(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Animal ecology"

/^ /■ 



TEXT-BOOKS OF ANIMAL BIOLOGY 

Edited by Julian S. Huxley, M.A. 

FuUerian Professor of Physiology, Royal Institution; 
Honorary Lecturer, King's College, London. 



ANIMAL ECOLOGY 




TEXT-BOOKS OF ANIMAL 
BIOLOGY 

Edited by Julian S. Huxley 

By 



COMPARATIVE PHYSIOLOGY. 

L. T. HOGBEN, 

Other volumes m pfeparation. 

VERTEBRATE MORPHOLOGY. 
C. R. DE Beer. 



By 



EXPERIMENTAL ZOOLOGY. By Julian 
S. Huxley. 

ANIMAL MORPHOLOGY, WITH ESPE- 
CIAL REFERENCE TO THE IN- 
VERTEBRATA. By W. Garstang. 






ANIMAL ECOLOGY 



BY 

CHARLES ELTON 



WITH AN INTRODUCTION BY 

JULIAN S. HUXLEY, M.A. 

FULLERIAN PROFESSOR OF PHYSIOLOGY, ROVAL INSTITUTIOV 



NEW YORK 

THE MACMILLAN COMPANY 
1927 



PRINTED IN GREAT BRITAIN BY 

WILLIAM CLOWES AND SONS, LIMITED 

LONDON AND BECCLES. 



TO MY BROTHER 

GEOFFREY ELTON 




AUTHOR'S PREFACE 

Ecological methods can be applied to many different branches 
of animal biology. For instance, they may be employed in the 
study of evolution and adaptation. Most of the work done 
so far under the name of ecology has been concerned with 
this side of the subject, and has been summed up to some 
extent by Borradaile (The Animal and its Environment) and 
Hesse (Tiergeographie auf Okologische Grundlage). Or they 
may be appHed to the study of the normal and abnormal life 
of cells in the bodies of animals, as has been so strikingly done 
by Morley Roberts {Malignancy and Evolution) ; or again they 
may be used in the study of man, when they form the sciences 
of sociology and economics (as exemplified by Carr- Saunders 
in The Population Problem)^ or of the relations between the two 
sexes, as illustrated by the work of EHot Howard and J. S. 
Huxley on bird-habits. The present book is chiefly concerned 
with what may be called the sociology and economics of animals, 
rather than with the structural and other adaptations possessed 
by them. The latter are the final result of a number of pro- 
cesses in the lives of animals, summed up over thousands or 
millions of years, and in order to understand the meaning and 
origin of these structures, etc., we must study the processes and 
not only their integrated results. These latter are adequately 
treated in the two works mentioned above, and although of 
great intellectal interest and value, a knowledge of them throws 
curiously little Ught on the sort of problems which are en- 
countered in field studies of living animals. I have laid a good 
deal of emphasis on the practical bearings of many of the ideas 
mentioned in this book, partly because many of the best 
observations have been made by people working on economic 



viii AUTHOR'S PREFACE 

problems (most of whom, it may be noted, were not trained 
as professional zoologists), and partly because the principles 
of animal ecology are seldom if ever mentioned in zoological 
courses in the universities, in spite of the fact that it is just 
such knowledge which is required by any one who is brought 
up against practical problems in the field, after he leaves the 
university. Ecology is a branch of zoology which is perhaps 
more able to offer immediate practical help to mankind than 
any of the others, and in the present rather parlous state of 
civilisation it would seem particularly important to include 
it in the training of young zoologists. Throughout this book 
I have used analogies between human and animal communities. 
These are simply intended as analogies and nothing more, 
but may also help to drive home the fact that animal interrela- 
tions, which after all form the more purely biological side of 
ecology, are very complicated, but at the same time subject 
to definite economic laws. 

Finally, I wish to point out that no attempt has been made 
to provide a handbook containing references to all the ecological 
work that has ever been done, since such a work would be 
both exhausting to read, and useless when it had been read. 
I have simply taken the various principles and ideas and 
illustrated them by one or two examples. 

I am indebted to Mr. O. W. Richards for a great deal of 
help and criticism. Many of the ideas in this book have been 
discussed with him, and gained correspondingly in value, and 
in particular his extensive knowledge of insects has been 
invaluable in suggesting examples to illustrate various points. 
The list of works dealing with British insects was compiled 
with his aid. 

I am also indebted to Professor J. S. Huxley for much help- 
ful advice, and I have to thank Dr. T. G. Longstaff, Dr. K. S. 
Sandford, and Captain C. R. Robbins, for allowing me to use 
photographs, all taken in rather out-of-the-way parts of the 
world (Spitsbergen, Egyptian desert, and Burma, respectively). 

CHARLES ELTON. 

University Musewm, 
Oxford. 



EDITOR'S INTRODUCTION 

When I decided to undertake the editing of this series of 
volumes, I had a perfectly definite idea in mind. Biological 
science has been of late years growing and expanding at a 
prodigious rate. As a result, teachers of zoology and also of 
botany — but I shall confine myseff to Animal Biology — have 
had to face the gravest difficulties in regard to their curriculum. 
The first difficulty is a purely quantitative one : now that the 
subject has invaded so many new fields, how to stuff this 
tenfold bulk of knowledge into the brains of students in the 
same time as before. The second difficulty concerns the 
relative value of the different biological disciplines. Shall 
Comparative Morphology continue in the future to dominate 
the undergraduate's learning period, with Genetics, Cytology, 
Entwicklungsmechanik, Animal Behaviour, Systematics, Dis- 
tribution, Ecology, Histology, Comparative Physiology, and 
Evolution tacked or thrown on here and there like valances or 
frills or antimacassars ? or can it and should it renounce its 
pretensions and become one of a society of equals ? 

It is to my mind more important to attempt an answer to 
this second question first. I do not believe that comparative 
morphology has the right to demand the lion's share of the 
students' time and energy. That it at present obtains that 
lion's share is due almost entirely to historical reasons — to the 
fact that zoological departments grew up while comparative 
anatomy and morphology were the most fruitful and the most 
interesting lines of attack in zoology. 

Those who uphold the present system tell us that morph- 
ology is the foundation of zoology, the backbone of the subject, 
and that it is impossible or unprofitable to embark on subjects 

ix 



X EDITOR^S INTRODUCTION 

like comparative physiology or developmental physiology or 
ecology until the student has gained a general knowledge of 
the main types of the animal kingdom. I am quite prepared 
to admit that morphology is the backbone of zoology. On 
the other hand, I am not at all sure that it is the foundation of 
our science ; I should be more disposed to confer that title 
upon physics and chemistry. Hov^ever, all such doubts 
apart, the fact of being either a foundation or a backbone most 
emphatically does not call for the size-privileges at present 
accorded to morphology. We do not live in the foundations 
of our houses, nor are they larger than the superstructure. 
And as for backbones, it should not need more than a very 
elementary acquaintance wjth natural history to realise that 
an animal whose backbone weighed more than its muscles, 
nervous system and viscera combined, would be biologically 
very inefficient. 

As to the claim that other subjects can only be tackled 
after a morphological survey of the animal kingdom, this must 
be taken cum grano. It is in one sense obviously true, but in 
another completely false. It is false if the knowledge of 
morphology assumed is that detailed and intensive knowledge 
which is usually required for a zoological degree. It is true 
if we mean that a general survey of the main types of structure 
and development found among animals is a desirable pre- 
requisite to many other branches of biology. But such a 
survey can be given in a small fraction of the time now allotted 
to the morphological discipline ; what is more, if thus given, 
the wood will not be obscured by the trees, which is unfortu- 
nately too often the case {experto crede !) when the intensive 
and detailed system is practised. There are only about twelve 
phyla in the animal kingdom ; while the total number of groups, 
whether sub-phylum, class, sub-class or order, of which the 
budding zoologist need know the bare existence before he 
embarks on general biology, is certainly less than a hundred, 
and the characteristics of at least half of these he need only be 
acquainted with in the most superficial way, provided that he is 
well instructed in the ground- work of the phyla and sub-phyla. 

Another claim which I have often heard made is that the 



EDITOR'S INTRODUCTION xi 

majority of the subjects of general biology can be most profit- 
ably treated in relation to a thorough general morphological 
survey of the animal kingdom. This, I must confess, appears 
to me pedagogically a most pernicious doctrine. It was all 
very well so long as the other subjects remained scrappy. 
But once they have penetrated deep enough to acquire definite 
principles of their own, the defects of the method are revealed, 
since it is almost impossible to teach two not very closely 
related sets of principles simultaneously. This was soon 
recognised for subjects with a definite body of principles of 
their own, such as cytology or genetics ; but the old point 
of view too often fingers in respect of, for instance, ecology, 
or systematic and faunistic studies. 

The remedy, in my opinion, is to drop the whole notion 
of having a single main course around which, like his para- 
phernalia around the White Knight, the remainder of the 
subjects are hung. There should be a series of courses of 
approximately equal " value," each covering one of the main 
fields of biological inquiry, each stressing a different set of 
principles, so that the student will at the close have seen his 
science from the greatest possible number of different angles. 
As a preliminary programme, I should suggest about ten such 
courses. For instance : (i) vertebrate morphology, stressing 
the principles of comparative anatomy ; (2) vertebrate embry- 
ology, stressing the principles of development, including 
organogeny and histogenesis ; (3) the invertebrates and lower 
chordates, stressing both comparative anatomy and embry- 
ology, so as to bring out the divergencies and various grades 
of animal fife ; (4) cytology and histology ; (5) genetics ; 
(6) developmental physiology, including the effects of function 
upon structure, regeneration, dedifferentiation, tumour- 
formation, etc., as well as what is usually called experimental 
embryology ; (7) faunistic zoology and ecology, bringing out 
the types of environment, and of adaptations to various environ- 
ments, as well as the *' animal sociology and economics " 
covered by ecology in the narrower sense ; (8) comparative 
physiology ; (9) animal behaviour ; and (10) evolution, 
including some treatment of the principles of systematics. 



xii EDITOR'S INTRODUCTION 

I do not wish to imply that each of these courses should 
require the same number of lectures, still less the same amount 
of practical work. But I do claim that each of them is in a 
certain real sense of equal importance, since each of them, if 
properly taught, can impart its own characteristic point of 
view ; and I claim that each of these points of view is of equal 
importance to any one claiming the title of biologist. 

It would be perfectly possible to add to the list : Economic 
Zoology and Historical Zoology at once occur to the mind. 
It would be equally possible to arrange for a divergence in 
specialisation in a student's last year, some choosing the more 
physico-chemical, some the more biological side of the subject. 
But such details must depend on the experience gained as 
teaching adapts itself to the growth of the subject, and also 
upon local conditions. 

Thus what I had in mind in arranging for this series, was 
to make an attempt to cover these separate fields, each field 
being handled on approximately the same scale. Some 
modern developments have been so well treated in recent years 
that I have not thought it worth while to enter into useless 
competition with the admirable existing works. That is 
eminently the case with genetics, in which the volumes by Crew, 
Morgan, Jones, and others already carry out what was in my 
mind. 

Other fields have also been covered, but covered too well. 
Parker and Haswell, Adam Sedgwick, MacBride's Invertebrate 
Embryology and Graham Kerr's Vertebrate Embryology, are 
books of reference. To ask undergraduate students to read 
through such works is merely to give them mental indigestion, 
though they may obviously be used with great profit in con- 
junction with lecture-notes and short text-books. 

T. H. Huxley, that great scholar and man of science whose 
name I am proud to bear, wrote one text-book of Vertebrate, 
another of Invertebrate Anatomy. It is text-books of that 
size and scope at which this series aims, in which the detail 
shall be used to illustrate the principles, but no deadening 
attempt made at a completeness which in any case must remain 
unattainable. 



EDITOR'S INTRODUCTION xiii 

Finally, there remain subjects which are of such recent 
growth that their principles have never yet been treated in a 
comprehensive way. Such, for instance, are developmental 
and comparative physiology, animal behaviour, and ecology. 
From the point of view of the rapid growth and expansion of 
general biology, it is these subjects which it is at the present 
moment most important to summarise in brief text-books, 
since otherwise the multifarious knowledge which we have 
already attained regarding them remains locked up in scattered 
papers, the property of the specialist alone. 

The present volume deals with a much misunderstood and 
often underrated subject. If we leave out Hesse's Tiergeo- 
graphie auf okologischen Gnmdlage, which deals with faunas 
and major habitats and animal adaptations rather than with 
ecology sensu stricto, hardly any books dealing with the subject 
have been published since Shelford's fine pioneering work 
of 1913, Animal Communities of Temperate America. 

The subject is also so new and so complex that it is only 
of recent years that principles have begun to emerge with any 
clearness. It has further suffered from taking over too whole- 
heartedly the concepts of plant ecology and applying them 
directly to animals instead of seeing whether the difference 
between animal and plant biology did not of necessity intro- 
duce a difference in the principles governing animal and plant 
ecology. 

Mr. Elton, ever since I had the good fortune to have him 
as my pupil at Oxford, has been largely occupied with the 
problems of animal ecology and the quest for guiding principles 
in the subject. He has been fortunate in having field ex- 
perience in the Arctic, where the ecological web of life is 
reduced to its simplest, and complexity of detail does not hide 
the broad outlines. He has also been fortunate in early be- 
coming preoccupied with the subject of animal numbers ; or, 
I should rather say, he early showed characteristic acumen in 
seeing the fundamental importance of this problem. He is 
finally fortunate in having an original mind, one which refuses 
to go on looking at a subject in the traditional way just because 
it has always been looked at in that way. The result, it 



xiv EDITOR'S INTRODUCTION 

appeared to me as I read through his manuscript, is an 
illuminating and original book, the first in which the proper 
point of view of animal ecology has yet been explicitly stated. 

I will take but one example, and that from Mr. Elton's 
pet subject, the regulation of animal numbers. 

Men of science do not escape the usual human weakness 
of regarding facts in a naive and superficial way until some 
special stimulus to deeper analysis arises. I suppose that 
most professional biologists think of the relation of carni- 
vores to herbivores, preyer to preyed-upon, almost wholly 
in the light of the familiar metaphor of enemies ; and of 
the relation between the two as being in some real way like 
a battle. The ecologist, however, speedily arrives at the 
idea of an optimum density of numbers, which is the most 
advantageous for the animal species to possess. He then 
goes on to see by what means the actual density of popula- 
tion is regulated towards the optimum ; and finds that in 
the great majority of cases the existence of enemies is a 
biological necessity to the species, which without them 
would commit suicide by eating out its food-supply. To 
have the right " enemies," though it can hardly be spoken 
of as an adaptation, is at least seen to be a biological 
advantage. 

Ecology is destined to a great future. The more advanced 
governments of the world, among which, I am happy to say, our 
own is coming to be reckoned, are waking up to the fact that the 
future of plant and animal industry, especially in the tropics, 
depends upon a proper application of scientific knowledge. 
Tropical Research Stations, like those at Trinidad and Amani ; 
special investigations, like that into the mineral salt require- 
ments of cattle in equatorial Africa ; schemes for promoting 
the free flow of experience and knowledge from problem to 
problem and from one part of the world to another, such as 
were outlined in the report of the Research Committee of the 
Imperial Conference — all these and more will be needed if 
man is to assert his predominance in those regions of the 
globe whose climate gives such an initial advantage to his 
cold-blooded rivals, the plant pest and, most of all, the insect. 



EDITOR'S INTRODUCTION xv 

To deal with these problems, a cry is going up for economic 
entomologists, mycologists, soil biologists, and the rest. Ad hoc 
training in these and similar subjects is being given at various 
centres, and special laboratories are being erected for research 
in the separate branches. Valuable results are being achieved : 
but the general biologist is tempted to ask v^hether in the quest 
for specific knowledge and specific remedies it is not being 
forgotten that behind all the detail there is to be sought a body 
of general principle, and that all these branches of study are 
in reality all no more and no less than Applied Ecology. The 
situation has many points of resemblance to that which ob- 
tained in medicine in the last half of the nineteenth century. 
Then, under the magic of the germ- theory and its spectacular 
triumphs, medical research on disease was largely concentrated 
upon the discovery of specific " germs " and their eradication. 
But as work progressed, the limitations of the mode of attack 
were seen. Disease was envisaged more and more as a 
phenomenon of general biology, into whose causation the 
constitution and physiology of the patient and the effects 
of the environment entered as importantly as did the specific 
parasites. 

So it will be with the control of wild life in the interest 
of man's food-supply and prosperity. The discovery of the 
tubercle bacillus has not led to the eradication of tuberculosis : 
indeed it looks much more Hkely that this will be effected 
through hygienic reform than through bacteriological know- 
ledge. In precisely the same way it may often be found 
that an insect pest is damaging a crop ; yet that the only 
satisfactory way of growing a better crop is not to attempt 
the direct eradication of the insect, but to adopt improved 
methods of agriculture, or to breed resistant strains of the crop 
plant. In other words, a particular pest may be a symptom 
rather than a cause ; and consequently over-specialisation in 
special branches of applied biology may give a false optimism, 
and lead to waste of time and money through directing atten- 
tion to the wrong point of attack. 

The tropical entomologist or mycologist or weed- controller 
will only be fulfilling his functions properly if he is first and 



xvi EDITOR'S INTRODUCTION 

foremost an ecologist : and I look forward to the time when 
all the present ad hoc branches of applied biology will be 
unified in relation to laboratories of pure and applied ecology. 
I will give but one example of the value of ecological 
knowledge and the ecological outlook in these matters. It is 
a familiar fact that serious plagues of mice, rats, and other 
rodents occur from time to time in various parts of the world, 
often causing a great deal of material damage. At the moment 
that I write these lines, the newspapers record a rodent plague 
in California so serious that all crops are in danger over a 
considerable section of the State. Readers of Mr. Elton's 
book will discover that these violent outbreaks are but special 
cases of a regular phenomenon of periodicity in numbers, which 
is perfectly normal for many of the smaller mammals. The 
animals, favoured by climatic conditions, embark on reproduc- 
tion above the mean, outrun the constable of their enemies, 
become extremely abundant, are attacked by an epidemic, 
and suddenly become reduced again to numbers far below the 
mean. When such a number-maximum is so accentuated as 
to become a plague, remedial measures are called for locally, 
and large sums of money may be spent. Eventually the normal 
epidemic breaks out and the plague abates. The organisers 
of the anti-rodent campaign claim the disappearance of the 
pest as a victory for their methods. In reality, however, it 
appears that this disappearance is always due to natural causes, 
namely, the outbreak of some epidemic ; and that the killing 
off of the animals by man has either had no effect upon the 
natural course of events, or has delayed the crisis with the 
inevitable effect of maintaining the plague for a longer period 
than would otherwise have been the case 1 In the latter event, 
it would actually have been a better counter-measure to do 
nothing at all than to spend time and money in fruitless killing. 
If remedial measures are to be desired, they must be of some 
special sort. Either they must encourage the development 
of the epidemic, as by introducing infection among the wild 
population of the pest species ; or they must aim at reducing 
reproduction, as in the Rodier anti-rat campaign, where after 
trapping, only females are killed and all males liberated once 



EDITOR'S INTRODUCTION xvii 

more ; or they must be aimed at the general ecological 
status of the species, making it more difficult for it to live 
and reproduce, as has in another sphere been accomplished 
by drainage and cultivation v^ith regard to the malarial 
mosquito. 

I recommend Mr. Elton's book to biologists as a valuable 
and original contribution to pure science, and as a fresh 
foundation for applied zoology. 

JULIAN S. HUXLEY. 

February, 1927. 




CONTENTS 



PAGE 

Preface vii 

Introduction by Professor Julian S. Huxley, M.A. ix 

Contents xix 

List of Plates xx 

List of Diagrams in the Text . . . . xxi 

CHAPTER 

I. Introduction i 

II. The Distribution of Animal Communities . . 5 

III. Ecological Succession 18 

IV. Environmental Factors 33 

V. The Animal Community 50 

VI. Parasites 71 

VII. Time and Animal Communities .... 83 

VIII. The Numbers of Animals . . ... . .101 

IX. Variations in the Numbers of Animals . . .127 

X. Dispersal 146 

XI. Ecological Methods 162 

XII. Ecology and Evolution 179 

Conclusion 188 

List of References . . ^ 192 

Index 201 



30404 



LIST OF PLATES 

bATE 

I. A climax tropical forest in Burma . 
n. A limestone desert 

III. (a) A typical sketch of high arctic dry tundra . 
(^) Drifting pack-ice (Spitsbergen Archipelago) 

IV. {a) A semi-permanent pond in Oxfordshire 
(d) Zonation of habitats on the sea-shore . 

V. {a) Zonation of plant communities on the Lancashire sand 
dunes ....... 



FACING PAGE 



(d) Zones of vegetation on the edge of a tarn . 

VI. (a) Animal community in the plankton of a tarn 

(d) Effect of " rabbit pressure" on the Malvern Hills 

VII. (a) A guillemot cliff in Spitsbergen . . . . 
(d) Adelie penguin rookery on Macquarie Island 

VIII. A "cemetery" of walruses on Moffen Island 



14 

22 
22 

30 
30 

40 

40 

54 

54 

102 

102 

106 



XX 



LIST OF DIAGRAMS IN THE TEXT 

FIG, PAGE 

1. Zonation of corals on a reef in the New Hebrides. . . 15 

2. Stages in ecological succession on Oxshott Common 



3. Food relations of the herring to the North Sea plankton 

community ........ 

4. Food-cycle among the animals on Bear Island 

5. Food-cycle on young pine-trees on Oxshott Common 

6. Food-cycle of tapeworm through rabbit and fox 

7. Food-cycle of tapeworm through fish .... 

8. Food-chains of parasites and carnivores 

9. Effect of man on polar bears and seals 

10. Food-relations of tsetse, dragonfly, and bee-eater . 

11. Periodic fluctuations in numbers of lemmings 

12. Fluctuations in the numbers of Canadian mammals 

13. Diagram of food-chains and sizes of animals in a Canadian 

animal community ....... 



58 
58 
66 

76 
76 
80 
121 
122 
134 
136 

178 



ANIMAL ECOLOGY 



CHAPTER I 

INTRODUCTION 

" Faunists, as you observe, are too apt to acquiesce in bare descriptions 
and a few synonyms ; the reason for this is plain, because all that may 
be done at home in a man's study, but the investigation of the life and 
conversation of animals is a concern of much more trouble and difficulty, 
and is not to be attained but by the active and inquisitive, and by those 
that reside much in the country." — Gilbert White, 1771. 

I. Ecology is a new name for a very old subject. It simply 
means scientific natural history. To a great many zoologists 
the word '' natural history " brings up a rather clear vision 
of parties of naturalists going forth on excursion, prepared 
to swoop down on any rarity which will serve to swell the local 
list of species. It is a fact that natural history has fallen into 
disrepute among zoologists, at any rate in England, and since 
it is a very serious matter that a third of the whole subject of 
zoology should be neglected by scientists, we may ask for 
reasons. The discoveries of Charles Darwin in the middle 
of the nineteenth century gave a tremendous impetus to the 
study of species and the classification of animals. Although 
Linnaeus had laid the foundation of this work many years 
before, it was found that previous descriptions of species were 
far too rough and ready, and that a revision and reorganisation 
of the whole subject was necessary. It was further realised 
that many of the brilliant observations of the older naturalists 
were rendered practically useless through the insufficient 
identification of the animals upon which they had worked. 
Half the zoological world thereupon drifted into museums 

X B 



2 ANIMAL ECOLOGY 

and spent the next fifty years doing the work of description 
and classification which was to lay the foundations for the 
scientific ecology of the twentieth century. The rest of the 
zoologists retired into laboratories, and there occupied their 
time with detailed work upon the morphology and physiology 
of animals. It was an age of studying whole problems on 
many animals, rather than the whole biology of any one 
animal. The morphologist does not require the identification 
of his specimens below orders or families or perhaps (in extreme 
cases) genera. The physiologist takes the nearest convenient 
animal, generally a parasite or a pet of man, and works out his 
problems on them. The point is that most morphology and 
physiology could be done without knowing the exact name of 
the animal which was being studied, while ecological work 
could not. Hence the temporary dying down of scientific 
work on animal ecology. 

2. Meanwhile a vast number of local natural history societies 
burst into bloom all over Britain, and these bent their energies 
towards collecting and storing up in museums the local animals 
and plants. This work was of immense value, as it provided 
the material for classifying animals properly. But as time 
went on, and the groundwork of systematics was covered and 
consolidated, the collecting instinct went through the various 
stages which turn a practical and useful activity into a mania. 
At the present day, local natural history societies, however 
much pleasure they may give to their members, usually perform 
no scientific function, and in many cases the records which are 
made are of less value than the paper upon which they are 
written. Miall commented on this fact as long ago as 1897 
when he said : " Natural history ... is encumbered by 
multitudes of facts which are recorded only because they are 
easy to record." ''^^ Such is the history of these societies. 
Like the bamboo, they burst into flower, produced enormous 
masses of seed, and then died with the effort. But however 
this may be, it is necessary for zoologists to reaHse that the 
work of the last fifty years has made field work on animals a 
practical possibility. It was of little use making observations 
* The small numbers refer to the bibliography at the end of this book. 



INTRODUCTION 3 

on an animal unless you knew its name. Scientific ecology 
was first started some thirty years ago by botanists, who 
finished their classification sooner than the zoologists, because 
there are fewer species of plants than of animals, and because 
plants do not rush away when you try to collect them. Animal 
ecologists have followed the lead of plant ecologists and copied 
most of their methods, without inventing many new ones of 
their own. It is one of the objects of this book to show that 
zoologists require quite special methods of their own in order 
to cope properly with the problems which face them in animal 
ecology. 

3. When we take a broad historical view, it becomes evident 
that men have studied animals in their natural surroundings 
for thousands of years — ever since the first men started to 
catch animals for food and clothing ; that the subject was 
developed into a science by the briUiant naturalists before and 
at the time of Charles Darwin ; and that the discoveries of 
Darwin, himself a magnificent field naturalist, had the remark- 
able effect of sending the whole zoological world flocking 
indoors, where they remained hard at work for fifty years or 
more, and whence they are now beginning to put forth cautious 
heads again into the open air. But the open air feels very cold, 
and it has become such a normal proceeding for a zoologist to 
take up either a morphological or physiological problem that 
he finds it rather a disconcerting and disturbing experience to 
go out of doors and study animals in their natural conditions. 
This is not surprising when we consider that he has never had 
any opportunity of becoming trained in such work. In spite 
of this, the work badly needs doing ; the fascination of it lies 
in the fact that there are such a number of interesting problems 
to be found, so many to choose from, and requiring so much 
energy and resource to solve. Adams says : " Here, then, is a 
resource, at present largely unwt)rked by many biologists, 
where a wealth of ideas and explanations lies strewn over the 
surface and only need to be picked up in order to be utilised 
by those acquainted with this method of interpretation," ^^ 
while Tansley, speaking of plant ecology, says : " Every 
genuine worker in science is an explorer, who is continually 



4 ANIMAL ECOLOGY 

meeting fresh things and fresh situations, to which he has to 
adapt his material and mental equipment. This is con- 
spicuously true of our subject, and is one of the greatest 
attractions of ecology to the student who is at once eager, 
imaginative, and determined. To the lover of prescribed 
routine methods with the certainty of ^^^ safe ' results the study 
of ecology is not to be recommended." ^^^ 




CHAPTER II 

THE DISTRIBUTION OF ANIMAL COMMUNITIES 

Each habitat (i) has living in it a characteristic community of animals ; 
(2) these can be classilEied in various ways and (3) their great variety 
and richness is due to the comparative specialisation of most species of 
animals. (4) It is convenient to study the zonation of such com- 
munities along the various big gradients in environmental conditions, 
such as that from the poles to the equator, which (5) shows the dominating 
influence of plants upon the distribution of animals, in forming special 
local conditions and (6) by producing sharp boundaries to the habitats 
so that (7) animal communities are more sharply separated than they 
would otherwise be. This is clearly shown by (8) the vertical zones 
of communities on a mountain-side, which also illustrate the principle 
that (9) the members of each community can be divided into those 
*' exclusive " to and (10) into those *' characteristic " of it, while the 
remaining species, which form the bulk of the community, occur in 
more than one association. (11) Other vertical gradients are that of 
light in the sea and (12) that of salts in water. (13) Each large zone 
can be subdivided into smaller gradients of habitat, e.g. water-content 
of the soil, and (14) these again into still smaller ones, until we reach 
single species of animals, which in turn can be shown to contain gradients 
in internal conditions supporting characteristic communities of parasites. 
(15) In such ways the differences between communities can be classified 
and studied as a preliminary to studying the fundamental resemblances 
amongst them. 

I. One of the first things with which an ecologist has to 
deal is the fact that each different kind of habitat contains a 
characteristic set of animals. We call these animal associations, 
or better, animal communities, for we shall see later on that 
they are not mere assemblages of species living together, but 
form closely-knit communities or societies comparable to our 
own. Up to the present time animal ecologists have been 
very largely occupied with a general description and classifica- 
tion of the various animal habitats and of the fauna living in 
them. Preliminary biological surveys have been undertaken 
in most civilised countries except England and China, where 
animal ecology lags behind in a pecuKar way. In particular 

5 



6 ANIMAL ECOLOGY 

we might mention the work of the American Bureau of 
Biological Survey (which was first started in order to study 
problems raised by the introduction of the English sparrow 
into the United States), and of other institutions in that country, 
and similar surveys undertaken under the initiative of the late 
Dr. Annandale in India. It is clearly necessary to have a list 
of the animals in different habitats before one can proceed to 
study the more intricate problems of animal communities. 
We shall return to the question of biological surveys in the 
chapter on '' Methods." 

2. Various schemes have been proposed for the classifica- 
tion of animal communities, some very useful and others com- 
pletely absurd. Since, however, no one adopts the latter, 
they merely serve as healthy examples of what to avoid, namely, 
the making of too many definitions and the inventing of a host 
of unnecessary technical terms. It should always be remem- 
bered that the professional ecologist has to rely, and always 
will have to rely, for a great many of his data, upon the obser- 
vations of men like fishermen, gamekeepers, local naturalists, 
and, in fact, all manner of people who are not professional 
scientists at all. The hfe, habits, and distribution of animals 
are often such difficult things to ascertain and so variable from 
time to time, that it will always be absolutely essential to use 
the unique knowledge of men who have been studying animals 
in one place for a good many years. It is a comparatively 
simple matter to make a preliminary biological survey and 
accumulate lists of the animals in different communities. 
This preliminary work requires, of course, great energy and 
perseverance, and a skilled acquaintance with the ways of 
animals ; but it is when one penetrates into the more intimate 
problems of animal life, and attempts to construct the food- 
cycles which will be discussed later on, that the immensity of 
the task begins to appear and the difficulty of obtaining the 
right class of data is discovered. It is therefore worth em- 
phasising the vital importance of keeping in touch with all 
practical men who spend much of their lives among wild 
animals. To do this effectually it is desirable that ecology 
should not be made to appear much more abstruse and difficult 



DISTRIBUTION OF ANIMAL COMMUNITIES 7 

than it really is, and that it should not be possible to say that 
" ecology consists in saying what every one knows in language 
that nobody can understand." The writer has learnt a far 
greater number of interesting and invaluable ecological facts 
about the social organisation of animals from gamekeepers 
and private naturalists, and from the writings of men like 
W. H. Hudson, than from trained zoologists. There is some- 
thing to be said for the view of an anonymous writer in Nature, 
who wrote : " The notion that the truth can be sought in 
books is still widely prevalent and the present dearth of 
illiterate men constitutes a serious menace to the advancement 
of knowledge." Even if this is so, it is at the same time true 
that there is more ecology in the Old Testament or the plays 
of Shakespeare than in most of the zoological textbooks ever 
published ! 

All this being so, there seems to be no point in making 
elaborate and academic classifications of animal communities. 
After all, what is to be said for a scientist who calls the 
community of animals living in ponds the '' tiphic asso- 
ciation," or refers to the art of gardening as " chronic 
hemerecology " ? 

3. It is important, however, to get some general idea of the 
variety and distribution of the different animal communities 
found in the world. The existence of such a rich variety of 
communities is to be attributed to two factors. In the first 
place, no one animal is sufficiently elastic in its organisation 
to withstand the wide range of environmental conditions 
which exist in the world, and secondly, nearly all animals tend 
during the course of evolution to become more or less specialised 
for life in a narrow range of environmental conditions, for by 
being so specialised they can be more efficient. This tendency 
towards specialisation is abundantly shown throughout the 
fossil record and is reflected in the numerous and varied 
animal communities of the present day. Primarily, there is 
specialisation to meet particular climatic and other physical 
and chemical factors, and secondarily, animals become adapted 
to a special set of biotic conditions — food, enemies, etc. The 
effect of the various environmental factors upon animals is a 



8 ANIMAL ECOLOGY 

subject which will be followed up separately in the next 
chapter. 

4. One way of giving some idea of the range of different 
animal habitats, and of the communities living in them, is to 
take some of the big gradients in environment and show how 
the communities change as we pass from one end to the other. 
The biggest of these is the gradient in temperature and light 
intensity between the poles and the equator, which owes its 
existence to the globular shape of the world. At the one 
extreme there are the regions of polar ice-pack, with their 
peculiar animal communities living in continuous dayhght 
during the summer and continuous darkness in the winter, and 
with corresponding abrupt seasonal changes in temperature. 
At the other end of the scale we have equatorial rain forest 
with a totally different set of animals adapted to life in a 
continuously hot climate, and in many cases in continuous 
semi- darkness, in the shade of the tropical trees. An animal 
like Bosman's Potto sees less light throughout the year than the 
Arctic fox. In between these extremes we have animal com- 
munities accustomed to a moderate amount of heat and light. 

5. These examples serve to introduce a very important 
idea, namely, the effect of vegetation upon the habitats and 
distribution of animals. 

Although a tropical rain forest partly owes its existence to 
the intense sunlight of the tropics, yet inside the forest it is 
quite dark, and it is clear that plants have the effect of trans- 
lating one climate into another, and that an animal living in 
or under a plant community and dependent on it, may be 
living under totally different climatic conditions from those 
existing outside. Each plant association therefore carries 
with it, or rather in it, a special local " climate " which is 
peculiar to itself. Broadly speaking, plants have a blanketing 
effect, since they cut off rain, and radiant energy like light 
and heat. Their general effect is therefore to tone down the 
intensity of any natural climate. At the same time they 
reduce the amount of evaporation from the soil surface, and 
so make the air damper than it otherwise would be. 

