., 36 : 1392-1402.

MezuzuHerin, V. А., 1964а. “Yavlenie denelya i ego vozmozhnoe
ob’yasnenie”’ (““The Denel Phenomenon and Its Possible Explanation’),
Acta Theriol., 8 : 95-114. |

MezyuzHerin, У. А., 19645. “Osobennosti sezonnoi i geograficheskoi
izmenchivosti zemleroek-burozubok i ikh znachenie v evolyutsii dannoi
отирру” (“Characteristics of Seasonal and Geographical Variability of
Shrews and Their Significance in the Evolution of this Group’), Sb.
Voprosy Vnutrividovoi Izmenchivosti Mikroevolyutsiya, рр. 76-77. Sverdlovsk.

MezuzuHerin, У. A., 1965. “О smene adaptatsii у protsesse evolyutsii
(па primere года Sorex, Insectivora, Mammalia)” [““On Changes in Adapta-
tion in the Process of Evolution (with an Example from the Genus
Sorex, Insectivora, Маттайа)”], Materialy ool. Soveshch. po Probleme
Biologicheskie Osnovy Rekonstruktsii-Fauny Yuzhnoi Кот Evropecheskot Chasti
SSSR, рр. 76-78. Kishimev.

Micuurin, L. N., 1965. “О nekotorykh morfologicheskikh osbennostyakh
dikikh sovernykh olenei poluostrova Taimyr’’ (“Оп Certain Morpholo-
gical Characters of Northern Wild Deer of the Taimyr Island”), <ool.
Kh., 44 : 1396-1404.

330 Bibliography

Moment, С. B., 1962. “ВеНежуе Selection : А Possible Answer to an
Old Puzzle,” Scrence, 136 : 263-272.

MUuKHAMEDGALIEV, Е. M., 1953. “О biodinamike legkikh’’ (“Оп the Bio-

dynamics of Lungs’), Trudy Alma-Atinskogo Zoovetinstituta, 7 : 159-169.

Murray, К. F., 1965. ‘Population Changes during the 1957-1958 Vole

(Microtus) Outbreak in California,” Ecology, 46 : 163-171.

NAEVDAL, GUNNAR, 1966. ‘Protein Polymorphism Used for Identification

of Harp Seal Populations,” Bergen Univ. Aarbok, Mat. Naturvit. Ser.

Nakat, J., and Т. Suipa, 1948. “Sinus Hairs of the Sei Whale (Balaenopteris

borealis),”’ Sci. Repts. Whales Res. Inst., Tokyo, 1 : 41-49.

Naumov, М. Р., 1941. “‘Lastonogie okhotskogo morya’’ (Pinnipeds of the

Okhotsk Sea”’), Uch. zap. Mosk. Gos. Ped. Inst., 24 : 19-72.

Naumov, N. P., 1945. ““Geograficheskaya izmenchivost’, dinamika
chislennosti 1 evolyutsii zhivotnykh” (“Geographical Variability,
Dynamics of Number and Evolution of Animals’), Zh. Obshch. Biol.,
Oi: 3 7202.

Naumov, №. Р., 1958. ‘“‘Vzaimodeistvie so sredoi edinichnykh organizmov.
1 populyatsii zhivotnykh” (“Interaction of Individual Organisms and
Populations with the Environment’’), Filosof. Vopr. Estestv., 1 : 289-308.

Naumov, М. P., 1963. “‘Biologicheskie makrosistemy”’ (““Biological Macro-
systems’’), Priroda, 1963 (5) : 22-29. ats

Naumov, М. P. and С. У. Nixow’sku, 1962. “О nekotorykh obshchikh
zakonomernostyakh dinamiki populyatsii zhivotnykh” (“Золе General
Patterns of Population Dynamics of Animals’’), Zool. Zh., 41 : 1132-1141.

NAZARENKO, Yu. I., and A. V. УАвгокох, 1962. ‘‘Otsenka metoda ucheta
chislennosti belomorskogo lysuna i soobrazheniya o sostoyanii ego
zapasov’ (“Evaluation of a Method of Calculation of the Numbers of
the White Sea Lisun and a Consideration of the State of the Population’’),
Kool. Zh., 41 : 1864-1871.

NEHRING, A., 1884. ‘‘Fossile Pferde aus Deutschen Diluvial-Ablagerungen
; . ‚’ Landwirtshaft Fahrb., 13 : 81-160. [Cited from Long (1968)].
Nemoto, Т., 1959. ‘‘Food of Baleen Whales with Reference to Whale

Movements,” Sci. Repts. Whales Res. Inst., Tokyo, 14 : 149-205.

Newman, H. H., 1913. ‘“The Modes of Inheritance of Aggregates of
Meristic (Integral) Variates in the Polyembryonic Offspring of the Nine-
banded Armadillo,” 7. Exper. Zool., 15 : 145-192.

Newson, J., and О. Curry, 1962. ‘Hemoglobin Levels, Growth and
Survival in Two Microtus Populations,” Ecology, 53: 733-738.

Мког’зки, С. V., 1955. “‘Ob izmenchivosti organizmov”’ (“Оп the Vari-
ability of Organisms’’), Zool. <h., 34 : 723-733.

Nrxow’sku, С. V., 1961. “О prichinakh fluktatsii chislennosti губ” (“Оп
the Reasons of Fluctuations of Numbers of Fish?’), Vopr. Ikhtiol., Vol. 1,
4(21) : 659-665.

Bibliography 331

Мкот’зки, С. V., 1963a. Vid i vidoobrazovanie (Species and Speciation).
Moscow, ‘Znanie’. ;

Nixov’sku, С. V., 1963b. “О biologicheskikh osnovakh matematicheskogo
modelirovaniya dinamiki populyatsii ryb” (“On the Biological Basis of ©
а Mathematical Model for the Dynamics of a Fish Population”), Vopr.
Tkhtiol., 3 : 591-609.

\ког?зки, С. V., and У. A. РжкотеуА, 1958. “О prisposobitel’nom
gnachenii amptitudy izmenchivosti vidovykh priznakov 1 svoistv
organizma”’ (“On the Adaptive Significance of the Amount of Vari-
ability of Traits of Species and the Nature of an Organism’’), Zool. Zh.,
37 : 972-988. :

Nixoro, Z. S., 1964. ‘‘Izmenenie stroeniya populyatsii nad deistviem
otbora у sluchae sverkhdominirovaniya” (“Change in Population
Structure under the Influence of Selection in Case of Overdominance’’),
Bull. Mosk. Obshch. Ispyt. Prir. (Biol.), 19 : 5-21.

Nixutn, P. С., 1937. ‘‘Nablyudeniya nad _ lastonogimi Okhotskogo 1
Yaponskogo morei’’ (“Observation on Pinnipeds of the Okhotsk and
Japan Seas”), Izv. Тай Nauchno-Issled. Inst. Ryb. Khoz. 1 Okeanogr.
(Vladivostok), 10*: 49-59.

Nisntwakt, M., Тозню Kasuya, Т. Kamia, Т. ТовдуАМА, and
М. Nakajima, 1965. Sci. Rept. Whales Res. Inst., Tokyo, 19 : 65-90.
Nissen, H. W., and A. H. Влезем, 1964. “The Eruption of the Permanent

Dentition of the Chimpanzee,” Amer. 7. Phys. Anthropol., 22 : 285-294.

Norv, H. J., 1963. ‘‘Quantitative Untersuchungen am Mus musculus
domesticus Rutty, 1772,” Zool. Anz., 170 : 311-335.

Nuzupin, М. Г., 1956. ‘“‘Vozglyadi Darvina 1 Michurina na rol’ sredi
у izmenchivosti organizmov” (“The Views of Darwin and Michurin
on the Role of Environment in the Variability of Organisms’’), Izv. Akad.
Nauk SSSR, Ser. Biol., 1956 (2) : 18-28.

Ovum, E., 1959. Fundamentals of Ecology. Philadelphia : Saunders.

Oetrexinc, B., 1923. “Оп the Morphological Significance of Certain
Cranio-vertebral Variations,’ Anat. Rec., 25 : 339-353.

Ocawa, T., and T. Kamia, 1957. “А Case of the Cachalot with Protruded
Rudimentary Hind Limbs,” Scz. Repts. Whales Res. Inst., Tokyo, 12:
197-208.

Oenev, S. I., 1931. Zveri Vostochnoi Evropy & Severnor Али (Mammals of
Eastern Europe and Northern Asia). Vol. 2, Khishchnye (Carnivores).
Moscow : Glavnauka-Gosudarstvennoe Izdatelstvo. (Translated 1962,
Jerusalem: Israel Program for Scientific Translations).

Oenev, 5. Т., 1935. Zveri SSSR i prilezhashchikh этап : HI. Khishchnye 2
lastonogie (Mammals of the USSR and Adjacent Countries: ПТ. Carni-
vores and Seals). Moscow : Izdatelstvo Akad. Nauk SSSR.

Ocnev, S. I., 1942. ‘Problemy sistematiki : I. Znachenie Morfologii

332 Bibliography

dlya_ sistematicheskikh issledovanii” (“Problems of Taxonomy: I.
Significance of Morphology for Taxonomic Research”), Zool. Zh.,
21 : 281-283. |

Отемоу, Yu. M., 1941. “О vliyanii predshestvuyushie istorii vida na
ego dalneishee razvitie’’ (“On the Influence of the Present History of
the Species on Its Future Development’), Dokl. Akad. Nauk SSSR,
31 : 157-160.