Looking at the matter very broadly, in the far north there 



PLATE I 




Aerial photograph of a chmax tropical forest in Burma (taken by Captain C R. 
Robbins). Each lump in the photograph represents a forest tree some two 
or three hundred feet high. A ridge runs diagonally across the photo, and 
in the upper right-hand corner ther^ are two white patches, which are 
landslides. 



DISTRIBUTION OF ANIMAL COMMUNITIES 9 

is very sparse vegetation, which does not always cover the 
ground at all completely. As we go south the vegetation 
becomes more dense and higher until we reach a zone with 
scattered trees. These are separated by wide intervals owing 
to the fact that the soil is too shallow (being frozen below a 
certain depth) to allow of sufficient root development except 
by extensive growth sideways. Then we find true forests, but 
still not very luxuriant. Finally, there are the immense rain 
forests of the equatorial belt. The gradation consists essentially 
first of a gradual filling up of the soil by roots, and then a cover- 
ing up of the surface by vegetation. 

6. There is a further important way in which plant 
communities affect animals. When we look at two plant 
communities growing next to one another, it is usually notice- 
able that the junction between the two is comparatively sharply 
marked. Examples of this are the zones of vegetation round 
the edge of a lake or up the side of a mountain. The reason 
for this sharp demarcation between plant communities is 
simple. Plants are usually competing for light, and if one 
plant in a community manages to outstrip the others in its 
growth it is able to cut off much of the light from them, and 
it then becomes dominant. This is the condition found in 
most temperate plant communities. Examples are the common 
heather {Calluna), or the beech trees in a wood, or the rushes 
{Juncus) in a marshy area. The process of competition is not 
always so simple as this, and there may be all manner of com- 
plicated factors affecting the relative growth of the competing 
species, but the final battle is usually for light. In some cases 
the winning species kills off other competitors, not by shading 
them, but by producing great quantities of dead leaves which 
swamp the smaller plants below, or by some other means. 
The main phenomenon of dominance remains the same. 
Now at the junctions of two plant communities there is also a 
battle for light going on, and it resolves itself mainly into a 
battle between the two respective dominants. Just as within 
the community, so between two different dominants, no 
compromise is possible in a battle for light. If one plant 
wins, it wins completely. Now every plant has a certain set 



10 ANIMAL ECOLOGY 

of optimum conditions for maximum growth, and as con- 
ditions depart from this optimum, growth becomes less efficient. 
Since each dominant has different optimum conditions, there 
is always a certain point in an environmental gradient 
where one dominant, and therefore one community, changes 
over fairly abruptly into another. There may be originally 
a regular and gradual gradient in, say, water-content of the 
soil, as from the edge of a lake up on to a dry moor; but the 
existence of dominance in plants causes this to be transformed 
into a series of sharply marked zones of vegetation, which to 
some extent mask the original gradient, and may even react 
on the surroundings so as to convert the conditions themselves 
into a step-like series. 

7. It is clear, then, that because green plants feed by means 
of sunlight, the boundaries of their communities tend to be 
rather sharply defined ; and since we have seen that each plant 
community carries with it a special set of " climatic " con- 
ditions for the animals living in it, the rather sudden difference 
in conditions at the edges of the plant communities will be 
reflected in the animals. This means that the species of animals 
will tend to be subdivided into separate ones adapted to different 
plant zones, instead of graded series showing no sudden 
differences. It also means that animal communities are made 
much more distinct from one another than would be the case 
if they were all living in one continuous gradient in conditions, 
or in a series of open associations of plants like arctic fjaeldmark 
(stony desert). It would be infinitely more difficult to study 
animal associations if this were not the case, for we should 
not have those convenient divisions of the whole fauna into 
communities which are so useful for working purposes. It is 
sometimes assumed in discussions on the origin of species 
that the environmental conditions affecting animals are always 
in the form of gradients. It is clear that such is by no means 
always the case. 

8. As has been mentioned above, the abrupt transitions 
between plant communities are particularly well seen on the 
sides of mountains, where there are vertical zones of vegetation 
corresponding on a small scale to the big zones of latitude. 



DISTRIBUTION OF ANIMAL COMMUNITIES ii 

In America they are usually referred to as " life zones," and 
the existence of great mountain ranges in that country is one 
of the reasons why ecology has attracted more attention there. 
In England, where the mountain ranges are in the north, we 
do not see the impressive spectacle of great series of vegetation 
zones which have so much attracted the American ecologist. 
This vertical zonation is most striking in the tropics, where, 
within the same day, one may be eating wild bananas at sea- 
level and wild strawberries on the mountains. One of the 
best descriptions of this phenomenon is given by A. R. Wallace 
in the account of his travels through Java.^^ 

The work of botanists has given us fairly clear ideas about 
the distribution of zones of vegetation, but we are still in great 
ignorance as to the exact distribution and boundaries of the 
animal communities in these Hfe zones, and of their relation 
to the plants. A good deal of work has been done by Americans 
upon certain groups of animals (chiefly birds and mammals), 
and in particular may be mentioned the extremely fine account 
of the Yosemite region of the Sierra Nevada by Grinnell and 
Storer,^o i^ which are given accurate data of the distribution 
of vertebrates in relation to life zones, together with a mass of 
interesting notes on the ecology of the animals. 

9. There is a further important point in regard to the 
distribution and composition of animal communities. If we 
take the community of animals living, say, in the Canadian zone, 
we should find that a definite percentage are confined to that 
zone, and in fact that the distribution of some of the animals 
is strictly determined by the type of vegetation. These 
species we speak of as " exclusive " to that community. The 
game-birds found in Great Britain afford good examples of 
this. The ptarmigan {Lagopus mutus) lives in the alpine 
zone of vegetation, while the red grouse (Lagopus scoticus) 
replaces it at lower levels on the heather moors. Another 
bird, the capercaillie (Tetrao urogallus), lives in coniferous 
woods, while the pheasant {Phasianiis colchicus) occurs chiefly 
in deciduous woods. Finally, the common partridge {Perdix 
perdix) comes in cultivated areas with grassland, etc. We 
see here examples of birds which are exclusive to certain plant 



12 ANIMAL ECOLOGY 

associations, and we may also note that they are all living a 
similar life as regards food and general habits. Each associa- 
tion has some kind of large vegetarian bird, although the 
actual species is different in each case. Another well-known 
example is the common grass-mouse {Microtus agrestis) which, 
except when it is extremely abundant and " boils over " into 
neighbouring habitats, is chiefly found living underground in 
grassland, where it feeds on the roots of the grass. A great 
number of vegetarian insects are attached to one species of 
plant, and if that plant only occurs in one association, the 
animal is also limited in the same way. The oak {Quercus 
robur) supports hundreds of insects peculiar to itself, and if 
we include parasites the number will be far greater. 

10. Continuing our survey of one zone, we should find 
that there are certain species which occur in particularly large 
numbers there, although they are not exclusively confined to 
it. These we call '* characteristic " species. A good example 
of this type is the long-tailed mouse (Apodemus sylvaticus) 
which occurs in woods, but is not confined to them. Trapping 
data for one area near Oxford showed that 82 per cent, of 
specimens were caught in woods, while 18 per cent, occurred 
outside in young plantations and even occasionally in the open. 
Here the animal is not so strictly limited to one habitat as, 
for instance, Microtus^ but we are quite justified in calling it a 
wood-mouse in this district. Thorpe 1^2 has described some 
of the exclusive and characteristic British birds with reference 
to plant associations. 

In most cases in which we have any complete knowledge 
(and they are few) it is found that these two classes of animals — 
the exclusives and the characteristics — may often form only 
a comparatively small section of the whole community, that 
there are many species of animals which range freely over 
several zones of vegetation, either because they are not limited 
by the direct or indirect effects of the vegetation, or because 
they can withstand a greater range of environment than the 
others. As an example of this type of distribution we may 
take the common bank vole {Evotomys glareolus) which, in 
contrast to the Apodemus mentioned above, comes both in 



DISTRIBUTION OF ANIMAL COMMUNITIES 13 

woodland and in wood margin, shrub communities, and young 
plantations. The actual figures for comparison with Apodemus 
are as follows : 47 per cent, of the specimens in woods and 
53 per cent, outside, chiefly in shrub or young tree habitats. 

It may be, in fact, often rather an arbitrary proceeding to 
split up the animals living on, say, a mountain-side into com- 
munities corresponding to the exclusive and characteristic 
species of each life zone, and it should be realised quite clearly 
and constantly borne in mind when doing field work, that 
many common and important species come in more than one 
zone. Richards, after several years' study of the animals of 
an English heath, says : " The commonest animals in a plant 
community are often those most common elsewhere." ^^^* At 
the same time it is probably true that animals living in several 
zones of vegetation show a marked tendency to have their 
limits of distribution coinciding with the edges of the plant 
zones. This is only natural in view of the step- like nature of the 
gradient in environment produced by the plant communities. 

II. Another important vertical gradient is that found in 
the sea and in fresh- water lakes, and this is caused by the 
reduction in the amount of light penetrating the water as 
the depth increases. This gradient shows itself both in the 
free-living communities (plankton) and in those living on the 
bottom (be?ithos). As we go deeper down the plants become 
scarcer owing to lack of light, until at great depths there are 
no plants at all, and the animals living in such places have to 
depend for their living upon the dead bodies of organisms 
falling from the well-lighted zone above, or upon each other. 
There is the same tendency for the plants to form zones as 
on land, and one of the most interesting things about marine 
communities is the fact that certain animals which have 
become adapted to a sedentary existence compete with the 
plants (seaweeds of various kinds)- and in some cases com- 
pletely dominate them. The reason for this is that in the sea, 
and to a lesser extent in fresh water, it is possible for an animal 
to sit still and have its food brought to it in the water, while 
on land it has to go and get it. Web-spinning spiders are 
almost the only group of land animals which has perfected a 



14 ANIMAL ECOLOGY 

means of staying in the same place and obtaining the animals 
carried along in the air. In the tropics certain big spiders 
are actually able to snare small birds in their webs. In the 
sea an enormous number of animals sit still in one place and 
practically have their food wafted into their mouths. Indeed, 
food is probably not usually a limiting factor for such animals, 
and competition is for space to sit on. Hence it is that we 
find these animals behaving superficially like plants. Over 
great areas of the tropical seas (dependent probably on certain 
temperature conditions of the sea, or upon the plankton living 
in such waters) corals almost completely replace seaweeds on 
the seashore and shallow waters, where they feed like other 
animals on plankton organisms or upon small organic particles 
in the water. Corals on a reef usually form zones, each 
dominated by one or more species, just as in plants. The 
zonation is apparently determined by gradients in such factors 
as surf-action, amount of silt in the water, etc. Amongst the 
corals grow various calcareous algae which resemble them 
very closely in outward appearance. As we go farther from 
the equator, plants become relatively more and more abundant 
on the shores of the oceans, but even in Arctic regions certain 
groups of animals, e.g. hydroids, may form zones between 
other zones composed of seaweeds. ^^^ 

12. There is another vertical gradient in conditions which 
is clear-cut and of universal occurrence. This is the gradient 
in salt content of the water from mountain regions down to 
sea- level. Through the action of rain all manner of substances 
are continually being washed out of the rocks and soil. These 
pass into streams and rivers and accumulate temporarily in 
lakes and ponds at various levels. But since the salts are 
always being washed down, we find that on the whole the higher 
we go the purer is the water. Exceptions must be made to 
this rule in the case of places which have the higher parts of 
their mountain ranges composed of very soluble rocks, or 
in places like Central Asia, which have high plateaux on 
which many salt lakes develop. But, on the whole, we can 
distinguish an upland or alpine zone of waters containing few 
salts and often slightly acid in reaction. Lower down, the 





WTmrn 




1 ^ 


in 1926) shows 
d had no rain for 
-ipply of dew) anc 
, etc. 




V-^' i 


Desert 
ality ha 
the SI 

insects 




^■' 


ford in the Eastern ICgyptian 

1 rock escarpments. This loc 

vegetation (probably owing t 

th a good many birds, lizards, 




^- 


S. Sand 
xnd and 
an open 
ether wi 






1 (taken by Dr. K. 
esert with blown sr 
ind yet supported 
f gazelles, ibex, tog 




^,^ * 


f. <D T- ba 




^*^ 


togr 
;ton 
yeai 
stin 




.i»^_^ 


I'he phol 
limes 
four 
consi 


u. 


:i 


h 


■< 


< 


-< l-H 


^_ 


^ 


a 


^ 





DISTRIBUTION OF ANIMAL COMMUNITIES 15 

rivers and standing waters contain considerably more salts 
(e.g. places like the Norfolk Broads or the meres of Lancashire 
and Cheshire). Then there is a rather sudden increase in 
the steepness of the gradient through brackish water lagoons 
and estuaries to the sea itself. The sea, having been there 



Kllepora . ichotema 
n. Iruncota 

Sxitnphytum hirhinv 
S poimcSuJrk 

Heliopora corrulia. 
Seriafopora. hyiOrix 



I 



ft}^ri|fOfiA23NE|''^)J'|^|A«OPORA 2DNE| SCLEROPHYTUM ZONE \8Q\UilR 



ZONES 



F • paUida. 
Fav\tiii abdita 
Coiioitnea pet^nafc, 
Pavona dianai 
P vca\a/is 



Montipcr* ranoio. 



M. hopida 

M. foliosa. 

Acropora .smifhi 

A. hOLfnA 

A qufteki 
A 




pKaraoiuiipcijriKrftia 



Qofiopora. ttn^udtns 
fbrt«4 froqOKX 

Hahwtdicn- »p 

LthetHunnvtn hrmo o-uJtoceo. 



i 



The thickness of the black stripes indicates the abundance of each species 
at various distances from "the shore. 

Fig. 1. — Zonation of corals (together with a few calcareous algae) on a 
reef in the New Hebrides, showing that the phenomenon of dominance 
exists among corals, just as among plants. The left-hand side of the 
diagram is the shore of the island, while the right-hand side is the outer 
edge of the reef. (From Baker .^°*) 



much longer than the inland lakes and ponds, contains enor- 
mously more salts than the latter, but really it is only one end 
of a gradient which started high up in the mountains. There 
are well-marked different associations of animals in all these 
types of waters. Of course, other factors than salt content 



i6 ANIMAL ECOLOGY 

are important (particularly temperature), but the salt content 
itself is undoubtedly very important as a controlling factor, 
since it acts not only directly but also by affecting the hydrogen 
ion concentration of the water. 

13. Within each of the big zones which owe their existence 
to major differences in climate there are numerous smaller 
gradients in outer conditions, each of which gives rise to a 
series of more or less well-marked associations of plants and 
animals. These gradients are caused by local variations in 
soil and climate, or by biotic factors such as grazing by animals. 
One obvious example is the gradient in the amount of water in 
the soil. At one end we may find the animal community of a 
dry heather moor, and at the other the community of free- 
floating and free-swimming animals which form the plankton 
of a lake. Between these two extremes there would be zones 
of marsh, reed swamp, and so on, each with a distinctive set of 
animals. These various zones are due to the fact that at one 
end of the gradient there is much soil and practically no water 
(at any rate in summer), while at the other there is much water 
and very little soil, the proportion of soil and water gradually 
changing in between. 

14. We can carry the subdivision of animal communities 
further and split up one ordinary plant association, like an oak 
wood, into several animal habitats, e.g. tree-tops, tree-trunks, 
lower vegetation, ground surface, and underground, and we 
should find that each of these habitats contained an animal 
community which could be treated to some extent at least as a 
self-contained unit. Again, each species of plant has a number 
of animals dependent upon it, and one way of studying the 
ecology of the animals would be to take each plant separately 
and work out its fauna. Finally, each animal may contain 
within its own body a small fauna of parasites, and these again 
can be split up into associations according to the part of the 
body which they inhabit. If we examine the parasites of a 
mouse, for instance, we find that the upper part of the in- 
testine, the lower part, the caecum, the skin, the ears, each have 
their peculiar fauna. 

It is obviously impossible to enumerate all the different 



DISTRIBUTION OF ANIMAL COMMUNITIES 17 

gradients in the environment and all the different communities 
of animals which inhabit them. One habitat alone, the edge 
of a pond, or the ears of mammals, would require a whole 
book if it were to be treated in an adequate way. The aim 
of the foregoing sketch of the whole subject is to show that the 
term " animal community " is really a very elastic one, since 
we can use it to describe on the one hand the fauna of equatorial 
forest, and on the other hand the fauna of a mouse's caecum. 

For general descriptions of the animal communities of 
the more important habitats, the reader may be referred to a 
book on animal geography by Hesse,^^^ and to a more recent 
book by Haviland.^^^ 

15. The attention of ecologists has been directed hitherto 
mainly towards describing the differences between animal 
communities rather than to the fundamental similarity between 
them all. The study of these differences forms a kind of 
animal ethnology, while the study of the resemblances may be 
compared to human sociology (soon to become social science). 
As a matter of fact, although a very large body of facts of the 
first type has been accumulated, few important generalisations 
have as yet been made from it. So much is this the case that 
many biologists view with despair the prospect of trying to 
learn anything about ecology, since the subject appears to them 
at first sight as a mass of uncoordinated and indigestible 
facts. It is quite certain that some powerful digestive juice 
is required which will aid in the assimilation of this mass of 
interesting but unrelated facts. We have to face the fact that 
while ecological work is fascinating to do, it is unbearably dull 
to read about, and this must be because there are so many 
separate interesting facts and tiny problems in the lives of 
animals, but few ideas to link the facts together. It seems 
certain that the key to the situation lies in the study of animal 
communities from the sociological point of view. This branch 
of ecology is treated in Chapter V., but first of all it is necessary 
to say something about the subject of ecological succession — 
an important phenomenon discovered by botanists, since it 
enables us to get a fuller understanding of the distribution 
and relations of animal and plant communities. 

c 



CHAPTER III 

ECOLOGICAL SUCCESSION 

A number of changes ( i ) are always taking place in animal communities, 

(2) one of the most important of which is ecological succession, which 

(3) causes plant associations to move about slowly on the earth's surface, 
and (4) is partly due to an unstable environment and partly to plant 
development which typically consists (5) of a sere of associations starting 
with a bare area and ending with a climax association. (6) Each 
region has a typical set of seres on different types of country which 
(7) may be studied in various ways, of which the best is direct observation 
of the changes as in (8) the heather moor described by Ritchie or (9) the 
changes following the flooding and redraining of the Yser region 
described by Massart or (10) a hay infusion ; but (11) indirect evidence 
may be obtained as in the case of Shelford's tiger beetles. (12) The 
stages in succession are not sharply separated and (13) raise a number 
of interesting problems about competition between species of animals, 
which (14) may be best studied in very simple communities. (15) In 
the sea, succession in dominant sessile animals may closely resemble 
that of land plants, while (16) on land, animals often control the direction 
of succession in the plants. Therefore (17) plant ecologists cannot 
afford to ignore animals, while a knowledge of plant succession is 
essential for animal ecologists. 

I. We have spoken of animal communities so frequently in 
the last chapter that the reader may be in danger of becoming 
hypnotised by the mere word " community " into thinking 
that the assemblage of animals in each habitat forms a com- 
pletely separate unit, isolated from its surroundings and quite 
permanent and indestructible. Nothing could be farther 
from the true state of affairs. The personnel of every com- 
munity of animals is constantly changing v^ith the ebb and 
flow of the seasons, with changing weather, and a number of 
other periodic rhythms in the outer environment. As a 
result of this it is never possible to find all the members of 
an animal community active or even on the spot at all at any 
one moment. To this subject we shall return in the chapter 
on the Time Factor in Animal Communities, since its 

iS 



ECOLOGICAL SUCCESSION 19 

discussion comes more suitably under the structure of animal 
communities than under their distribution. 

2. There is another type of change going on in nearly all 
communities, the gradual change known as ecological succession, 
and with this we will now deal. Such changes are sometimes 
huge and last a long time, like the advance and retreat of ice 
ages with their accompanying pendulum swing from a tem- 
perate climate with beech and oak forests and chaffinches, to 
an arctic one with tundra and snow buntings, or even the 
complete blotting out of all life by a thick sheet of ice. They 
may, on the other hand, be on a small scale. Mr. J. D. Brown 
watched for some years the inhabitants of a hollow in a beech 
tree, and the ecological succession of the fauna. At first 
an owl used it for nesting purposes, but as the tissues of the 
tree grew round the entrance to the hollow it became too small 
for owls to get into, and the place was then occupied by nesting 
starlings. Later the hole grew smaller still until after some 
years no bird could get in, and instead a colony of wasps 
inhabited it. The last episode in this story was the complete 
closing up of the entrance-hole. This example may sound 
trivial, but it is an instance of the kind of changes which are 
going on continuously in the environment of animals. 

3 . If it were possible for an ecologist to go up in a balloon 
and stay there for several hundred years quietly observing the 
countryside below him, he would no doubt notice a number of 
curious things before he died, but above all he would notice 
that the zones of vegetation appeared to be moving about 
slowly and deliberately in different directions. The plants 
round the edges of ponds would be seen marching inwards 
towards the centre until no trace was left of what had once 
been pieces of standing water in a field. Woods might be 
seen advancing over grassland or heaths, always preceded by 
a vanguard of shrubs and smaller trees, or in other places 
they might be retreating ; and he might see even from that 
height a faint brown scar marking the warren inhabited by the 
rabbits which were bringing this about. Again and again 
fires would devastate parts of the country, low-lying areas 
would be flooded, or pieces of water dried up, and in every 



20 ANIMAL ECOLOGY 

case it would take a good many years for the vegetation to 
reach its former state. Ahhough bare areas would constantly 
be formed through various agencies, only a short time would 
elapse before they were clothed with plants once more. 

There are very few really permanent bare areas to be met 
with in nature. Rocks which appear bare at a distance are 
nearly always covered with lichens, and usually support a 
definite though meagre fauna, ranging from rotifers to eagles. 
Apparently barren places like lakes contain a huge microscopic 
flora and fauna, and even temporary pools of rain-water are 
colonised with almost miraculous rapidity by protozoa and 
other small animals. 

4. It is the exception rather than the rule for any habitat 
to remain the same for a long period of years. Slow geological 
processes, like erosion and deposition by rivers and by the 
sea, are at work everywhere. Then there are sudden disasters, 
like fires, floods, droughts, avalanches, the introduction of 
civilised Europeans and of rabbits, any of which may destroy 
much of the existing vegetation. There is a third kind of 
change which is extremely important but not so obvious, and 
is the more interesting since its movements are orderly and 
often predictable. This is the process known as the develop- 
me?it of plant communities. Development is a term used by 
plant ecologists in a special technical sense, to include changes 
in plant communities which are solely or largely brought 
about by the activities of the plants themselves. Plants, like 
many animals, are constantly moulting, and the dead leaves 
produced accumulate in the soil below them and help to form 
humus. This humus changes the character of the soil in such 
a way that it may actually become no longer suitable for the 
plants that five there, with the result that other species come 
in and replace them. Sometimes the seedlings of the dominant 
plant (e.g. a forest tree) are unable to grow up properly in the 
shade of their own parents, while those of other trees can. This 
again leads to the gradual replacement of one community by 
another. 

When a bare area is formed by any of the agencies we have 
mentioned, e.g. the changing course of a river, it is first 



ECOLOGICAL SUCCESSION 21 

colonised by mosses or algae or lichens ; these are driven out 
by low herbs, which kill the pioneer mosses by their shade ; 
these again may be followed by a shrub stage ; and finally 
a woodland community is formed, with some of the earlier 
pioneers still living in the shade of the trees. 

This woodland may form a comparatively stable phase, and 
is then called a climax association, or it may give way to one or 
more further forest stages dominated by different species of 
trees in the manner described above. It is not really possible 
to separate development of communities from succession 
caused by extrinsic changes, such as the gradual leaching out 
of salts from the soil or other such factors unconnected with 
the plants themselves. The important idea to grasp is that 
plants react on their surroundings and in many cases drive 
themselves out. In the early stages of colonisation of bare 
areas the succession is to a large extent a matter of the time 
taken for the different plants to get there and grow up ; for 
obviously mosses can colonise more quickly than trees. 

5. In any one region the kind of climax reached depends 
primarily upon the climate. In high Arctic regions succession 
may never get beyond a closed association of lichens, contain- 
ing no animals whatsoever. In milder Arctic regions a low 
shrub climax is attained, while farther south the natural 
climax is forest or in some cases heath, according to whether 
the climate is of a continental or an oceanic type. Sometimes 
ecological succession is held up by other agencies than climate 
and prevented from reaching its natural climax. In such 
cases it is a common custom to refer to the stage at which it 
stops as a sub-climax. A great deal of grassland and heath 
comes under this heading, for further development is pre- 
vented by grazing animals, which destroy the seedlings of the 
stage next in succession. An area of typical heather moor 
in the New Forest was fenced off for several years from grazing 
ponies and cattle by its owner, with the immediate result 
that birches and pines appeared by natural colonisation, and 
the young pines, although slower in growth than the birches, 
will ultimately replace them and form a pine wood. Here 
grazing was the sole factor preventing normal ecological 



S* 



22 



ANIMAL ECOLOGY 



succession. The same thing is well known to occur in a great 
many places when heather or grass is protected, the important 
animals varying in different places, being usually cattle, 
sheep, horses, or rabbits, or even mice. 

6. We begin to see that the succession of plant communities 
does not take place at random, but in a series of orderly stages, 
which can be predicted with some accuracy. The exact type 
of communities and the order in which they replace one another 
depend upon the climate and soil and other local factors, such 
as grazing. It is possible to classify different series of stages 
in succession in any one area, the term " sere " being used to 
denote a complete change from a bare area in water or soil up 

Pine WOOD 

'Before. fcUfng Aftzr felling 






Boc; Societies y 




kMouniA 



CONSOCIES 

MOLINIA -JoNCUS. 
ECOXOME 

4- 

JUNCUS. 



Mixed Wood 

,1 ( loeaUy) 



.jBrhCH' 



I 



s.i--.-_r.^ 



Assoc I es" 



^ z^Sphaqnuh 

Boq 

■^ C+t^ier has occu-T-rci or m /brogrts^ 
•> "ProbAble JroM observations (sccTiKt) 



*/Newoo'D 



Fig. 2. — The diagram shows the stages in ecological succession follow- 
ing colonisation of damp bare areas formed by felling of a pine wood on 
Oxshott Common. The succession is different on the drier areas. (From 
Summerhayes and Williams.^ 2°) 

to a climax like pine wood {cf. Fig 2). Each type of soil, etc., 
has a different type of '' sere " which tends to develop upon it, 
but they all have one character in common : bare areas are 
usually very wet or very dry, and the tendency of succession 
is always to establish a climax which is living in soil of an 
intermediate wetness — a type of vegetation called " meso- 
phytic," of which a typical example is an oak wood. Thus a 
dry rock surface gains ultimately a fairly damp soil by the 
deposition of humus, while a water-logged soil is gradually 
raised above the water-level by the same agency, so that there 
tends to appear a habitat in which the exjpenditure by plants 
and by direct loss from the soil is suitably balanced by the 



PLATE III 




(a) A typical stretch of high arctic dry tundra, inhabited by reindeer, arctic fox, 
etc. (The photograph was taken in August, 1924, by Dr. K. S. Sandford, 
on. Reindeer Peninsula, North Spitsbergen.) 




{/>) Drifting pack-ice near North-East Land (Spitsbergen Archipelago), with the 
bearded seal {Erignathus barbatus) lying on a fioe. (Photographed .by 
Mr. J. D. Brown, July, 1923.) 



ECOLOGICAL SUCCESSION 23 

income of water, and extremes of environment are avoided. 
This is, of course, only a rough generalisation and applies 
especially to temperate regions, but it explains v^hy w^e often 
get seres on very different kinds of bare areas converging 
towards the same final climax. 

7. The account of this subject given above is necessarily 
brief, and a much fuller account is given by Tansley in his 
book Practical Plant Ecology}^ which is essential to the work 
of all animal ecologists. Clements has treated the whole 
subject in stupendous detail in his monograph Plant Sue- 
cesston,^^ which is illustrated by a very fine series of photographs 
of plant communities. 

Let us now consider a few examples of succession in animal 
and plant communities. It is clearly impracticable to take 
more than a few species as examples of changes in whole 
communities, and naturally the exclusives afford the most 
striking ones. There are several ways in which animal suc- 
cession can be studied. The best way is to watch one spot 
changing over a series of years and record what happens to 
the fauna. This is the method least practised, but the most 
likely to lead to productive results, since we stand a good 
chance of seeing how the structure of the communities is 
altered as one grades into another. Yapp 3i says : '* We may 
perhaps regard the organisms, both plants and animals, occupy- 
ing any given habitat, as woven into a complex but unstable 
web of life. The character of the web may change as new 
organisms appear on the scene and old ones disappear during 
the phases of succession, but the web itself remains." It is 
just the changes in this " web " about which we know so little 
at present, and that is why study of the actual changes will 
always be the most valuable. 

8. One of the most interesting and clear-cut examples of 
succession, recorded by Ritchie,^^^ is so striking that it has 
been often quoted, and is worth quoting again here. He 
describes the manner in which a typical heather moor in the 
south of Scotland, with its normal inhabitant, the red grouse 
{Lagopus scoticus), was converted in the short space of fifteen 
years into a waste of rushes and docks, inhabited by a huge 



24 ANIMAL ECOLOGY 

colony of black-headed gulls {Larus ridibundus), and then in 
about ten years turned back again into a moor like the original 
one. These events were brought about by the arrival of a 
few pairs of gulls which nested there for the first time in about 
1892. The gulls were protected by the owner, and after 
fifteen years they had increased prodigiously until there were 
well over 3 ,000 birds nesting. The occupation of the ground 
by gulls, with its accompanying manuring and trampHng of 
the soil, caused the heather to disappear gradually and to 
give way to coarse grass. The grass was then largely replaced 
by rushes (Junciis), and the latter ultimately by a mass of 
docks (Rumex). At the same time pools of water formed 
among the vegetation and attracted numbers of teal {Anas 
creccd). The grouse meanwhile had vanished. Then pro- 
tection of the gulls ceased, and their numbers began to decrease 
again, until in 1917 there were less than sixty gulls nesting, 
the teal had practically disappeared, and the grouse were 
beginning to return. In fact, with the cessation of *' gull 
action " on the ground the place gradually returned to its 
original state as a heather moor. As Ritchie remarks, there 
must have been a huge number of similar changes among the 
lower animals which also would be profoundly affected by 
the changes in vegetation. 

9. Another striking story is that told us by Massart,^^ who 
studied the changes wrought during the war by the flooding of 
parts of Belgium in the Yser district. Here the sea was allowed 
by the Belgian engineers to inundate the country in order to 
prevent the advance of the German army. The sea-water 
killed off practically every single plant in this district, and all 
available places were very soon colonised by marine animals 
and plants, space being valuable in the sea. When the country 
was drained again at the end of the war, ecological succession 
was seen taking place on a generous scale. At first the bare 
" sea-bottom " was colonised by a flora of salt-marsh plants, 
but these gave way gradually to an almost normal vegetation 
until in many places the only traces of the advance and retreat 
of the sea were the skeletons of barnacles (Balanus) and 
mussels (Mytilus) on fences and notice-boards, and the presence 



ECOLOGICAL SUCCESSION 25 

of prawns {Palcemonetes varians) left behind in some of the 
shell holes. 

10. Ecological succession may easily be studied experi- 
mentally by making a hay infusion in water and leaving it 
exposed to the air for several weeks. Bacteria are the first 
things to become abundant, since they live upon the decaying 
vegetable matter. Then various protozoa appear, and it is 
possible to see a whole animal community being gradually 
built up, as each new species arrives and multiplies and fits 
into its proper niche. In a hay infusion the bacteria are 
followed by small ciHate protozoa of the Paramecium type, 
which subsist upon bacteria and also by absorbing substances 
in solution and in suspension in the water. Then there are 
larger hypotrichous ciliates, which prey upon bacteria and 
also upon the smaller ciliates. Eventually the whole culture 
may degenerate owing to the exhaustion of food material for 
the bacteria and therefore for the animals dependent on them. 
On the other hand, green plants, in the form of small algae, 
may arrive and colonise the culture. These will be able to 
subsist for a long time, and may change the character of the 
whole community by providing a different type of food. 
Succession in hay infusions is particularly fascinating, since it 
can be studied anywhere, and does not last over a very long 
time. 

11. Another method of determining the course of animal 
succession is to work from a knowledge of the succession 
relations of the plant communities (gained either from direct 
observation or from deduction and comparison with other 
districts) and then work out the animal communities of each 
plant zone. It is then possible to say in a general way what 
animals will replace existing ones when succession does occur. 
This was done by Shelford,20 who studied the tiger beetles 
of the genus Cicindela (carnivorous ground beetles of variegated 
colours) in a sere of plant communities on the shores of Lake 
Michigan. The lake-level has been falling gradually of late 
years, and there can be seen all stages in succession on the 
bare areas left behind on the shores. On the lake margin was 
Cicindela cuprascens^ whose larvae live in wettish sand. Young 



26 ANIMAL ECOLOGY 

cottonwoods colonise this ground, and another species of tiger 
beetle (C lepida) then replaces the first one. In the old cotton- 
woods, where conditions are different with grass and young 
pine seedlings, C. formosa took the place of C lepida. Then 
with the formation of a dominant pine community on the 
ridges still another species, C. scutellaris^ replaces the previous 
one and continues to live on into the next stage in succession, 
a black oak community ; but it gradually becomes scarcer 
with the change to white oak, until a stage is reached with no 
tiger beetles at all. With the following red oak stage there 
arrives C sexguttata^ which persists afterwards in clearings, 
but when the climax association of beech and maple has been 
reached tiger beetles again disappear altogether. Here there 
can be distinguished at least eight stages in the development of 
plant associations, and five different species of tiger beetle, 
none of which come in more than two plant zones. 

12. It is important to note that we have to deal in this case 
with a genus of animals which tends to form species which 
are exclusive or confined to one or two plant associations. 
In England there seems to be the same tendency among the 
species of Cicindela. But this kind of strict Hmitation to 
plant associations is probably rather unusual, especially among 
carnivorous animals. Exception must be made in the case of 
some of the great host of herbivorous animals (in particular 
insects) which are attached to one species of plant only. But 
even in these cases the plant itself, and therefore the animal, 
is not usually confined to one plant association. In practice, 
succession in animal communities is an infinitely more com- 
plicated affair. One reason for this is the great lag of animals 
behind plants, due to their different powers of dispersal. 
Another reason is that the survival of only a few of the earlier 
plants in the later stages will enable a great number of animals 
to hang on also, and these of course cannot be separated in 
their inter-relations with the newcomers. For instance, on 
areas in the south of England where pine woods have grown 
up over and largely replaced heather (Calluna), there are still 
many patches of heather growing in the more open places, 
and these have been found to contain the typical heather 



ECOLOGICAL SUCCESSION 27 

communities of animals, showing that the animals are in this 
case affected rather by the food, shelter, etc., provided by the 
heather than by the general physical " climate " produced by 
the pine wood.^^^ It is when we try to work out the food 
relations of the animals that the presence of small patches of 
earher pioneer animals in a climax association becomes such a 
complicating factor. In fact, succession (at any rate in animals) 
does not take place with the beautiful simplicity which we 
could desire, and it is better to realise this fact once and for 
all rather than to try and reduce the whole phenomenon to a 
set of rules which are always broken in practice ! The present 
state of our knowledge of succession is very meagre, and this 
ignorance is to a large extent due to lack of exact knowledge 
about the factors which Hmit animals in their distribution and 
numbers. The work done so far has necessarily been re- 
stricted to showing the changes in exclusive species of one 
genus, or in the picking out of one or two salient features in 
the changes as an indication of the sort of thing that is taking 
place. Work like that done by Shelf ord, and observations like 
those of Ritchie and Massart, make it quite clear that succession 
is an important phenomenon in animal Hfe ; the next stage of 
the inquiry is the discovery of the exact manner in which 
succession affects whole communities. 

13. It will be as well at this point to remind the reader that 
most of the work done so far upon animal succession has been 
static and not dynamic in character ; that the cases in which 
the whole thing has been seen to happen are few in number 
and, although extremely valuable and interesting, of necessity 
incompletely worked out. 

Given a good ecological survey of animal communities and 
a knowledge of the local plant seres, we can predict in a general 
way the course of succession among the animals, but in doing 
so we are in danger of making a good many assumptions, and 
we do not get any clear conception of the exact way in which 
one species replaces another. Does it drive the other one out 
by competition ? and if so, what precisely do we mean by 
competition ? Or do changing conditions destroy or drive 
out the first arrival, making thereby an empty niche for another 



28 ANIMAL ECOLOGY 

animal which quietly replaces it without ever becoming *' red 
in tooth and claw " at all ? Succession brings the ecologist 
face to face with the whole problem of competition among 
animals, a problem which does not puzzle most people because 
they seldom if ever think out its implications at all carefully. 
At the present time it is well known that the American grey 
squirrel is replacing the native red squirrel in various parts of 
England, but it is entirely unknown why this is occurring, 
and no good explanation seems to exist. And yet, more is 
known about squirrels than about most other animals. In 
ecological succession among animals there are thousands of 
similar cases cropping up, practically all of which are as 
little accounted for as that of the squirrels. There is plenty 
of work to do in ecology. 

14. It is probable that accurate data about the succession 
in animal communities will be most easily and successfully 
obtained by taking very small and limited communities living 
in peculiar habitats, for it is here that the number of species 
is reduced to reasonable proportions. Experience has shown 
that the general study of animal communities is best carried 
out on simple communities such as those of Arctic regions or 
of brackish water. Habitats in which succession can be studied 
quickly and conveniently are the dead bodies of animals, the 
dead bodies of plants {e.g. logs, or fungi), the dung of mammals, 
marine timber, temporary pools, and so on. The ecologist 
will be able to find a number of such habitats wherever he is ; 
and they all contain on a small scale similar communities to 
those found in woods or lakes. In such places succession is 
always in progress and, what is more important, in quick 
progress. The writer has found that it is almost impossible 
to make even a superficial study of succession in any large 
and complicated community, owing to the appaUing amount of 
mere collecting which is required, and the trouble of getting 
the collected material identified. When one has to include 
seasonal changes throughout the year as well, the work becomes 
first of all disheartening, then terrific, and finally impossible. 
Much of this mental strain can be avoided by choosing simple 
communities, and the results as contributions to the general 



ECOLOGICAL SUCCESSION 29 

theory of succession are probably just as valuable. In any 
case, it is desirable that botanists and animal ecologists should 
cooperate in such studies, and in most cases a team of several 
people is required for the proper working out of the animals. 
It is probable, however, that in the simpler cases one man could 
do very valuable work. 

15. We have not spoken so far of succession in the sea. 
Owing to the peculiar importance of sessile animals on the 
shores and slopes of the sea, it is not uncommon to find plant 
and animal succession becoming rather closely related. 
Wilson ^2 watched the succession on bare areas on the shores 
of California at La Jolla, and found that the pioneers were 
colonial diatoms which formed the first community. These 
were followed by an association of colonial hydroids (mostly 
Obelia)y and the latter were then replaced by a seaweed (Ecto- 
carpus), which became dominant after some four months. 
Further stages were foreshadowed, and it appeared that the 
whole would ultimately develop into a cHmax association of 
other seaweeds (chiefly kelp). The interesting thing about 
this sere is the fact that the second stage in succession is 
formed by sessile animals, and is sandwiched between two 
plant stages. The zonation of animals and plants in the inter- 
tidal zone in other regions is often an alternation of animal 
and plant dominance, e.g. Balanus or Mytilus and seaweeds, 
in the temperate regions. On coral islands succession may 
consist almost entirely of a series of animal zones with only an 
occasional plant zone formed of calcareous algae.io* In such 
cases it is perfectly legitimate to refer to the most abundant 
animal as the " dominant " species, but in most animal com- 
munities the term has little meaning owing to the different 
methods of feeding adopted by animals and plants. It is 
perhaps better to avoid the use of the term " dominant " in 
the cases of animals except for sessile aquatic species, since 
dominance implies occupying more space or getting more light 
than other species. 

16. We have just pointed out that animals play an important 
part in ecological succession in the sea ; but it should also be 
realised that they also have very important effects in a different 



30 ANIMAL ECOLOGY 

sort of way upon plant succession on land. Herbivorous 
animals are often the prime controlling factor in the particular 
kind of succession which takes place on an area. Farrow ^* 
says : "It is thus seen that variation in the intensity of the 
rabbit attack alone is sufficient to change the dominant type 
of vegetation in Breckland from pine woodland to dwarf 
grass-heath vegetation through the phases of Calluna heath 
and Carex arenaria^ and that for each given intensity of rabbit 
attack there is a certain associated vegetation." He showed 
that there were definite zones of vegetation round each rabbit 
colony and even each hole, since the distance from the colony 
resulted in different intensities of attack. The curious fact 
also appears that Carex arenaria will dominate Calluna when 
both are eaten down intensely. It is the relative intensity 
of attack that matters. The example quoted here is only one 
of many which could be given, although the situation has not 
been worked out so ingeniously or fully for any other animals 
or place. 

17. Hofman's work ^^ has made it probable that the in- 
fluence of rodents in burying seeds of conifers may sometimes 
determine the type of succession which starts after a forest 
fire. The Douglas Fir {Pseudotsuga taxifolia) is a dominant 
tree over large parts of the Cascade and Coast region of 
Washington and Oregon. When the seed crops are light 
they are largely destroyed by an insect {Megastigmus spermo- 
trophns) and by rodents. But when there is an unusually 
heavy crop large quantities of surplus seed are collected and 
stored in caches by the rodents. Since in many cases the 
animals do not return to the caches the seeds remain there for 
a good many years, and if there is a fire in the forest or the 
trees are cut down, large quantities of the stored seed germinate. 
More of the Douglas fir seed than of other species of trees 
such as hemlock and cedar is cached by rodents, so that the 
Douglas has accordingly an advantage in the early stages of 
succession. The same thing holds good for the white pine ; 
its seeds are much eaten by rodents, which gather them and 
store them in the ground. But here other factors such as 
germination come in and affect the succession. 



PLATE IV 




a) A semi-permanent pond in Oxfordshire (permanent except in drought years, 
when it dries up completely). Various successive zones of aquatic and 
marsh plants lead finally to grazed grassland on the right, and to elm 
trees on the left. 




{d) Zonation of habitats on the sea-shore at low tide of Norway (Svolvaer), 
showing gradation from sea to land, through seaweeds [Fuci/s, etc.), 
drift-line, shingle, rocks, up to pasture or birchwoods. 



ECOLOGICAL SUCCESSION 31 

18. One more example may be given of the way in which 
animals and plants are intimately bound up together in 
ecological succession. Cooper ^^ has studied the relation of 
the white pine blister rust to succession in the New England 
and the Adirondacks. This disease is a very important one, 
and the most critical point in its Hfe-history^ from the point of 
view of controlling it, is its occurrence at one stage on the 
various species of wild gooseberries (Ribes), Cooper found 
that the distribution of Ribes was to an important extent 
affected by the distribution of fruit-eating birds, and that 
changes in the character of the birds during ecological suc- 
cession resulted in a failure of the gooseberry seeds to spread 
in sufficient numbers to establish many new plants, after a 
certain stage in succession had been reached. The normal 
succession in places where the original forest had been cleared 
and then allowed to grow up again is as follows : first a stage 
with rank grass and weeds ; this is followed by a shrub 
stage, in which plants like raspberry, blackberry, juniper, etc., 
are important. At this point the species of RibeSy though 
relatively unimportant, reach their maximum abundance. 
The shrub stage is followed by one with trees, of which the 
most important are white pine, aspen, birch, and maple. 
Finally a climax of other species may be reached. But the 
important point is the appearance of the first trees, for at this 
stage the bird fauna changes considerably and the number of 
fruit-eating, or at any rate gooseberry- eating, birds diminishes 
suddenly. The Ribes is able to exist in the shade of these 
later forest stages of succession, but cannot produce fruit in 
any quantity, and so, unless birds bring in new seeds from out- 
side, the Ribes is bound to die out ultimately. As we have seen, 
the corresponding changes in the bird species prevent any 
large amount of seed getting into the forest after these stages, 
and this reacts upon the rust. 

19. It is obvious that a knowledge of animals may be of 
enormous value to botanists working on plant succession. 
At the same time it is necessary for the animal ecologist to 
have a good general knowledge of plant succession in the 
region where he is working. It enables him to classify and 



32 ANIMAL ECOLOGY 

understand plant communities, and therefore animal com- 
munities, much more easily than he can otherwise do, by 
providing an idea which links up a number of otherwise rather 
isolated facts. Furthermore, as will be shown later on, suc- 
cession of communities, which is nothing more than a migration 
of the animals' environment, plays an important part in the 
slow dispersal of animals. Again, as we have pointed out, 
the intricate problems of competition between different species 
of animals can be studied to advantage in a series of changing 
communities. Tansley's Types of British Vegetation i^ is the 
standard work in which the plant associations of Britain are 
described, and in many cases the probable lines of succession 
in the associations are given also. It is essential to have access 
to this book if one is carrying out any extensive work in the way 
of preliminary surveys or studies in ecological succession. 



CHAPTER IV 

ENVIRONMENTAL FACTORS 

The ecologist is (i) concerned with zohat animals do, and with the factors 
which prevent them from doing various things ; (2) the study of factors 
which Hmit species to particular habitats Hes on the borderHnes of so 
many subjects, that (3) he requires amongst other things a slight know- 
ledge of a great many scattered subjects. (4), (5), (6), (7) The ecology 
of the copepod Eurytemora can be taken as a good example of the methods 
by which such problems may be studied, and illustrates several points, 
e.g. (8) the fact that animals usually have appropriate psychological 
reactions by which they find a suitable habitat, so that (9) the ecologist 
does not need to concern himself very much with the physiological 
limits which animals can endure ; and (10) the fact that animals are 
not completely hemmed in by their environment, but by only a few 
limiting factors, which (11) may, however, be difficult to discover, 
since (12) the factor which appears to be the cause may turn out to be 
only correlated with the true cause. (13), (14) Examples of limiting 
factors to the distribution of species are : hydrogen ion concentration 
of water ; (15) water supply and shelter ; (16) temperature ; (17) 
food plants ; and (18) interrelations with other animals. The last 
subject is so huge and complicated, that it requires special treatment — 
as the study of animal communities. 

I. It will probably have occurred to the reader, if he has got 
as far as this, that rather Uttle is known about animal ecology. 
This is, of course, all to the good in one way, since one of the 
most attractive things about the subject is the fact that it is 
possible for almost any one doing ecological work on the right 
lines to strike upon some new and exciting fact or idea. At 
the same time it is often rather difficult to know what are the 
best methods to adopt in tackling the various problems which 
arise during the course of the work. This is particularly true 
of the branch of ecology dealt with in the present chapter. 
Much of the work that is done under the name of ecology 
is not ecology at all, but either pure physiology — i,€, finding 
out how animals work internally — or pure geology or me- 
teorology, or some other science concerned primarily with the 

33 D 



34 ANIMAL ECOLOGY 

outer world. In solving ecological problems we are con- 
cerned with what animals do in their capacity as whole, living 
animals, not as dead animals or as a series of parts of animals. 
We have next to study the circumstances under which they 
do these things, and, most important of all, the limiting factors 
which prevent them from doing certain other things. By 
solving these questions it is possible to discover the reasons for 
the distribution and numbers of different animals in nature. 

2. It is usual to speak of an animal as living in a certain 
physical and chemical environment, but it should always be 
remembered that strictly speaking we cannot say exactly where 
the animal ends and the environment begins — unless it is 
dead, in which case it has ceased to be a proper animal at all ; 
although the dead body forms an important historical record 
of some of the animal's actions, etc., while it was alive. The 
study of dead animals or their macerated skeletons, which has 
to form such an important and necessary part of zoological 
work, and which has bulked so largely in the interest of 
zoologists for the last hundred years, has tended to obscure 
the important fact that animals are a part of their environment. 
There are numerous gases, liquids, and solids circulating 
everywhere in nature, the study of which is carried on by 
physicists, chemists, meteorologists, astronomers, etc, ; certain 
parts of these great systems are, as it were, cut off and formed 
into little temporary systems which are animals and plants, 
and which form the objects of study of physiologists and psycho- 
logists. Ecological work is to a large extent concerned with 
the interrelations of all these different systems, and it must be 
quite clear that the study of the manner in which environmental 
factors affect animals lies on the borderline of a great many 
different subjects, and that the task of the ecologist is to be a 
sort of Haison officer between these subjects. He requires a 
slight, but not superficial, knowledge of a great many branches 
of science, and in consequence must be prepared to be rather 
unpopular with experts in those sciences, most of whom will 
view him with all the distaste of an expert for an amateur. 
At the same time, although his knowledge of these other sub- 
jects can only be slight, owing to the absolute impossibility of 



ENVIRONMENTAL FACTORS ^5 

learning them thoroughly in the time at his disposal, he has 
the satisfaction of being able to solve problems, often of great 
practical importance, which cannot possibly be solved by having 
a profound grasp of only one field of science. Many people 
find it rather a strain if their work includes more than one or 
two " subjects," and this is probably the reason why many 
who start doing ecological work end up by specialising on one 
of the ready-made subjects with which they come in contact 
in the course of their ecological researches. It is quite true 
that one is frequently held up by the absence of existing data 
about some problem in geology or chemistry, and it may be 
necessary to turn to and try to solve it oneself. It is also true 
that the division of science into water-tight compartments 
is to be avoided like the plague. At the same time there are 
problems in the reactions of animals with their environment 
which call for a special point of view and a special equipment, 
and one of the most important of these is a slight knowledge of 
a number of different subjects, if only a knowledge of whom 
to ask or where to look up the information that is required. 

3. Suppose one is studying the factors limiting the dis- 
tribution of animals living in an estuary. One would need to 
know amongst other things what the tides were (but not the 
theories as to how and why they occur in a particular way) ; 
the chemical composition of the water and how to estimate 
the chloride content (but not the reasons why silver nitrate 
precipitates sodium chloride) ; how the rainfall at different 
times of the year affected the muddiness of the water ; some- 
thing about the physiology of sulphur bacteria which prevent 
animals from living in certain parts of the estuary ; the names 
of common plants growing in salt-marshes ; something about 
the periodicity of droughts (but not the reasons for their 
occurrence). One would also have to learn how to talk 
politely to a fisherman or to the man who catches prawns, 
how to stalk a bird with field-glasses, and possibly how to 
drive a car or sail a boat. Knowing all these things, and a 
great deal more, the main part of one's work would still be 
the observation and collection of animals with a view to 
finding out their distribution and habits. Having obtained 



36 ANIMAL ECOLOGY 

the main data about the animals, one would be faced by a 
number of inexplicable and fascinating cases of limited dis- 
tribution of species, upon which would have to be focussed 
every single scrap of other knowledge which might bear on 
the problems. Having crystallised the problems it is often 
desirable to carry out a few simple experiments ; but more 
often, careful observation in the field will show that the ex- 
periment has already been done for one in nature, or by some 
one else — unintentionally. 

4. Let us take an example which will give some idea of the 
way in which the factors Hmiting the distribution of animals 
may be studied. To continue the same line of thought as 
before, we will consider Eurytemora lacinulata, a copepod 
crustacean which normally inhabits weak brackish water in 
lagoons and estuaries. In an estuarine stream near Liverpool 
which has been studied, E. lacinulata is absent from the upper 
fresh- water parts, as also from fresh- water ponds in the neigh- 
bourhood, but begins to occur in the region which is in- 
fluenced for a short time only by salt water at the height of 
the tides. In this part of the stream the environment consists 
of fresh water at low tides, brackish water (e.g. up to a salinity 
of 4 (chlorides) per mille) at high tide, with a gradation in 
between. Further downstream, under more saUne conditions, 
the Eurytemora again disappears. Near this stream there is 
a moat surrounding an old duck decoy, and filled with brackish 
water whose salinity fluctuates slightly about 4 per mille, 
and in which Eurytemora lives and flourishes. It is clea;* 
that in this region E. lacinulata lives in weak brackish water 
up to about 4 per mille, or in alternating weak brackish and 
fresh water, but neither in permanent fresh water nor in more 
saline water. We are justified in making the hypothesis that 
E. lacinulata requires a certain optimum salinity, any wide 
deviation from which is unfavourable to it. 

5. Now it has been found, on the other hand, that in the 
Norfolk Broads this species lives in the fresh-water broads 
and is almost entirely replaced in the brackish ones by an 
allied species, E. affinis. Furthermore, there are a certain 
number of curious records of the occurrence of E. lacinulata 



ENVIRONMENTAL FACTORS 37 

in inland fresh-water ponds which contain no unusual amount 
of salt in the water. For instance, it lives permanently in a 
small pond in the Oxford Botanic Gardens. Therefore 
fresh water by itself is no bar to the survival of this animal 
(if we assume, as we are doing here, that there are not several 
different physiological races of this species, which have 
different requirements in the matter of salinity). What is 
the cause of its absence from the fresh-water parts of the Liver- 
pool stream ? The chief way in which the fresh and brackish 
parts of the stream differ, apart from the salinity, is in flora 
and fauna, the upper part having different associations of 
animals from the lower. So perhaps there is some biotic 
limiting factor. This idea is supported by outside evidence. 
If we examine the cases mentioned above of its occurrence 
inland, we notice several remarkable things about them. In 
the first place, the ponds in which Eury femora occurs are few, 
and these few are all artificial and fairly new, mostly under 
twenty years of age. In the second place, it is absent from all 
older ponds, many of which are very similar to the newer ones 
in fauna, flora, and general conditions. But these older ponds 
do differ in one important respect : they all contain another 
copepod, Diaptomus gracilis, which can be shown to have bad 
powers of dispersal, and which therefore does not reach a new 
pond for some time, usually not less than thirty or forty 
years. The ponds containing Diaptomus have in them no 
Eurytemoray and the ponds containing Eurytemora have no 
Diaptomus. We can explain the latter fact by the bad dispersal 
powers of the Diaptomus, but how are we to explain the former } 
It appears that there must be some form of competition (see 
p. 27) between the two species, in which Diaptomus is success- 
ful. At this point we turn for further evidence to an experiment 
carried out with great care (although unintentionally) by the 
Oxford Corporation. It is always exhilarating to find an 
experiment already done for you on a large scale, and in nature, 
instead of having to do it yourself in a laboratory. There is a 
large pond in the Oxford waterworks in which Diaptomus 
gracilis occurs abundantly, but from which Eurytemora is 
absent. A few yards away there is a series of filter-beds 



38 ANIMAL ECOLOGY 

which are dried out and cleaned every few weeks to remove 
the conferva which grows upon them. Diaptomiis does not 
live in the filter-beds because its eggs cannot withstand being 
dried up, but there are swarms of Eurytemora lacinulata taking 
its place. In this case the possibiHty of dispersal causing the 
diflferent distribution is removed, and we have practically a 
controlled experiment. 

6. The upshot of all this is that E. lacinulata appears to be 
limited in its inland distribution to ponds which do not contain 
Diaptomus. Returning to the original stream near Liverpool, 
it now seems very likely that there are biotic factors at work 
limiting the distribution of the Eurytemora, since we have 
knocked out salinity as a possibiHty, and shown that competi- 
tion is probably an important factor elsewhere. Now, as 
Diaptomus does not occur in the stream, some other animal 
must be acting as limiting factor or competitor. So far the 
inquiry has not been pushed further, but the example has been 
given in order to show that it pays to make the investigation 
as wide as possible in order to find what is probably the best 
line to follow up. In this case one would concentrate upon the 
animal interrelations of Eurytemora, knowing that there was 
a good chance of solving the problem along that hne. 

7. There would still remain the question of the limiting 
factor or factors of Eurytemora at the other end of its range. 
This has not been studied on E, lacinulata^ but some work 
has been done upon E. raboti, and this will serve to illustrate 
methods, although of course no conclusions based on one 
species must ever be applied to another. (There have been 
not infrequently cases in which some one has repeated a 
man's work in order to confirm his experiments, using for his 
material not the original species but a closely allied one, and 
has been able to contradict the conclusions of the first man — 
not unnaturally, when we realise how enormously closely 
allied species may differ in their physiological reactions.) 

Eurytemora raboti inhabits weak brackish water in tidal 
lagoons on the coasts of Spitsbergen, and sometimes when 
these become cut oflF from the sea by rising of the land, the 
copepods survive successfully in the relict lagoons, which then 



ENVIRONMENTAL FACTORS 39 

contain only fresh water. The only factor preventing this 
species from living in other fresh-water ponds appears to be 
bad powers of dispersal, or at any rate difficulty in establishing 
itself from a few eggs carried by accidental dispersal. At the 
other end of its habitat it disappears at a certain point where 
the salinity at high tides is over about 9 per mille (chlorides). 
To find out whether high salinity itself was the limiting factor 
an experiment was tried, and it was found that the animals 
died in a salinity of 13 per mille after about two hours, while 
they were quite unharmed in a salinity of 9 per mille after 
more than three hours. The experiment showed with some 
certainty that salinity was in fact the cause of its limited dis- 
tribution in estuarine regions. -^^ 

8. When one is studying limiting factors, it is really more 
important to have a nodding acquaintance with some of the 
things which are going on in the environment, than to know 
very much about the physiology of the animals themselves. 
This statement may sound odd, and is certainly rather in 
opposition to much of the current ecological teaching, but 
there is a perfectly good reason for making it. Most animals 
have some more or less efficient means of finding and remain- 
ing in the habitat which is most favourable to them. This 
may be done by a simple tropism or by some elaborate instinct. 
Examples of animals which employ the first type of response 
are the white butterflies Pieris rapce and brassicce, which select 
the leaves of certain species of cruciferous plants for the pur- 
pose of laying their eggs, in response to the chemical stimulus 
of mustard oils contained in the leaves of the plants. i^^ An 
example of the complicated type is the African lion, which 
chooses its lair v^th great attention to a number of different 
and rather subtle factors. It is usually supposed that animals 
choose their habitats merely by avoiding all the places which 
are physiologically dangerous to them, in the same way that 
a Paramecium turns away from certain kinds of chemical 
stimuli in the water in which it is swimming. This is true in 
one sense ; but the stimuli which lead an animal to keep away 
from the wrong habitat are not usually capable of doing any 
direct harm to it, and are much more in the nature of warning 



40 ANIMAL ECOLOGY 

signals which indicate to it that if it goes much further into 
this unsuitable habitat, or remains there too long, the results 
will be dangerous. For instance, the fresh-water amphipod 
Gamfnarus pulex is a nocturnal animal, which lives under stones 
and plants during the day, its habitat being determined by a 
simple negative reaction to light. But light is not in itself 
harmful to Gammarus (except possibly in a very early stage 
of development), and the reason for living in the dark is 
almost certainly because of the protection which it gets from 
enemies, since the animal has no effective means of defending 
itself against enemies except by swimming away or hiding. 
Often, however, the signs by which animals choose their 
habitats are not warnings of danger to the animal itself, but 
have the effect of keeping it out of places in which it could 
not breed or bring up its young successfully. 

9. The point which we are trying to make is that most 
animals are, in practice, limited in their distribution by their 
habits and reactions, the latter being so adjusted that they 
choose places to live in, which are suitable to their particular 
physiological requirements or to their breeding habits. The 
latter may often be much more important than the former : 
the willow wren, Phylloscopus trochilus, and the chiff chaff, 
P. colly bita, both range through similar kinds of vegetation, 
and do not appear to be affected directly by the physical or 
biotic conditions at different levels of a wood or in different 
plant associations. At the same time, the habitats for 
breeding are very markedly different, the willow wren 
choosing places where the ground vegetation is low, while 
the chiffchaff occurs in woods which have grown up and in 
which the undergrowth has correspondingly changed. ^^^ We 
might put the matter this way : every animal has a certain 
range of external conditions in which it can live successfully. 
The ultimate Hmits of environment are set by its physiological 
make-up ; if these limits are reached the animal will die. 
It is therefore undesirable that the animal should run the risk 
of meeting such dangerous conditions, and it has various 
psychological reactions which enable it to choose, to a large 
extent, the optimum conditions for life. The animal is not 



PLATE V 




((/) Zonation of plant communities on the Lancashire sand-dunes. The upper 
parts are covered by marram [Psamiiia], the lower parts by dwarf 
willow {Salix refens). Next comes marsh on the edge of the pond and 
finally aquatic plants, such as Chara, under the water. In the back- 
ground can be seen the effects of secondary erosion bv wind. 




[b) Zones of vegetation on the edge of a tarn on a peat moor in tlie English Lake 
District. From left to right are {a) floating leaved plants (water-lily), 

[b) 'Teedswamp " of Carex mixed with floating pondweed [Potamogeton), 

[c) marsh with Carex, grading into sweet gale [Myrica] marsh, [d] 
bracken [Pteris), or coniferous woods. 



ENVIRONMENTAL FACTORS 41 

usually occupying the extreme range of conditions in which 
it could survive, since at the limits it is not so efficient, and 
because the actual habitat is usually still further limited by 
the breeding requirements. In other words, it is usually 
possible to use the psychological reactions of animals as an 
indication of their physiological " abilities," and to that extent 
it is possible to solve ecological problems without knowing a 
great deal about the physiological reasons why certain con- 
ditions are unsuitable. We simply assume that they are 
unsuitable from the fact that animals avoid places where they 
occur. (This adjustment is presumably brought about by 
the process of natural selection acting over very long periods, 
since animals which chose a habitat which turned out to be 
unsuitable would inevitably die or fail to breed successfully 
unless they could leave it in time.) By making the assumption 
that animals are fairly well adapted to their surroundings we 
certainly run a risk of making serious mistakes in a few cases, 
because owing to the lag in the operation of natural selection 
(see p. 185) animals are not by any means always perfectly 
adapted to their surroundings. But the rule is useful in a 
general way, and may prevent an ecologist from getting side- 
tracked upon purely physiological work which, however 
interesting and valuable in itself, does not throw any light on 
the ecological problems which he set out to solve. 

10. We may conclude this general discussion of the relation 
of animals to their environment by referring more particularly 
to the important general idea of limiting'^ factors. Animals 
are not completely hemmed in by their environment in any 
simple sense, but are nearly always prevented from occupying 
neighbouring habitats by one or two limiting factors only. 
For instance, in the example given above, Eurytemora lacinulata 
was quite capable of living in fresh water, but was prevented 
from doing so by some other quite different factor. It was 
adapted to live in low salinities, but was unable to realise this 
power completely. This may be said to apply to all animals. 
They are limited in any one direction by one or two factors, but 
are otherwise quite able to survive a wider range of conditions. 
Crayfish would be able to live successfully in many rivers 



42 ANIMAL ECOLOGY 

where they are not found at present if the latter contained a 
sufficiently high content of calcium carbonate, a substance 
which the crayfish needs for the construction of its exo- 
skeleton.106 in this case, a study of the oxygen requirements 
of the crayfish might throw practically no light on the reasons 
for its distribution. There are many instances of herbi- 
vorous animals which follow their food-plant into widely 
different climates from those to which they are accustomed, 
when the plant itself spreads into a new country. It may be 
taken as a rule that animals are never fully utilising all their 
possibilities, owing to the presence of a few limiting factors. 
In fact, an animal is limited by the things at which it is least 
efficient, and if these disabilities are removed it can immediately 
occupy a new range of habitat. We might put this in another 
V way and say that in order to occupy a new environment the 
animal has only to alter one or two of its psychological or 
physiological characteristics. This idea is of some importance 
from the point of view of theories of evolution and adaptation. 
II. We may now consider some examples of the way in 
which environmental factors limit the distribution of wild 
animals. Very frequently when a particular species is being 
studied, it is found that the distribution of the animal is cor- 
related with some definite feature in its environment, but it 
must not be assumed from this evidence that that factor is 
the actual limiting one which is at work. This was clearly 
seen in the case of Eurytemora, in which salinity at first sight 
appeared to be the limiting condition in its distribution, 
whereas the factor was really a different one which happened to 
be correlated with the salinity. This principle is of very wide 
application in ecological work. The habitat of any animal 
can only be accurately described in terms of actual limiting 
factors, but in practice we have to describe it in a very rough 
way by means of other factors which are correlated with the 
really important ones. When we say that an amphibian lives 
in caves, that is only a way of indicating that it lives under a 
set of environmental conditions of light, heat, and food, which 
can be roughly described as cave conditions. In such a case 
careful investigation might show that the real determining 



ENVIRONMENTAL FACTORS 43 

factor was something quite different from any of these, e.g. 
high lime-content of the water. There are so many environ- 
mental factors at work, and these are so often closely inter- 
woven in their action that it may be difficult to say exactly 
which one is playing the vital part in determining an animal's 
distribution. For instance, the humidity of the air depends 
upon a number of things, such as the temperature, wind, and 
rainfall, while these in turn may be controlled by the type of 
vegetation or the degree of exposure. 

Suppose we take Silpha quadripunctata, a beetle which 
occurs in England almost entirely in oak woods. Since it is 
carnivorous, the factors which actually limit it to this type of 
wood might be any out of a very large selection, e.g. food, 
humidity, temperature, the right colour of background, or 
breeding conditions involving the same or other factors. 
Again, the northern limit of the mouse Apodetnus sylvaticiis in 
Norway coincides roughly with latitude 62° N., in other 
words, certain temperature conditions. But another factor 
which is correlated also with latitude is the length of the night, 
and since this species of mouse is nocturnal it might be very 
well limited by this factor, since in latitude 62° the night only 
lasts five hours in the height of summer, and there is conse- 
quently only a very short period of darkness in which the mouse 
can feed. Without further work upon the general ecology of 
Apodemiis we cannot say which of these factors is the important 
one limiting its range, or whether either of them is the real 
reason. For this reason it is fairly useless to make elaborate 
" laws of distribution " based entirely upon one factor like 
temperature, as has often been done in the past. It is too 
crude a method. 

12. In field work there is not usually time or opportunity 
for studying the real limiting factors of species ; and so descrip- 
tion of habitats resolves itself into an attempt to record any 
condition with which a particular species appears to be con- 
stantly associated, even if it is a condition which is only 
correlated with the real limiting factor or factors, and has no 
significance in the life of the animal. If this is done it is 
possible for other ecologists to get at least a clear idea of the 



44 ANIMAL ECOLOGY 

exact type of habitat frequented by the animal, and perhaps 
carry the work still further. There is a species of spider 
(Leptyphantes sohrius) living in Spitsbergen, which illus- 
trates this method. Leptyphantes sobrius inhabits a great 
many types of country, occurring high up on the sides of 
mountains, or on lowlands, or even on the sea- shore, both in 
damp and dry places, and it does not appear to be very con- 
stantly associated with any particular species of animal or 
plant. At first sight we might describe its distribution as 
universal, but careful notes on the habitat of every specimen 
collected show that there is one feature in its distribution which 
is absolutely constant. It invariably comes in places where 
there are rather unstable patches of stones, in stream gullies 
or on screes, or on the shores of ponds and lakes or of the sea. 
If there is a patch of soil in one of its typical habitats which 
has become stabilised through the agency of plants and is no 
longer liable to slide away or be washed away during the spring 
snow-melting, then Leptyphantes disappears, or at any rate 
its numbers will be relatively very small, and other species of 
spiders (such as Typhochrestus spetsbergensis) will take its place. 
It is worth while in a case like this to be able to define the 
habitat of the species by some definite factor which is always 
associated with the animal's distribution, even though, as 
with Leptyphantes, we do not know in the least what the im- 
mediate reason for this distribution can be. But it should 
always be borne in mind that such descriptions of habitat 
are only a temporary expedient to assist in discovering the 
real habitat, wliich must be defined in terms of limiting 
factors. These rough habitat- descriptions are also of great 
use, since they enable one to make records which assist in the 
working out of the animal communities of different general 
habitats. 