‘‹ОтеЕмоу, Yu. M., 1961. ‘“‘Nekotorye problemy evolyutsionnoi genetiki i
Darvinizma” (“Some Problems of Evolutionary Genetics and
Darwinism”), Moscow and Leningrad: Izdatelstvo Akad. Nauk
SSSR.

ОтЕмоу, Yu. M., and I. 5. Кнвматз, 1939. ‘““Transformatsiya genotipa v
prirodnykh popyulyatsiyakh D. melanogaster” (““Transformation of the
Genotype in Natural Populations of D. melanogaster”), Dokl. Akad. Nauk
SSSR, 24 : 972-975.

Omura, H., 1964. “А Systematic Study of the Hyoid Bones in the Baleen
Whale,” Sci. Repts. Whales Res. Inst., Tokyo, 18 : 149-170.

Ortov, У. N., 1961. ‘“‘Vozrostnaya izmenchivost’ cherepa kulana (Equus
hemionus РаП.)” [‘‘Growth Variability of the Skull of the Kulan (Equus
hemionus Pall.)”’], Zool. <h., 40 : 592-601.

Озвовм, D. J., 1964. ‘Notes on the Moles of Turkey,” 7. Mammal.,
45: 127-129.

Озвовх, Н. F., 1915. “Origin of Single Characters as Observed in Fossil
and Living Animals and Plants,’’ Amer. Natur., 49 : 193-240.

Ovcuinikova, N. A., 1962. “‘Izmenenie otnositel’nogo уеза nadpoche-
chnikoy и mlekopitayushchikh у eksperimental’nykh usloviyakh”
(‘Change in the Relative Weights of Adrenals of Mammals under Experi-
mental Conditions”), Dokl. П Nauchnot Konf. Molodykh Spetsialistov-
biologov, pp. 79-85. Sverdlovsk.

Ovcuinikova, М. A., 1964. “Biologicheskie osobennosti nominal’nogo i
severnogo podvidov polevki ekonomiki 1 В gibridov’ (“Biological
Characteristics of the Nominal and Northern Subspecies of Field Voles
and Their Hybrids’’), Sb. Vopr. Vnutrividovoi Izmenchivosti 1 Mikroevolutsiya,
pp. 89-90. Sverdlovsk.

Ovcuinikova, №. A., 1966 “О nekotorykh morfologicheskikh izmeneni-
yakh gryzunov У zrelom vozroste” (“Оп Some Morphological Changes
in Adult Rodents”), Trudy Vsesoyuzn. Soveshchnii Vnutrivid Izmenchivost
Nazemnykh Pozvonochnykh Zhivotnykh 1 Mikroevolyutsiya, pp. 199-201.
Sverdlovsk.

Paaver, К. L., 1964. “О vekovoi izmenchivosti subfossilnykh popyulyatsii
mlekopitayushchikh у pribaltike”’ (“Оп the Temporal Variability of
Subfossil Mammal Populations in the Baltic Region”), Bull. Mosk.
Obshch. Ispyt. Рит. (Biol.), 19 : 83-95.

Bibliography 333

Paaver, К. L., 1965. Formirovanie teriofauny 1 izmenchivost? mlekopitayush-
chikh pribaltiki v Golotsene (The Formation of the Mammal Fauna and
Mammalian Variability in the East Baltic Region in the Holocene).
Tartu : Akad. Nauk Estonskoi SSR. 497 pp.

PacxarpD, В. L., 1960. “Speciation and Evolution of the Pigmy Mice,
Genus Baiomys,” Univ. Kansas Publ. Mus. Nat. Hist., 9 : 579-670.

Packarp, В. L., and H. Garner, 1964. ‘“АтБогеа! Nests on the Golden
Mouse in Eastern ‘Texas,” JF. Mammal., 45 : 369-374.

Pautus, М., 1964. “Etude ostéographique et osteométrique sur deux
Grampus griseus С. Cuvier 1812, echoues au brusc (уаг.) en 1887 et
1895,” Bull. Mus. Histoire Nat. Marseille, 24 : 81-122.

Pavuinin, У. N., 1962. “О morphologicheskoi opredelennosti popyulyatsii
khishchnykh mlekopitayushchikh” (“Оп the Morphological Identifica-
tion of Populations of Carnivorous Mammals”), Trudy Isnt. Biol. Uralsk
Fil. Akad. Nauk SSSR, 29 : 93-119.

Pavuinin, У. N., 1963. ТофоРзки зобоР (The Sable of Tobol). Ibid.,
Wolkia4s

Pavuinin, V. N., 1964. ‘“‘Vnutrividovaya izmenchivost’ zhivotnykh 1
okhotniche khozyaistvo” (“Intra-population Variability of Animals
and the Hunting Industry”), Sb. Vopr. Vnutrividovor Izmenchivosti .

t Mikroevolyutsiya, pp. 102-105. Sverdlovsk. :

*Pavuinin, V. N., 1965. “ОБ areale i morfologii lesnykh kunits Tyumenskoi
oblasti” (“Оп the Habitat and Morphology of Wild Martens of the
Tyumensk Region’’), Trudy Inst. Biol., Uralsk Fil. Akad. Nauk SSSR,
38 : 41-52.

Pawuik, M., 1967. ‘Uber die Variabilitat der prabasialen Kyphose bei
der Gattung Cercopithecus,’ Folia Primatol., 5 : 201-212.

Peart, R., 1939. The Natural History of Populations. London : Oxford
Univ. Press.

Pearson, K., and A. С. Davin, 1924. “Оп the Biometric Constants of the
Human Skull,” Biometrika, 16 : 328-363. [Cited from Long (1968)].

Pearson, О. P., 1958. “А Taxonomic Revision of the Rodent Genus
Phyllotis,’ Univ. California Publ. Zool., 56 : 391-496.

Peck, Е. D., and .P. В. Sawin, 1950. “Morphogenetic Studies of the
Rabbit : VIII. Genetic Variations in the Sternum as Determined by
the Interaction of General and of Regionally Specific Growth Factors,”’
п ВАР cools 11 335-357.

Pett, Е. L. von, and P. J. H. Bree, 1962. ‘‘Notizen uber die Waldmaus,
Apodemus sylvaticus (Linnaeus, 1758) von der niederlandischen Insel
Terschelling,” <. Saugetierkunde, 27 : 222-228.

Prsuev, T., and С. Grorcirv, 1961. “Studies in the Taxonomy of Some
Populations of Apodemus sylvaticus (Linnaeus, 1758) in Bulgaria,” Acta
Theriol., 5 : 185-202.

334 Bibliography

Ретвоу, В. M., 1961. “К _ sistematike 1 vozrostnoi izmenchivosti surka
Menzbira (Marmota menzbiert Kasch)”’ [“Оп the Taxonomy and Growth
Variability of Marmot (Marmota menzbieri Kasch)’’], Zool.. Zh., 40:
93-105.

Рниллрз, J. С., 1920. “Skull Measurements in the Northern Virginia
Deer,” 7. Mamm., Г: 130-133.

Prxuarev, С. A., 1941. “Tyuleni Yugo zapadnoi chasti Okhotskogo
morya”’ (‘Seals of Southwest Part of the Okhotsk Sea’), Izv. Tikh
Nauchno-Issled. Inst. Ryb. Khoz. 1 Okeanogr. (Vladivostok), 20 : 68-76. —

Ртоснтмки, М. A., 1960. Bzometriya (Biometrics). Novosibirsk : Izdatel’-
stvo Sibirsk. Otdel. Akd. Nauk SSSR.

Роквоузки, A. V., 1962. “Individual’naya izmenchivost’ skorosti
polovogo sozrevaniya samok stepnoi pestrushki (Lagurus lagurus)”
(“Individual Variability in the Rate of Sexual Maturity in Female
Lagurus lagurus’), Trudy Inst. Biol. Uralsk Fil. Akad. Nauk SSSR, 29 :
023,

Роквоузки, А. V., У. 5. Smrrnov and Ъ. 5. Scowartz, 1962. ‘‘Kolorimetri-
cheskoe izuchenie izmenchivosti okraski gryzunoy у eksperimental’nykh
usloviyakh У svyazi $ problemoi gibridnykh popyulyatssii’’ (““Colorimetric
Study of Variability of Coloration of Rodents under Experimental
Conditions in Connection with the Problem of Hybrid Populations’’),
Maids) 29: 15-26.

РотудАкоу, С. D., 1961. ‘“‘Prisposobitel’noe znachenie izmenchivosti
priznakov 1 svoistv popyulyatsii губ” (““The Adaptive Significance of
Traits and the Nature of Fish Populations’), Trudy Soveshch. Ikhtiol
Kommissti, 13 : 158-172.

Ротухдкоу, С. D., 1962. ‘‘Prisposobitel’naya vzaimosvyaz’ izmenchivosti
popyulyatsii ryb s usloviyami pitaniya”’ (“Adaptive Relations of Vari-
ability of Fish Populations with Feeding Conditions”), Trudy Inst. Morfol.
Khivotnykh im. А. N. Severtsova, 42 : 5-63.

Ponomarey, A. L., 1944. ‘“‘Reaktsiya nekotorikh khishchnykh (Mustelidae)
na gradient temperatury” [“Веасйоп of Some Carnivora (Mustelcdae)
to a Temperature Gradient’’], Zool. <h., 23 : 51-55.

Popov, Г. А., 1959. “‘Osobennosti polovogo dimorfizma tyulenya-
khokhlacha”’ (‘Characteristics of the Sexual Dimorphism of the Hooded
Seal’), Trudy | Konf. Mosk. Morf. Labor., Moscow. .

Popov, L. A., 1961. “Маемау К obshchei morfologii khokhlacha
Grenlandskogo morya’”’ (“Materials on the General Morphology of
the Greenland Seal’’), Trudy Soveshch. Ikhtiol. Котпиззи, 12 : 180-191.

Popov, У. А., 1959. ‘“Materialy po ikologii norki (Mustela vison Br.) 1
rezultaty akklimatizatsi ee у Tatarskoi SSSR”’ [‘‘Data on the Ecology
of Minks (Mustela vison Br.) and the Results of Its Acclimatization in
the Tatar USSR”’], Trudy Kaz. Fil. Akad. Nauk SSSR, 2 : 3-140.

Bibliography 335

Popov, У. A., 1960. Mlekopitayushchie Volzhsko-hamskogo kraya : Nasekomy-
adnie, rukokril’e, gryzuny (Mammals of the Volga-Kamsk Region : Insec-
tivores, Pinnipeds, Rodents). Kazan’ : Izdatelstvo Akad. Nauk SSSR.

Popova, Е. T., 1941. ‘“‘Rezultaty metiztisii kuchugurovskikh ovets
linkol’nami i puti dalneishego napravleniya selektsionnoplemennoi
raboty s metisami”’ (“Results of Crossing Kuchugurov Ewes with Lin-
colns and the Direction of Future Trends in Selection and Breeding
Work with Crossbreeds”), Trudy Mosk. Zootekhn. Inst., 1 : 186-210.

Ротегоу, У. A., 1963. “К kharakteristike shonnokh skoplenii khokhlacha
у Grinlandskom morye’’ (“Оп the Density of the Hooded Seal in the
Greenland Зеа”), Sb. Rabot Sev. Otd. Polyarnogo Inst. Ryb. Khoz. 1 Okeanogr.
(Arkhangelsk), 1962, pp. 53-63.

‚Ротегоу, У. A., and A. У. YasBLoxov, 1966. “Моуое у morfologicheskoi
kharakteristike khokhlacha (Cystophora cristata Erxl.)” [“New Data on
the Morphological Characters of the Hooded Seal (Cystophora cristata
В cools СА 45: 1222-1226.

Рвдкдзн, I., 1964. ‘“‘Cyto-morphology of Adrenal Gland of Indian Desert
Gerbil,” 7. Anat. Soc. India, 13 : 76-79. ;

Praxkasu, [., and С. С. Комвклвмт, 1962. ‘‘Eco-toxicology and Control
of Indian Desert Gerbil, Meriones hurriane (Jerdon) : I. Feeding Behavior,
Energy Requirements, and Selection of Bait,” 7. Bombay Natur. History
Soc., 59 : 800-806.

Prout, T., 1962. “The Effect of Stabilizing Selection on the Time of
Development in’ Drosophila melanogaster,” Genet. Res., 3.

Prout, Т., 1964. ‘‘Observations on Structural Reduction in Evolution,”
Amer. Naturalist, 48 : 239-249.
PSenicKa, P., and J. Jurin, 1964. ‘“‘Pozorovani 100 pripadu utvareni

circulus arteriosus (Willisi) и opice Macaca mulatta,” Ceskosl. Morfol.,
12 : 321-326.

Pucex, Zp., 1956. ‘‘Untersuchungen iiber die Veranderlichkeit des
Schadels im Lebenszyklus von Sorex araneus araneus L.,” Ann. Univ. М.
Curie-Slodowska, Sect. C, 9 : 163-208.

Pucex, Zp., 1962. ‘“The Occurrence of Wormian Bones (Osstcula vormiana)
in Some Mammals,” Acta Theriol., 6 : 1-51.
Pucex, Zp., 1963. ‘Seasonal Changes in the Braincase of Some Re-

presentatives of the Genus Sorex from the Palearctic,” 7. Mammal.,
44 : 523-536.

Pucex, Zp., and С. Млвкоу, 1964. “Зеазопа! Changes in the Skull of
the Common Shrew from Bulgaria,” Acta Theriol., 9 : 363-366.

Рововт, К. G., 1965. “Effect of Sexual Activity on the Adrenal Gland
in the Northern Palm Squirrel, Funambulus pennanti Wroughton,”
Experientia, 19 : 1-4.

Raicu, P., 1960. ‘Die gerichtete Variabilitat und die Anpassungsfatigkeit

336 Bibliography

der Organismen,” Arb. Fragen Evolution, Jena, 1959, pp. 113-120.

Rasmussen, B., 1957. “Exploitation and Protection of the East Greenland
Seal Herds,” Norsk. Hvalfangs-tidende, 46 : 45-59.

Rasmussen, B., 1960. “Om klappmyssbestanden i det nordlige At-
lanterhav,” Fisk. og Hevet., 1: 1-23.

Rasmussen, B., and T. Orirstanp, 1964. “Norwegian Tagging of Harp
Seals and Hooded Seals in North Atlantic Waters,” Fiskdirector Skr.,
1343-595: |

Клозсн, К. L., 1963. ‘Geographic Variation in Size in North American
Brown. Bears, Ursus arctos L., as Indicated by Condylobasal Length,”
Canad. 7. Zool., 41 : 33-45.

Razorenova, A. P., 1952. ‘‘Vozrostnaya izmenchivost’ u krsnokhvostnoi
polevki (Clethrionomys)” [“Асе Variability of the Red-tailed Vole
(Clethrionomys)”?], Bull. Mosk. Obshch. Ispyt. Prir. (Biol.), 57: 23-28.

Reapine, А. J., 1966. ‘‘Effect of Maternal Environment on the Behavior
of Inbred Mice,”’ 7. Compar. and Physiol. Psychol. , 62 : 437-440.

Reep, 5. С., 1936. “Нагейр in the House Mouse : I. Effect of the External
and Internal Environments,” Genetics, 21 : 339-360.

Reini8, St., 1964. “‘Kvantitativni analisa morfologichych zmen у mozkove
kure у prubehu ontogeneza krysy,”’ Ceskosl. Morfol., 12 : 283-291.

Remane, A., 1960. “Die Beziehungen zwischen Phylogenie und Onto-
genie,” ool. Anz., 164 : 306-337.

Remang, A., 1961. “‘Gedanken zum Problem : Homologie und Analogie,
Pre-adaptation und Parallelitat,’ <ool. Anz., 166 : 447-465.

Rosertson, Е. W., and Е. С. К. Reeve, 1952. “Heterozygosity, Environ-
mental Variation, and Heterosis,’’ Nature, 170 : 286.

Rosson, С. 5., and О. W. Ricuarps, 1936. The Variation of Animals in
Nature. New York : Longman and Green.

Rocinsku, YA. Ya., 1954. “‘Velichina izmenchivosti izmeritel’nykh
priznakov cherepa 1 nekotorie zakonomernosti ikh korrelyatsii u
cheloveka”’ (‘““Magnitude of Variability of Measurable Traits of the
Skull and Some Regularities of Their Correlations in Man’’), Uch.
Zap. Moskovskit Gos. Univ., No. 16.

Rocinsku, Ya. УА., 1959. “О Nekotorykh rezul’tatakh primeneniya
kolichestvennogo metoda К izucheniyu morfologicheskoi izmenchivosti”’
(“Some Results of Using the Quantitative Method to Study Morphologi-
cal Variability”), Arkhiv Anat. Gistol. Embriol., 36 : 83-89.

Косихзки, YA. Ya., 1960. “‘Nekotorye zakony izmenchivosti i korelyatsii
izmeritel’nykh priznakov u cheloveka i grugikh mlekopitayushchikh”
(“Some Rules of Variability and Correlation of Measurable Traits
of Man and Other Mammals’), Antropologicheskie Dokl. na VI Mezhdunor.
Kongr. Antropol. 1 Etnogr. Nauk, рр. 23-28. Izdatelstvo Moskovskii Gos.
Univ. i

Bibliography 337

Roxirsku, P. F., 1961. Osnovy variatsionnoi statistiki dlya biologov (Funda-
mentals of Variation Statistics for Biologists). Minsk: Izdatelstvo
Belorus. Gos. Univ.

Вокитзки, Р. F., 1964. “К voprosu о matematicheskikh metodakh analiza
vnutrividovykh razlichii” (“On the Question of Mathematical Methods
of Analysis of Intra-species Variation”), Materialy Soveshch. Sovrem. Probl.
Izucheniya Dinamtki Chislennosti, рр. 86-87.

Вомдзноу, О. D., 1931. “Ob usloviyakh ‘ravnovesiya’ у popyulyatsii”
(“On the Conditions of ‘Balance’ in a Population”), <h. Eksperim.
Biol., 8 : 442-454.

Romasuoy, О. D., and Е. О. мА, 1943. “ОБ aberrativnoi izmenchivosti
у popyulyatsiyakh pyshnykh zverei” (“On Aberrant Variability in
Populations of Fur Animals”), Zh. Obshch. Biol., 4 : 286-312.

Romer, A. S., 1960. Phylogeny and Morphology. Lecture, XIII Annual
Meeting Soc. Study of Evolution. |

Roos, Т. B., and В. М. SHackELForD, 1955. ‘“‘Some Observations on the
Gross Anatomy of the Genital System and Two Endocrine Glands
and Body Weights in the Chinchilla,” Anat. Rec., 123 : 301-311.