13. Although we have said that a knowledge of physiology 
is not always necessary for the solution of ecological problems, 
it is sometimes of use, especially in cases where the animals 
with which we are dealing belong to the class which finds its 
habitat by broadcasting large numbers of young ones, so that 
only those survive which find a suitable spot. This is true of 



ENVIRONMENTAL FACTORS 45 

many marine animals, and of groups like spiders on land. 
Vallentin ^^^ states that on the Falkland Islands the waters are 
acid owing to the peaty nature of the soil, and that in winter 
this acid water runs down on to the shore and rots the shells of 
the mussels {Mytilus) which grow in vast numbers in some 
places. It is obvious that acidity is a limiting factor to an 
animal whose shell is made of calcium carbonate. This 
becomes an important factor with fresh -water molluscs . Waters 
with a hydrogen-ion concentration of less than 6 seem to be 
deficient in molluscs. But here again we are in danger of 
only establishing a correlation, since such molluscs (as also the 
crayfish) also require a certain minimum salt-content of water 
in order to build up their shells properly. In these cases the 
ideal thing is to combine very wide field observations with 
laboratory experiments in which one factor is varied at a time, 
the rest being kept constant. This was done by Saunders "^^ 
in the case of the large protozoan Spirostomum ambiguutn. 
Studies in the field showed that Spirostomum was most active 
at a pH (hydrogen-ion concentration) of about 7-4, and 
occurred most commonly in such habitats. This was 
confirmed by experiments which showed that it died in 
water of about pH 8 or at a pH below 6, the optimum 
condition being 7-4. When placed in a gradient of pH it 
migrated into the region of 7-4 (but for some reason it did 
this only in the light, regardless, however, of the direction 
from which the Hght was coming). This piece of work is an 
almost ideal example of the way to set about the study of 
limiting factors : field observation on occurrence and activity 
followed by experiments on the limits of endurance of the 
animal, the whole being clinched by putting the animal in a 
one-factor gradient in order to see what its reactions were. 

14. MacGregor ^2 has worked out the effects of hydrogen- 
ion concentration upon the distribution of certain mosquitoes 
in the south of England. There is one species {Finlaya 
geniculata) which lives as a larva in tree holes containing small 
amounts of very acid water. The normal pH of these tree- 
hole waters is well below 4-4. In experiments it was found 
that the larva develops normally in jars of water whose pH 



46 ANIMAL ECOLOGY 

was below 4-6. But when the pH was allowed to rise, the 
larvae died or suspended their development. This was also 
the case with another tree-hole mosquito {Anopheles plumbeus) 
which flourished in acid and not in alkaline water. On the 
other hand, two species of mosquitoes {Anopheles hifurcatus 
and maculipennis) normally found in alkaline ponds with a 
pH of about 8*4, died when brought up in acid water of 4-4. 
The death was partly due to the fact that the water fungus 
Saprolegnia, which is one of the greatest potential enemies of 
mosquito larvae, flourishes in acid but not in alkaline water, so 
that A. hifurcatus and maculipennis are not normally attacked 
by it in their alkaline ponds. As soon as the water is made 
acid Saprolegnia can grow and attack the larvae. Finlaya does 
not normally get attacked by Saprolegnia ^ but when the water 
is made slightly more alkaline than usual the lowered resistance 
of the larvae gives the Saprolegnia an advantage and causes it 
to become dangerous to the mosquito. 

15. In tropical countries, water-supply is often the chief 
limiting factor in the ranges of many wild birds and mammals. 
In the Burmese forests the occurrence of elephants, buffalo, 
tiger, panther, sambur, barking deer, pig, wild cat, monkeys, 
etc., is determined in the dry season by the proximity of 
water-holes, or, in the case of rooting and digging animals, 
such as pigs and moles, by the softness of the ground, which 
in turn depends on the water-supply.* In Mesopotamia the 
black partridge {Francolinus vulgaris) is never found more than 
a hundred yards from water (which it requires for drinking 
purposes). ^^ The desert quail {Lophortyx gambeli) of Cali- 
fornia is similarly limited by water-supply to areas of scrub 
by rivers."^ 

Shelter is sometimes the determining factor in distribution. 
In North America the round-tailed ground squirrel {Citellus 
tereticaudus), which inhabits deserts, is confined to sandy 
places which have bushes, since it is diurnal and therefore 
exposed to the rigours of the hot sun, which it avoids by running 
from one bush to the next. The big desert kangaroo-rat 
{Dipodomys deserti), which inhabits similar sandy places, is not 

* This information was supplied by Capt. C. R. Robbins. 



ENVIRONMENTAL FACTORS 47 

limited by the presence of bushes, since it comes out at night ; 
but it requires deep sand in order that its burrows may be 
sufficiently cool during the day. The reaUty of these limiting 
factors is shown by the fact that both rodents die or are ex- 
tremely affected if exposed to the sun for any length of time.^'' 

16. In many cases temperature has a direct effect in limiting 
the activities or in controlling the development of animals. 
Austin 37 showed that the larvae of the house fly {Musca 
domestica) died at temperatures of 105° F. or over, and that 
if piles of horse manure in which there were fly larvae living 
were close-packed, the heat generated by decay inside the 
heap was sufficient to kill off the larvae except at the edges, 
and that they actually migrate away from the hotter parts 
towards the periphery. Temperature-regulated animals like 
mammals may also, of course, be severely affected by heat, as 
shown by the fact that nearly all desert mammals are nocturnal ; 
while the Alaskan fur seal, which is normally adapted to the 
cold foggy climate of the Pribiloff Islands, suffers a great deal 
if the temperature rises above 46° to 48° F. At this tem- 
perature they show great distress when moving about and 
fighting, while at 55° to 60° they lie about motionless except 
that they all fan themselves vigorously with their flippers ."^^ 

17. Since we do not intend in this book to do more than 
describe certain principles and methods in ecology, illustrated 
by examples, we need not enumerate any more examples of 
the action of physical and chemical factors in limiting the 
distribution of animals. We have not, however, mentioned 
any cases of biotic limiting factors, and we shall therefore 
conclude this chapter by mentioning a few out of the enormous 
number which exist. It is one of the commonest things in 
nature to find a herbivorous animal which is attached solely to 
one plant either for food, or for breeding purposes, or for both. 
It is hardly necessary to quote examples from amongst insects, 
since they are so numerous. Such Hmited choice of plants for 
food or nesting is not so commonly found among the higher 
vertebrates, but there are some interesting examples. The 
beaver {Castor fiber) used to occur all over North America from 
the Colorado River and Northern Florida right up to Alaska 



48 ANIMAL ECOLOGY 

and Labrador, and its northern limit coincides exactly with the 
northern limit of the aspen {Populus tremuloides)^ which is its 
favourite and, in those regions, most easily available food. 
The southern range of the beaver overlaps that of the aspen 
enormously, and in that part of America the beaver uses, or 
once used, other trees instead. "^^ This example shows the 
way in which factors are often only limiting to distribution in 
one part of an animal's range. 

1 8. A species may depend on a plant and an animal, both 
of which may act as limiting factors in its distribution. The 
elf owl {Micropallas zvhttneyi), which inhabits parts of the 
deserts of CaHfornia and Arizona, is only found in places 
where the giant cactus {Cereiis giganteiis) grows, since it nests 
exclusively in that plant. But it is also dependent upon two 
species of woodpecker {Centurus uropygialis and Colaptes 
chrysoides mearnsi) which also nest in the cactus, and whose 
old nesting-holes are used by the owl. Buxton 20 says : 
** Less strictly dependent upon the cactus and woodpeckers 
are a screech owl {Otus asio gilmani)^ a sparrow-hawk {Falco 
sparverius), a flycatcher {Myiarchus c. cinerascens)^ and other 
birds. None of these are confined to the area inhabited by 
the Giant Cactus, but they all inhabit that area, and within it 
they all use old woodpecker holes as nesting sites. Frequently 
a single trunk of the Giant Cactus contains nests of one or 
other woodpecker, and also of one of the birds which use the 
old woodpecker holes. Honey-bees also use these excavations 
as hives. One must remember that the number of living 
creatures which eventually depend upon the Giant Cactus 
includes the scavengers in the birds' nests and bees' nests, 
the insects, few though they may be, which devour it or fre- 
quent its blossoms, and many others. All these organisms 
depend upon the growth of Cereiis giganteiis for their existence 
in certain areas." 

19. The example that has just been given illustrates the 
difference between an environmental factor which acts on an 
animal in the ordinary way, and one which acts as a limiting 
factor. The woodpeckers are both affected by the presence 
of the giant cactus, and yet not, as species, absolutely dependent 



ENVIRONMENTAL FACTORS 49 

upon it, whereas the elf owl is. The other birds mentioned 
are not strictly dependent upon the woodpecker, although its 
presence affects them enormously, while the elf owl is entirely 
linked up in its distribution with the woodpecker holes. It 
further shows that the working out of biotic limiting factors is 
far more difficult and complex than similar work on physical 
and chemical factors, since there are such elaborate inter- 
relations between the different species. We are therefore 
compelled to study more closely the relations between different 
animals, and this leads us on to a consideration of animal 
communities. 




CHAPTER V 



THE ANIMAL COMMUNITY 

" The large fish eat the small fish ; the small fish eat the water insects ; 
the water insects eat plants and mud." 
" Large fowl cannot eat small grain." 
" One hill cannot shelter two tigers." — Chinese Proverbs. 

Every animal is (i, 2) closely linked with a number of other animals living 
round it, and these relations in an animal community are largely food 
relations. (3) Man himself is in the centre of such an animal com- 
munity, as is shown by his relations to plague-carrying rats and (4) to 
malaria or the diseases of his domestic animals, e.g. liver-rot in sheep. 
(5) The dependence of man upon other animals is best shown when he 
invades and upsets the animal communities of a new country, e.g. the 
white man in Hawaii. (6) These interrelations between animals appear 
fearfully complex at first sight, but are less difficult to study if the follow- 
ing four principles are realised : (7) The first is that of Food-chains 
and the Food-cycle. Food is one of the most important factors in the 
life of animals, and in most communities (8) the species are arranged 
in food-chains which (9) combine to form a whole food-cycle. This 
is closely bound up with the second principle, (10) the Size of Food. 
Although animals vary much in size, any one species of animal only 
eats food between certain limits of size, both lower and (11) upper, which 
(12) are illustrated by examples of a toad, a fly, and a bird. (13) This 
principle applied to primitive man, but no longer holds for civilised 
man, and (14) although there are certain exceptions to it in nature, it 
is a principle of great importance. (15) The third principle is that of 
Niches. By a niche, is meant the animal's place in its community, its 
relations to food and enemies, and to some extent to other factors also. 
(16), (17), (18), (19), (20) A number of examples of niches can be given, 
many of which show that the same niche may be filled by entirely different 
animals in different parts of the world. (21) The fourth idea is that 
of the Pyramid of Numbers in a community, by which is meant the 
greater abundance of animals at the base of food-chains, and the com- 
parative scarcity of animals at the end of such chains. (22) Examples 
of this principle are given, but, as is the case with all work upon animal 
communities, good data are very scarce at present. 

I. If you go out on to the Malvern Hills in July you will 
find some of the hot limestone pastures on the lower slopes 
covered with ant-hills made by a Httle yellow ant {Acanthomyops 
flavus). These are low hummocks about a foot in diameter, 

50 



THE ANIMAL COMMUNITY 51 

clothed with plants, some of which are different from those of 
the surrounding pasture. This ant, itself forming highly 
organised colonies, is the centre of a closely-knit community 
of other animals. You may find green woodpeckers digging 
great holes in the ant-hills, in order to secure the ants and their 
pupae. If you run up quickly to one of these places, from which 
a woodpecker has been disturbed, you may find that a robber 
ant {Myrmica scabrinodis) has seized the opportunity to carry 
off one of the pupae left behind by the yellow ants in their 
flight. The latter with unending labour keep building up 
the hills with new soil, and on this soil there grows a special 
set of plants. Wild thyme {Thymus serpylhim) is particularly 
common there, and its flowers attract the favourable notice 
of a red-tailed bumble-bee (Bombus lapidarius) which visits 
them to gather nectar. Another animal visits these ant-hills 
for a different purpose : rabbits, in common with many other 
mammals, have the peculiar habit of depositing their dung in 
particular spots, often on some low hummock or tree-stump. 
They also use ant-hills for this purpose, and thus provide 
humus which counteracts to some extent the eroding effects 
of the woodpeckers. It is interesting now to find that wild 
thyme is detested by rabbits as a food,^^^ which fact perhaps 
explains its prevalence on the ant-hills. There is a moth 
{Pempelia subornatella) whose larvae make silken tubes among 
the roots of wild thyme on such ant-hills ; then there is a 
great army of hangers-on, guests, and parasites in the nests 
themselves ; and so the story could be continued indefinitely. 
But even this slight sketch enables one to get some idea of 
the complexity of animal interrelations in a small area. 

2. One might leave the ants and follow out the effects of 
the rabbits elsewhere. There are dor-beetles (Geotrupes) 
which dig holes sometimes as much as four feet deep, in wliich 
they store pellets of rabbit- dung for their own private use. 
Rabbits themselves have far-reaching effects upon vegetation, 
and in many parts of England they are one of the mqst im- 
portant factors controlling the nature and direction of ecological 
succession in plant communities, owing to the fact that they 
have a special scale of preferences as to food, and eat down 



52 ANIMAL ECOLOGY 

some species more than others. Some of the remarkable 
resuhs of " rabbit action " on vegetation may be read about in 
a very interesting book by Farrow, i^ Since rabbits may influ- 
ence plant communities in this way, it is obvious that they have 
indirectly a very important influence upon other animals also. 
Taking another Hne of investigation, we might follow out the 
fortunes and activities of the green woodpeckers, to find them 
preying on the big red and black ant (Formica rufa) which 
builds its nests in woods, and which in turn has a host of other 
animals linked up with it. 

If we turned to the sea, or a fresh- water pond, or the inside 
of a horse, we should find similar communities of animals, 
and in every case we should notice that food is the factor which 
plays the biggest part in their lives, and that it forms the 
connecting link between members of the communities. 

3. In England we do not realise sufficiently vividly that 
man is surrounded by vast and intricate animal communities, 
and that his actions often produce on the animals effects which 
are usually quite unexpected in their nature — that in fact man 
is only one animal in a large community of other ones. This 
ignorance is largely to be attributed to town life. It is no 
exaggeration to say that our relations with the other members 
of the animal communities to which we belong have had a big 
influence on the course of history. For instance : the Black 
Death of the Middle Ages, which killed off more than half the 
people in Europe, was the disease which we call plague. Plague 
is carried by rats, which may form a permanent reservoir of 
the plague bacilli, from which the disease is originally trans- 
mitted to human beings by the bites of rat fleas. From this 
point it may either spread by more rat fleas or else under certain 
conditions by the breathing of infected air. Plague was still 
a serious menace to Hfe in the seventeenth century, and finally 
flared up in the Great Plague of London in 1665, which swept 
away some hundred thousand people. Men at that time were 
still quite ignorant of the connection between rats and the 
spread of the disease, and we even find that orders were given 
for the destruction of cats and dogs because it was suspected 
that they were carriers of plague. ^^^ And there seemed no 



THE ANIMAL COMMUNITY 53 

reason why plague should not have continued indefinitely to 
threaten the lives of people in England ; but after the end of 
the seventeenth century it practically disappeared from this 
country. This disappearance was partly due to the better 
conditions under which people were living, but there was also 
another reason. The dying down of the disease coincided 
with certain interesting events in the rat world. The common 
rat of Europe had been up to that time the Black or Ship Rat 
{R. rattus), which is a very effective plague-carrier owing to 
its habit of living in houses in rather close contact with man. 
Now, in 1727 great hordes of rats belonging to another species, 
the Brown Rat (R. norvegicus), were seen marching westwards 
into Russia, and swimming across the Volga. This invasion 
was the prelude to the complete occupation of Europe by 
brown rats.^'^ Furthermore, in most places they have driven 
out and destroyed the original black rats (which are now 
chiefly found on ships), and at the same time have adopted 
habits which do not bring them into such close contact with 
man as was the case with the black rat. The brown rat went 
to live chiefly in the sewers which were being installed in some 
of the European towns as a result of the onrush of civilisation, 
so that plague cannot so easily be spread in Europe nowadays 
by the agency of rats. These important historical events 
among rats have probably contributed a . great deal to the 
cessation of serious plague epidemics in man in Europe, 
although they are not the only factors which have caused a 
dying down of the disease. But it is probable that the small 
outbreak of plague in Suffolk in the year 19 10 was prevented 
from spreading widely owing to the absence of very close 
contact between man and rats."^! We have described this 
example of the rats at some length, since it shows how events 
of enormous import to man may take place in the animal world, 
without any one being aware of them. 

4. The history of malaria in Great Britain is another 
example of the way in which we have unintentionally interfered 
with animals and produced most surprising results. Up to 
the end of the eighteenth century malaria was rife in the low- 
lying parts of Scotland and England, as also was liver-rot in 



54 ANIMAL ECOLOGY 

sheep. No one in those days knew the causes or mechanisms 
of transmission of either of these two diseases ; but at about 
that time very large parts of the country were drained in order 
to reclaim land for agricultural purposes, and this had the 
effect of practically wiping out malaria and greatly reducing 
liver-rot — quite unintentionally ! We know now that malaria 
is caused by a protozoan which is spread to man by certain 
blood-sucking mosquitoes whose larvae live in stagnant water, 
and that the larva of the liver-fluke has to pass through one 
stage of its life-history in a fresh-water snail (usually Limncea 
truncatula). The existence of malaria depends on an abun- 
dance of mosquitoes, while that of liver-rot is bound up with 
the distribution and numbers of the snail. With the draining 

of land both these animals disappeared or became much 
rarer.isb 

5. On the whole, however, we have been settled in this 
country for such a long time that we seem to have struck a 
fairly level balance with the animals around us ; and it is because 
the mechanism of animal society runs comparatively smoothly 
that it is hard to remember the number of important ways in 
which wild animals affect man, as, for instance, in the case 
of earthworms which carry on such a heavy industry in the 
soil, or the whole delicately adjusted process of control of the 
numbers of herbivorous insects. It is interesting therefore 
to consider the sort of thing that happens when man invades 
a new country and attempts to exploit its resources, disturbing 
in the process the balance of nature. Some keen gardener, 
intent upon making Hawaii even more beautiful than before, 
introduced a plant called Lantana camaray which in its native 
home of Mexico causes no trouble to anybody. Meanwhile, 
some one else had also improved the amenities of the place 
by introducing turtle-doves from China, which, unlike any of 
the native birds, fed eagerly upon the berries of Lantana. The 
combined effects of the vegetative powers of the plant and 
the spreading of seeds by the turtle-doves were to make the 
Lantana multiply exceedingly and become a serious pest on 
the grazing country. Indian mynah birds were also intro- 
duced, and they too fed upon Lantana berries. After a few 



PLATE VI 




(a) A typical animal community in the plankton of a tarn in the English Lake 
District. Three important key-industry animals are shown : Diapto7nus, 
Daph/iia, and Bosinina. 




{p) Effect of " rabbit pressure " on grass and furze on the Malvern Mills. The 
plants are closely nibbled by rabbits. The white web on the furze bush 
was constructed by a minute mite {Erythrceus regalis, Koch, var.) 
which was present in enormous numbers. 



# 



THE ANIMAL COMMUNITY 



55 



years the birds of both species had increased enormously in 
numbers. But there is another side to the story. Formerly 
the grasslands and young sugar-cane plantations had been 
ravaged yearly by vast numbers of army- worm caterpillars, 
but the mynahs also fed upon these caterpillars and succeeded 
to a large extent in keeping them in check, so that the outbreaks 
became less severe. About this time certain insects were 
introduced in order to try and check the spread of Lantana, 
and several of these (in particular a species of Agromyzid 
fly) did actually destroy so much seed that the Lantana began 
to decrease. As a result of this, the mynahs also began to 
decrease in numbers to such an extent that there began to 
occur again severe outbreaks of army- worm caterpillars. It 
was then found that when the Lantana had been removed 
in many places, other introduced shrubs came in, some of 
which are even more difficult to eradicate than the original 
Lantana?^ 

6. It is clear that animals are organised into a complex 
society, as complex and as fascinating to study as human 
society. At first sight we might despair of discovering any 
general principles regulating animal communities. But care- 
ful study of simple communities shows that there are several 
principles which enable us to analyse an animal community 
into its parts, and in the light of which much of the apparent 
complication disappears. These principles will be considered 
under four headings : 

A. Food-chains and the food-cycle. 

B. Size of food. 

C. Niches. 

D. The pyramid of numbers. 

Food-chains and the Food-cycle 

7. We shall see in a later chapter what a vast number of 
animals can be found in even a small district. It is natural 
to ask : *' What are they all doing ? " The answer to this is 
in many cases that they are not doing anything. All cold- 
blooded animals and a large number of warm-blooded ones 



56 ANIMAL ECOLOGY 

spend an unexpectedly large proportion of their time doing 
nothing at all, or at any rate, nothing in particular. For 
instance, Percival ^~^ says of the African rhinoceros : *' After 
drinking they play . . . the rhino appears at his best at night 
and gambols in sheer lightness of heart. I have seen them 
romping like a lot of overgrown pigs in the neighbourhood 
of the drinking place." 

Animals are not always struggling for existence, but when 
they do begin, they spend the greater part of their lives eating. 
Feeding is such a universal and commonplace business that 
we are incHned to forget its importance. The primary driving 
force of all animals is the necessity of finding the right kind 
of food and enough of it. Food is the burning question in 
animal society, and the whole structure and activities of the 
community are dependent upon questions of food-supply. 
We are not concerned here with the various devices employed 
by animals to enable them to obtain their food, or with the 
physiological processes which enable them to utilise in their 
tissues the energy derived from it. It is sufficient to bear in 
mind that animals have to depend ultimately upon plants for 
their supplies of energy, since plants alone are able to turn 
raw sunlight and chemicals into a form edible to animals. 
Consequently herbivores are the basic class in animal society. 
Another difference between animals and plants is that while 
plants are all competing for much the same class of food, 
animals have the most varied diets, and there is a great diver- 
gence in their food habits. The herbivores are usually preyed 
upon by carnivores, which get the energy of the sunlight at 
third-hand, and these again may be preyed upon by other 
carnivores, and so on, until we reach an animal which has no 
enemies, and which forms, as it were, a terminus on this food- 
cycle. There are, in fact, chains of animals linked together 
by food, and all dependent in the long run upon plants. We 
refer to these as " food-chains," and to all the food-chains in 
a community as the " food-cycle." 

8. Starting from herbivorous animals of various sizes, 
there are as a rule a number of food-chains radiating outwards, 
in which the carnivores become larger and larger, while the 



THE ANIMAL COMMUNITY 57 

parasites are smaller than their hosts. For instance, in a 
pine wood there are various species of aphids or plant-lice, 
which suck the juices of the tree, and which are preyed on by 
spiders. Small birds such as tits and warblers eat all these 
small animals, and are in turn destroyed by hawks. In an 
oak wood there are worms in the soil, feeding upon fallen leaves 
of plants, and themselves eaten by thrushes and blackbirds, 
which are in turn hunted and eaten by sparrow-hawks. In the 
same wood there are mice, one of whose staple foods is acorns, 
and these form the chief food of the tawny owl. In the sea, 
diatoms form the basic plant food, and there are a number of 
Crustacea (chiefly copepods) which turn these algae into food 
which can be eaten by larger animals. Copepods are living 
winnowing fans, and they form what may be called a *' key- 
industry " in the sea. The term " key-industry " is a useful 
one, and is used to denote animals which feed upon plants 
and which are so numerous as to have a very large number of 
animals dependent upon them. This point is considered 
again in the section on " Niches." 

9. Extremely little work has been done so far on food- 
cycles, and the number of examples which have been worked 
out in even the roughest way can be counted on the fingers of 
one hand. The diagram shown in Fig. 3 shows part of a 
marine plankton community, which has been studied by 
Hardy ,102 ^nd which is arranged to show the food-chains leading 
up to the herring at different times of the latter's life. To 
complete the picture we should include the dogfish, which 
attacks the herring itself. Fig. 4 shows the food- cycle on a 
high arctic island, and is chosen because it is possible in such a 
place to work out the interrelations of its impoverished fauna 
fairly completely. 

At whatever animal community we look, we find that it is 
organised in a similar way. Sometimes plants are not the 
immediate basis of the food-cycle. This is the case with 
scavengers, and with such associations as the fauna of tem- 
porary fresh-water pools and of the abyssal parts of the sea 
where the immediate basic food is mud and detritus ; and the 
same is true of many parasitic faunas. In all these cases, which 



58 



ANIMAL ECOLOGY 



ADULT HERRING 




Fig. 3. — Diagram showing the general food relations of the herring to 
other members of the North Sea plankton community. Note the effect of 
herring size at different ages upon its food. (From Hardy.^^^) 



KirriwAKG 
cuiulemots 

FULMA(\ PErO.EL 
UTTue AUK 



INOKTHEFW e.ll>afK 
LOM<;-rAil.e3> jMrfCK 




Fr^J«» Wo-ter 



X 



T/\Rl>(<;HA2>A 

OLICOCHAET-A 

NfiMATOiA 



»--rw<. t.i.tf< 



Fig. 4. — Food-cycle among the animals on Bear Island, a barren spot 
in the arctic zone, south of Spitsbergen. (The dotted lines represent 
probable food relations not yet proved.) The best way to read the 
diagram is to start at " marine animals "and follow the arrows. (From 
Summerhayes and Elton. ^^) 



THE ANIMAL COMMUNITY 59 

are peculiar, the food-supply is of course ultimately derived 
from plants, but owing to the isolation of the animals it is 
convenient to treat them as a separate community. 

Certain animals have succeeded in telescoping the par- 
ticular food-chain to which they belong. The whale-bone 
whale manages to collect by means of its sieve-like apparatus 
enough copepods and pteropods to supply its vast wants, and 
is not dependent on a series of intermediate species to produce 
food large enough for it to deal with effectively. This leads 
us on to a more detailed consideration of the problem of 

Size of Food 

10. Size has a remarkably great influence on the organisa- 
tion of animal communities. We have already seen how 
animals form food-chains in which the species become pro- 
gressively larger in size or, in the case of parasites, smaller in 
size. A little consideration will show that size is the main 
reason underlying the existence of these food-chains, and that 
it explains many of the phenomena connected with the food- 
cycle. 

There are very definite limits, both upper and lower, to the 
size of food which a carnivorous animal can eat. It cannot 
catch and destroy animals above a certain size, because it is 
not strong or skilful enough. In the animal world, fighting 
weight counts for as much as it does among ourselves, and a 
small animal can no more tackle a large one successfully than 
a light-weight boxer can knock out a trained man four stone 
heavier than himself. This is obvious enough in a broad 
way ; spiders do not catch elephants in their webs, nor do 
water scorpions prey on geese. Also the structure of an 
animal often puts limits to the size of food which it can get 
into its mouth. At the same time a carnivore cannot subsist 
on animals below a certain size, because it becomes impossible 
at a certain point to catch enough in a given time to supply 
its needs. If you have ever got lost on the moors and tried 
to make a square meal off bilberries, you will at once see the 
force of this reasoning. It depends, however, to a large 
extent on the numbers of the prey : foxes find it worth while 



6o ANIMAL ECOLOGY 

to live entirely on mice in the years when the latter are very 
abundant, but prey on larger animals like rabbits at other 
times. 

11. It is thus plain that the size of the prey of carnivorous 
animals is limited in the upward direction by its strength and 
ability to catch the prey, and in the downward direction by the 
feasibility of getting enough of the smaller food to satisfy its 
needs, the latter factor being also strongly influenced by the 
numbers as well as by the size of its food. The food of every 
carnivorous animal lies therefore between certain size limits, 
which depend partly on its own size and partly on other factors. 
There is an optimum size of food which is the one usually eaten, 
and the limits actually possible are not usually realised in prac- 
tice. (It is as well to point out that herbivorous animals are 
not strictly limited by the size of their plant-food, except in 
special cases such as seed-eating birds, honey-collecting 
insects etc., owing to the fact that the plants cannot usually 
run away, or make much resistance to being eaten.) We have 
very little information as to the exact relative sizes of enemies 
and their prey, but future work will no doubt show that the 
relation is fairly regular throughout all animal communities. 

12. Three examples will serve to illustrate the part played 
by size. There lives in the forests round Lake Victoria a kind 
of toad which is able to adjust its size to the needs of the 
moment. When attacked by a certain snake the toad swells 
itself out and becomes puffed up to such an extent that the 
snake is quite unable to cope with it, and the toad thus achieves 
its object, unlike the frog in ^sop's fable.^'^'^ Carpenter ^^ has 
pointed out another curious case of the importance of size in 
food. The tsetse fly {Glossina palpalis)^ whose ecology was 
studied by him in the region of Lake Victoria, can suck the 
blood of many mammals and birds, in which the size of the 
blood corpuscles varies from 7 to iS^ut, but is unable to suck 
that of the lungfish, since the corpuscles of the latter (41^ in 
diameter) are too large to pass up the proboscis of the fly. 
A third case is that noticed by Vallentin ^^^ in the Falkland 
Islands. He found that the black curlew {Hcematopus quoyi) 
ate limpets {Patella cenea) on the rocks at low tide, but was only 



THE ANIMAL COMMUNITY 6i 

able to dislodge those of moderate size, not usually more than 
45 millimetres across. 

13. These are three rather curious cases of what is a 
universal phenomenon. Man is the only animal which can 
deal with almost any size of food, and even he has only 
been able to do this during the later part of his history. 
It appears that the very early ancestors of man must have 
eaten food of a very limited range of size — such things as 
shellfish, fruits, mushrooms, and small mammals. Later on, 
man developed the art of hunting and trapping large animals, 
and he was thus able to increase the size of his food in the 
upward direction, and this opened up possibilities of obtaining 
food in greater bulk and variety. After the hunting stage 
came the agricultural stage, and this consisted essentially in 
the further development of the use of large animals, now in a 
domesticated state, and in the invention of means of dealing 
with foods much smaller than had previously been possible, 
by obtaining great quantities of small seeds in a short time. 
All other animals except man have their food strictly confined 
within rather narrow limits of size. The whale-bone whale 
can feed on tiny Crustacea not a thousandth of its bulk, while 
the killer whale can destroy enormous cuttle-fish ; but it is 
only man who has the power of eating small, large, and medium- 
sized foods indiscriminately. This is one of the most im- 
portant ways in which man has obtained control over his 
surroundings, and it is pretty clear that if other animals had 
the same power, there would not be anything like the same 
variety and specialisation that there is among them, since the 
elaborate and complex arrangements of the food-cycles of 
animal communities would automatically disappear. For the 
very existence of food- chains is due mainly to the fact that any 
one animal can only live on food of a certain size. Each stage 
in an ordinary food-chain has the effect of making a smaller 
food into a larger one, and so making it available to a larger 
animal. But since there are upper and low^er limits to the 
size of animals, a progressive food-chain cannot contain more 
than a certain number of links, and usually has less than five. 

14. There is another reason why food-chains stop at a 



62 ANIMAL ECOLOGY 

certain point ; this is explained in the section on the Pyramid 
of Numbers. Leaving aside the question of parasites at 
present, it may be taken as a fairly general rule that the enemy 
is larger than the animal upon which it preys. (This idea is 
contained in the usual meaning of the word " carnivore.") 
But such is not invariably the case. Fierceness, skill, or some 
other special adaptation can make up for small size. The 
arctic skua pursues and terrorises kittiwake gulls and compels 
them to disgorge their last meal. It does this mainly by naked 
bluff, since it is, as a matter of fact, rather less in weight than 
the gull, but is more determined and looks larger owing to a 
great mass of fluffy feathers. In fact, when we are dealing 
v^dth the higher animals such as birds, mammals, and the 
social ants and bees, the psychology of the animals very often 
plays a large part in determining the relative sizes of enemies 
and their prey. Two types of behaviour may be noticed. 
The strength of the prey and therefore its virtual size may be 
reduced ; this is done by several devices, of which the com- 
monest are poison and fear. Some snakes are able to paralyse 
and kill by both these methods, and so can cope with larger 
animals than would otherwise be possible. Stoats are able 
to paralyse rabbits with fear, and so reduce the speed and 
strength of the latter. It is owing to this that the stoat can 
be smaller than its prey. The fox, which does not possess this 
power of paralysing animals with fear, is considerably larger 
than the rabbit. The second point is that animals are able 
to increase their own effective size by flock tactics. Killer 
whales in the Antarctic seas have been seen to unite in parties 
of three or four in order to break up the thick ice upon which 
seals, their prey, are sleeping.^^^ Wolves are another example. 
Most wolves are about half the linear size of the deer which 
they hunt, but by uniting in packs they become as formidable 
as one very large animal. The Tibetan wolf, which eats small 
gazelles, etc., hunts singly or in twos and threes.^^^ On the 
other hand, herbivores often band together in flocks in order 
to increase their own powers of defence. This usually means 
increased strength, but other factors come in too. Ants have 
achieved what is perhaps the most successful solution of the 



THE ANIMAL COMMUNITY 63 

size problem, since they form organised colonies whose size 
is entirely fluid according to circumstances. Schweitzer ^8 
noted a column of driver ants in Angola march past for thirty- 
six hours. They are able by the mass action of their terrible 
battalions to destroy animals many times their own size 
{e.g. whole litters of the hunting dog ^"^), and at the same 
time can carry the smallest of foods. 

It must be remembered, therefore, that the idea of food- 
chains of animals of progressively larger size is only true in a 
general way, and that there are a number of exceptions. 
Having considered the far-reaching effects of size on the organi- 
sation of animal communities, we are now in a position to 
consider the subject of 

Niches 

15. It should be pretty clear by now that although the 
actual species of animals are different in different habitats, 
the ground plan of every animal community is much the same. 
In every community we should find herbivorous and carni- 
vorous and scavenging animals. We can go further than this, 
however : in every kind of wood in England we should find 
some species of aphid, preyed upon by some species of lady- 
bird. Many of the latter live exclusively on aphids. That is 
why they make such good controllers of aphid plagues in 
orchards. When they have eaten all the pest insects they just 
die of starvation, instead of turning their attention to some 
other species of animal, as so many carnivores do under similar 
circumstances. There are many animals which have equally 
well-defined food habits. A fox carries on the very definite 
business of killing and eating rabbits and mice and some kinds 
of birds. The beetles of the genus Stenus pursue and catch 
springtails (Collembola) by means of their extensile tongues. 
Lions feed on large ungulates — in many places almost entirely 
zebras. Instances could be multiplied indefinitely. It is 
therefore convenient to have some term to describe the status 
of an animal in its community, to indicate what it is doing 
and not merely what it looks like, and the term used is '* niche." 
Animals have all manner of external factors acting upon them — 



64 ANIMAL ECOLOGY 

chemical, physical, and biotic — and the " niche " of an animal 
means its place in the biotic environment, its relations to food 
and enemies. The ecologist should cultivate the habit of 
looking at animals from this point of view as well as from the 
ordinary standpoints of appearance, names, affinities, and past 
history. When an ecologist says '' there goes a badger " he 
should include in his thoughts some definite idea of the animal's 
place in the community to which it belongs, just as if he had 
said *' there goes the vicar." 