Rossotimo, О. L., 1962. “О vnutrividovoi izmenchivosti krasnoi polevki
(Clethrionomys rutilus Pall.)” [‘‘On the Intra-species Variability of фе.
Red Vole (Clethrionomys rutilus Pall.)”], Zool. Zh., 41: 443-452.

Russet, W. М. S., 1958. ‘‘Evolutionary Concepts in Behavioral Science,”
General Systems Yearbook, 3 : 18-28.

Samnt Girons, М. С., and P. J. H. van Bree, 1962. ‘Recherches sur la
systématique de Apodemus sylvaticus (Linnaeus, 1758) en Afrique du
М га,” Mammalia, 26: 478-488.

SarycHEva, Т. G., 1948. “Оруё primenenia graficheskovo metoda у
izuchenii izmenchivosti produktid’ (“An Experiment on the Use of
Graphical Methods in the Study of Variability of Productids”), 125.
Akad. Nauk SSSR, Ser. Biol., 1948 (2): 205-212.

Sauer, М. E., and Cu. Т. Ruse, 1946. ‘The Number of Nerve Cells in
Mysenteric and Submucous Plexuses of the Small Intestine of the Cat,”
Anat. Rec., 96 : 373-382.

Sawin, Р. B., 1937. ‘‘Preliminary Studies of Hereditary Variations in the
Axial Skeleton on the Rabbit,” Anat. Rev., 69 : 407-428.

Sawin, P. B., 1945. ‘Morphogenetic Studies of the Rabbit : I. Regional
Specificity of Hereditary Factors Affecting Homeotic Variations in
the Axial Skeleton,” 7. Exper. Zool., 100 : 301-329.

ScuerFer, У. B., 1958. Seals, Sea Lions and Walruses. Oxford: Oxford
University Press.

Scuerrer, V. B., 1962. ‘‘Pelage and Surface Topography of the Northern
Fur Seal,” North Amer. Fauna, 64 : 1-206.

Scuerrer, V. B., 1964. “Наш Patterns in Seals (Pinnipedia),” 7. Morphol.,

338 Bibliography

115: 291-8303.

SCHMALHAUSEN, I. I. (SHMAL’GAUZEN), 1935. ‘‘Opredelenie osnovnykh
ponyatii i metodika issledovaniya rosta’’ (“Determination of Basic
Concepts and Methods of Investigations of Growth’’), Sb. Rost zhivotnykh,
pp. 8-60. Moscow, Leningrad : Biomedgiz.

ScHMALHAUSEN, I. JI. (SHMAL’GAUZEN), 1939. Puti i zakonomernostz
evolyutsionnogo protsessa (Paths and Regularities of Evolutionary Processes).
Moscow, Leningrad : Izdatelstvo Akad. Nauk SSSR.

SCHMALHAUSEN, I. I. (SHMAL’GAUZEN), 1940. “‘Izmenchivost’ 1 smena
adaptivnykh norm У protsesse evolyutsii” (“Variability and Changes of
Adaptive Norms in the Process of Evolution”), Zh. Obshch. Biol., | : 9-24.

ScHMALHAUSEN, I. I. (SHMAL’GAUZEN), 1943. ‘“Temp evolyutsii 1 faktory
ego opredelyayushchie” (“Ге Rate of Evolution and Factors Deter-
mining It”), <h. Obshch. Biol., 4 : 253-285.

ScHMALHAUSEN, I. I. (SHMAL’GAUZEN), 1946. Faktory evolyutsu (Factors of
Evolution). Moscow, Leningrad: Izdatelstvo Akad. Nauk SSSR.
(Translated 1949, Philadelphia : Blakiston).

SCHMALHAUSEN, I. I. (SHMAL’GAUZEN), 1947. Osnovy sravnitel’not anatomit
(Fundamentals of Comparative Anatomy). Moscow : Izdatelstvo
‘Sov. Nauka’. |

SCHMALHAUSEN, I. I. (SHMAL’GAUZEN), 1958. “‘Kontrol’ i regulyatsiya у
evolyutsii’? (‘Control and Regulation in Evolution”), Bull. Mosk.
Obshch. Ispyt. Prir. (Biol.) 13 (5) : 93-121.

SCcHMALHAUSEN, I. I. (SHMAL’GAUZEN), 1959. “‘Perspektivy primeneniya
tochnykh metodov dlya izucheniya faktorov evolyutsii” (‘Prospects
for the Use of Exact Methods for the Study of Evolutional Factors’’),
Vestnik Leningradsku Gos. Univ., 1959 (9) : 108-118.

ScHMALHAUSEN, I. I. (SHMAL’GAUZEN), 1965. “Evolyutsiya у svete
kibernetiki” (“Evolution in the Light of Cybernetics”), Problemy
Kibernetiki, 13 : 195-199.

Зенмшгу, О. ]., 1971. “‘Population Variation in Dipodomys ordu from Wes-
tern Texas,” 7. Mammal., 52: 108-120.

Scuuttz, A. H., 1917. ‘Variations in Man and Their Evolutionary
Significance,” Amer. Naturalist, 60 : 297-323.

Scuutz, A. H., 1926. ‘‘Studies on the Variability of Platyrrhine Monkeys,”
F. Morphol., 7 : 286-304.

ScuuLttz, A. H., 1958. “Cranial and Dental Variability in Colobus
Monkeys,”’ Proc. Zoal. Soc. London, 130 : 79-105.

ScHuMACHER, С. Н., Е. Wotrr, and Е. Jurzi, 1965. “Quantitative
Untersuchungen tber das postnatale Organwachstum des Goldhamsters
(Mesocricetus auratus Wrth),” Morphol. fb. : 1. Korpergewicht-Herz,
107 : 550-567. II. Lunge-Leber-Milz, 108: 18-40. ПТ. Thymus-
Schilddriise, 108: 41-66. ТУ. Besprechung der Ergebnisse, 108: 123-138.

Bibliography 339

Scuwartz, S. 5. (SHwartz), 1959. ‘“МеКоогуе voprosy problemy vida
и nazemnykh pozvonochnykh zhivotnykh” (‘Some Questions on the
Species Problem in Terrestrial Vertebrate Animals”), Trudy Inst. Biol.
Uralsk Fil. Akad. Nauk SSSR, Vol. 11.

ScHwartz, 5. 5. (SHwartTz), 1959. ‘‘Nekotorye zakonomernosti ekologi-
cheskoi obuslovlennosti interernykh osobennostei nazemnykh pozvo-
nochnykh zhivotnykh” (“Some Regularities of Ecological Relationships
of Inner Characters of Terrestrial Vertebrate Animals’), Ibid., 14:
113-177. ;

ScHwartz, 5. 5. (SHwaRtTz), 1962. ‘“‘Izuchenie korrelyatsii morfologiches-
kikh osobennostei gryzunov so skorostyu ikh rosta у svyazi $ nekotorymi
voprosami vnutrividovoi sistematiki” (“Study of the Correlation of
Morphological Characters of Rodents with Their Rate of Growth in
View of Some Questions of Intra-species Taxonomy’’), Ibid., 29 : 5-14.

Scuwartz, 5. S. (SHwARTZ), 1962a. ‘‘Morfologicheskie i ekologicheskie
osobennosti zemlerosek na krainem severnom predele ikh rasprostrane-
niya” (“Morphological and Ecological Peculiarities of Shrews on the
Extreme Northern Borders of Their Distribution”), Ibid., 29 : 45-51.

Scuwartz, 5. 5. (SuHwartz), 1963. ‘“Vnutrividovaya izmenchivost’
mlekopitayushchikh i metody ee izucheniya”’ (‘‘Intra-species Variability
of Mammals and Methods of Its Study”’), Zool. Zh., 42 : 417-433.

Scuwartz, 9. 5. (SHwarrz), 1964. ‘‘Ekspiremental’nye metody
issledovaniya \У teoreticheskoi sistematike” (“Experimental Methods
of Investigation in Theoretical Taxonomy’), Sb. Vopr. Vnutrividovor
Izmenchivosti, pp. 146-152. Sverdlovsk.

Scuwartz, S. S. (SHwartz), 1965. ‘“Uchenie о mikroevolyutsii 1
teoreticheskie voprosy sistematiki ptits’” (“The Study of Micro-evolution
and Theoretical Problems of the Taxonomy of Birds”), Sb. Sovremennye
Problemy Ornitologii, рр. 65-86. Frunze, Izdatelstvo “Пит”.

ScHwartz, S. S. (SHwarrz), К. I. Kopein and A. У. Роквоузки, 1960.
“Sravnitel’noe izuchenie nekotorykh biologicheskikh osobennostei
polevok Microtus gregalis gregalis Pall., М. g. major Ogn. 1 ikh pomesei”
[Comparative Study of Some Biological Characteristics of Field Voles
(Microtus gregalis gregalis Pall., М. g. major Ogn) and Their Crosses” |,
Moule ie 3911912956.

Scuwartz, S. S. (SHwartz), А. У. Роквоузкт, У. С. [5тенемко, У. С.
ОтЕмуеу, N. А. OvtscHinnicova, and О. А. Р]АзтогеЕУА, 1964.
“Biological Peculiarities of Seasonal Generations of Rodents, with Special
Reference to the Problem of Senescence in Mammals,” Acta Theriol.,
8 : 11-43. |

Scorkowskl, Ер., 1958. “Die dreiartige Variabilitat im der Natur,”
Arch. Tierzucht, 1 : 283-293.