1 6. The niche of an animal can be defined to a large extent 
by its size and food habits. We have already referred to the 
various key-industry animals which exist, and we have used 
the term to denote herbivorous animals which are sufficiently 
numerous to support a series of carnivores. There is in every 
typical community a series of herbivores ranging from small 
ones (e.g. aphids) to large ones (e.g. deer). Within the herbi- 
vores of any one size there may be further differentiation 
according to food habits. Special niches are more easily 
distinguished among carnivores, and some instances have 
already been given. 

The importance of studying niches is partly that it enables 
us to see how very different animal communities may resemble 
each other in the essentials of organisation. For instance, 
there is the niche which is filled by birds of prey which eat 
small mammals such as shrews and mice. In an oak wood 
this niche is filled by tawny owls, while in the open grassland 
it is occupied by kestrels. The existence of this carnivore 
niche is dependent on the further fact that mice form a definite 
herbivore niche in many different associations, although the 
actual species of mice may be quite different. Or w^e might 
take as a niche all the carnivores which prey upon small 
mammals, and distinguish them from those which prey upon 
insects. When we do this it is immediately seen that the 
niches about which we have been speaking are only smaller 
subdivisions of the old conceptions of carnivore, herbivore, 
insectivore, etc., and that we are only attempting to give more 
accurate and detailed definitions of the food habits of animals. 

17. There is often an extraordinarily close parallelism 



THE ANIMAL COMMUNITY 65 

between niches in widely separated communities. In the 
arctic regions we find the arctic fox which, among other 
things, subsists upon the eggs of guillemots, while in winter 
it relies partly on the remains of seals killed by polar bears. 
Turning to tropical Africa, we find that the spotted hyaena 
destroys large numbers of ostrich eggs, and also lives largely 
upon the remains of zebras killed by lions.12^ The arctic fox 
and the hyasna thus occupy the same two niches — the former 
seasonally, and the latter all the time. Another instance is 
the similarity between the sand-martins, which one may see 
in early summer in a place like the Thames valley, hawking 
for insects over the river, and the bee-eaters in the upper part 
of the White Nile, which have precisely similar habits. Both 
have the same rather distinct food habits, and both, in addition, 
make their nests in the sides of sand cliffs forming the edge 
of the river valleys in which they live. (Abel Chapman ®^° 
says of the bee- eaters that *' the whole cliff- face appeared 
aflame with the masses of these encarmined creatures.") 
These examples illustrate the tendency which exists for 
animals in widely separated parts of the world to drift into 
similar occupations, and it is seen also that it is convenient 
sometimes to include other factors than food alone when 
describing the niche of any animal. Of course, a great many 
animals do not have simple food habits and do not confine 
themselves religiously to one kind of food. But in even these 
animals there is usually some regular rhythm in their food 
habits, or some regularity in their diverse foods. As can be 
said of every other problem connected with animal com- 
munities, very little deliberate work has been done on the 
subject, although much information can be found in a scattered 
form, and only awaits careful coordination in order to yield a 
rich crop of ideas. The various books and journals of orni- 
thology and entomology are like a row of beehives containing 
an immense amount of valuable honey, which has been stored 
up in separate cells by the bees that made it. The advantage, 
and at the same time the difficulty, of ecological work is that 
it attempts to provide conceptions which can link up into some 
complete scheme the colossal store of facts about natural 



66 



ANIMAL ECOLOGY 



history which has accumulated up to date in this rather hap- 
hazard manner. This applies with particular force to facts 
about the food habits of animals. Until more organised 
information about the subject is available, it is only possible 
to give a few instances of some of the more clear-cut niches 
which happen to have been worked out. 

1 8. One of the biggest niches is that occupied by small 
sap-suckers, of which one of the biggest groups is that of the 
plant-lice or aphids. The animals preying upon aphids form 
a rather distinct niche also. Of these the most important are 
the coccinellid beetles known as ladybirds, together with the 
larvae of syrphid flies (cf. Fig. 5) and of lacewings. The niche 




Moths 



^Caterpillars] 



llchneumons | 



jfUJrph^s 

y- > | LadyblrdB| — - ^Spiders | 

JI ^ -^ 



I Secretion} 



Dlsser 
Wasps 



iPsammochareaj 



Fig. 5. — Food-cycle on young pine-trees on Oxshott Common. 
(From Richards. ^^) 

in the sea and in fresh water which is analogous to that of 
aphids on land is filled by copepods, which are mainly diatom- 
eaters. This niche occurs all over the world, and has a number 
of well-defined carnivore niches associated with it. If we take 
a group of animals like the herbivorous grass-eating mammals, 
we find that they can be divided into smaller niches according 
to the size of the animals. There is the mouse niche, filled 
by various species in different parts of the world ; the rabbit 
niche, of larger size, filled by rabbits and hares in the palae- 
arctic region and in North America, by the agouti and viscacha 
in South America, by wallabies in Australia, and by animals 
like the hyrax, the springbuck, and the mouse deer ^^ in Africa. 
In the same way it can be shown that there is a special niche of 
carnivorous snakes which prey upon other snakes — a niche 
which is filled by different species in different countries. In 



THE ANIMAL COMMUNITY 67 

South America there is the mussarama, a large snake four or 
five feet in length, which is not itself poisonous, but preys 
exclusively upon other snakes, many of which are poisonous, 
being itself immune to the venoms of lachesis and rattlesnake, 
but not to colubrine poisons. In the United States the niche 
is filled by the king-snake which has similar habits, while in 
India there is a snake called the hamadryad which preys upon 
other (in this case non-poisonous) snakes.®^* 

19. Another widespread niche among animals is that 
occupied by species which pick ticks off other animals. For 
instance, the African tick-bird feeds entirely upon the ticks 
which live upon the skin of ungulates, and is so closely de- 
pendent upon its mammalian " host " that it makes its nest 
of the latter's hair {e.g. of the hartebeest).^^^ In England, 
starlings can often be seen performing the same office for sheep 
and deer. A similar niche is occupied on the Galapagos Islands 
by a species of scarlet land-crab, which has been observed 
picking ticks off the skin of the great aquatic lizards {Ambly- 
rhynchus).^^^ Another niche, rather analogous to the last one, 
is that occupied by various species of birds, which follow 
herds of large mammals in order to catch the insects which are 
disturbed by the feet of the animals. Chapman ^^^ saw 
elephants in the Sudan being followed by kites and grey 
herons ; Percival i^s says that the buff-backed egret follows 
elephants and buffalo in Kenya for the same purpose ; in 
Paraguay ^^^ there are the Aru blackbirds which feed upon 
insects disturbed by the feet of cattle ; while in England 
wagtails attend cattle and sheep in the same way. 

20. There is a definite niche which is usually filled by earth- 
worms in the soil, the species of worm differing in different 
parts of the world. But on coral islands their place may be 
largely taken by land- crabs. Wood- Jones ^^^^ states that on 
Cocos-Keeling Island, coconut husks are one of the most 
important sources of humus in the soil, and in the rotting 
husks land- crabs (chiefly of the genus Cardiosoma) make 
burrows and do the same work that earthworms do in our own 
country. (There are as a matter of fact earthworms as well 
on these islands.) On the coral reefs which cover such a large 



68 ANIMAL ECOLOGY 

part of the coast in tropical regions, there is a definite niche 
filled by animals which browse upon the corals, just as herbi- 
vorous mammals browse upon vegetation on land. There are 
enormous numbers of holothurians or sea-cucumbers which 
feed entirely in this way. Darwin ^o gives a very good descrip- 
tion of this niche. Speaking also of Cocos-Keeling Island, he 
says : 

" The number of species of Holothuria, and of the in- 
dividuals which swarm on every part of these coral-reefs, is 
extraordinarily great ; and many ship-loads are annually 
freighted, as is well known, for China with the trepang, which 
is a species of this genus. The amount of coral yearly con- 
sumed, and ground down into the finest mud, by these several 
creatures, and probably by many other kinds, must be immense. 
These facts are, however, of more importance in another point 
of view, as showing us that there are living checks to the growth 
of coral-reefs, and that the almost universal law of ' consume 
and be consumed,' holds good even with the polypifers forming 
those massive bulwarks, which are able to withstand the force 
of the open ocean." 

This passage, besides showing that the coral-eating niche 
has a geological significance, illustrates the wide grasp of 
ecological principles possessed by Darwin, a fact which con- 
tinually strikes the reader of his works. We have now said 
enough to show what is meant by an ecological niche, and how 
the study of these niches helps us to see the fundamental 
similarity between many animal communities which may 
appear very different superficially. The niche of an animal 
may to some extent be defined by its numbers. This leads us 
on to the last subject of this chapter, 

The Pyramid of Numbers 

21. *' One hill cannot shelter two tigers." In other and 
less interesting words, many carnivorous animals, especially 
at or near the end of a food-chain, have some system of terri- 
tories, whereby it is arranged that each individual, or pair, or 
family, has an area of country sufficiently large to supply its 
food requirements. Hawks divide up the country in this way, 



THE ANIMAL COMMUNITY 69 

and Eliot Howard's work ^ has shown that similar territory 
systems play a very important part in the lives of warblers. 
We can approach the matter also from this point of view : 
the smaller an animal the commoner it is on the whole. This 
is familiar enough as a general fact. If you are studying the 
fauna of an oak wood in summer, you will find vast numbers 
of small herbivorous insects like aphids, a large number of 
spiders and carnivorous ground beetles, a fair number of 
small warblers, and only one or two hawks. Similarly in a 
small pond, the numbers of protozoa may run into millions, 
those of Daphnia and Cyclops into hundreds of thousands, 
while there will be far fewer beetle larvae, and only a very few 
small fish. To put the matter more definitely, the animals 
at the base of a food-chain are relatively abundant, while those 
at the end are relatively few in numbers, and there is a pro- 
gressive decrease in between the two extremes. The reason 
for this fact is simple enough. The small herbivorous animals 
which form the key -industries in the community are able to 
increase at a very high rate (chiefly by virtue of their small 
size), and are therefore able to provide a large margin of 
numbers over and above that which would be necessary to 
maintain their population in the absence of enemies. This 
margin supports a set of carnivores, which are larger in size 
and fewer in numbers. These carnivores in turn can only 
provide a still smaller margin, owing to their large size which 
makes them increase more slowly, and to their smaller numbers. 
Finally, a point is reached at which we find a carnivore {e.g. 
the lynx or the peregrine falcon) whose numbers are so small 
that it cannot support any further stage in the food-chain. 
There is obviously a lower limit in the density of numbers of 
its food at which it ceases to be worth while for a carnivore to 
eat that food, owing to the labour and time that is involved 
in the process. It is because of tliese number relations that 
carnivores tend to be much more wide-ranging and less 
strictly confined to one habitat than herbivores. 

22. This arrangement of numbers in the community, the 
relative decrease in numbers at each stage in a food-chain, is 
characteristically found in animal communities all over the 

Lu , L I 8 R A R Y % 



70 ANIMAL ECOLOGY 

world, and to it we have applied the term " pyramid of num- 
bers." It results, as we have seen, from the two facts (a) that 
smaller animals are preyed upon usually by larger animals, 
and (b) that small animals can increase faster than large ones, 
and so are able to support the latter. 

The general existence of this pyramid in numbers hardly 
requires proving, since it is a matter of common observation 
in the field. Actual figures for the relative numbers of different 
stages in a food-chain are very hard to obtain in the present 
state of our knowledge. But three examples will help to 
crystallise the idea of this " pyramid." Birge and Juday ^2 
have calculated that the material which can be used as food 
by the plankton rotifers and Crustacea of Lake Mendota in 
North America weighs twelve to eighteen times as much as they 
do. (The fish which eat the Crustacea would weigh still less.) 
Again, Mawson^s estimated that one pair of skuas (Megalestris) 
on Haswell I. in the Antarctic regions, required about fifty to 
one hundred Adelie penguins to keep them supplied with 
food (in the form of eggs and young of the penguins) ; while 
Percival '^^^ states that one lion will kill some fifty zebras per 
year, which gives us some idea of the large numbers of such a 
slow-breeding animal as the zebra which are required to produce 
this extra margin of numbers. 



CHAPTER VI 

PARASITES 

As opposed to carnivores, parasites are (i) much smaller than the hosts 
upon which they prey, but (2) feed in essentially the same way as carni- 
vores, the chief difference being that the latter live on capital and the 
former on income of food ; but (3) a complete graded series can be 
traced between typical parasites and typical carnivores, both among 
animals which eat other ones, and (4) among those animals which live 
by robbing other animals of their food. (5) We can therefore apply 
the same principles to parasites as were applied to carnivores in the last 
chapter, making certain alterations as a result of the different size- 
relations of the two classes of animals. (6) The food-cycle acts as an 
important means of dispersal for internal parasites, so that they often 
have two, or (7) more, hosts during their life-histories, and (8) this means 
of dispersal is also used to some extent by external parasites. Animals 
which suck blood or plant-juices often play an important part in such 
life-histories. (9) In food-chains formed by parasites, the animals at 
each stage become smaller, and at the end of a chain are so small that 
bacteria are often found to become important organisms at that point. 
(10) The parasitic hymenoptera occupy a rather special position in the 
food-cycle. (11) When parasites and carnivores are both included in 
the same scheme of food-cycles, the latter become very complex, as 
is (12) shown by an example ; but (13) in practice, a number of parasites 
can be considered as forming part of their host, as far as food is concerned, 
although when numbers are being studied the parasites must be treated 
separately. 

I . If you make a list of the carnivorous enemies and of the 
parasites of any species of animal, you will see (although 
they are so obvious that they easily escape notice unless 
pointed out) certain curious facts about the sizes of the two 
classes of animals relative to their prey or host. For instance, 
a frog would have for enemies such animals as otters, herons, 
or pike, which would be anything up to fifty times the size 
of the frog, while the parasites would be animals like flat- 
worms, nematodes, or protozoa, which would be five hundred 
or five thousand times smaller than the frog. The same thing 
applies to any other animal. A mouse is preyed on by 

71 



72 ANIMAL ECOLOGY 

hawks, owls, foxes, and weasels on the one hand, and by lice, 
fleas, ticks, mites, tapeworms, and protozoa on the other. 
Small oligochaete worms in the arctic soil are eaten by purple 
sandpipers and parasitised by protozoa, while earthworms 
in England are eaten by moles and shrews and thrushes and 
toads, and parasitised by protozoa and nematodes. In fact, 
most animals have a set of carnivorous animals much larger 
than themselves, and a set of parasitic enemies much smaller 
than themselves, and usually there are very few enemies 
of intermediate size. In all these cases we are, of course, 
speaking of the size relative to their prey or host. There are 
perfectly good reasons for this size- distribution. Most car- 
nivores are able to overpower and eat their prey by virtue of 
their larger size and greater strength (with exceptions noted in 
the last chapter). On the other hand, parasites must necessarily 
be much smaller than their hosts, since their existence depends, 
either temporarily or permanently, upon the survival of the 
host, and for this reason the parasite cannot exceed a certain 
size without harming its host too much. 

2. It is very important to realise quite clearly that most 
parasites are in their feeding habits doing essentially the same 
thing as carnivores, except that while the carnivore destroys 
its prey, the parasite does not do so, or at any rate does not do 
so immediately or completely. A parasite's existence is usually 
an elaborate compromise between extracting sufficient nourish- 
ment to maintain and propagate itself, and not impairing too 
much the vitality, or reducing the numbers of its host, which 
is providing it with a home and a free ride. In consequence 
of this compromise, a parasite usually destroys only small 
portions of its host at a time, portions which can often be 
replaced fairly quickly by regeneration of the tissues attacked. 
Or it may exploit the energies of its host in more subtle ways, 
as when it subsists on the food which the host has collected 
with great expenditure of time and energy. The difference 
between the methods of a carnivore and a parasite is simply 
the difference between living upon capital and upon income ; 
between the habits of the beaver, which cuts down a whole 
tree a hundred years old, and the bark-beetle, which levies a 



PARASITES 73 

daily toll from the tissues of the tree ; between the burglar and 
the blackmailer. The general result is the same, although the 
methods employed are different. 

3. Although the relative sizes of carnivores and parasites 
are so markedly different in the vast majority of cases, it is 
really rather difficult to draw the Hne in all cases between the 
two classes of animals, since there are a number of species 
which combine to some extent the characteristics of both 
types. It is possible and interesting to trace a complete 
series between the two extremes. A hookworm lodged in the 
intestine of a mammal is a typical parasite, destroying the 
tissues of its host, and, in the adult stage, entirely confined to 
it. A louse is also clearly a typical parasite, although it can 
walk about from one host to another. Fleas, however, are less 
constant in their attendance upon their host, since they are 
able to live for some days at a time without feeding, and may 
be found walking about at large, in the open or in nests of their 
host. When we come to blood-sucking flies it is quite diffi- 
cult to know whether to class them as parasitic or carnivorous. 
Their habits and size may be the same as those of fleas, and 
the time which is spent with the host may be no less than that 
of the flea, but they are less closely attached to any one host. 
This time varies considerably : a complete series could be 
traced between species which spend a great deal of their time 
in the company of their host-animal {e.g. the horse-fly (Hippo- 
hosca equina) which lives close to its host), and others which 
lead highly independent lives {e.g. the Tabanids, which are 
generally active fliers and sit about waiting for some animal 
to pass). 

4. We can find a similar graded series amongst animals 
which live by robbing other animals of their food. The adults 
of certain Filaria worms live embedded in the muscles of their 
hosts. The larvae of these worms live in the blood or the 
lymph- vessels of the host. Then there are tapeworms, which 
inhabit the intestine and absorb food which has previously 
been made soluble by the digestive juices of the host. Other 
animals {e.g. nematode worms), often living in the same place 
as the tapeworms, exist by eating solid food particles, which 



74 ANIMAL ECOLOGY 

are either undigested or only partly digested. The next stage 
in this series is the crocodile bird, which sits inside the open 
mouth of the crocodile and picks bits of food from amongst 
the teeth of its " host." The arctic skua employs similar but 
more drastic methods, when it chases kittiwake gulls and 
terrorises them into yielding up their last meal, which is skil- 
fully caught by the skua before reaching the water below. 
Another method employed by some animals is to take small 
bits out of their food-animal, without actually destroying it, just 
after the manner of the hookworm or the filaria. For instance, 
Lortet ^^ says : " The fishes of the lake of Tiberius [Tiberias], 
very good to eat, serve as a pasturage for the myriads of crested 
grebes {Podiceps cristatus) and of pelicans. Frequently the 
grebes snatch at the eyes of the chromids, and with one stroke 
of their long sharp beaks lift out as cleverly as would a skilful 
surgeon the two eyeballs and the intro-orbital partition. These 
unhappy fish, now blind, of which we have taken numerous 
examples, have thus the entire face perforated by a bloody 
canal which cicatrises rapidly. It is only the larger individuals 
who are thus operated on by the grebes, for, not being able to 
avail themselves of the entire fish, these voracious birds take 
the precaution to snatch only the morsel of their choice." 
Put less poetically, this means that the bird is able to carry 
on the ordinary carnivorous method of destroying the whole 
of its prey as long as the latter is below a certain size, but when 
it grows above this limit new methods are adopted, which 
closely resemble those of a true parasite. A further stage in 
the series which we have been tracing is the arctic fox, which, 
although an ordinary carnivorous animal in summer, when it 
eats birds and lemmings, often travels out on to the frozen 
sea-ice in winter and there accompanies the polar bear and 
subsists on the remains of seals killed by the bear, and upon 
the dung of the latter. The bear does all the work, and the fox 
gets a share of the proceeds. From this point it is only a short 
distance to a true carnivore, like the polar bear itself, which 
is, after all, only living by exploiting the energies of the seal. 
5. To imagine that parasites are unique in exploiting the 
activities and food-products of their hosts is to take a very 



PARASITES 75 

limited view of natural history. It is common to find parasites 
referred to as if they were in some way more morally oblique 
in their habits than other animals, as if they were taking some 
unfair and mean advantage of their hosts. If we once start 
working out such " responsibilities " we find that the whole 
animal kingdom lives on the spare energy of other species or 
upon plants, while the latter depend upon the radiant energy 
of the sun. If parasites are to occupy a special place in this 
scheme we must, to be consistent, accuse cows of petty larceny 
against grass, and cactuses of cruelty to the sun. Once we 
take a broad view of animal interrelations it becomes quite clear 
that it is best to treat parasites as being essentially the same as 
carnivores, except in their smaller size, which enables them to 
live on their host. In other words, the resemblances between 
the two classes of animals are more important than the 
differences. 

6. We will now turn to a consideration of the place of para- 
sites in the food-cycle of any animal community, and the ways 
in which the food-cycle affects them. One of the greatest 
questions which has to be solved by many parasites is what 
to do when their host dies, as it is bound to do sooner or later. 
This applies with especial force to internal parasites like 
flatworms, which have become so specialised to a life of passive 
absorption in the dark that they are unable to take any active 
steps to deal with the situation created by the death of their 
host. It is here that the food-cycle comes in and plays an 
important part. Probably the commonest death for many 
animals is to be eaten by something else, and as a result we find 
that a great many parasites pass automatically with the prey 
into the body of its enemy, and are then able in some way to 
occupy the new host. Let us take the case of a tapeworm which 
lives as a young larva or bladderworm in the muscles of a rabbit. 
When the rabbit is eaten by some enemy, say a fox, some of 
the bladderworms pass unharmed into the intestine of the fox, 
and there continue their development and grow up into adult 
tapeworms ; and, in this way, the problem created by the 
death of the first host is solved. But foxes being also mortal, 
the tapeworm has to get back again into the rabbit before the 



76 



ANIMAL ECOLOGY 



fox dies, and this is also brought about by the food-cycle. 
For the tapeworm produces vast numbers of eggs which pass 
out with the excretory products of the fox ; some of these eggs 
contaminate the vegetation which the rabbit is eating, or in 
some other way get in with its food, and are then able ultimately 
to grow up into more bladderworms in the body of the rabbit. 
The diagram in Fig. 6 shows the way in which the food-cycle 



PLANTS 
--EggD-- 



^RABBIT 

Larvae- 



+»POX 
Adult 



Fig. 6. 

acts as a means of conveyance for the parasites, throughout 
its life- cycle. 

7. This was a fairly simple case. There is always this 
tendency for parasites to get transferred from one stage in a 
food-chain to the next, like passengers on a railway. Many 
parasites get out at the first station — in other words, they have 
a direct Hfe-history, with no alternate host. An example of 
this is a tapeworm {Hymenolepis) which occurs in mice, and 
which has the larval and adult stages in the same host, although 
in different parts of the body.^^ Or the parasite may get out 
at the second station, like the rabbit tapeworm described above. 
Or again, it may travel as far as a third host. The broad tape- 
worm {Diphyllohothrium latum), which occurs occasionally 
in man, causing severe ansemia, gets into the gut of some 
small fresh-water copepod {e.g. Cyclops strenuus or Diaptomus 



WATER- 



'COPEPOD- 



, Egg ^ — Larva ->- Larva— ^— 




Fig. 7. 

gracilis) with the food of the latter, is then eaten by fish in 
which it exists in the form of bladderworm cysts, and is finally 
eaten by some carnivore or by man.^* The diagram (Fig. 7) 
sums up this cycle. 



PARASITES 77 

Perhaps the most striking examples of the way in which 
parasites may pass along food-chains are afforded by the worms 
of the genus Echinorhynchus y which in some cases go on being 
transmitted from one host to another until they end up in 
the animal which forms the end of a whole chain, and so can 
get no farther, and indeed probably never get back again to 
their first host. They are like passengers who forget to get 
out at the right station and travel by mistake on to the terminus. 
8. There is often a similar tendency amongst ectoparasites 
to become transferred from their first host to its enemy. This 
is well known to be the case amongst some species of fleas. 
Ceratophyllus sciurorum, which lives commonly upon the 
squirrel and dormouse, occurs also occasionally on the pine- 
marten, weasel, and stoat, while C. walkeri comes commonly 
on the bank- vole, weasel, and stoat.^^ 

All the cases in which parasites are transmitted from one 
host to another by blood-sucking insects are also examples 
of the role played by the food-cycle in the lives of parasites. 
There are parallel cases of plant parasites being carried from 
plant to plant by insects. The protozoan PhytomonaSy which 
is found very widely in the tissues of various species of spurge 
(Euphorbia), mainly in warmer countries, is carried from one 
plant to another, in some places at least, by insects. It has 
been shown that Phytomonas davidi, which occurs in Euphorbias 
in Portugal, is carried from plant to plant by a bug, Steno- 
cephalus agilis, which lives upon the juices of the spurge, and 
in which the Phytomonas has a definite life-history stage. ^^* 

Since the food-cycle is so important in determining many of 
the possible modes of transference of parasites from one host 
to another, it is plain that biological surveys carried out along 
food-cycle lines would be of great value in narrowing the field 
of inquiry when the Hfe-history of any particular parasite is 
being studied. 

9. Having shown the relation which exists between parasites 
and food-chains formed by herbivorous and carnivorous 
animals, we may now turn to a consideration of food-chains 
among parasites themselves. Just as the carnivores in a 
food-chain usually become progressively larger and larger, 



78 ANIMAL ECOLOGY 

so of course the animals in a parasite food-chain become 
gradually smaller and smaller. And just as the carnivores 
become fewer in numbers, so do the parasites become 
usually more numerous. Let us take some examples. Fleas 
are parasitic upon birds and mammals, and many fleas in 
turn are parasitised by protozoa of the genus Leptomonas.^^^ 
One squirrel might support a hundred fleas or more, and 
each flea might support thousands or hundreds of thousands 
of Leptomonas. The pyramid of numbers in such a case 
is inverted. Many other examples could be given. Egyptian 
cattle have ticks (Hyalomma), which in turn carry inside 
them a protozoan {Crithidia hyalommce)}'^^ Apparently there 
are never very many stages in such food-chains of para- 
sites. The reason for this is that the largest parasite is not 
very big, and any hyperparasite living on or in this must be 
very much smaller still, so that the fifth or sixth stage in the 
chain would be something about the size of a molecule of 
protein ! Actually, bacteria, although they are not animals, 
may be conveniently included in parasite food-cycles, and in 
many cases they form the last link in the chain. The plague 
bacillus lives in the flea of the rat in warm countries (also 
in the rat itself), the flea lives on the rat, and the rat lives to 
some extent " parasitically " on mankind. Recent work upon 
bacteriophages seems to show that there are sometimes still 
smaller organisms (or something with the power of multipli- 
cation) which have a controlling effect on the numbers of the 
bacteria themselves. So possibly the last link in some para- 
site food-chains may be formed by such bacteriophages. 

10. There has been a good deal of discussion by ento- 
mologists as to the exact position of the Parasitic Hymenoptera 
with respect to parasitism. These insects lay their eggs in the 
eggs, larvae, or later stages of various other insects (and in some 
other animals too), and these eggs grow into larvae which live 
as true parasites in the bodies of their hosts, which they 
eventually destroy, the hymenoptera finally emerging in the 
form of free-living insects. An animal like an ichneumon, 
therefore, combines the characteristics of a parasite (in its 
larval stage) and of a free-living animal (in its adult stage), 



PARASITES 



79 



the latter being carnivorous, or herbivorous, or in some cases 
not feeding at all. For this reason the Parasitic Hymenoptera 
have often been referred to as " parasitoids." There are 
numerous other kinds of animals which have alternate free- 
living and parasitic stages in their life-history, but not many 
of them in which the adult is free-living as it is in these 
hymenoptera. It is usually the larva which undertakes the 
task of finding a new host, where this is done by active 
migration. 

As Richards ^^^ has pointed out, insects like ichneumons, 
braconids, and chalcids, do not have directly a very big effect 
upon the food-cycle in a community, since they are merely 
turning the tissues of their host into hymenopteran tissue, 
and very often the enemies of the host eat the adult hymeno- 
ptera as readily as they would eat the original host if it had 
survived. Of course it does make a certain amount of differ- 
ence ; for one thing, there is naturally a great loss of energy 
in the process of turning host into parasite, and therefore 
the activity of the hymenoptera reduces the available food- 
supply to some extent. 

II. It has been necessary in this chapter, as well as in the 
last, to speak continually of food-chains as if they commonly 
consisted of simple series of species, without taking into 
account any of the complications found in actual practice. 
This is of course far from the truth, because the food-relations 
of animals are extremely complicated and form a very closely 
and intricately woven fabric — so elaborate that it is usually 
quite impossible to predict the precise effects of twitching 
one thread in the fabric. Simple treatment of the subject 
makes it possible to obtain a glimmering of the principles 
which underlie the superficial complication, although it must 
be clearly recognised that we know at present remarkably 
little about the whole matter. One of the most important of 
these principles is that the sizes, and in particular the relative 
sizes, of the various animals which live together in a com- 
munity, play a great part in their lives, and partly determine 
the effects which the various species will have upon one another. 
One important effect we have mentioned — that the food- 



8o ANIMAL ECOLOGY 

cycle provides a means of dispersal for parasites, a means 
which is very commonly employed by them. The most 
striking effects, however, are concerned with the numbers 
of animals ; but this subject must be reserved for another 
chapter. 

12. When the food- relations of parasites and carnivores 
to other species are combined into a common food-cycle 
scheme, the amazingly complex nature of animal interrela- 
tions is seen. The diagram in Fig. 8 is intended to give some 
idea of the sort of thing which is met with ; but it is to be under- 
stood that the diagram is only illustrative and does not claim 
to have anything but a general basis of truth, i.e. it does not 

I FLEA I > I PROTOZOA I 

I 
„ I 



I GRASS I 



LAMB 



Placental 



» |WOLF[ 

*- ' |BrRDLICE] J^^ 

^■^ 1 TICK h » I STARLING \ ^- >{cH} 

I PFtPTOZOA I ^v 

ISPIROCHAETESI 



mm 



I WARBLE FLyI |PR0T0Z0A| 



I FLATWORM | | ROUNDWORM! 

sk 

I roundwormI 

Fig. 8. 

represent any particular community in real life, and ** all the 
characters in the story are entirely fictitious." If there were 
plenty of real examples to choose from, it would be much 
better and more satisfactory ; but there are not. It is this 
very fact — the lack of properly worked-out examples — ^which 
makes it important to try and point out the type of problem 
which requires solution. The diagram illustrates several 
points. For one thing it shows that the food-chains do not 
always go on from carnivore to carnivore, or parasite to para- 
site. In this case the tick which is parasitic upon the sheep 
is eaten by the starling, which is a carnivore, the latter in turn 
by an animal larger than itself (the cat), and the cat in turn by 
fleas, and the fleas by protozoa. In this way, when parasites 
and carnivores are considered together it is seen that there may 



PARASITES 8i 

be a great many more stages in the history of the food than 
when parasites alone or carnivores alone are considered. 
In this example, in which mutton is converted, amongst other 
things, into flea-protozoa, there are six stages involved, whereas 
the simple carnivore chain from the sheep to the wolf has only 
two stages, and the simple parasite chains only two or three 
at the most. The reason for the greater number of stages 
possible in the former case is simply that the size-relationships 
of the animals make it possible for the chain to continue longer 
without reaching either the upper or lower size limits of 
animals. 

13. A great many ectoparasites have no very important 
direct effects upon the food-cycle in general, since they are 
eaten either by their hosts {e.g. birds and bird-lice) or else they 
are eaten together with their host by the enemy of the latter 
{e.g. copepods containing worm-larvae by fish). In such cases 
the parasite and its host act as one unit for food purposes. 
(Of course the circulation of the parasite in this way may 
ultimately have very important effects upon the numbers of 
animals in the community.) It is for this reason that 
parasites can very often be ignored in practice, when one 
is making out the first rough scheme of food-relations in 
an animal community, although the control of numbers can 
only be understood by bringing in the parasites too. For 
instance, in the arctic tundra food-cycle shown in Fig. 4, 
on p. 58, the parasites do not play a very important part, 
and probably it is the exception rather than the rule for 
parasites to form a large independent food-supply for any 
other animal. There are, however, a number of ectopara- 
sites {e.g. ticks) occurring upon mammals, which form an 
important article of diet for certain species of birds. There 
is one rather interesting instance of tliis sort of thing, recorded 
by Wilkins ^4 from the antarctic regions. On Elephant 
Island there is in summer a colony of nesting Gentoo Penguins 
{Pygoscelis papiid) and haunting these colonies are a certain 
number of birds called Paddies {Chionis alba). The Paddies 
hve largely upon parasitic nematode worms which pass out 
from the intestine of the penguins with their excreta. In 

G 



82 ANIMAL ECOLOGY 

winter the Paddies become very thin, owing to the absence 
of the penguins upon which they depend in summer. 

These instances show that it is not possible to neglect the 
existence of parasites as a food-supply for other animals, but 
that they do not usually act as such except m the sense ot 
being eaten at the same time as their host, by the latter's 
enemy. 



CHAPTER VII 

TIME AND ANIMAL COMMUNITIES 

Many of the animals in a community (i) never meet owing to the fact that 
they become active at different times. This is because (2) the environ- 
ment is subject to a number of rhythmical changes which (3) result in 
corresponding variations in the nature of the animal communities at 
different times. (4) There is the day and night rhythm which affects 
both free-living animals and (5) some parasites. (6) This rhythm miay 
be of practical importance, e.g. from its influence upon blood-sucking 
insects and (7) is most strongly marked in deserts, but (8) there is really 
very little detailed information about day and night communities. 
(9) Some of the changes in the fauna are caused by migration, as in the 
vertical strata of a wood or in the plankton. (10) There is not always a 
very sharp limit between day and night communities. (11) In polar 
regions there is no night fauna, while (12) in the tropics the latter is 
very rich. (13) Other rhythms are those of the tides and (14) of weather 
(caused by the passage of depressions) which (15) produce variations 
in the composition of the active animal communities, e.g. those of dry 
and (16) wet conditions, which (17) may override the day and night 
rhythm. (18) Weather changes also have important effects on blood- 
sucking insects. (19) Then there is the annual cycle of the seasons 
which (20) has a particular interest from its relation to bird migration 
and (2t) to changes both in the food-habits of animals and (22) in the 
particular species occupying any one niche. (23) These rhythmical 
changes in communities enormously increase the difficulty of studying 
the latter completely, so that (24) it is advisable to choose extremely 
simple ones in order to work out the principles governing animal com- 
munities in general. (25) Larger pulsations in climate chiefly affect 
the numbers of animals, and so we are led on to the next chapter. 

I. One of the commonest rodents of the South African veld 
is the gerbille, springhaasrot, or rooiwitpens (Taterona loben- 
gula)y which lives sociably in warrens in sandy country where 
there is plenty of sweet grass and bulbs to eat. Quite often it 
makes its network of burrows in places already occupied by 
two species of carnivorous animals, the yellow mongoose 
{Cynictis pencillata) and the suricat {Suricator suricator). But 
although the rodents and the carnivores live in close contact 
with one another, actually using to some extent the same system 

83 



84 ANIMAL ECOLOGY 

of underground runways, they do not usually clash in any way 
in their activities ; for while the gerbilles come out exclusively 
at night, leaving their burrows after sunset and returning always 
before dawn, the mongooses and suricats feed only during the 
day, and retire to earth at night.'^^ It appears that a mongoose 
never attacks a gerbille under ordinary circumstances because 
the two creatures do not usually meet : they have different 
hours for business. It is only when the gerbilles are smitten by 
an epidemic of some disease like plague, at which times they 
wander out of their holes in the daytime, that they are attacked 
and eaten by the mongoose. This last fact has a practical 
importance, since the South African plague investigator is 
able by examining the excreta of the mongoose to find out with 
tolerable certainty whether the gerbilles have been dying of 
plague, a fact which is rather difficult to establish easily in 
any other way. If the mongoose excreta contain gerbille fur, 
then there is strong evidence of epidemic amongst the latter.'^^ 

In this case the alternation of day and night has the effect 
of separating almost completely two animals which live in 
the same place, and although the phenomenon and its results 
happen to have an important practical bearing, it is only one 
example among thousands which might be given, all of which 
go to show that the phenomenon is of general occurrence in 
nearly all animal communities. 

2. The environment even in the same place is always 
changing rhythmically and more or less violently ; some of 
these changes being regular, like the alternation of day and 
night, of high and low tides, or the annual succession of the 
seasons, while others are more irregular, like the fluctuations 
in weather from day to day and week to week. These changes 
all leave a corresponding impress upon the arrangement and 
composition of animal communities. Just as animals tend to 
become specialised for life in certain places, so also most 
of them are active only at certain times. There are various 
ways of meeting the onset of unfavourable conditions. If 
the latter last only for a short time, the animal may merely 
retire to some hiding-place or become inactive wherever it 
happens to be at the moment. Every one must have noticed 



TIME AND ANIMAL COMMUNITIES 85 

the extraordinary effect upon insects when a passing cloud 
covers the sun. The drop in temperature slows down their 
movements or actually stops them altogether. The return of 
the sun starts them all off at high speed once more. It is 
worth while to watch a big ant-hill under such conditions. In 
the sun the whole place swarms with hurrying ants, carrying 
sticks, caterpillars, or each other, with restless energy. When 
it gets suddenly cooler they all stop working fast and do every- 
thing vdth painful slowness. The larvse of a species of locust 
which periodically undertakes great migrations in the Northern 
Caucasus has similar reactions, which have been worked out 
rather carefully by Uvarov.^s On the first stages of the 
journey the larvae march along on the ground in great droves 
(they do not grow their wings until a later stage in their travels) ; 
but they never march at night, and if the temperature falls 
below about 13° to 15° C. their movements cease and they 
have to stop wherever they happen to be. In the same way 
they will stop for only a tiny passing cloud. It is interesting 
to note that there is also a higher limit of temperature above 
which they will not continue to march, so that they halt some- 
times in the middle of the day. We see, then, that this locust 
is able to remain active only under certain optimum conditions 
of temperature, and many other examples of the same kind of 
thing could be given. 

3. With many animals the coming of nightfall has precisely 
the same effect as a cloud over the sun, but the stoppage of 
work is longer and may require the taking of more elaborate 
precautions. And in addition we find that another set of 
animals adapted to a different set of conditions comes out and 
takes the place of the others. If the changes in conditions are 
greater, or last longer, many animals migrate away altogether to 
a more suitable locality or else tide over the bad period in some 
special way ; for instance, by renouncing all outside feeding 
and living upon their own fat like a hibernating marmot. 

Animal communities are therefore organised into a series V 
of separate smaller communities, each of which is in action at 
a different time. There are " day and night shifts," wet and 
dry weather sets of animals, communities of winter and 



86 ANIMAL ECOLOGY 

summer, and so on. It would be wrong to get the impression 
that these time-communities are quite separate from each 
other. Such is obviously not by any means the case. The 
point is that the main community changes in personnel to a very 
large extent at different times ; although the changes are not 
complete, they are very considerable, and in some cases 
{e.g. day and night) there may be very few species of animals 
which live in both communities. The community of animals 
living in one place still remains a definite and fundamental 
unit, since its periodic changes are regular and characteristic. 
We might compare the place-community to an elaborate piece 
of machinery (e.g. a motor-car), which still remains one unit 
although there are several more or less independent special 
mechanisms contained inside it. 

We will now consider some of the time- communities in 
more detail. 

4. Day and Night.- — In woods, the separation of animals 
into day and night species may be very easily studied. Take 
an English oak wood as an example. In the daytime there 
would be, among a host of other animals, birds like sparrow- 
hawks, blackbirds, thrushes, woodpeckers, and also bank- 
voles, weasels, butterflies, bees, and ants. At sunset there is 
often a short pause when the diurnal animals have gone to 
rest or begun to think of doing so, and the nocturnal ones 
have not yet got up full steam. Then there would begin to 
appear the forerunners of a host of night animals : nightjars, 
owls, moths, bats, long-tailed field mice, etc. Not only is one 
kind of animal replaced by another, but one kind of food-chain 
is replaced by another, and certain niches which are unused 
by any animal during the day become occupied at night. The 
weasel — bank-vole industry is changed into a tawny-owl — 
wood-mouse industry. The woodpecker — ant connection has 
no equivalent at night, while the moth — nightjar-or-bat chain 
is almost unrepresented by day. In fact, one food-cycle is 
switched off and another starts up to take its place. With 
the dawn the whole thing is switched back again. The two 
communities of animals are not completely separate, owing to 
the fact that some animals are not so particular about their 



TIME AND ANIMAL COMMUNITIES 87 

time of feeding ; or else they come out chiefly at dusk and so 
form a transition from one to the other. Also many animals 
have regular habits which do not correspond exactly with day 
and night, owing to the fact that the thing controlling them 
is not light or heat but something else, such as rain or other 
weather conditions. 

5. Such day and night changes are not found in free-living 
animals only, but also exist among parasites of mammals, and 
probably of birds too. Owing to the fact that most mammals 
sleep either by day or by night there exist corresponding 
rhythmical changes inside their bodies, especially in tem- 
perature. Both in birds and mammals the body is slightly 
colder during sleep than when they are awake. This rhythm 
depends entirely upon the activity of the animal, since nocturnal 
birds like owls have the normal rhythm reversed {i.e. they are 
warmer at night), and this in turn can be reversed by changing 
the conditions under which they live so as to cause the birds 
to come out by day and sleep by night. Now there are certain 
round-worms (nematodes) parasitic in man which show the 
effects of the sleep rhythm in a very remarkable way. The 
first species (Filaria bancrofti) lives as an adult in the 
lymphatic glands of man in tropical countries, but its larvae 
live in the blood. In the daytime these larvae retire to the 
inner parts of the body, mostly to the lungs ; but at night they 
issue forth into the peripheral circulation, appearing first 
about five to seven in the evening, reaching a maximum about 
midnight, and disappearing again by about seven or eight in the 
morning. This rhythm can be reversed if a person stays up 
all night and sleeps in the day, which shows that the nematode's 
activity is affected by rhythmical changes in the conditions of 
the body like those which we have described above. Another 
species of Filaria (called Loaloa)J[i2iS larvae which live in the 
blood of man, but unlike the other species these larvae come 
out only in the day, disappearing at night. It is stated that 
this periodicity is not affected by reversal of sleep, but pre- 
sumably it must originally have been caused by some rhythm 
in the bodily environment.-^^^ A third species has larvae in 
the blood which occur in the peripheral circulation equally 



88 ANIMAL ECOLOGY 

by day or by night. The habits of these larval worms have 
a very important bearing upon the means of transmission 
from one man to another ; for F, hancrofti is transmitted 
by blood-sucking mosquitoes which fly at night, while Loaloa 
is now known to be transmitted by Tabanid flies {Chrysops 
dimidiatus and silaria) which bite by day.^^ 

6. There are a good many other instances of the daj or 
night habits of blood-sucking insects having an immense 
influence upon the spread of disease. This has been especially 
well shown by Carpenter in the course of his studies upon 
sleeping sickness in the Lake Victoria region in Africa. One 
big problem was to find out which animal acted as an important 
reservoir of sleeping sickness from which human beings 
might become infected. The matter was to some extent 
simplified by the habits of the tsetse fly (Glossina palpalis)^ 
which carries the trypanosome of the disease from one host 
to another. The tsetse is diurnal in habits, and so there are 
various animals which it never comes across at all in the normal 
course of events. Certain potential enemies are avoided owing 
to this ; for it is preyed upon neither by bats which come 
out at night, nor by tree-frogs, which do not feed except at 
night .^^ On the islands of Lake Victoria the most important 
reservoir animals are the tragelaph (a species of marsh-haunt- 
ing antelope which comes out to feed just when the flies are 
" on the bite ") and the hippopotamus, which, although mainly 
nocturnal, comes out about half an hour before sunset and so 
is just in time to be bitten by the flies. 

7. The most violent fluctuations in light, temperature, and 
humidity are probably those found in deserts, where a man 
may be nearly dead with heat in the middle of the day and 
nearly freezing at night. Often the conditions are so severe 
that small rodents (which are usually rather sensitive to a dry 
atmosphere) are able to come out only at night. 

A good account of these changes is given by Buxton in his 
fascinating and scientific book Animal Life in Deserts? and 
the subject has been further studied by Williams ^i in a series 
of papers on the climate of the Egyptian desert. Williams 
found, Hke most biologists who are engaged upon intensive 



TIME AND ANIMAL COMMUNITIES 89 

ecological work, that the routine observations taken by meteoro- 
logists were not always of much use in the study of animals. 
Their observations are taken at rather arbitrary times and 
under extremely unnatural conditions, and are therefore often 
of little value to the ecologist. To take a simple example, 
meteorological screens are usually fixed at a height of 4 feet 
from the ground and the instruments in them record the 
climate at a height where comparatively few animals live. 
Furthermore, very few animals live in the open at that height — 
except cows and zebras and children and storks and certain 
hovering insects. What we have said is particularly true 
when the communities of animals at night and in the day are 
being worked out. As a matter of fact, the only kind of data 
which are of any use in the solving of this kind of ecological 
problem are accurate charts of temperature, humidity, and rain- 
fall, obtained from continuously recording instruments placed 
actually in the habitat which is being studied. In most places 
only the first of these is available, and even that may be absent. 
8. A careful study of the changes in external conditions 
during the day and night with reference to corresponding 
changes in the activities of animals is very badly wanted, for 
our ignorance of the matter is profound. It is remarkable to 
reflect that no one really knows why rabbits come out to feed 
only at certain times, and on different times on different days. 
Weather and diurnal changes are no doubt partly responsible, 
but there our knowledge ends. And yet rabbits are common 
animals and of great practical importance, and millions of 
people have watched their habits. We do not know whether 
Hght, temperature, humidity, or something else determines 
the appearance and retirement of animals at certain times. 
About the food-relationships of nocturnal animals we know 
less than about those of animals which come out in the day, 
and that is to say we know pathetically little. And, after all, 
it is quite as important to have information about the factors 
which limit animals in time as those which limit them in their 
spatial distribution, from whatever point of view we regard 
the question, whether from that of evolution or of wider 
problems in ecology. 



go ANIMAL ECOLOGY 

V 9. Most animals have more or less definite migratory 

movements during the twenty-four hours of day and night, 
and in some cases these are regular and rhythmical, but not 
necessarily correlated exactly with light and darkness. The 
result of these movements is to alter the composition of animal 
communities in any one place. Sanders and Shelford ^^ found 
that among the animals of a pine wood in North America there 
was a certain amount of diurnal migration up and down 
in a vertical direction. For instance, one species of spider 
(Tetragnatha lahoriosa) was to be found among low herbs at 
4.30 a.m. and among shrubs at 8.30 a.m., while another species 
(Theridium spirale) occurred in trees at 4.30 p.m. and in herbs 
at 8.30 p.m. Many insects (especially flies) occurred at 
different heights in the vegetation, depending upon the time 
of day. There exist similar vertical migrations among plankton 
animals in fresh-water lakes and in the sea. A number of 
species of, e.g., Crustacea come nearer to the surface during the 
night. 

10. We may repeat here that the distinction between day 
and night communities is not necessarily a very sharp one, 
and that there are a number of animals which come out both 
by day and by night {e.g. the common black bear of North 
America ^8), and others whose time-limits are determined by 
other factors {e.g. the slugs mentioned later on). The length 
of dusk varies throughout the year ; in England it is longest 
at midsummer and midwinter, and shortest in spring and 
autumn. Again, the amount of light at night is tremendously 
influenced by the state of the moon and the occurrence of 
cloudy weather. In fact, the distinction between day and night 
communities may turn out to be less marked than we might at 
first sight suppose ; but enough has been said to show that the 
alternation of day and night communities is a very important 
phenomenon and that it affects animal society profoundly in 
nearly all parts of the world. 

1 1 . In the polar regions there is no such alternation of day 
and night except during the spring and autumn ; and, since 
at these times the temperature is too low or the ground too 
snowy to support much animal life, the species living there are 



TIME AND ANIMAL COMMUNITIES 91 

nearly all typical daylight ones. And these form a permanently 
working community which lives in continuous daylight during 
the summer, and may in some cases have very little rest for 
three months — at any rate, as a population. Conversely, below 
a certain depth in the sea, or in big lakes, and in subterranean 
waters, and inside the bodies of animals, there is continuous 
darkness, so that the animals living there also form homo- 
geneous and permanent communities. Sometimes, however, 
the bodies of animals reflect the rhythm of their outer environ- 
ment and cause corresponding differences in their parasite 
fauna, as in the case of the Filarias already described. 
Probably the most conservative, smooth- working, and per- 
fectly adjusted communities are those living at a depth of 
several miles in the sea ; for here there can be no rhythms in 
the outer environment, such as there are on land. 

12. As we pass from the poles to the equator the night 
fauna begins to appear and becomes gradually more elaborate 
and important, until in such surroundings as are found in a 
tropical forest it may be more rich and exciting and noisy than 
the daylight fauna. Alexander von Humboldt ^°^ gives a 
good idea of this. Camping in the Amazon forests in the 
early nineteenth century, he wrote : " Deep stillness prevailed, 
only broken at intervals by the blowing of the fresh-water 
dolphins. . . . After eleven o'clock such a noise began in the 
contiguous forest, that for the remainder of the night all sleep 
was impossible. The wild cries of animals rung through the 
woods. Among the many voices which resounded together, the 
Indians could only recognise those which, after short pauses, 
were heard singly. There was the monotonous, plaintive cry 
of the Aluates (howling monkeys), the whining, flute-like notes 
of the small sapajous, the grunting murmur of the striped 
nocturnal ape {Nyctipithecus trivirgatus, which I was the first 
to describe), the fitful roar of the great tiger, the Cuguar or 
maneless American lion, the peccary, the sloth, and a host 
of parrots, parraquas (Ortalides), and other pheasant-like 
birds." 

In temperate countries the night-Hfe of animals is by no 
means so abundant or complex. This may be partly due to 



92 ANIMAL ECOLOGY 

the fact that whereas the tropical day and night are always 
twelve hours long throughout the whole year, the night in, 
say, the south of England is only eight hours long at mid- 
summer, and the day, therefore, sixteen hours. A consideration 
of the time-factor in animal communities opens up a number of 
interesting lines of inquiry ; we have considered day and night 
in some detail, as it is a clear-cut phenomenon and a fair 
amount is known about it in a general way. We may now 
turn to the subject of tidal variation. 

13. Tides. — In the intertidal zone on the sea-shore there 
is a marked division of the animals into those which come out 
or become active at high tide when covered with water and 
those which appear at low tide. The former group is of course 
the bigger, and forms the main part of the population, con- 
sisting of typical marine species. We can only give a few 
examples of individual cases, owing to lack of fully worked-out 
data on the subject. Since most of the dominant animals 
depend for their living upon plankton organisms in the water, 
they simply close down at low tide and start feeding again 
when covered by water at high tide. But there are a number 
of important animals commonly found on the shore at low tide 
■ — ^mostly birds such as waders, and these differ as to actual 
species according to the type of shore habitat. It will often 
be noticed that shore birds divide themselves up into rough 
zones when they are feeding, some feeding at the edge of the 
water, others nearer the shore, while others haunt the upper 
part of the shore near the drift- line. Again, there are different 
species found on mud-fiats, sandy shores, and rocky coasts. 
These differences in habitat of the birds are no doubt correlated 
with differences in the food, etc., in the various shore habitats. 
Besides birds, there are a number of insects which live between 
tide-marks, e.g. Anurida maritima^ which during high water 
hides in crevices in the rocks, surrounded by a bubble of air, 
and comes out to feed at low tide.^*^ The number of insects, 
mites, etc., which behave like this increases as we go towards 
the higher parts of the shore, since in these places they do 
not have to withstand such a long immersion in the sea during 
high tides. 



TIME AND ANIMAL COMMUNITIES 93 

Sometimes the effect of the tidal rhythm is overridden by 
that of light and darkness. It appears that many of the corals 
which form reefs in the tropics only become active and feed 
at night, closing down during the day.^°'^^ The times at which 
they can feed at night will sometimes be conditioned by the 
state of the tides. 

14. The Weather. — Most people are aware nowadays that 
variations in the weather from day to day are caused by the 
rather irresponsible movements of centres of high pressure 
(anticyclones) and centres of low pressure (depressions or 
storm-centres) in the atmosphere. It is customary to speak 
as if the controlling factors in weather were the depressions ; 
but this is only a convention which owes its origin to the 
natural belief that anything which disturbs our peace and 
happiness by bringing bad weather is an actively interfering 
agent, probably the Devil himself. In reahty, the two kinds 
of pressure areas and their complex relations are equally 
important, but for convenience we talk in terms of depressions. 
Changes in weather are associated with depressions travelling 
over the country, and although the actual path of depressions 
cannot yet be predicted with certainty, there is a perfectly 
definite and predictable series of events which accompanies 
their passage. Generally speaking (in England), a depression 
produces a zone of rainy weather in its front, and a still larger 
zone of cloudy weather including and extending beyond the 
rain-area, while in its rear there is fine weather again. Owing 
to various complicated factors this ideal series of events is by 
no means always realised in practice, but the sequence is true 
on the whole. These changes are accompanied by corre- 
sponding changes in temperature and humidity of the air, and 
by variations in the muddiness, hydrogen-ion concentration, 
etc., of fresh water. 

15. These cycles of weather are of varying length, but are 
usually of the order of a few days or a week or two, so that they 
fall in periodicity between tidal or diurnal changes, and the 
annual cycle of the seasons. They have important effects on 
animals. Many species are restricted in their activity to certain 
types of weather. For instance, most mammals avoid rain 



94 ANIMAL ECOLOGY 

because it wets their fur, and, by destroying the layer of warm 
air round their bodies, upsets their temperature regulation 
and makes them liable to catch cold. Mice tend to stay at 
home when it is raining hard, and the badger has to he in the 
sun to dry himself if he happens to get wet. Birds are not 
quite so much affected by weather conditions, since the 
architecture and arrangement of their feathers usually act as 
a more efficient run-off for rain. Many insects also necessarily 
stop work during wet weather owing to the danger of getting 
their wings wet, or to the drop in temperature often associated 
with the rain. 

1 6. There are, on the other hand, certain animals which 
come out only when it is wet (either when is is raining or when 
the ground is damp after rain). Slugs form a good example 
of this class of animal. This rule does not apply to all slugs, 
for there are some species which always Hve in damp places, 
as under vegetation ; it holds good mainly for certain of the 
larger wide-ranging slugs. A record of the activity of slugs 
was kept for some weeks in woods near Oxford, when the 
writer was trapping mice. The slugs visited the traps for the 
bait, and every morning the number of slugs was counted. 
The number of slugs walking abroad fluctuated greatly, and 
appeared to be determined mainly by the rain or dampness of 
the ground. These results were confirmed by casual observa- 
tions on the general occurrence of slugs on different days. 
On some days slugs might be seen practically waiting in queues 
trying to get into the mouse- traps, while on other days the 
latter would be entirely deserted. 

17. These examples serve to illustrate the general idea that 
animal communities vary according to the weather conditions, 
and that the variations follow a comparatively regular sequence, 
although the actual times and periods of the cycle are irregular. 
We have, broadly speaking, communities of fine weather, of 
wet weather, and of drying-up weather, but they grade into 
one another to a large extent. One significant thing is that 
the weather-cycle may entirely override that of day and night, 
as in the case of certain slugs. One big black slug with a grey 
stripe down its back {Limax cinereo-niger) comes out in wet 



TIME AND ANIMAL COMMUNITIES 95 

conditions only, and is more or less unaffected by the light or 
darkness. 

18. The effect of weather upon the habits of animals has 
a certain practical importance, since weather conditions control 
the habits of many blood-sucking flies, and their disposition 
to bite people. The tsetse fly {Glossina palpalis), which 
conveys sleeping sickness to man by its bites, is entirely a 
diurnal feeder, but also shows a marked tendency to bite more 
on some days than on others. Carpenter *^ says : *' The time 
when they are most eager to feed is early in a morning after 
a little rain, when the sun is hardly through the clouds, and it 
is close and still," while, on the other hand, '* if one wishes 
to avoid being bitten [the time] is in the middle of a day 
on which there is a fresh breeze, cloudless sky, and brilliant 
sun." 

19. Seasons of the Year. — It is unnecessary to say very 
much about this subject since, although its importance in 
ecology is immense, the general facts of annual changes and 
their effects upon the fauna are so well known as to require no 
underlining. In temperate regions the annual changes in 
plants and animals are primarily caused by variations in the 
amount of heat and light received from the sun. In sub- 
tropical countries this is also true, but there are often in addition 
very big variations in the rainfall, which are much more 
abrupt and regular than in more temperate climates. In the 
actual equatorial belt, with rain-forest, the temperature may 
be practically the same all the year round, while rain may be the 
only important climatic factor in which there are marked 
annual changes. In the Arctic regions the winter is so cold 
that active life among land animals (other than warm-blooded 
ones) is confined to the short summer season. It seems 
probable, then, that the greatest richness in variety of com- 
munities found at different times of year is in the temperate 
and subtropical regions, although the actual profusion of species 
is not so great as in the tropics. 

20. The difference between winter and summer in a country 
hke England is sufficiently great to change the animal com- 
munities to a considerable extent. This is a matter of every- 



96 ANIMAL ECOLOGY 

day experience, but the details of the changes, and the reasons 
for them, are at present little understood. A study of seasonal 
changes in the fauna leads us on directly to a number of 
problems, one of the biggest of which is that of bird migration — 
a vast subject which presents us with a number of smaller 
problems which are essentially ecological. The arrival of 
certain groups of birds in the spring, e.g. warblers, and their 
departure again in autumn, are linked up with the summer 
outburst of insect life, which in turn depends directly upon the 
rise in temperature, and indirectly upon temperature and light 
changes acting through plants. For instance, one of the 
biggest key-industries in many animal communities on land is 
that formed by aphids which suck the juices of plants. Many 
small birds depend for their food either directly upon aphids, 
or indirectly upon them through other animals. The aphids 
which form this food-supply are only abundant during the 
height of summer (June, July, and August), and thus their 
seasonal occurrence has enormously important effects upon the 
birds. So far the attention of ornithologists has been directed 
to the accumulation of facts about the actual dates and routes 
of migrations. This work has resulted in the setting of a 
number of problems, and the asking of a number of questions. 
A further advance which will throw light upon the ultimate 
reasons for the migration-behaviour of birds must be sought 
along ecological lines, and will only be attained by a careful 
study of the relations of each species to the other animals and 
plants amongst which they move in nature, and upon which 
they are vitally dependent. 

21. Since the biological environment is constantly shifting 
with the passage of the seasons, it follows that the food habits 
of animals often change accordingly. In the case of many 
higher animals, a different food is required for the young. 
The adult red grouse feeds upon the shoots of heather, 
but the young eat almost entirely insects and other small 
animals. The food of adult animals also changes in a regular 
way, especially when they are omnivorous or carnivorous. 
Niedieck'^2 describes how the big brown bear of Kamskatka 
varies his diet as the seasons pass. When he comes out in 



TIME AND ANIMAL COMMUNITIES 97 

spring from the snow-hole in which he has been hibernating, 
he has at first to eat seaweed, and a little later on may be seen 
actually grazing. In the middle of June the salmon start 
to come up the rivers from the sea, and from this time onward 
salmon forms the bear's staple diet. In August he also eats 
large quantities of wild peas which are then abundant, and in 
September, berries. Finally, in the late autumn, he goes and 
digs up ground-marmots (susliks) in the hills. Having 
accumulated enough fat to last him through the winter he 
retires into hibernation again and lives on it in a comatose 
condition until the following spring. 

Many foods are available only at certain times of the year, 
and this results in the formation of certain temporary niches 
which may be at the same time world-wide in their distribution. 
For instance, in Britain the raven (Cormis corax) feeds in 
spring upon the placentae, or afterbirths, of sheep, ^*^ and on 
the Antarctic ice-pack MacCormick's Skua (Megalestris 
MacCormicki) eats the afterbirths of the Weddell seals. ^-^ 

22. On the other hand, the difference of physical and 
chemical conditions may be so great that the same niche is 
filled by different species, often of the same genus, at different 
times of the year. There are a number of copepods of the 
genus Cyclops which live commonly in ponds, feeding upon 
diatoms and other alg^e. In winter we find one species, 
Cyclops strenuus, which disappears in summer and is replaced 
by two other species, C. ftiscus and C. albidus, the latter, 
however, disappearing in the winter. There are also, however, 
some species of Cyclops which occur all the year round, e.g. 
C. serratulus and viridis. Instances of this kind could be 
multiplied indefinitely, not only from Crustacea, but also from 
most other groups of animals, and many will occur to any 
one who is interested in birds. 

23 . We started with the conception of an animal community 
organised into a complicated series of food-chains of animals, 
all dependent in the long run upon plants, and we showed 
that each habitat has its characteristic set of species, but at 
the same time retains the same ground plan of social organisa- 
tion. When the factor of time is introduced it is immediately 

H 



98 ANIMAL ECOLOGY 

seen that each place has several fairly distinct communities 
(distinct in characteristic composition of species, not in the 
sense of possessing entirely different species) which come out 
and transact their business of feeding and breeding at different 
times. We have further seen that the changes in the environ- 
ment v^hich cause this division into communities at different 
times are in many cases regular and rhythmical, so that it is 
possible to classify the latter into definite types — day and night, 
high and low tide, wet and dry weather, winter and summer, 
and so on. In spite of the comparatively regular nature of 
these changing communities, they make the study of this side 
of ecology excessively complicated, and it is almost impossible 
to work out the food-cycles, etc., of any ordinary well- developed 
community of animals with anything remotely approaching 
completeness. The field worker is faced with masses of pre- 
liminary routine work in the way of collecting, etc., with little 
chance of getting on with a more fundamental study of the 
problems he is continually coming across. The incredibly 
intricate and complex nature of a fully developed community 
of animals is a really serious obstacle which has to be faced, 
especially as most ecologists are unHkely to be able to obtain 
the help of more than one or two others. 

24. It is therefore desirable, and in fact essential, that any 
one who intends to make discoveries about the principles 
governing the arrangement and mode of working of animal 
communities, should look round first of all with great care, 
with the idea of finding some very simple association of animals 
in which the complications of species and time changes are 
reduced to a minimum. The arctic tundra forms just such 
a habitat. Here time-changes are practically ruled out {i.e. 
there is only one community in each habitat, and it forms a 
comparatively homogeneous unit). Experience has shown 
that it is quite possible for one person to study with reasonable 
ease the community-relations of arctic animals, in a way that 
would be entirely impossible in some more complicated place 
like a birch wood.^^ In our own latitudes simple communities 
may also be found in certain rather peculiar habitats such as 
brackish water and temporary pools ; and it seems certain that 



TIME AND ANIMAL COMMUNITIES 99 

our knowledge of the social arrangements of animals will be 
most successfully and quickly advanced by elaborate studies 
of simple communities rather than superficial studies of com- 
plicated ones. 

In order to avoid misunderstanding it should be pointed 
out that the latter type of work is of great value and interest in 
other ways, above all for the light that it throws upon other 
aspects of ecology, e.g. distribution of species, or ecological 
succession. The statements made above apply only to eco- 
logists who intend to study the principles of social relations 
among animals, and it must not be thought that the value 
of general biological surveys is in any way depreciated. The 
latter will always form a most important part of ecological 
work in the field. 

25. If we followed the subject of time- communities to its 
logical conclusion (which happily we shall not do, since it 
would involve a consideration of astronomy and the causes of 
ice ages, and finally a discussion of the evolution of man), we 
should have to consider the larger periodic variations in climate 
from year to year, which undoubtedly exist, even though 
opinions may differ as to their exact cause and periodicity. 
For instance, in England there were severe droughts in 1899, 
191 1, and 1 92 1. In the same way there have been extremely 
wet years, or very cold winters (see p. 131), all of which have 
enormous effects upon wild animals. These periodic variations 
in the climate and weather have chiefly an influence upon the 
numbers of animals by encouraging or discouraging their 
increase, and therefore to some extent their distribution. 

26. The exact limits of the ranges of a number of animals 
are constantly shifting backwards and forwards, ebbing and flow- 
ing as the outer conditions change, and as the numbers of each 
species increase or decrease. We. understand at present little 
about the precise causes of these fluctuations in range ; but 
although the immediate influence at work may often be biotic, 
many of these changes are no doubt ultimately referable to 
short-period cHmatic pulsations, whether irregular or regular. 
For instance, in certain years there are great influxes into the 
British Isles of various animals not normally found there, or 



100 ANIMAL ECOLOGY 

only rarely. Well-known examples are the crossbill, the arctic 
skua, the sand-grouse, the clouded yellow butterfly, and various 
marine fish. These variations in the numbers of animals, or the 
arrival of entirely strange species, have a definite effect upon the 
species-composition of animal communities, but they are less 
important than such smaller rhythms as the seasons of the year. 
A consideration of this question leads us on naturally to con- 
sider the numbers of animals, and the means by which these 
numbers are regulated. This is a very big subject. It is 
also a very interesting one, and less is known about it than 
about almost any other biological subject. The study of 
animal numbers will form in future at least half the subject 
of ecology, and even in the present state of our knowledge 
it seems worth while to devote two chapters to it. This we 
shall accordingly do. 



CHAPTER VIII 



THE NUMBERS OF ANIMALS 



The subject of this chapter is an extremely important one, (i) but at present 
only quite a small amount is known about it. (2) Most people do not 
realise how immense are the numbers of wild animals now, and even 
more in the past ; but (3) in some parts of the world there are still huge 
numbers of the larger animals, which enable us to adjust our ideas on 
the subject. Examples among birds are : (4) aquatic birds on the 
White Nile, (5) guillemots in the Arctic, and (6) penguins in the Antarctic 
regions ; while examples among mammals are : (7) zebras in Africa, 
and, comparatively recently, bison in America, passenger pigeons in 
America, whales and walruses in the Arctic, and tortoises on the Gala- 
pagos Islands, and (8) the animals recently protected in various countries. 
(9) These examples, from amongst birds and mammals, enable us to form 
some idea of the colossal numbers of smaller animals (e.g. springtails) 
everywhere. (10) Another way of realising the large numbers capable 
of being reached by animals is to take cases of sudden and almost 
unchecked increase, e.g. among mice, (11) springboks, (12) insects, 
water-fleas, or protozoa, or (13, 14) epidemic parasites. (15) Such 
"plagues" are not uncommon under natural conditions, but are 
(16, 17) especially striking when animals are introduced by man into 
new countries. (18) This enormous power of multiplication is not 
usually given full rein, owing to the fact (19, 20) that every animal 
tends to have a certain optimum density of numbers, which (21) depends 
among other things upon the special adaptations of each species, and 
which (22, 23) applies both to herbivores and carnivores. (24) The 
existence of an optimum density of numbers leads us to inquire how the 
numbers of animals are regulated. (25) The food-cycle structure of 
animal communities forms one of the most important regulating 
mechanisms, (26, 27, 28) enemies being more important than food-supply 
as a direct limiting check on numbers of most animals ; while repro- 
ductive limitation is also of great importance but is more or less fixed 
by heredity, whereas the other checks are very variable. (29) We 
lack, at present, the necessary precise data for working out the dynamics 
of a whole community, and (30, 31) it is therefore not a simple matter 
to predict the effects of variation in numbers of one species upon those 
of other ones in the same community. (32) The existence of alternative 
foods for carnivores has an important influence in maintaining balanced 
numbers in a community. (33) Animals at the end of food-chains 
(i.e. with no carnivorous enemies) employ special means of regulating 
their numbers, e.g. (34) by having territory, (35) a system found widely 
among birds, and (36) probably equally widely among manimals. 
(37) Some carnivorous animals have means of getting food without 
destroying their prey, and may therefore encourage increase instead of 
limiting the numbers of the latter. 

I. In the two chapters which follow we shall point out some 
of the more important things about the numbers of animals 



102 ANIMAL ECOLOGY 

and the ways in which they are regulated, and show how a 
great many of the phenomena connected with numbers owe 
•their origin to the way in which animal communities are 
arranged and organised, and to various processes going on in 
the environment of the animals. We shall first of all try to 
give a picture of the enormous numbers of animals, both of 
individuals and species, that there are in every habitat ; then 
we shall describe the great powers of increase which they 
possess. This leads on to the question of what is the desirable 
density of numbers for different animals ; and it will be seen 
that the whole question of the optimum number for a species 
is affected by the unstable nature of the environment, which is 
always changing, and furthermore by the fact that practically 
no animals remain constant in numbers for any length of 
time. We have further to inquire into the effects of varia- 
tions in numbers of animals, and into the means by which 
numbers are regulated in animal communities. The final 
conclusion to which we shall come is that the study of animal 
numbers is as yet in an extremely early stage, but that it is one 
of profound importance both theoretically and practically, 
and one which can best be studied through the medium of 
biological surveys and study of animal communities, from the 
point of view of food-cycles and time- communities. 