SEALANDER, J. A., 1961. ‘Hematological Values in Deer Mice in Relation

340 Bibliography —

to Botfly Infection,” 7. Mammal., 44 : 57-60.

SEARLE, А. G., 1954. “Сепейса| Studies on the Skeleton of the Mouse :
IX. Causes of Skeletal Variation within Pure Lines,” 7. Genet., 52 :
68-102.

SEARLE, А. G., 1954a. ‘‘Genetical Studies on the Skeleton of the Mouse :
XI. The Influence of Diet Variation within Pure Lines,” 7. Genet., 52 :
413-424.

SEGAL’, А. N., 1962. “Opyt perevoda i akklimatizatsii у Karelii
tundrovykh olenei iz murmanskoi oblasti’”’ (““Experiment in Rearing and
Acclimatization of Deer in the Tundra of the Murmansk Region’’),
Severny: olen о Karel’skoi ASSR, pp. 58-80. Moscow : Izdatelstvo Akad.
Nauk SSSR. |

SEKHLYANU, V., 1960. “О Primenenii matematiki у biologii i meditsine”’
(“On the Use of Mathematics in Biology and Medicine”’), Vopr. filosofit,
1960 (10) : 100-109.

SERAFINSKI, W., 1965. “The Sub-specific Differentiation of the Central
European House Mouse (Mus musculus L.) in the Light of Their Ecology
and Morphology,” Ecolog. Polska, Ser. A, 13 : 304-335.

SERGEANT, D. E., 1963. “Нагр Seals and the Sealing Industry,’ Canad.
Audubon, 25 (2) : 29-35.

ЗЕМЕВТРОЕЕ, A. N., 1931 (1927). “Uber die Beziehungen zwischen der
Ontogenese und Phylogenese der Tiere,” Jen. Ztschr., 63 : 51-180.

Severtsov, A. N., 1912 (1945). ‘‘Etyudi po teorii evolyutsii”’ (“Studies
on the Theory of Evolution’’), Sobr. Soch., Vol. III, рр. 19-216. Moscow,
Leningrad : Izdatelstvo Akad. Nauk SSSR.

Severtsov, A: N., 1914. Soveremennye zadachi evolyutsionnor teoriu
(Current Problems of Evolutionary Theory). Moscow : Izdatelstvo
Akad. Nauk SSSR.

Severtsov, А. N., 1939. Morfologicheskie zakonomernosti evolyutsi: (Morpho-
logical Regularities of Evolution). Moscow, Leningrad : Izdatelstvo

‚ Akad. Nauk SSSR.

SHARMAN, G. B., H. J. Евтн, and J. Н. САглву, 1964. “Growth of the
Pouch, the Young, Tooth Eruption, and Age Determination in the Red
Kangaroo, Megaleia rufa,’ CSIRO Wildlife Res., 9 : 20-49.

SHaw, А. M., 1929. “Variations in the Skeletal Structure of the Pig,”
Sci. Agric., 10 : 23-27.

SHNOLL’, 5. E., 1964 (Ed.). Buologicheskie chasy. Predislovie К russkomu
izdaniyu (Biological Clocks. Introduction to the Russian Edition),
pp. 5-10. Izdatelstvo ‘Mir’.

Super, G. Г., and У. К. Гудмко, 1940. “Раппуе ob izmenchivosti geologi-
cheski, drevnykh i molodykh form nasekomykh” (“Data on Variability
of Geologically Old and Young Forms of Insects’), <h. Obshch. Biol.,
LQ):

Bibliography 341

Suustov, A. Р., and A. У. УАвгокоу, 1967. “Sravnitel’no-morfologi-
cheskaya kharakteristika grenlandskogo i polosatogo tyulenei (materialy
К popyulyatsionnomu analizu lastonogikh)”’ [“‘Comparative Morphologi-
cal Characteristics of the Greenland and Striped Seals (Materials on the
Population Analysis of Seals)”’], Trudy Polyar. Inst. Ryb. Khoz. 1 Okeanogr.,
21 : 51-59.

Smporkin, У. I., and М. У. Микнагку, 1964. “Kolichestvennoe Sootno-
shenie belkovykh fraktsii syvorotki krovi и dvukh vidov serykh polevok”’
(“Quantitative Correlation of Protein Fractions of Blood Plasma in
Two Species of Gray Field Voles’’), Sb. Voprosy Vnutrividovor Izmenchivosti
t Mikroevolyutsiya, pp. 119-120. Sverdlovsk.

Sicmunp, L., 1964. “Ваайуе Wachstum und intraspezifische Allometrie
der Grossmausohr (Myotis myotis Borkh.),” Acta Univ. Carolinae Buol.,
3 : 235-303.

SIKORSKA-PivovsKA, Z., 1965. ‘““The Variability of Structure of the
Lacrimal Bone in the Lacrimal Sac Fossa of Primates,” ool. Polon.,
15 : 147-158.

ЭТМРЗОМ, С. G., 1944. Tempo and Mode т Evolution. New York : Columbia
Univ. Press.

Simpson, С. G., 1948. Tempy а formy evolyutsii (Tempo and Mode in Evolu-
tion). Moscow : Izdatelstvo Inostrannoi Literatury.

Simpson, Ск. G., 1951. ‘“The Role of the Individual in Evolution,” 7. Wash.
Acad. Sci., 31 : 1-20.

Simpson, С. G., 1953. The Major Features of Evolution. New York:
Columbia Univ. Press.

Smmpson, (С. G., and А. Rog, 1939. Quantitative Zoology. New York :
McGraw-Hill.

SINSKAYA, Е. N., 1948. Dinamika vida (Dynamics of the Species). Moscow :
Sel’khozgiz.

SKAREN, Оотемт, 1964. “Variation of Two Shrews, Sorex unguiculatus
Dobson and 5. a. araneus, Г.” Ann. Zool. Fennica, 1 : 94-124.

Skoczen, Sr., 1962. “Аое Structure of Skulls of the Mole, Talpa europaea
(Linnaeus, 1758) from the Food of the Buzzard (Buteo buteo),” Acta
Theriol., 6 : 1-9.

Skopakorr, Cristo von, 1965. ‘Uber die Variabilitat der Abzweigung
der A. subclavia und ihrer Hauptaste,” Anat. Anz., 116 : 73-91.

Sutentsov, М. M., 1939. “О rudimentakh zadnykh lastov и chornomors-
Кого del’fina D. delphis” [On the Rudimentation of the Posterior
Flippers of the Black Sea Dolphin (D. delphis)’’], Zool. Zh., 18 : 361-366.

Stentsov, М. M., 1955. Kitoobraznye dal’nevostochnykh more: (Cetaceans
of the Far East Seas). Vladivostok.

Stentsov, М. M., 1961. ‘‘Rezul’taty vzveshivaniya krupnykh i melkikh
kitoobraznykh dobivaemykh na Dal’nem Vostoke”’ (‘“Weights of Large

342 Bibliography

and Small Cetaceans Found in the Far East’’), Trudy Inst. Morfol. Zhivot.
Akad. Nauk SSSR.

SLYPER, Е. J., 1962. Whales. London: Hutchinson Publ.

Smirnov, N. А., 1927. “Diagnostical Remarks about Some Seals (Phocidae)
of the Northern Hemisphere,” Troms Mus. Aarb., 48 : 1-23.

SmirNOoV, N. A., 1927. “‘Issledovanie nad belomorskim. tyulenyem.
Noveishie nablyudeniya nad belomorskim lysunom’’ (Studies on White
Sea Seals : Recent Observations on White Sea Seals’’), Zzv. Otd. Prikl.
Ikhtiol., 6: 5-27.

SMIRNOV, М. A., 1935. ‘“‘Morskie zveri arkhticheskikh уогег” (‘Sea
Animals in the Arctic Seas’), vert Arktiki, Leningrad : Izdatelstvo
‘“Glavsevmorput’.

омтвмох, V.S., 1962. “К taksonomicheskoi kharakteristike pestsa Yamala
1 Стетапаи?” (“Оп Taxonomic Characters of the Arctic Fox from Yamel
and Greenland”), Trudy Inst. Biol., Uralsk. Fil. Akad. Nauk USSR, 29 :
71-80.

Sokal, К. R., 1962. “Some Stages in the Development of the Concept
of Natural Selection,” Univ. Kansas Sct. ВиЦ., 42 (Suppl.) : 129-151.

Soka, К. R., and P. H. A. SNeatH, 1963. Principles of Numerical Taxonomy.
San Francisco : Freeman.

SoxoLoy, Е. A., 1941. ‘“‘Emkost’ pishchevaritel’nogo trakta 1 pokazateli
perevarivaemosti и pushnykh zverei’’ (“Volume of the Digestive Tract
and the Digestive Indices in Fur Animals’), Trudy Mosk. Xootekhn.
Inst., |: 72-83.

Soxo.ov, I. I., 1960. Morfologiya porod domashnykh ovets (Morphology of
Domestic Sheep Breeds). Moscow, Leningrad: Izdatelstvo Akad.
Nauk SSSR.

Soxo.ov, I. I., and V. L. RasHex, 1961. ‘‘Razvitie zubov i cherepa kak
pokazatel’ vozrosta и saigi (Saiga tatarica)”’ (“The Development of Teeth
and Skull as Growth Parameters in Sazga tatarica’’), Bull. Moskov. Obshch.
Ispyt. Prir. (Biol.), 66 (6) : 77-98.

SPARKES, К. S., and О. T. АваАкАкт, 1967. ‘“‘Intra-subspecific and Inter-
subspecific Chromosomal Polymorphism in Peromyscus maniculatus
(Deer Mouse), Cytogenetics, 5 : 411-418.

Spitz, F., 1963. “Croissance en élévage de la longeur. téte et corps chez
le campagnol des champs (Microtus arvalis Pallas) de San-Michel-en-
р Herm (Vendée), Mammalia, 27 : 111-118.

ЗтЕвви\$, С. L., 1950. Variation and Evolution in Plants. New York :
Columbia Univ. Press.

STECHER, К. M., 1961. ‘‘Numerical Variation in the Vertebrae of the
Prjevalesky Horse,’ Mammalia, 25 : 192-194.

Stein, С. H. W., 1963. ‘“‘Anomalien der Zahnzaht und ihre geographische
Variabilitat bei Insectivora, I. Maulwurf, Talpa europaea L.,” Mitt. Zool.

Bibliography 343

Mus. Berlin, 39 : 223-244.

Stockuaus, K., 1965. ‘‘Metrische Untersuchungen an Schadéln von
Wolfen und Hungen,” Хей. Zool. Syst. Evolutionsforschung, 3: 157-258.
[Cited from Long (1968) ].

Straka, F., 1963. “Ха nyakoi osobennosti у morfologiyata na lalugera
(Citellus citellus) у Bulgariya” (“Оп the Morphological Characteristics
of Bulgarian Citellus citellus’), Priroda, 12 (4) : 100-101.

SrrRanbskov, H. H., 1942. “Skeletal Variations in Guinea Pigs and Their
Inheritance,” 7. Mammal., 23 : 65-75. [Cited from Long (1968)].
Straus, W. L., Jr., 1962. “‘The Mylohyoid Groove in Primates,” Bzblioth.

Primatol., 1 : 197-216.

Strocanovy, 5. U., 1948. ‘‘Sistematika krotovykh (Talpidae)” (“Taxonomy
of Talpidae”), Trudy Zool. Inst. Akad. Nauk SSSR, 8 : 286-405.

StucREN, B., and М. Popovicu, 1961. ‘“‘Analiz izmenchivosti vneshnykh
priznakov zherlyanok Rumenii”’ (“Analysis of Variability of External
Traits of Zherlyanika in Rumania’’), Zool. <h., 41 : 569-576.

StuRTEVANT, A., 1961. Genetics and Evolution. San Francisco : Freeman.

SturTEVANT, A., and Tu. Doszansxy, 1936. “Geographical Distribution
and Cytology of ‘Sex Ratio’ in Drosophila pseudoobscura,” Genetics, 21 :
475-510.

Suryosui, Akira, 1964. ‘‘Arteries of the Gluteal Region in Macacus
cyclopsis,’ Acta. Med. Nagasakiensis, 8 : 49-62.

Sumner, Е. B., 1909. “Зоше Effects of External Conditions upon the White
Mouse, 7. Exper. 200. 7: 97-155.

Sumner, Е. B., 1915. “ботае Studies of Environmental Influence, Heredity,
Correlation and Growth in the White Mouse,” 7. Exper. Xool., 18 :
325-432.

Sumner, Е. B., 1920. ‘‘Geographic Variation and Mendelian Inheritance,”’
j. Exper. Zool., 30 : 369-402.

Sumner, Е. B., 1921. ‘‘Desert and Lava-dwelling Mice and the Problem
of Protective Coloration in Mammals,” 7. Mammal., 2 : 75-86.

Sumner, F. B., 1923. “‘Results of Experiments in Hybridizing Sub-species
of Peromyscus,’ F. Exper. Zool., 38 : 245-292.

Sumner, Е. В., 1924. “The Stability of Sub-species under Changed
Conditions of Environment,’ Amer. Naturalist, 58 : 475-481.

Sumner, Е. B., 1926. “Ап Analysis of Geographic Variation in Mice
of the Peromyscus polionotus Group from Florida and Alabama,” 7.
Mammal., 7 : 149-184.

Sumner, F. B., 1927. “Linear and Colorimetric Measurements of Small
Mammals,” 7. Mammal., 8 : 177-206.
Sumner, Е. В., 1928. ‘‘Observations on the Inheritance of a Multi-factor

Color Variation in White-footed Mice (Peromyscus), Amer. Naturalist,
62 : 193-206.

344 Bibliography

Sumner, Е. B., 1932. “Сепейс, Distributional and Evolutionary Studies
of the Sub-species of Deer Mice (Peromyscus) ,” Bibl. Genet., 9 : 1-106.
Suroviak, J., 1965. “Influence of UV Radiation on the Post-embryonal
Development of the Testes of the White Mouse (Mus musculus Г.)

Vest. Ceskosl. Spolec. Zool., 29 : 255-281.

`Сулмотев, О. K., J. A. Gavan, and W. М. Turner, 1963. “Molar Tooth
Size Variability in African Monkeys,” Human Biol., 35 : 104-122.

Taxayosu, Ino, and Sapao [знтсАкт, 1966. “Comparison of Hemoglobin
Quantities in Two Inbred Strains of Mice and Reciprocal Hybrids,”
Nat. Inst. Animal Health Quart., 6 : 101-108.

Тлгвот, Lez M., and J. 5. С. McCutticu, 1965. “Weight Estimations
for East African Mammals from Body Measurements,” 7. Wildlife
Manag., 29 : 84-89.

Tararinova, О. M., 1964. “Osobennosti okraski mekha trekh podvidov
stepnoi pestrushki” (“Characters of the Color of the Fur of Three
Sub-species of Gerbil”), Sb. Voprosy Vnutrividovot Izmenchivosti 1 Mikroevo-
lyutsiya, pp. 125-126. Sverdlovsk. |

Trerenv’ey, P. V., 1957. “О primenimosti ponyatiya ‘podvid’ у izuchenti
vnutrividovoi izmenchivosti” (“Ор the Use of the Concept of ‘Sub-
species’ in the Study of Intra-species Variability”), Vestnik Leningradsku
Gos. Univ. 1957 (21) [Ser. Biol. 1957 (4)] : 78-79.

TERENT’EV, Р. V., 1959. “‘Metod korrelyatsionnykh pleyad” (“The Method
of Correlation Pleiades’’), Vestnik Leningradski Gos. Univ. 1959 (9):
137-141. ;

Tuorpe, W. H., 1940. “Ecology and the Future of Systematics,’ in
The New Systematics (J. S. Huxley, ed.), pp. 341-364. Oxford : Clarendon
Press.

Тткномтвох, Е. A., 1961. “‘Raspredelenie i migratsii tyulenei у vodakh
Dal’nego vostoka” (“Distribution and Migration of Seals in the Waters
of the Far East USSR”’), Trudy Soveshch. Ikhtiol. Kommissii, 2 : 204-207.

TimirvazeEv, К. A., 1937. Darvinizm 1 selektsiya (Darwinism and Selection).
Moscow : Sel’khozgiz.

TimoreEv-Ressovsku, М. У. (TimMoFEEFF-REssovsky), 1927. “‘Genetische
Analyse einer freilebenden Drosophila melanogaster Population,” Roux.
Arch. Entwicklungsmech. Organsmen, 109 : 70-109.

TimoreEev-Ressovski, М. У. (TtMoFEEFF-Ressovsxy), 1939. “Genetik
und Evolutionforschung,” Verh. Deutsch. Zool. Ges., 41 : 157-169.

TimoreEv-Ressovsku, М. У. (TimoFreeFF-REssovsky), 1958. ‘‘“Mikroevoly-
utsiya, elementarnye yavleniya, material 1 faktory mikroevolyutsionnovo
protsessa”’ (‘“Micro-evolution, Elementary Phenomena, Materials,
and Factors of Micro-evolutional Process’), Bot. Zh., 43 : 317-336.

TIMOFEEV-__--sovsku, М. У. (TIMoFEEFF-REssovsky.), 1965а. “О mikro i
makrofilogenti и рооууКБ _ perekestnooplodotvoryayushchikhsya :

Bibliography 345

organizmov” (“On Micro- and Macro-phylogeny in Cross-fertilizing
Organisms”), Trudy Inst. Biol., Uralsk Fil. Akad. Nauk SSSR, 44 : 5-10.

TimoreEy-Ressovsku, М. У. (Timoreerr-Ressovsky), 1965b. “КЮ teorii
vida” (“Оп the Theory of Species’), Ibid., 44 : 11-26.

TimorgEEy-Ressovsku, М. V. (Timoreerr-Ressovsky), and Yu. М.
SVIREZHEV, 1966. “Ob adaptatsionnom polimorfizme у popyulyat-
siyakh Адайа bipunctata L.” (“On the Adaptive Polymorphism ‘т
Populations of Адайа bipunctata L.’’), Problemy Kibernetiki, 16 : 137-146.

Томилм, A. G., 1954. “Prisposobitel’nye Иру otryada kitoobraznykh”
(“Adaptive Types of the Order Cetacea’’), Zool. Zh., 33 : 667-692.

Tomitin, A. G., 1957. Kitoobraznye zveri SSSR (Cetaceans of the USSR).
Vol. IX. Moscow : Izdatelstvo Akad. Nauk SSSR.

Твтве, О. E., and L. Prez, 1963. “Воду Composition of the Kangaroo
(Macropus зр.).” Austral. 7. Zool., 11 : 273-289.