2. It is rather difficult to realise what enormous numbers 
of animals there are everywhere, not only in species but in 
number of individuals of each species. The majority of 
animals, especially in this part of the world, are small and 
inconspicuous — ^it is estimated that at least half the species in 
the animal kingdom are insects — and their presence is therefore 
not very obvious. If you ask the ordinary person, or even 
the average naturalist, how many animals he thinks there are 
in a wood or a pond or a hedge, his estimate is always sur- 
prisingly low. Two boys of rather good powers of observation 
who were sent into a wood in summer to discover as many 
animals as they could, returned after half an hour and re- 
ported that they had seen two birds, several spiders, and some 
flies — that was all. When asked how many species of all kinds 
of animals they thought there might be in the wood, one 



PLATE VII 




{a) A guillemot cliff in Spitsbergen. The black streaks represent small flocks of 
guillemots flying away from their nesting places, disturbed by a gunshot. 
Each streak represents from ten to thirty birds. (Photographed by Dr. 
K. S. Sandford, 192^). 




(d) Adelie penguin rookery on Macquarie Ibland. Amongst the penguins can be 
seen small areas of tussock grass and rock. (From Sir Douglas Mawson, 
The Home of the Blizzard, by permission.) 



THE NUMBERS OF ANIMALS 103 

replied after a little hesitation " a hundred," while the other 
said " twenty." Actually there were probably over ten 
thousand. An exactly similar result is obtained with classes 
of zoological students who are taken out on field work ; and it 
is simply due to the fact that it requires a good deal of practice 
to find animals, most of which are hard to see or live in hidden 
places. It is well known that specialising in one particular 
group of animals enables a man to spot animals which any one 
else would miss. It is therefore necessary to allow an 
enormous margin on this account, when one is trying to get 
an idea of the actual numbers of animals in the countryside. 

3. Living as we do in a world which has been largely 
denuded of all the large and interesting wild animals, we are 
usually denied the chance of seeing very big animals in very 
big numbers. If we think of zebras at all, we think of them as 
" the zebra " (in a zoo) and not as twenty thousand zebras 
moving along in a vast herd over the savannahs of Africa. 
To correct this picture of the numbers of big animals, which 
were so much greater everywhere in the past, to accustom the 
mind to dealing with large numbers, and to help in forming 
some conception of the colossal abundance which is reached 
everywhere still by the smaller animals, we shall describe 
some of the places where animals of a convincing and satis- 
factory size may still be seen in enormous and even stagger- 
ing numbers. These places are mostly rather out of the 
way and owe the persistence of their rich fauna either to the 
existence of natural barriers such as an unhealthy climate 
or an unproductive soil, or to the sensible game-preserving 
methods of the natives. 

4. In the lower reaches of the White Nile, between the 
vast swamp of the Sudd and Khartoum, there exist countless 
numbers of water-fowl, many of them large birds as big as a 
man. For days it is possible to sail past multitudes of storks, 
herons, spoonbills, cranes, pelicans, darters, cormorants, 
ibises, gulls, terns, ducks, geese, and all manner of wading 
birds like godv/its and curlews. Abel Chapman ^^^ describes 
the scene as follows : " The lower White Nile, as just stated, 
is immensely broad and its stream intercepted by low islands 



I04 ANIMAL ECOLOGY 

and sand-banks diWded one from another by shallo\^'s, 
oozes, and back\vaters. At inten'als these natural sanctuaries 
are so completely caq)eted ^^-ith water-fowl as to present an 
appearance of being, as it were, ttssellutcd ^^ith li\-ing creatures, 
and tliat over a space of perhaps half a niile and sometimes 
more. These feathered armies are composed not only of ducks 
and geese but also of tall cranes, herons » and storks, marshalled 
rank beyond rank in sembLmce of squadrons of cavalr\'.'' 
This amazingly rich bird-life has been photographed by 
Bengt Berg,^'- and the reader may be referred to his book, the 
illustrations of which \\i\\ show that the description given 
above is not in tlie least exaggerated. 

5. It would be a mistake to imagine that such an abundance 
of birds is only to be met \Wth in tropical regions. In the .\rctic, 
where continuous daylight throughout the summer encourages 
a rich hars-est of diatoms and other ph)-toplankton near the 
surface of the sea, upon which is based an equally rich com- 
munit}- of plankton animals, there are still to be found in some 
places stupendous numbers of sea-fowl. Guillemots are 
particularly abundant, for they breed in great colonies on the 
sides of steep sea-cliifs, where their black and white costume 
makes them ver\- conspicuous. The ^^Titer is thinking of one 
clitf in particular on which the birds could be seen sitting 
packed close together on even- ledge of rock up to a height of 
a thousand feet or more. When a gun was tired a few odd 
hundred thousand or million birds would tly oil in alarm, 
without, however, noticeably affecting the nimibers still to be 
seen on the cliff. The photograph in Plate VII (ti) \rill give 
some idea of the number of birds ti\ing oil the clilT, and 
therefore a remote idea of the number not fl>*ing olf. In the 
photograph each streak of black represents not one bird, 
but a small tiock of ten to thirt}'. The noise made by 
these multitudinous egg-layers resembled that produced by 
an expectant audience in a vast opera house, t\Nittering and 
rustling its programmes. 

6. In the southern hemisphere the penguin rookeries 
afford an example similar to that of the guillemots in the north ; 
but the penguins spread out their colonies over the ground 



THE NUMBERS OF ANIMALS 105 

and not up the side of cliffs. If one examines the photographs 
of Adelie penguin rookeries given in antarctic books of travel 
(e.g. by Mawson ^^ for Macquarie Island), one can get a vivid 
idea of the numbers of birds involved. Imagine several 
million short gentlemen in dress clothes (tails) standing about 
in a dense crowd covering several square miles of otherwise 
barren country (see photo in Plate VII (b)). Viewed from a 
height they look like gravel spread uniformly over the land, 
with dark patches at intervals to mark the areas of tussock 
grass, which stand out as islands in the general ocean of 
penguins. 

In both these latter examples^ — the penguins and the 
guillemots — the birds represent the numbers from a very 
large feeding area concentrated in one place for breeding 
purposes ; and to this extent they do not give a fair idea of the 
normal density of the population. 

7. If we turn again to Africa, not this time to the rivers 
but to the open grass plains and savannahs, we shall find in 
some places vast herds of hoofed animals — zebra, buffaloes, 
and many kinds of antelopes. These are sometimes amazingly 
abundant. Percival ^^® records seeing a herd of zebras in close 
formation which extended for over two miles, and other 
observers have recorded similar large numbers in the case of 
other animals. Alexander Henry j^*^ describing the abundance 
of American buffalo in one place in 1 801, wrote in his journal : 
" The ground w^as covered at every point of the compass, as 
far as the eye could reach, and every animal was in motion." 
A hundred years later the bison was reduced to a small herd 
kept in a national park.^'^ The fate of the American bison is 
only one example of the way in which advancing civilisation 
has reduced or exterminated animals formerly so characteristic 
and abundant. The bison has practically gone ; the passenger 
pigeon has completely vanished ; but in 1869 a single town 
in Michigan marketed 15,840,000 birds in two years, while 
another town sold 11,880,000 in forty days.^^ The Arctic 
seas sw^armed with whales in the sixteenth century, but with 
the penetration of these regions by Dutch and EngHsh whalers 
the doom of the whales was sealed, and in a hundred and 



io6 ANIMAL ECOLOGY 

fifty years they had nearly all disappeared, while a similar 
fate is now threatening those of the southern hemisphere. In 
the photograph in Plate VIII can be seen the skulls and bones 
of a huge colony of walruses on Moffen Island, up in latitude 
80° N., which were slaughtered as they lay there, by these 
same early explorers. On the Galapagos Islands there is a 
similar cemetery of giant tortoises, of which only the shells are 
left to mark their former abundance. ^'^^^ Almost everywhere 
the same tale is told — former vast numbers, now no longer 
existing owing to the greed of individual pirates or to the 
more excusable clash with the advance of agricultural 
settlement. 

8. It is not much use mourning the loss of these animals, 
since it was inevitable that many of them would not survive 
the close settlement of their countries. The American bison 
could not perform its customary and necessary migrations 
when railways were built across the continent and when the 
land was turned into a grain-producing area.^^ Our object 
is rather to point out that the present numbers of the larger 
wild animals are mostly much smaller than they used to be, 
and that the conditions under which the present fauna has 
evolved are in that respect rather different from what one might 
imagine from seeing the world in its present state. At the 
same time there is in many cases no reason why animals should 
be reduced in numbers or destroyed to the extent that they 
have been and still are. From the purely commercial point 
of view it often means that the capital of animal numbers is 
destroyed to make the fortune of a few men, and that all possible 
benefits for any one coming later are lost. Enlightened govern- 
ments are now becoming alive to this fact, and measures are 
being taken to protect important or valuable animals. Thus 
the fur seal on the Alaskan Islands, which was in some danger 
of being gradually exterminated, has increased greatly under 
protection. Since 1910 killing has been prohibited on the 
Pribiloff Islands, except by Federal agents, and the herd of 
seals had increased from 215,000 in 1910 to 524,000 in 1919.21 
Again, the Siberian reindeer has been introduced into Alaska, 
with similar favourable results in repopulating the country 




V. c 






OJ 



is 

^ '6 

.§ c 
(u .;2 

o 

o ^O 

3 C 







OJ 


(fl 




^ 


ri 


"n 


!i: 


!1) 






o 


^ 
^ 


c 
o 


c 


o 




o 


j~ 





2 



Oh 



THE NUMBERS OF ANIMALS 107 

with animals. A little over a thousand animals were brought 
over in the period from 1891 to the present day, and the 
multiplication of these, under semi-wild conditions, has resulted 
in a great increase. It was estimated that there were 200,000 
of them in 1922.*^ It seems, then, that man is beginning to 
rectify some of his earlier errors in destroying large and 
interesting animals, and that the future will in certain regions 
show some approach to the original condition of things before 
man began to become over-civilised. 

9. When we turn to the smaller animals such as insects, 
worms, etc., we find that there are not very many accurate 
data about the density of their numbers, but it may be safely 
said that the numbers of most species are immensely great, 
reaching figures which convey little meaning to most people. 
Censuses which have been taken of the soil fauna at the Rotham- 
sted Experimental Station give some idea of the density of 
numbers reached. ^^* In an acre of arable land there were 
estimated to be over 800,000 earthworms (these figures being 
obtained by taking a series of small samples, making complete 
counts, and then estimating the total number in an acre). In 
a similar plot of arable land there were nearly three million 
hymenoptera, one and a half million flies, and two and a 
third million sp ringtails. These census figures bring out 
the interesting fact that many groups of small animals which 
are usually ignored or unfamiliar to zoologists bulk larger 
in numbers than other groups which have received a very 
great deal of attention. Such groups are the springtails, of 
which there were about two and a third millions, while of 
lepidoptera there were only thirty thousand individuals. 

10. We have attempted to give some idea of the great num- 
bers of many large animals in the past ; they are still to be found 
at the present day in secluded parts of the earth. By thinking 
in terms of large, interesting, and even spectacular species, 
it is possible to accustom the mind to dealing with the vast 
numbers in which the smaller, less noticeable, but none the 
less important forms are nearly everywhere found. We may 
also look at the matter from another point of view. We can 
consider what would happen if any one species were allowed 



io8 ANIMAL ECOLOGY 

to multiply unchecked for several years. Although various 
interesting calculations have been made about what w^ould 
happen if unlimited increase took place, and alarming pictures 
have been drawn of an earth entirely peopled with elephants 
so closely packed together that they would be unable to sit 
down except on each others' knees, there are as a matter of 
actual fact a number of wild animals which habitually multiply 
for short periods almost at the maximum rate which is theo- 
retically possible. It is therefore unnecessary to give rein 
to the imagination in this matter, since we can obtain actual 
examples. Some small mammals increase in numbers for 
several years at a very high speed until they reach such an 
immense abundance that malignant epidemic diseases break 
out and wipe out the major part of the population, and those 
which are left start again on another cycle of increase. Mice 
do this. Such an over-increase of mice has been described 
very vividly by Holinshed,^^ who wrote of a mouse plague in 
1581 : " About Hallontide last past, in the marshes of Danesy 
Hundred, in a place called South Minster, in the county of 
Essex . . . there sodainlie appeared an infinite number of 
mice, which overwhelming the whole earth in the said marshes, 
did sheare and gnaw the grass by the rootes, spoyling and taint- 
ing the same with their venimous teeth, in such sort that the 
cattell which grazed thereon were smitten with a murraine and 
died thereof." In 1907 another such *' plague " occurred in 
Nevada, during which 15,000 out of 20,000 acres of alfalfa 
were completely destroyed.^^ The natural increase of the 
field-mice {Microtus) was so terrific that the ground was in 
many places riddled with holes for miles. A Frenchman, 
describing a similar outburst in Europe, said that the ground 
was so perforated with holes as to resemble a sieve. In Nevada 
it was estimated that there were some 3,000 birds of prey and 
carnivorous mammals at work in the " plague " district, that 
these would be destroying about a million mice or more every 
month, and that this made no appreciable difference to the 
numbers. Again, one and a half million mice were killed in 
a fortnight in one district in Alsace during a great out- 
break in 1822,11^ while during the mouse plague in 19 17 in 



THE NUMBERS OF ANIMALS 109 

Australia 70,000 mice were killed in one stackyard in an 
afternoon ! ^7 

11. In former years great hordes of a small antelope 
called the springbok or trekbok used to occur periodically 
in South Africa.^ These appeared at intervals from the 
region of the Kalahari Desert, and in some of their migra- 
tions they marched south into the settled districts, doing great 
damage to crops on their travels. Eye-witnesses of these 
migrations have described the fantastically large numbers of 
animals taking part in them. One observer, after careful 
estimation, thought that there were half a million animals in 
sight at one moment, and it could be shown that the area 
covered by the whole migrating horde occupied a space of 
country one hundred and thirty-eight by fifteen miles. Even 
though they were not equally dense throughout, there must 
have been a good many ! Another says : *' One might as 
well endeavour to describe the mass of a mile-long sand dune 
by expressing the sum of its grains in cyphers, as to attempt 
to give the numbers of antelopes forming the living wave that 
surged across the desert in 1892 and broke like foam against 
the western granite range. I have stood on an eminence 
some twenty feet high, far out on the plains, and seen the 
absolutely level surface, as wide as the eye could reach, covered 
with resting springbucks, whilst from over the eastern horizon 
the rising columns of dust told of fresh hosts advancing." ^ 

12. The results of unchecked increase are also seen in a 
striking way in the big migrations of locusts and butterflies 
which have been recorded in various parts of the world. 
(In some of these cases the unusually large numbers have 
probably been due to local concentration into migratory 
swarms, rather than to over-increase of the population by 
breeding. In the majority of cases, however, there is almost 
certainly over-population, caused either by excessive increase 
of animals or by unusual scarcity of food, etc.) 

13. There are numerous cases in which similar outbursts 
in numbers among still smaller animals have been sufficiently 
large to attract notice. In Switzerland the railway trains are 
said to have been held up on one occasion by swarms of 



no ANIMAL ECOLOGY 

collembola or sp ringtails, which lay so thickly on the lines as 
to cause the wheels of the engines to slip round ineffectually 
on the rails. The fact that individual springtails are usually 
about one-twentieth of an inch long will give some idea of the 
numbers involved. Again, a huge multiplication of water- 
fleas (Cladocera) took place in the Antwerp reservoirs in 1896 ; 
the numbers were so serious that six men had to work night 
and day removing the water-fleas by straining the water 
through wire gauzes. It was estimated that ten tons of water- 
fleas were taken out — that is, two and a half times the weight 
of a large hippopotamus.^^ In the sea there are sometimes 
" plagues " of protozoa. Peridinians (e.g. Gonyaulax) some- 
times turn the sea to the colour of blood with their vast 
numbers, oiT the coast of India, of California, and of 
Australia.S2 They may be so numerous as to remove most 
of the free oxygen from the water, so that the fish die from 
suffocation. Gran once found that the water in Christiania 
Fjord was milky with a species of coccosphere {Pontosphcera 
Huxleyi)y which is a microscopic plant; and estimated that 
there were five to six million per litre.^^ 

14. Finally, there are the diseases caused by various para- 
sitic animals ; these are nothing more than a breaking away of 
parasites from the control of the host and increasing at an 
enormous speed. For example, malaria in the blood, and 
sleeping-sickness, and all such diseases are the result of over- 
increase of parasites, just as mouse-plagues are the result of 
over-increase in mice. The most striking epidemic diseases 
are of course caused by bacteria or by invisible " viruses,'* 
but they illustrate the same idea. 

15. We started to describe these examples of enormous 
multiplication in wild animals in order to emphasise the 
tremendous powers of increase possessed by them and by all 
animals. Any species, if given the opportunity, is capable of 
increasing in the same alarming way as the mice, the locusts, or 
the Gonyaulax ; and as a matter of fact most species probably 
do so occasionally, producing plagues which are rather sudden 
in onset, and which are terminated by disease or some other 
factors, or else are relieved by migration during which the 



THE NUMBERS OF ANIMALS iii 

animals mostly perish. It is not a rare or exceptional thing for 
a species to break out of control of its normal checks ; and we 
shall have to return to this subject again later on, since we shall 
see that " plagues " of animals are an inevitable consequence 
of the way in which animal communities are arranged and of 
the great instability of the environment. 

1 6. Many of the most striking cases of sudden increase in 
animals occur when a species is introduced into a country 
strange to it, in which it does not at first fit harmoniously, 
often with disastrous results to itself or to mankind. 

The most familiar example of such an introduction is that 
of the rabbit in Australia. It was also introduced into New 
Zealand, where it multiplied so excessively as to eat down and 
destroy the grass over wide areas, so that many thousands of 
sheep died from starvation. ^"'^ In the same way the Gipsy 
Moth (Lymantria dispar) was introduced into America from 
Europe, where it became for some years one of the more 
serious pests in forests, owing to its great increase and 
spread. It has been shown recently that the increase was due 
to the absence of its normal parasites, which keep down the 
numbers in Europe. The introduction of these parasites into 
America appears to have acted as an effective check, reducing 
the numbers to reasonable proportions. ^^ 

17. When an animal spreads rapidly in this way upon 
being introduced into a new country, there is usually a definite 
sequence of events, which is rather characteristic. At first 
the animal is unnoticed for several years, or else is highly 
prized as forming a link with the home country. Thus the 
starling was introduced into New Zealand by acclimatisation 
societies bent upon brightening the country with British 
birds.^^^ The next stage is that the animal may suddenly appear 
in the dimensions of a plague, often accompanied by a migra- 
tion, as when huge armies of rats marched over New Zealand 
in the early days.^^^ The starling was instrumental in spread- 
ing the seeds of the common English blackberry in New 
Zealand and has been undoubtedly one of the biggest factors 
in the production of the blackberry plague there. This has 
resulted in the formation of thickets of blackberry covering 



112 ANIMAL ECOLOGY 

the country for miles, making agriculture impossible, and in 
some places forming a danger to lambs, since the latter get 
caught inextricably on the thorns of the blackberry plants. 
Finally, after a good many years there is often a natural dying 
down of the plague. This is in most cases not due to the 
direct efforts of mankind in killing off the pests, but appears 
rather to be due to the animal striking a sort of balance with 
its new surroundings, and acquiring a set of checks which act 
fairly efficiently. Thus the rabbits in New Zealand now 
apparently have periodic epidemics, which reduce the popula- 
tion. A parallel case among plants is that of the Canadian water 
weed (Elodea or Anacharis canadensis), which was introduced 
into Europe by an enthusiastic botanist during the nineteenth 
century, and subsequently spread for some years like wild- 
fire, choking up rivers and lakes. After a certain time it 
appeared to lose its great multiplying power, and has now 
settled down to be a normal and innocuous member of the 
flora. The fame of Anacharis is still so great, however, that a 
certain town council in Britain, faced with a plague of '' water- 
bloom " in one of their lakes (water-bloom being caused by 
species of blue-green algae increasing abnormally in the water), 
hopefully stocked the lake with swans in order to eat down the 
Anacharis which was living there quite harmlessly. There are 
now a great many swans there. 

1 8. We have seen that animals possess extremely high 
powers of increase, which sometimes have a chance of being 
realised with results which are often very remarkable. Such 
high powers of increase do not merely reside in animals which 
have very large broods or breed very fast : there is no animal 
which could not (theoretically) increase enormously, given 
sufficient time and opportunity. Not more than fifty years 
at the maximum, or in most cases not more than two or three 
years, are required to achieve this result. For instance, the 
opossum (Trichosurus vulpecula), which was introduced from 
Australia into New Zealand, has only one young one every 
year, yet in some places it has increased alarmingly. 
Thomson ^'^^ says that the black-tailed wallaby introduced on 
to Kawau Island " ate out most of the vegetation, and starved 



THE NUMBERS OF ANIMALS 113 

out most of the other animals, being assisted in this by the 
hordes of opossums. They came out at night in the fields, 
grazing like sheep, and in the summer went into the garden, 
stripping it of fruit and vegetables." 

As Hewitt has pointed out, the converse of this is also true, 
and great abundance is no criterion that a species is in no 
danger of extinction. Just as an animal can increase very 
quickly in a few years under good conditions, so on the other 
hand it may be entirely wiped out in a few years, even though 
it is enormously abundant. The argument that a species is 
in no danger because it is very common, is a complete fallacy ; 
but is very often brought forward quite honestly, especially 
by people who have a financial interest in destroying the 
animals. One might mention the case of whales in the 
southern hemisphere. 

19. The examples of upsets in the normal balance of 
numbers which we have described bring us up against the 
question : what is the desirable density of numbers for any 
one species (" desirable " being used in the teleological sense 
of that density which will in the long run give the best chances 
of survival for the species) ? The question of the desirable 
number on a given area has received a great deal of attention 
from people studying the ecology of human beings. It is 
found that there is an optimum density of numbers for any 
one place and for people with any particular standard of 
skill. 6 To take a simple case : when a man is running a 
farm he cannot afford to employ more than a certain number 
of men on it, since after a certain point the income he gets 
from the farm begins to diminish. It does not pay to put in 
more than a certain amount of work as long as the standard 
of skill remains the same. If a new invention or a new idea 
opens up new lines of production, then it becomes possible 
to employ more men with advantage, but not more than a 
certain number. On the other hand, it does not pay the owner 
to employ less than a certain number of men if he is to get 
the maximum return from his outlay. If there are too few 
men working, the maximum production is not reached. 

20. Let us see whether the idea of optimum numbers 

I 



114 ANIMAL ECOLOGY 

applies to wild animals, and whether the analogy with man can 
be followed up. If we go into the question carefully, it soon 
becomes clear that there is an optimum density in numbers 
for any one species at any one place and time. This optimum 
number is not always the same and it is not always achieved, 
but in a broad way there is a tendency for all animals to strike 
some kind of mean between being too scarce and too abundant. 
As examples we may take the domestic cat in two of its wilder 
moments. Some years ago a schooner was run ashore on the 
coast of Tristan da Cunha, a remote island in the Southern 
Atlantic, and some of the ship rats were able to get ashore and 
colonise the island. In a few months they bred and increased 
excessively until they became quite a plague, even attacking 
and eating rabbits on the island. The inhabitants accordingly 
introduced some cats with the praiseworthy idea of extinguish- 
ing the rats. But the rats were so very much more numerous 
that they killed off the cats instead. ^^ In this case there were 
not enough cats. They were overpowered by weight of 
numbers. 

The other example is also about a small island, called Ber- 
lenga Island, off the coast of Portugal. *" This place supports 
a lighthouse and a Hghthouse-keeper, who was in the habit of 
growing vegetables on the island, but was plagued by rabbits 
which had been introduced at some time or other. He also 
had the idea of introducing cats to cope with the situation, 
which they did so effectively that they ultimately ate Up every 
single rabbit on the island. Having succeeded in their object 
the cats starved to death, since there were no other edible 
animals on the island. In this case there were too many cats. 

21 . If we follow up further the analogy with human density 
of population, it becomes clear that every animal tends to have 
a certain suitable optimum which is determined mainly by the 
habits and other characteristics of the species in question. 
But these are continually changing during the course of evolu- 
tion, and any such change is liable to cause a corresponding 
alteration in the optimum density of numbers. For instance, 

* This incident was related to me by Mr. W. C. Tate, the well-known 
authority on Portuguese birds, and is published here with his permission. 



THE NUMBERS OF ANIMALS 115 

if the cats on Tristan da Cunha had possessed poison fangs 
like a cobra they might have been able to maintain themselves 
with a small population. On the other hand, if the cats on 
Berlenga Island had possessed chloroplasts like Euglena^ they 
might have been able to exist permanently, without eating out 
their food-supply. Again, the density possible for a species 
depends partly upon the size of the animals. Given the same 
food-supply and other things being equal, a small species can 
be more abundant than a large one. This has a certain im- 
portance in ecology, since there are a great many examples of 
species in the same genus, and with the same sort of food 
habits, differing in size to a very marked extent. The common 
shrew {Sorex araneus) and the pygmy shrew {Sorex minutus), 
the small and large cabbage white butterflies (Pieris rupee and 
brasstcce), the smooth newt {Molge vulgaris) and the crested 
newt {Molge cristatd), are instances. Of course if the size- 
differences are too great they often automatically involve 
different food habits, so that the two animals cannot be 
compared closely. 

22. The principle of optimum density applies equally to 
any herbivorous animal. Under normal circumstances the 
numbers of deer are kept down by two big factors — enemies 
and disease. Recently the deer in a sanctuary in Arizona were 
left to themselves for some years. Owing to the absence of 
their usual carnivorous enemies {e.g. cougars or wolves) they 
increased so much that they began to over-eat their food-supply, 
and there was a serious danger of the whole population of deer 
starving or becoming so weakened in condition as to be unable 
to withstand the winter successfully. The numbers were 
accordingly reduced by shooting, with the result that the re- 
maining herds were able to regain their normal condition.^i 
Here it was clear that the absence ^of their usual enemies was 
disastrous to the deer, that the former are in fact only hostile 
in a certain sense, in so far as they are enemies to individual 
deer ; for the deer as a whole depend on them to preserve their 
optimum numbers and to prevent them from over-eating their 
food-supply. 

23. One more example may be given. Carpenter, when 



ii6 ANIMAL ECOLOGY 

carrying out a survey of the islands on Lake Victoria in order 
to discover the distribution and ecology of the tsetse fly, noted 
that islands below a certain size did not support any flies at 
all, although the conditions for breeding and feeding (v^hich 
are well defined and regular) were otherwise apparently 
quite suitable.^* The explanation of this was probably that 
the fly population is subject to certain irregular checks upon 
numbers, and that any one population must be sufficiently 
large to survive these checks. There would not be a big 
enough margin of numbers on a very small island. 

There are suggestions of a similar state of affairs among 
certain protozoa. It has been found that if there are too few 
individuals in a culture they do not live so successfully,^! and 
this is also said to be true of cells growing in tissue- cultures. 
Again, it has been found that the minimum density is not the 
optimum density for a population of the fruit-fly Drosophila 
growing in the laboratory. ^^ It is quite probable that there 
are sometimes physiological or even psychological reasons 
controUing the desirable density of population, just as it is 
bad for most people to Hve alone, or, on the other hand, under 
too crowded conditions. But we do not know much about 
this matter among wild animals. It may be of very great 
importance in their lives and cannot be ignored as a possible 
factor affecting numbers. 

24. Before going on, it will be convenient to sum up what 
has been said so far about the numbers of animals. Most 
animals are more numerous than is usually supposed, and it 
is necessary to accustom the mind to deaHng with large actual 
numbers of individuals. One is the more likely to under- 
estimate the numbers of animals, owing to the destruction of 
the large and more conspicuous species which were formerly 
so much more abundant in many parts of the world, now 
occupied by industrial or agricultural civilisation. Descrip- 
tions of the enormous numbers in which these larger animals 
still exist in the more secluded parts of the world and 
of the former numbers of animals which are now rare 
or extinct, enable us to grasp to some extent the vast 
abundance of the smaller and more inconspicuous forms. 



THE NUMBERS OF ANIMALS 117 

That is the first point — the vast numbers of animals almost 
everywhere. 

We have further seen that all animals possess an extremely 
high power of increase, which if unchecked leads to over- 
increase on a large scale, so that a " plague " of one species or 
another is produced ; the best-known cases being mice and 
locusts. All animals are exerting a steady upward pressure in 
numbers, tending to increase, and they sometimes actually 
do so for a short time. That is the second point. We next 
considered the question of the desirable density of numbers 
for a species, and we saw that each species tends to approach a 
certain optimum density, neither too low nor too high, which 
is not the same at different times or in different places. If 
there are too few individuals the species is in danger of being 
wiped out by unusually bad disasters, and if the numbers are 
too great other dangers arise, the most important of which is 
the over-eating of the food-supply. The latter is always the 
ultimate check on numbers, but in practice other factors usually 
come in before that condition is reached. 

25. We now have to consider the regulation of numbers, 
the ways in which this desirable density of numbers is main- 
tained. How do animals regulate their numbers so as to 
avoid over-increase on the one hand and extinction on the 
other ? The manner in which animals are organised into 
communities with food-cycles and food-chains to some extent 
answers the question. As a result of the existence of pro- 
gressive food-chains, all species except those at the end of a 
chain are preyed upon by some other animals. Snails are 
eaten by thrushes, the thrushes by hawks ; fish are eaten by 
seals, seals by sea leopards, and sea leopards by killer whales ; 
and so on through the whole of nature. Most species usually 
have a number of carnivorous enemies, but in some speciaHsed 
cases may have only one. The latter condition is, however, 
extremely rare ; it is the commonest thing in the world to 
find a species preying exclusively upon another, but unusual 
for a species to have only one enemy. Every species has also 
a set of parasites living in or on it, which are often capable of 
becoming dangerous when they are very numerous. So, in a 



ii8 ANIMAL ECOLOGY 

general way, the food-cycle mechanism is in itself a fairly good 
arrangement for regulating the numbers of animals, and it 
works efficiently as long as the environment remains fairly 
uniform, or at any rate as long as its periodic pulsations con- 
tinue fairly steadily and regularly. If the balance of numbers 
in a community is upset by some sudden and unusual occur- 
rence, then the ordinary relations of carnivores and parasites 
to their prey are no longer effective in controlling numbers, and 
various results of a curious nature ensue. With the effects of 
irregularities in the surroundings of animals we shall deal more 
fully in the next chapter. We are here concerned chiefly with 
the ways in which the general system of food-cycles and food- 
chains in animal communities acts as a method of regulating 
numbers. 

26. It is plain enough that the amount of food available 
sets an ultimate limit to the increase of any animal ; but in 
practice, starvation seldom acts as a direct check upon numbers, 
although the possibility of it is always present. Instead we 
find that other factors, such as enemies of all kinds, usually 
keep numbers down well below the point which would bring 
the population in sight of starvation. There appear to be 
several good reasons for this. First of all, food, whether of 
an animal or plant nature, is not always available ; or, what 
comes to the same thing in the end, is not always increasing to 
keep pace with the needs of the animals requiring it ; so that 
the maximum numbers feasible for an animal at any moment 
are not only determined by the food-supply at that moment, 
but must be adjusted to the needs of the future. It would 
be an unworkable system for animals to live all the time up to 
the extreme limits of their food-supply, since no margin would 
be left for the times of scarcity which are always liable to 
occur. This can be well seen in the example of deer in 
Arizona quoted previously, where increase during the summer 
imperilled the food-supply for the following winter. It is one 
of the most obvious ideas to all stock-farmers that the number 
of cattle which can be kept on a given acreage is determined 
by the margin of food left over for the winter (when plant 
growth ceases), as well as by the immediate requirements of the 



THE NUMBERS OF ANIMALS 119 

animals during the summer. Another point is that over- 
eating of the food-supply usually results in the destruction of 
the entire population, irrespective of individual merits. There 
have been instances recorded of the various oak moths (such 
as Tortrix viridana) eating all the leaves of the trees upon which 
they were living and then simply dying of starvation, just as 
the cats did on Berlenga Island.-^^^ 

27. It is usual, therefore, to find that gregarious gluttony 
of the whole population is avoided by having various other 
checks which act in two ways, first by aflFecting the chances of 
reproduction or by Hmiting the number of young produced, 
and secondly by eliminating the animals themselves. The 
first method is often fixed more or less permanently by the 
hereditary constitution of each species ; but there are a great 
many cases known in which the weather affects mating or 
breeding, or in which climate or food-supply vary the number 
of young produced in a brood, or the number of broods born 
in a year. For instance, the short-eared owl (Asio flammeus) 
may have twice as many young in a brood and twice as many 
broods as usual, during a vole plague, when its food is extremely 
plentiful. ^^'^ But these variations in the reproductive capacity 
are small compared to the limits which are imposed by the 
constitution of the animals. 

The second result is brought about mainly by means of 
predatory enemies, carnivores or parasites, or both, not to 
speak of other checks such as climatic factors . These influences 
dispose all the time of a certain margin of the population, so 
that there are left a certain number of comparatively well-fed 
and, as it were, well- trained animals ; for these checks act 
selectively and probably have important effects on the quality 
as well as the quantity of the population. Starvation only 
comes in in various indirect ways, as by lowering the resistance 
to attack by carnivorous enemies or parasites, or to the weather, 
and so increasing the selective power of these agencies. 

28. The regulation of numbers of most animals would 
appear, therefore, to take place along the following lines. Each 
species has certain hereditary powers of increase, which are 
more or less fixed in amount for any particular conditions. It 



120 ANIMAL ECOLOGY 

is usually also kept down in numbers by factors which aflPect 
breeding, e.g. lack of breeding-sites, etc. It is further controlled 
by its enemies, and if these fail, by starvation. But the latter 
condition is seldom reached. There are a few species which 
seem to regulate their numbers almost entirely by limiting re- 
production, although they belong to groups which are normally 
controlled by carnivores. There is a species of desert mouse 
(Dipodomys merriani) which only has two young per year, 
that is to say, probably very few more than would be necessary 
to replace deaths in the population caused by old age or 
accident.-^^^ This, however, is unusual except in the case of 
animals at the end of a food-chain, with which we shall deal 
later. 

It has been necessary to speak in generalities, since so little 
is known at present about the rules governing the regulation 
of animal numbers. There are, however, a number of special 
separate phenomena which we shall pick out, and which will 
serve to illustrate the importance of animal interrelations, and 
the study of animal communities along the lines of food-cycles. 