Твоуев, J. R., 1965. ‘‘Neurosecretion in the Hibernating Bat,’ Anat.
cers Mole 77-9.

Твуовев, Г. F., 1937. “Izmenchivost’ cherepa chernomorskogo delfina”
(“Variability of the Skull of Black Sea Dolphin”), Bull. Moskov. Obshch.
Ispyt. Prir. (Biol.) 46 : 17-24.

Tsatxin, V. I., 1956. “Izmenchivost’ i sistematika turov zapadnogo
Kavkaza (Caprinae, Artiodactyla)” [‘“Variability and Taxonomy of the
tur (Caprinae, Artiodactyla) of the West Caucasus’’], Ibid., 60 : 17-33.

Тзатких, У. I., 1960. “‘Izmenchivost’ metapodii i ee znachenie Фуа
izucheniya krupnogo rogatogo skota drevnosti” (‘‘Variability of Meta-
podia and Its Significance in the Study of Ancient Cattle’’), Ibid., 45 :
109-126.

Tu Cun, 1963. “А. Femoralis and Its Branches in Macacus cyclopsis,”
Acta Med. WNagasakiensis, 7 : 87-114.

Tumurzuay, M., 1964. Sravnitel’noe izuchenie gistologii kozhi i razvitiya
volosyannykh follikullov u mongolskikh ovets (A Comparative Study of
the Histology of Skin and the Development of Hair Follicles in Mongo-
lian Sheep). Avtoref. Diss. na Soisk. Осй. St. Капа. Biol. Nauk., Unst.
Morf. Zhivot., Akad. Nauk SSSR, Moscow.

Ursin, E., 1956. “Geographical Variation in Apodemus sylvaticus and
Apodemus flavicollis (Rodentia, Muridae) in Europe, with Special Reference
to Danish and Latvian Populations,” Biol. Skr. Dansk. Vidensk. Selsk.,
8 (4) : 1-46.: '

Озндкоу, В. P., 1959а. ‘“‘Teploustoichivost’ tkanei-vidovoi priznak
poikilotermnykh zhivotnikh” (“Неа Resistance of Tissues—A Charac-
teristic of Poikilothermal Animals’’), Zool. Zh., 38 : 1292-1301.

ОзнАкоу, В. Р., 1959b. “Fiziologiya kletki i problema vida у zoologii’

(“Cell Physiology and the Problem of Species in Zoology’’), Tsitologiya,
| = 541-545.

346 Bibliography

Van VALEN, L., 1962. “А Study of Fluctuating Asymmetry,” Evolution,
IG) SAB alert

Van VALEN, L., 1962a. ‘Growth Fields in the Dentition of Peromyscus,”
Evolution, 16 : 272-277.

Van VALEN, L., 1963. “Selection in Natural Populations: Human
Fingerprints,’ Nature, 200 : 1237-1238.

VERHEYEN, К., 1969. ‘Adaptation to Environment,” Gerfaut, 49 : 235-237.

Vinocrapovy, A. P., 1964. “Provinces biogéochimiques et leur role dans
Pévolution organique,” Advances Organ. Geochem., pp. 317-337. Paris :
Pergamon Press.

Vorrio, PAAvo, 1962. “Color Variation in the Marten (Martes martes),”
Ann. Soc. Kool. Vanamo, 24 (6) : 1-14.

Vorontsov, М. N., 1958. “Znachenie izucheniya khromosomnykh
naborov dlya sistematiki mlekopitayushchikh”’ (“The Significance of
the Study of Chromosomal Units for the Taxonomy of Mammals’’),
Bull. Moskov. Obshch. Ispyt. Prir. (Biol.) 63 (2) : 5-36.

Vorontsov, М. N., 1961. “‘Neravnomernost’ tempov preobrazovaniya
organov pishchevaritel’noi sistemy и gryzunov i printsip kompensatsii
funktsii’’ (““Non-uniformity of the Rate of Formation of the Organs of
the Digestive System in Rodents and the Principle of Compensation
of Function’), Dokl. Akad. Nauk SSSR, 135 : 1494-1497.

Vorontsov, М. N., 1963. ‘‘Neravnomernost’ tempov preobrazovaniya
огоапоу 1 printsip kompensatsii funktsii’’ (““Non-uniformity in the Rate
of Reformation of Organs and the Principle of Compensation of
Functions’), Zool. Zh., 42 : 1289-1305.

Wappincton, С. H., 1942. “Canalization of Development and the
Inheritance of Acquired Characters,” Nature, 50 : 563-565.

WasILewsk1, W., 1952. “Badania nad morfologia Clethrionomys glareolus
glareolus Schreb.,’? Ann. Univ. М. Curie-Slodowska, Sect. С, 7 : 119-212.

М/Азтемзкь W., 1956. “Untersuchungen itiber die morphologische
Veranderlichkeit der Erdmaus Microtus agrestis Linne),”’ Ibid., 9:
261-305.

WasiLewski, W., 1956a. ‘‘Untersuchungen tiber die Veranderlichkeit des
Microtus oeconomus Pall. in Bialowieza,”’ Ibid., 9 : 355-384.
Wess, D. S., 1965. ‘““The Osteology of Camelops.” Bull. Los Angeles Co.

Mus. Sci., 1 : 1-54.

Wenk, P., 1964. “Die topographische Anatomie des lymphatischen
Systems der Baumwollratte Sygmodon hispidus Say et Ord.,” <. Versuchs-
tierkunde, 4 : 80-97.

Wuitaker, J. O., ]В., 1963. “А Study of the Meadow Jumping Mouse,
Xapus hudsonius (Zimmerman) in Central New York,” Ecol. Monographs,
33 : 215-254. [Cited from Long (1968) ].

WuitTAKeER, H. M., and М. J. Оеглму, 1969. “‘Variation in the Skull

Bibliography 347

of the Long-tailed Field Mouse, Apodemus sylvaticus, in Mainland Britain,”
fe Cool И: L472 157)

Wiruers, В. Е. J., 1964. ‘‘Morphological Correspondence and the Concept
of Homology,” Form and Strategy in Science (J. R. Gregg and F. T. C.
Harris, eds.), рр. 378-394. Dordrecht : D. Reidel.

Woop, А. J., T. McCowan, and М. J. Рамег, 1965. “Organ Weight-
Body Weight Relations in the Family Mustelidae, the Mink (Mustela
vison), Canad. 7. Kool., 43 : 55-68.

Wricut, S., 1926. ‘Effects of Age of Patterns on Characteristics of the
Guinea Pig,” Amer. Naturalist, 60 : 552-560.

Wricut, S.,° 1931. ‘Evolution in Mendelian Populations,” Genetics,
16 : 97-159.

Wricut, S., 1952. ‘““The Genetics of Quantitative Variability,” Quantitative
Inheritance, pp. 5-41. London : Agricultural Research Council.

Wricut, S., and Н. В. Chase, 1936. “Оп the Genetics of the Spotted
Pattern of the Guinea Pig,” Genetics, 21 : 758-787.

Улвгокоу, A. V., 1956. “Пиегезпауа funktsiya beloi okraski” (“An
Interesting BuncHON of White Coloration’’), Priroda, 1956 (6) : 92-94.

Улвгокох, A. V., 1958а. “О stroenii zubnoi sistemy i tipakh zubov u
kitoobraznykh” (“Structure of the Dental System and Types of Teeth
in Cetaceans’’), Bull. Moskov. Obshch. Ispyt. Prir. (Biol.) 63 (2) : 37-48.

Улвтокоу, A. V., 1958b. “К morfologii pishchevaritel’nogo така
zubatykh kitoobraznykh” (“Morphology of the Digestive Tract of
Toothed Whales’), Zool. Zh., 37 : 601-611.

Улвгокоу, А. V., 1959a. Morfologicheskie osobennosti belukhi Delphi-
napterus leucas) kak predstavitelya zubatykh kitoobraznykh [Morpho-
logical Characteristics of the White Whale (Delphinapterus leucas) as a
Representative of Toothed Whales]. Avtoref. Diss. па Soisk. Uch. St.
Капа. Biol. Nauk. Moscow.

Улвгокоу, A. V., 1959b. ‘“‘Nekotorye osobennosti stroeniya skeleta
belukhi kak predstavitelya zubatykh kitoobraznykh” (“Some Charac-
teristics of the Skeletal Structure of the White Whale as a Representa-
tive of Toothed Whales”), Trudy i konf. Melodykh nauch. бот. Мот}.
Labor., Moscow, рр. 137-140.

Улвгокоу, A. V., 196la. ‘“Funktsional’naya morfologiya organov
dykhaniya zubatykh kitoobraznykh” (“Functional Morphology of the
Respiratory Organs of Toothed Whales’’), Trudy Soveshch. po ekologu
г promyslu morskikh ОА рр. 79-87. Moscow : Izdatelstvo
Akad. Nauk SSSR.

Улвгокоу, A. V., 1961Ъ. ‘‘Nekotorye osobennosti mochepolovoi sistemy
kitoobraznykh”’ (“Some Features of the Urogenital System of Ceta-
ceans”), Bull. Moskov. Obshch. Ispyt. Prir. (Biol.), 46 (2) : 33-38.

Улвтокох, A. V., 1961с. “О znachenii stepeni razvitiya muzhskoi matki

348 Bibliography

kak taksonomicheskogo priznaka u kitoobraznykh” (“Significance of
the Degree of Development of the Male Uterus, as a Taxonomic Trait
in Cetaceans’’), Ibid., 46 (6) : 149-150.

Yastokxov, А. V., 19614. “Osbennosti stroeniya popyulyatsii kitoo-
braznykh kak evolusionnyi faktor’’ (‘“Features of the Population Structure
in Cetaceans as an Evolutionary Factor’’), Vsesoyuznoe Soveshch. ро Mleko-
ри., Vol. П : 111-112. Moscow : Izdatelstvo Moskovskii Gos. Univ.

Улвгокот, A. V., 1962. ‘‘Sud’ba grenlandskogo tyulenya” (“Fate of the
Greenland Seal’’), Priroda, 1962 (2) : 66-72.

Улвтокоу, А. V., 1963a. ‘“‘Nekotorye predvaritel’nye materialy К
morfologicheskoi kharakteristike grenlandskogo tyulenya Belogo 1
Grenlandskogo morei’”’ (“Some Preliminary Data on the Morphological
Characters of Greenland Seals of the White and Greenland Seas’),
Sb. Nauchno-Issled. Rabot Sev. Otd. Polyarnogo Inst. Ryb. Khoz 1 Okeanogr.
(Arkhangelsk), pp. 24-36.

Yasioxov, A. V., 1963b. ‘‘Materialy К morfologicheskoi kharakteristike
khokhlacha Grenlandskogo тогуа” (“Data on Morphological Charac-
ters of Hooded Seals of the Greenland Sea’’), Ibid., pp. 64-72.

Улвгокоу, A. V., 1963c. “Problema rudimentarnykh organov” (““The
Problem of Vestigial Organs’’), Zool. <h., 42 : 441-450.

Улвгокоу, A. V., 1963d. ‘‘Morfologicheskie kriterii у opredelenii
samostoyatel’nosti popyulyatsii nekotorykh lastnogikh” (“‘Morphological
Criteria in Determination of the Independence of Populations of Some
Pinnipeds”’), Tezisy Dokl. Vtorogo Vsesoyuzn. Soveshch. po Izucheniyu Mors-
kikh Mlekopitayushchikh, Leningrad, p. 50.

Улвгокох, A. V., 1963е. “О tipakh kitoobraznykh” (“Оп the Types of
Cetaceans’’), Bull. Moskov. Obshch. Ispyt. Prir. (Biol.), 68 (6) : 27-41.
Улвгокоу, A. V., 1964. ‘“‘Konvergentsiya ili parallelizm у razvitii
kitoobraznykh ?” (‘Convergence or Parallelism in the Evolution of

Cetaceans ?’’), Paleontol. Zh., 1946 (1) : 97-106.

Улвгокоу, A. V., 1964a. “О nekotorykh elementarnykh tendentsiyakh
izmenchivosti priznakov mlekopitayushchikh i popytka klassifikatsii
razlichnykh proyavlenii individual’noi izmenchivosti” (“On Some
Elementary Tendencies of Variability of Mammalian Traits and an
Attempt at Classification of Different Kinds of Individual Variability”’),
Sb. Voprosy Vnutrividovot Izmenchivosti 1 Mukroevolyutsiya,, pp. 153-155.
Sverdlovsk.

YasLoxov, A. V., 1968. ‘‘Popyulyatsionnaya morfologiya zhivotnykh”
(‘‘Population Morphology of Animals”), Zool. Zh., 47 : 1749-1765.
Улвгокоу, A.V., 1968. ‘‘Morfologicheskaya izmenchivost’ у popyulyatsi-

yakh mlekopitayushchikh”’ (‘Morphological Variability in Populations
of Mammals’’), Jtog: Nauki Zoologiya, 1966 : 48-85.
Улвгокох. А. V., 1970. ‘‘Morfologiya i mikroevolyutsiya’”’ (“Morphology

Bibliography 349

and Microevolution”), Zhur. Obshch. Biol., 31 : 3-13.

Улвгокоху, А. V., and Т. У. AnpREEvA, 1965. “Age Determination in
Baleen Whales (Mystacocett),” Nature, 205 : 412-413.

Улвгокоу, А. V., and У. Ya. Erin, 1965. “Analiz vnutrividovykh
razlichii v okraske tela mlekopitayushchikh (na primere grenlandskogo
tyulenya)”? [An Analysis of the Intra-species Differences in Body
Color of Mammals (with an Example from the Greenland Seal)”’],
Kool. Kh., 44 : 1094-1097.

Улвгокох, A. V., and V. Ya. Erin, 1968. “‘Izmenchivost’ morfologiches-
kikh priznakov nutrii Azerbaidzhana i Tadzhikistana” (“Variability of
Morphological Features of the Coypu in Azerbaidzhan and Tadzhikis-
tan’), Zool. Zh., 47 : 116-121.

_Улвгокох, А. V., and С. A. Kievezau’, 1964. “‘Vibrissy kitoobraznykh 1
lastonogikh, ikh raspredelenie, stroenie, znachenie” (‘‘Vibrissae of
Cetaceans and Pinnipeds : Their Distribution, Structure, and Import-
ance”), Sb. Morf. Osobennosti Vodnykh Mlekopitayushchikh, pp. 48-81.
Moscow : Izdatelstvo ‘Nauka’.

Улвгокоху, A. V., and О. Е. SERZHENT (SERGEANT), 1963. ‘‘Morfologiches-
kaya kharakteristika izmenchivosti kraniologicheskikh priznakov
grenlandskogo tyulenya”’ (“Morphological Characteristics of Variability
of Craniological Traits of the Greenland Seal’), Zool. Zh., 42 : 1857-1865.

ZARAPKIN, S. В., 1934. ‘‘Zur Phanoanalyse von geographischen Rassen
und Arten,” Arch. Naturgesch. (N. F.), 3 : 161-186.

ZARAPKIN, S. В., 1937. “Phanoanalyse von einigen Populationen der
Epilachna chrysomelina,’ Zschr. Indukt. Abstammungs- und Vererbungslehre,
И -1282-331:

ZARAPKIN, 5. R., 1939. | “Das Divergenzprinzip in der Bestimmung Кешег
systematischer Kategorien,” ГИ Internat. Kongr. Entomol., pp. 494-518.
Zavapsku, К. M., 1958. “К ponimaniyu ргостезза v organicheskoi
prirode” (“Оп the Understanding of Progress in Nature”), © Sb.

Razvitie v Ритой i Obshchestve, pp. 79-120. Leningrad.

Zavapsku, К. M., 1961. Uchenie о vide (The Concept of Species). Izdatel-
stvo Leningradskii Gos. Univ.

ZEJDA, ]., 1960. “The Influence of Age on the Formation of the Third
Upper Molar in the Bank Vole, Clethrionomys glareolus (Schreber, 1780)
(Mammalia, Rodentia),”’ Zool. Listy, 9 : 159-166.

Ге]рА, ]., 1965. ‘“‘Zur Variabilitat der Molarenwurzeln des Oberkiefers
vier Apodemus Arten (Mammalia), ” ©. Morphol. Okol. Tiere, 54 : 699-706.

Zemsku, У. А., 1960. “Byli li u kitoobraznykh zadnie konechnosti”
(“Did the Cetaceans Have Posterior Extremities ?”), Okhrana Prirody
2 Ozelenenie, 3 : 105-107.

Zemsku, У. A., and A. A. Berzin, 1961. “О redkom yavlenii atavisma и
kashalota (Physeter catadon Г..)” (“The Rare Phenomena of Atavism

350 Bibliography

in Physeter catadon L.”’), Nauchnye Dokl. Vysshei Shkoly, Seriya Biol. Nauki,
pp. 56-60.

ZHDANOV, О. A., and В. A. Nixiryux, 1964. “Vozrastnye izmeneniya
dlinny i vesa tela i okruzhnosti grudnoi kletki и sovremennogo
vozroslogo naseleniya ©. Moskvy’’ (“Growth Variation in Length and
Weight of the Body and the Circumference of the Thoracic Cavity in
the Modern Adult Population of Moscow’’), Arkhiv Anat. Gistol. Embriol.,
16 (3) : 28-42.

ZSCHIESCHE, W., and H. HeEtneoxe, 1964. “‘Biometrische Untersuchungen
ausseres Когрегтаззе au Mauseinzuchtstammen,” X. Versuchstkunde,
5 (1-3) : 78-87.

ZuBCHANINOVA, Е. V., 1962. “О geograficheskoi izmenchivosti burundu-
kov (Eutamias sibirica Laxm.)” (“Geographical Variability of Eutamias
sibirica Гахт.”), ©. Versuchstkunde, 5 (4) : 41-45.

enn
it

V4

в И в.

о. к

И КИ
Ра

(Contd. from front flap)

have been rediscovered or absorbed
into western science; others have not.
Yablokov’'s book documents part of
Schmalhausen’s approach with res-
pect to mammals and makes some
general theoretical advances.

In the present work the author has
put great emphasis on the study of
the characteristics of population
morphology in the Class Mammalia.
It includes every available fact on the
variability of different systems of
organs and the structure of mammals.
Based on an analysis of variability at
various taxonomic levels in the group
Pinnipedia an attempt has been made
to clarify the general characteristics
of variability of a group at different
taxonomic categories. A new scheme
of classification of the phenomenon
of variability has been suggested.
An attempt has been made to eluci-
date various general questions relat-
ed to the process of micro-evolution
operating at the level of a population.

er

ПАОЛО TNE

of
C