29. When we are dealing with a simple food-chain it is 
clear enough that each animal to some extent controls the 
numbers of the one below it. The arrangement we have 
called the pyramid of numbers is a necessary consequence of 
the relative sizes of the animals in the community. The smaller 
species increase faster than the large ones, so that they produce 
a sufficient margin upon which the latter subsist. These in 
turn increase faster than the larger animals which prey upon 
them, and which they help to support ; and so on, until a stage 
is reached with no carnivorous enemy at all. Ultimately it 
may be possible to work out the dynamics of this system in 
terms of the amount of organic matter produced and consumed 
and wasted in a given time, but at present we lack the accurate 
data for such calculations, and must be content with a general 
survey of the process. The effect of each stage in a food- 
chain on its successor is easy to understand, but when we try 
to estimate the effect of, say, the last species in the chain upon 
the first, or upon some other species several stages away, the 
matter becomes complicated. If A keeps down B, and B 



THE NUMBERS OF ANIMALS 121 

keeps down C, while A also preys on C, what is the exact 
effect of A upon C ? Two examples will show the sort of way 
in which this process works. 

30. For some years the great bearded seal or storkobbe 
{Erignathus harhatus) has been very intensively hunted and 
killed by Norwegians who go up every summer into the outer 
fringes of the ice-pack round about Spitsbergen. They seek 
the seals for the sake of their skins and blubber. The serious 
toll taken of their numbers can be gauged by the fact that one 
small sealing-sloop may bring back five thousand skins and 
sometimes many more in the course of a single summer. In 
spite of this steady drain on the numbers of seals the animals 
are, if an)rthing, more abundant than ever. This appears to 
be due to the fact that the Norwegian sealers also hunt and 
kill large numbers of polar bears, whose staple article of diet 
is the bearded seal, which they stalk and kill as they lie out on 
the pack-ice. By reducing the numbers of bears the sealers 
make up for their destruction of seals, since there are so many 
extra seals which would otherwise have been eaten by bears. 
The diagram in Fig. 9 sums up the situation which has just 



3£AL )-BEAR > 

I 



MAN 



Fig. 9. 

been described. In this case we know the results of man's 
interference, but they might very well have been different. 
For instance, if fewer bears had been killed, the seal numbers 
might have gone down considerably. On the other hand, if 
the same number of bears had been killed and more seals 
destroyed by the Norwegians, the seal numbers might also 
have gone down. The final result, as far as seals are con- 
cerned, depends entirely upon the relative numbers and de- 
structive powers of the species concerned. In this case a 
balance happens to have been struck. 

31. The second example illustrates the same point. The 
tsetse fly {Glossina palpalis) is preyed on in part of the Lake 
Victoria district by a small species of dragonfly {Cacergates 
leucosticta), and the latter is preyed on in turn by a larger 



122 ANIMAL ECOLOGY 

dragonfly. Both dragon-flies are eaten by various species of 
bee-eaters {Merops and Melittophagus)A^ Problem : what is 
the effect of the bee-eaters upon the numbers of tsetse flies ? 
The diagram in Fig. lo sums up the food-relations which we 



1 ^ 

TSETSE > DRAGONFLY >DRAGONFLy >BEE-EATER 

Fig. io. 

have described. It is clear that in this case we cannot say at 
a glance whether the birds are having a beneficial influence 
by helping to reduce tsetse flies, or the reverse. The result 
depends entirely upon the relative numbers of the species 
concerned in the matter, and upon a number of other things, 
such as the food preferences of the bee-eaters, the number of 
individuals eaten in a given time by the dragonflies, the rate 
of increase of the different species, and so on. But the example 
does show that each species will have some effect upon the 
numbers of the others, even though we cannot precisely define 
it without further investigation. In fact, no species in a 
community, unless it happens to live a very isolated life or be 
very rare, is without its effect upon numbers of the rest of the 
community, and that is why it is practically hopeless to reach 
any complete knowledge of the natural methods of regulation 
of numbers of an animal without doing a general biological 
survey, backed up later by some investigation of the food- 
cycle. 

32. It might be thought that there would be some danger 
of enemies doing too much in the way of controlling the 
numbers of their prey, so that the carnivore would run a 
risk of eating out its food-supply, and the prey of being 
exterminated or reduced below its lower limit of safety. There 
is, however, a natural method by which such a contingency is 
usually avoided, depending upon the fact that most carnivores 
do not confine themselves rigidly to one kind of prey ; so that 
when their food of the moment becomes scarcer than a certain 
amount, the enemy no longer finds it worth while to pursue 
this particular one and turns its attention to some other species 
instead. This process was pointed out by Hewitt, who gave 



THE NUMBERS OF ANIMALS 123 

as his example the goshawk {Accipiter atricapillus) in Canada, 
which preys ahernately on the varying hare (Lepus americanus) 
and upon grouse, according as one or the other is more 
abundant. In this way, whenever one species becomes for any 
reason scarce, the goshawk tends to eat more of the other and so 
allows the first one a certain amount of respite.^^ This switch 
arrangement is common enough in animal communities, and 
is probably an important factor in preventing the complete 
extermination of animals which happen for any reason to be 
at a rather low ebb of numbers {e.g. after an epidemic). In just 
the same way, the red fox in Canada preys on mice or varying 
hares according to their relative abundance. 

33. It will already have occurred to the reader that the 
animals which are at the end of food-chains — at the top of the 
pyramids — are in a peculiar position, since they have no 
further carnivorous animals present which might control their 
numbers, although they have of course parasites. These 
animals have in many cases evolved rather curious methods of 
regulating their numbers, of which we can only mention a few 
here. The Emperor Penguin is a large bird which forms the 
end of a long chain of marine animals (it appears to live chiefly 
on animals like fish and squids), and breeds in the heart of the 
very cold Antarctic winter. Since it has no serious enemies 
to control its numbers (nothing is known as to whether it has 
epidemics), it seems to depend chiefly on climatic factors to 
bring this about : or rather we should say that the only checks 
against which it has to produce extra numbers are climatic 
ones. One important thing is the cold, since the birds attempt 
to incubate their eggs and hatch their young at a temperature 
ranging below —70° F. and in severe blizzards. They are 
also destroyed by avalanches of snow, which cover them and 
cause desertion and freezing of the eggs. (Some birds were 
seen attempting to hatch out pieces of ice, which they had 
mistaken for their eggs.^-^) In other cases, animals at the end 
of food-chains may control their numbers by not breeding 
at all in some years. This appears to happen with the snowy 
owl, and probably with skuas, in certain years when food 
(especially lemmings) is scarcer than usual. Or again, the 



124 ANIMAL ECOLOGY 

reproduction of the carnivore may be always adjusted to such 
a low rate that there is hardly ever any danger of over- eating 
its food supply, and its numbers always remain relatively 
small. This is not so common, however, as other methods. 

34. The regulation of numbers of terminal animals is seen 
at its best in some of the birds and mammals which are either 
at or near the end of food- chains, e.g. hawks and tigers on the 
one hand, or warblers and insectivorous animals on the other. 
The fact that animals become less abundant as we pass from 
key-industry herbivores to the carnivores at the end of the 
chain makes it possible for animals at a certain point in the 
series (at a certain height in the number-pyramid) to limit their 
numbers by dividing up their country (and therefore the 
available food-supply) into territories each owned by one or a 
few individuals. For instance, in the EngHsh Lake District 
each buzzard or pair of buzzards requires a certain stretch 
of country to supply it with enough food, and the same 
applies to a bird like the Dartford warbler, which lives in the 
heather and furze heaths of Southern England. The division 
of country into territories is especially common at the beginning 
of the breeding season, when it is necessary not only to provide 
for the immediate needs of the animals, but also for the needs 
of the young which will appear later on. 

35. We owe our knowledge of the existence and nature of 
bird territory chiefly to Eliot Howard,^ whose remarkable 
studies on the subject, especially among the warblers, have 
opened up an entirely new field of ecological work. He showed 
that among birds like the willow wren (which migrates north- 
wards into England every spring) there is a very regular system 
of dividing up their habitat into parcels of land of roughly 
equal value. The arrangement is that the male birds arrive 
first in the early spring, before the females, and fight amongst 
themselves for territory ; and are then followed by the females, 
each one of which becomes attached to one male. Ultimately 
the nest is built and young produced and reared. At the end 
of the season the territories are given up and the birds go south 
again. We are not concerned here with the ways in which 
this territorial system in warblers and other birds is connected 



THE NUMBERS OF ANIMALS 125 

with courtship and nesting habits. These and other matters of 
great ecological interest may be found described and discussed 
in the works of Howard. It should be pointed out that 
territorial systems among birds are not always for the purpose 
of dividing up the food-supply in a suitable way. They may 
also be equally important in limiting numbers by being con- 
cerned with nesting-sites. 

36. Less is known about territory conditions in carnivorous 
mammals, but it is pretty clear that they do in many cases have 
territories, and for the same reasons as vv^ith hawks and warblers. 
It is well known that animals like the African lion or Indian 
tiger are few in numbers, and that each district will have one 
pair or one family living in it. We know practically nothing 
about the way in which such animals settle the size of their 
territory, and little enough about the extent to which the terri- 
torial system is found among mammals at all. There are 
indications that some herbivorous mammals limit their numbers 
to some extent in this way. For instance, Collett ^^^ says 
that the lemming does so in normal years. It is probable that 
insects like ants, which live in great towns, have some system 
of spacing out their colonies so as to avoid overcrowding. The 
whole subject requires more investigation before we can say 
exactly how important a part it plays in the lives of animals ; 
but Howard's work on birds is sufficient to show that it is 
certainly a very effective method of limiting numbers in some 
species, and it will probably be found to occur very widely 
among animals which do not possess any other convenient 
means of regulating numbers. 

37. We may conclude this chapter by referring to the 
animals which live by exploiting the work of other species, but 
do not actually eat them or destroy them in the process. The 
majority of parasites belong to this class. Such animals as 
tapeworms do not always harm their hosts very greatly, except 
that they divert a certain amount of food from its proper 
destination in the tissues of the host. In the same way fleas, 
if not too numerous, do not necessarily do much direct harm 
by withdrawing blood from their host, although they may 
accidentally spread the germs of disease in that way. Parasites 



126 ANIMAL ECOLOGY 

do not usually limit the numbers of their hosts except when 
the latter have increased unduly (as when fish die from tape- 
worm epidemics or grouse on overcrowded moors from nema- 
tode disease), or when the parasite has got into the wrong host 
(as with the trypanosome, which causes human sleeping-sick- 
ness). Most parasites exist by exploiting the work done by 
their hosts, without actually destroying them, just as a black- 
mailer takes care not to ask for too much money at one time. 
Besides parasites there are also certain animals which are not 
parasites (in the sense of living on their host), but which at the 
same time employ similar means of obtaining sustenance from 
their prey without destroying the latter. For instance, some 
species of ants keep farms of aphids which they visit for the 
purpose of getting drops of liquid containing food which has 
not been completely absorbed by the aphids themselves. 
These aphid farms may be carefully protected by the ants.^*^ 
Another example of avoiding killing the goose that lays the 
golden eggs is described by Beebe.^^^ On one of the Galapagos 
Islands there lives a scarlet rock-crab {Grapsus grapsus) 
which inhabits the lava rocks by the seashore. This crab is 
pursued by a species of blue heron {Butorides sundevalli) 
which catches the crab by the leg, upon which the crab breaks 
off the leg by autotomy, leaving it in the possession of the 
heron. Thus both animals get what they want — the heron 
its food and the crab its life. 

These cases of exploitation without destruction have been 
described in order to show that the food- cycle mechanism is 
not always effective in limiting numbers of animals. In such 
cases some other means must be employed. Sometimes 
another animal in the community, which still follows the old 
method of destruction, acts as a real control of the numbers 
of the species concerned. 



CHAPTER IX 

VARIATIONS IN THE NUMBERS OF ANIMALS 

The numbers of animals (i) never remain constant for very long, and 
usually fluctuate considerably, and often rather regularly, e.g. (2) many 
insects, (3) marine littoral animals, (4) protozoa in the soil, and elephants 
in the tropical forests. (5) The primary cause of these fluctuations is 
usually the unstable nature of the animals' environment, as is shown by 
the effects of periodic bad winters on the numbers of certain small 
birds. ((6) Fluctuations in numbers are of common occurrence among 
birds, but their causes are not usually known.) (7) Further instances 
of the irregular nature of the environment are cyclones in the tropics, 
and droughts affecting small ponds, but (8) such factors are not neces- 
sarily destructive, and may favour sudden increase of the animals. 
(9, 10) Some of the best data about fluctuations in numbers are from 
mammals, e.g. the lemming, whose fluctuations are partly controlled 
by climate and partly by periodic epidemics, (11) the latter occurring 
in a number of other small rodents, such as mice, (12) rabbits and hares, 
(13) marmots and muskrats ; but (14) the beaver forms an interesting 
exception to this rule. (15) These fluctuations in numbers affect the 
other animals dependent upon the rodents, and since (16) the length 
of the period of fluctuation depends chiefly on the sizes of the rodents 
concerned, the final effects upon animal communities are very complex. 
(17) Wild ungulates, at least in some cases, also have periodic epidemics. 
(18, 19) The occurrence of " plagues " of animals results from the 
structure of animal communities, and from the irregularities of the 
environment, and (20) especially from the irregularities occurring over 
wide areas, at the same time. (21) Among other results of these periodic 
changes in numbers are changed food habits, since (22) food preferences 
depend both on the quality and the quantity of the food. (23) The 
other habits of a species may also change with the variations in density 
of its numbers. 

I . So far we have been speaking as if the numbers of animals 
remained fairly constant. We have been describing the general 
mechanisms which assist in bringing about the optimum 
density of numbers for each species. In the present chapter 
we shall point out that practically no animal population remains 
the same for any great length of time, and that the numbers of 
most species are subject to violent fluctuations. The occa- 
sional " plagues " of animals already referred to are extreme 

127 



128 ANIMAL ECOLOGY 

examples of sudden variations in numbers. But the variations 
are usually less spectacular, although hardly less important. If 
we take any animal about which we possess a reasonable amount 
of information, we shall find that its numbers vary greatly 
from year to year. The common wasps (Vespa vulgaris and 
V. germanica) are very abundant in some summers, and very 
scarce in others. It is well known to collecting entomologists 
that there are very good and very bad years for butterflies. 
Although the data are scattered and have not so far been 
properly correlated, a study of the existing literature leaves 
no doubt that there is enormous annual variation in the numbers 
of insects. A typical example is recorded by Coward ,^2 who 
noted that the mottled umber moth {Hyhernia defoliaria) was 
particularly abundant on oak trees in Cheshire in 191 8 and 19 19, 
and especially in 1920, and caused much defoliation. At the 
same time leaf-galls, especially the spangle (caused by Neuro- 
terus lenticularis , a species of the Cynipidce), were unusually 
plentiful in 1920. He also noted that the acorn-crop was 
average in 191 8 and 1919, and failed entirely in 1920. These 
changes in the trees and the insects must have had an appre- 
ciable effect upon the other animals in the oak woods, since 
jays and wood-pigeons eat the acorns ; starlings, chaffinches, 
tits, and armies of warblers eat the caterpillars of the moth ; 
while pheasants and other birds eat the spangle galls. This 
example will give some idea of the way in which the numbers 
of insects are always shifting from year to year, and how the 
changes must inevitably affect a number of other animals 
associated with them, often causing the latter in turn to vary 
in numbers. 

2. In other countries similar variations in numbers occur. 
The sugar-cane froghopper of Trinidad fluctuates com- 
paratively regularly, with a period of four or five years. -^^^ 
The bad " cotton- worm " (weevil) outbreaks in the United 
States occur at intervals of about twenty-one years.^^^ Aphids 
occur very numerously in some summers, e.g. in 1836 there 
was a very big maximum of numbers in Cheshire, Derbyshire, 
and South Lancashire and Yorkshire, and in one place the 
swarms occurred over an area of twelve by five miles. ^^'^ 



VARIATIONS IN THE NUMBERS OF ANIMALS 129 

There were similar aphid swarms in parts of England in the 
year 1869,22 and doubtless many other ones which have 
escaped being recorded. The aphid increase sometimes 
affects the numbers of their normal enemies such as ladybirds 
and syrphid flies, which may attain vast numbers at an aphid 
maximum.22 

In some cases the fluctuations in numbers of insects are 
extraordinarily regular in their rhythm. The most remarkable 
of these are the cycles in numbers of cockchafers {Melolontha 
vulgaris and M. hippocastani) in Central Europe. Every few 
years these beetles appear in countless numbers, and the 
heavy damage which they do to crops and forest trees has 
directed much attention to the phenomenon, so that we possess 
very good records for a number of years back. In some 
districts there is a regular three-year cycle, in others a four- 
year cycle. The maximum numbers have in some places 
occurred exactly every three years for more than sixty years 
at a time.^^ There is also a fifteen-year period in the size of 
the maxima. 

3. Marine animals also are subject to considerable variations 
in numbers from year to year, and these are well shown by the 
careful notes made by Allee ^^ upon the marine littoral species 
in the neighbourhood of Woods Hole, Massachusetts. An 
echinoderm, Arhacia punctulata, was particularly abundant in 
191 7, but there were practically none in 1918. By 1919 the 
numbers had recovered again. The hydroid Tubularia, which 
usually dominates the wharf- pile association in early June, 
and dies away by early August, was so much influenced by the 
early season of 1919 that it had entirely gone by early July 
(instead of August) ; and in consequence many animals such 
as the molluscs Columhella and Lacuna, which feed upon 
Tuhulariay were very scarce that year, while the other members 
of the July Tubularia association were also scarce or absent 
(e.g. polychaetes of various genera). The failure of Tubularia 
to be present in the right month affected the numbers of all 
the other animals associated with it. 1920 was normal, but a 
similar upset took place again in 1921. 

4. The work which has been done upon soil protozoa at 

K 



130 ANIMAL ECOLOGY 

Rothamsted shows that there are similar fluctuations in 
numbers among these animals, but upon a much smaller scale. 
These variations are of the order of two days instead of two 
years, but the principle is the same.^^^ If we go to the 
other extreme it can be shown that animals as big as the 
Indian elephant are also subject to fluctuations in numbers, 
caused by epidemics ; but at very long intervals, of seventy to 
a hundred years. ^i^' i^*^ Whenever a group of animals or any one 
species is studied carefully over a series of years, it is found to 
vary in numbers in a more or less marked way. There is not 
space here to give all the evidence for this statement, but it 
seems to be true in practically all cases where there are any 
accurate data. Even when a species does not vary very much 
from year to year, it has in the vast majority of cases a marked 
\^ariation within each year, caused by its cycle of reproduction. 
Every year there is an annual increase in numbers from com- 
paratively few individuals, among such animals as protozoa, 
rotifers and water-fleas, many of which possess some well- 
defined means of increasing rapidly — by parthenogenesis, for 
instance. In fact, the numbers of very few animals remain 
constant for any great length of time, and our ideas of the 
workings of an animal community must therefore be adjusted 
to include this fact. It is natural to inquire why the numbers 
vary so much, and why, with all the delicate regulating 
mechanisms described in the last chapter, the community does 
not more successfully retain its balance of numbers. 

5. The chief cause of fluctuations in animal numbers is 
the instability of the environment. The climate of most 
countries is always varying, in some cases regularly — as in the 
case of the eleven-year cycles in temperature and the frequency 
of tropical cyclones associated with the sun-spot cycle ; or 
of snowfall in Norway, which has a very marked short perio- 
dicity of three or four years .23, 24 Qn the other hand, there 
are a number of irregular and so far unpredictable cycles, such 
as that of rainfall in England. The variations in climate 
affect animals and plants enormously, and since these latter 
are in intimate contact with other species, there are produced 
further disturbances which may radiate outwards to a great 



VARIATIONS IN THE NUMBERS OF ANIMALS 131 

distance in the community. Let us take the example of small 
birds in England. At intervals of ten or more years (and 
sometimes less) there have occurred in England very severe 
winters accompanied by continuous frost and snow for several 
months, which have killed off a very large proportion of the 
smaller birds each time. Such winters, resulting in the death 
on a large scale of small birds like thrushes, blackbirds, and 
tits, are recorded as having occurred in the years mi, 1115, 
1124, i335> 1407^ 1462, 1609, 1708, 1716,81a 1879,118 1917. 
Undoubtedly there are many more which have escaped record. 
In 1407 there was a very long and severe winter, with frost 
and snow during December, January, February, and March. 
Thrushes and blackbirds and many thousands of smaller birds 
died from hunger and cold. In 171 6 numbers of goldfinches 
and other species were destroyed. 

Recently there was a very severe winter in England (1916- 
17) which caused much death among birds and lowered the 
numbers so seriously that they have only just recovered again. 
Actually, the death of these little birds is probably due to 
starvation and not to cold acting directly ; for Rowan ^^ has 
shown that a small Canadian finch (Junco hyemalis) can, if 
provided with plenty of food, withstand blizzard temperatures 
ranging down to —52° F. As long as there is fuel, the body 
temperature can be kept up. It is probably the effect of 
frost upon the rest of the environment which is the serious 
factor in these cases. 

6. The net result of periodic bad winters is therefore a 
periodic fluctuation in the numbers of many small birds. 
Many species of birds vary in numbers from causes which are 
largely unknown. Baxter and Rintoul^s have studied for 
some years the fluctuations in numbers of breeding birds oh 
the island of May in Scotland. They say : " Frequently it is 
possible to explain an increase in numbers of a nesting species 
by some change in environment, such as new plantations 
which afford convenient nesting places. But on the Isle of 
May no such obvious alterations in environment have taken 
place. Nevertheless the species come and go there, they 
increase and decrease, and the reasons are not by any means 



132 ANIMAL ECOLOGY 

always easy to discover. . . . We have written this paper to 
draw attention to the variations which take place in a limited 
area. It is a line of investigation which we think would well 
repay greater study, and which, if pursued in other areas 
showing different conditions, might yield sufficient data to 
make it possible to draw definite conclusions." 

7. Irregular factors in the environment may sometimes 
act at long intervals but still have a tremendous effect on the 
numbers of animals. Wood-Jones ^^^^ says that the Cocos- 
Keeling Islands were visited by very severe cyclones in the years 
1862, 1876, 1893, and 1902. These cyclones wrecked the settle- 
ment on the islands and caused much destruction to plants 
and animals. Similar destructive effects are produced by 
occasional drought years, which completely dry up ponds of 
less than a certain depth, causing partial or complete extinction 
of many species of aquatic animals in the ponds. Each drought 
is followed by a period of recovery, and it may take several 
years to reach normal numbers once more. 

8. The irregular changes in the environment may be, on the 
other hand, favourable to the increase of animals, and so cause 
unusually large instead of unusually small numbers. For 
instance, the shipworm {Teredo navalis) was able to increase 
and spread in Holland in the years 1730-32, 1770, 1827, and 
1858-59, owing to dry summers rendering the fresh- water 
regions more saline than usual .^^ It is sometimes the plant 
environment which changes suddenly. In tropical regions, 
e.g. India, there are certain species of bamboos which flower 
only at long intervals, and often simultaneously over large 
areas. This flowering gives rise to enormous masses of seed, 
and the unusual food-supply allows various species of rodents 
to increase abnormally, and sometimes to reach the dimen- 
sions of a serious plague. ^^ 

9. Periodic fluctuations in numbers may be studied most 
easily in mammals, since we possess rather accurate and 
extensive data about certain species whose fluctuations are 
extraordinarily regular in their rhythm. The best-known 
cases are among rodents, and of these the most striking are the 
Norwegian lemming (Lemmus lentmus) and the Canadian 



VARIATIONS IN THE NUMBERS OF ANIMALS 133 

varying hare {Lepus americanus). The lemming has for 
hundreds of years attracted attention by its periodic appearance 
in vast numbers. A certain Ziegler, who wrote a treatise 
which was published in Strasburg in the year 1532, was the 
first to make any record on the subject. He said that when he 
was in Rome in 1522 he heard from two bishops of Nideros a 
story about a small animal called the *' leem " or '' lemmer," 
which fell down from the sky in tremendous numbers during 
showers of rain, whose bite was poisonous, and which died in 
thousands when the grass sprouted in the spring. The story 
about the downfall from the skies was " confirmed '* by 
Claussen in 1599, who brought forward new evidence of 
eye-witnesses who were *' reliable men of great probity." It 
was not until the lemming had been described fairly accurately 
by Olaus Wormius, in 1653, and by later writers, that the real 
truth became known. 33 The Norwegian lemming lives nor- 
mally on the mountains of Southern Norway and Sweden, 
and on the arctic tundras at sea-level farther north. Every 
few years it migrates down into the lowland in immense 
numbers. The lemmings march chiefly at night, and may 
traverse more than a hundred miles of country before reaching 
the sea, into which they plunge unhesitatingly, and continue 
to swim on until they die. Even then they float, so that their 
dead bodies form drifts on the seashore. This migration, a 
very remarkable performance for an animal the size of a small 
rat — MacClure called it " a diamond edition of the guinea- 
pig " — is caused primarily by over-population in their mountain 
home, and the migrations are a symptom of the maximum in 
numbers which is always terminated by a severe epidemic ; 
and this reduces the population to a very few individuals. 
After such a " lemming year " the mountains are almost 
empty of lemmings. 

10. Owing to the striking nature of these lemming maxima 
we possess records of nearly every maximum for a number of 
years back, which enable us to find out the exact periodicity 
of the pulsations in lemming numbers.^^' ^* In recent years 
the maxima have occurred every four years, while in the middle 
of the nineteenth century they sometimes occurred at rather 



134 



ANIMAL ECOLOGY 



shorter intervals of three or even two years. Now, by examin- 
ing the records of skins obtained by the Hudson's Bay Company 
in Canada, it is possible to find out when the lemming years 
have occurred in Arctic Canada. This is made possible by 
the fact that the arctic fox numbers depend mainly upon those 
of lemmings, since the latter are the staple food of the fox. 
The curve of fox skins, which shows violent and regular 
fluctuations with a period of three or four years, can therefore 
be used as an index of the state of the lemming population in 
different years. When we compare the lemming years in 




Fig. II. — The fluctuations in numbers of lemmings are very violent 
and very regular, and synchronise in Scandinavia and Canada. Curve A 
shows the number of arctic fox skins taken annually by the Hudson's Bay 
Company in Canada. Curve B shows the number of skins taken each year 
in the whole of Canada, from 1 920-1 924. Diagram C shows the lemming 
years in Canada deduced from the fox curves A and B. Diagrams D and E 
show the known dates of maximum ("lemming years") of lemmings in 
Norway and Greenland (the latter being incomplete). (From Elton. ^*) 

Canada and Norway the curious fact emerges that they 
synchronise almost exactly in the two countries, and there is 
little doubt that the numbers are controlled by some common 
factor, which can only be climate. The other records which 
exist show that Greenland and the islands of the Canadian 
Arctic Archipelago also have lemming maxima at the same 
time as the others. 

II. In the lemming we have therefore an animal which 
undergoes regular and violent fluctuations in numbers from 
year to year, which can be analysed into three main processes : 
first of all, the epidemic which kills them off when a certain 



VARIATIONS IN THE NUMBERS OF ANIMALS 135 

density of population is reached ; secondly, the natural 
tendency to increase which leads to recovery of their numbers 
after the epidemic ; and thirdly, the climatic factor which in 
some way controls this cycle. It might be thought that the 
lemming is exceptional among mammals in having such marked 
and regular fluctuations in its population. This is not the 
case. All small rodents, in all parts of the world for which 
there are any data, undergo periodic fluctuations of the order 
of three or four years. Wild rats, mice, hamsters, mouse- 
hares, gerbilles, etc., all appear to have this as their regular 
mechanism of number-regulation. The general fact has been 
noticed in a dozen different countries, e.g. England, Scandi- 
navia, Central Europe, France, Italy, Palestine, Siberia, 
Canada, United States, South Africa, and to some extent 
Brazil and India. The larger rodents also undergo similar 
fluctuations in which epidemics play an important part. 
Squirrels in North America, Europe, and Asia have periodic 
maxima in numbers separated by intervals of five to ten years. 
Some of these maxima are associated with huge migrations 
like those of the lemming. In 18 19 a vast army of grey 
squirrels swam across the Ohio River a hundred miles below 
Cincinnati ,101 while in 1897 a great swarm of the same sort 
passed through Tapilsk in the Ural Mountains : a solid army 
marched through for three days, only . stopping at night, 
and they also swam across the river.^s In Denmark ^7 and 
Norway ^Sc there are also periodic variations in squirrel 
numbers. 

12. Rabbits and hares and jack- rabbits are subject to 
violent fluctuations in numbers in most parts of the world 
{e.g. North America from Alaska to Utah and California ; and 
Siberia). The best-known example is that of the varying 
hare or snowshoe rabbit {Lepus americanus) ^ which, inhsihits 
Canada, and whose cyclical increase and decrease have long 
attracted attention owing to the effect that they have upon the 
numbers of valuable fur-bearing mammals which subsist on 
rabbits. The increase is partly due to the natural recovery after 
epidemics, and probably in part to climatic influences which 
speed up the rate of reproduction in certain years.^^ 



136 



ANIMAL ECOLOGY 



Soper 26 gives a good description of the difference between 
the time of maximum numbers and of the minimum which 
follows the epidemic. He says : " It so happened, that upon 
my first visit to the West in 191 2 the rabbit population was at its 
height. It was such a revelation after my eastern experiences, so 
startling, that the vividness of their abundance can never leave 
me. A certain brushy flat adjoining the White Mud River, 
south-west of Edmonton, yielded the initial surprise. It was 
grown to scrub willow, the common trembling aspen, and to 




Fig. 12. — Fluctuations in the numbers of Canadian mammals (taken 
from Hewitt,^ who used the records of the Hudson's Bay Company). The 
figures for any particular year represent the effect of increase in the previous 
year, since the animals are trapped in winter. There is sometimes a lag 
of more than one year, owing to various trapping factors. 



some extent with rank under- vegetation. The place was 
infested. I do not hesitate to say that over that tract of per- 
haps thirty acres hundreds of hares were found. October had 
come, without snow. The rabbits had already, wholly or in 
part, donned their snow-white livery of winter, and were 
consequently very conspicuous against the mellow brown of 
the autumn woods. At every turn during my ramble they 
popped up here and there and scurried for fresh cover. Not 
only in singles, which was astonishing enough, but often twos 
and even threes started up in wild alarm. '* This was the 



VARIATIONS IN THE NUMBERS OF ANIMALS 137 

maximum abundance. Later on he says : " Eventually, evidence 
of the inevitable decline arrives. Empire among the rabbits as 
elsewhere has its rise and fall, and then is swept away. A 
strange peril stalks through the woods ; the year of death 
arrives. An odd rabbit drops off here and there, then twos 
and threes, then whole companies die, until the appalling 
destruction reduces the woods to desolation. . . . One year 
(19 1 7) in the district of Sudbury, northern Ontario, the signs 
of rabbits were everywhere, but not a single rabbit could I 
start. It seemed incredible. Local inquiries disclosed that 
a little over a year before the Lepus population was beyond 
count. Now, as if by magic, they were gone. Needless to say, 
however, a few individuals survive the epidemic. These now, 
because of their paucity, are seldom encountered." 

13. Marmots, muskrats, and numerous other rodents also 
fluctuate with a more or less regular periodicity. Seton 1^ 
says " the muskrat's variation probably has relation chiefly to 
the amount of water, which, as is well known, is cyclic in the 
North-West." 

In the case of lemmings, some mice, the varying hare, and 
the muskrat, it is practically certain that the cycle of numbers 
is partly under the control of regular cycles in climate ; and 
this is also probably the case with many other rodents about 
which we have at present Httle information beyond the fact 
that their numbers do fluctuate. At the same time there is 
no reason why such wave-motions in the population should 
not be automatic, in so far as each epidemic is followed by a 
period of recovery which would culminate in another epidemic, 
and so on indefinitely. It seems possible that this may apply 
to the rodents in some countries which show no regular or 
synchronous over-increase. The mouse-plagues in France 
are often local and always irregular, although sometimes they 
may occur simultaneously over large areas. Here variations 
in climate probably play at some times a more important part 
than at others. But on the whole, from what little we know 
about the matter at present, it appears that periodic fluctuations 
in the numbers of rodents are primarily caused by the irregular 
behaviour of their environment. After all, any species must 



138 ANIMAL ECOLOGY 

either increase, remain constant, or decrease in numbers, and 
as there is always the danger of the numbers going down 
steadily if the balance is not struck exactly right, it seems 
reasonable to suppose that species would tend to be adapted to 
a steady increase in numbers, which is terminated at a certain 
point by disease or some other factor. 

14. It is interesting to consider at this point an animal 
whose peculiar habits and scheme of social existence make the 
study of its numbers of extraordinary interest. This animal 
is the beaver {Castor fiber). The beaver has been hunted and 
trapped and studied so intensively for the last hundred years, 
that we possess a great deal of reliable information about its 
ecology. The statistics of the Hudson's Bay Company show 
that the numbers of beaver undergo no very marked short- 
period variations ; although there was a general upward trend 
in number of skins during the early part of the nineteenth 
century, which was associated with the exploration and opening 
up of new parts of Canada. ^^ This was followed by a con- 
siderable drop, consequent upon exhaustion of the natural 
stock of beavers through settlement and over- trapping,, The 
reason for this comparative stability of the beaver population 
as a whole appears to be that it is almost entirely independent 
of short-period climatic variations. Since it uses the bark of 
trees for food the beaver is unaffected by annual variations 
of plant food-supply. It lives on capital and not on income — 
an almost ideal existence. Furthermore, although it is 
aquatic, the elaborate engineering feats which enable it to 
construct dams and houses (not to mention long canals for 
bringing food supplies from a distance) make it comparatively 
immune to the effects of annual variations in water-supply — 
unlike the muskrat, which is at the mercy of floods and droughts. 
Each colony of beavers lives in one place until the local supply 
of trees is exhausted, and when this takes place the animals 
apparently move on to some other locality. One would 
imagine that the beaver's habits, which cause it to live in 
isolated colonies, would make it difficult or impossible for any 
epidemic disease to spread throughout the population of a 
whole region (unless some alternative stage of a parasite were 



VARIATIONS IN THE NUMBERS OF ANIMALS 139 

carried by its enemies). This is actually the case. Epidemics 
are unknown among these animals, and, as a matter of fact, 
they have no very serious carnivorous enemies.''^ But local 
outbreaks of disease are probably an important check in 
numbers, since MacFarlane says,'^^ " j^ is not an uncommon 
occurrence for hunters to find one or more beavers dead of 
disease in their houses or ' washes.' " In the beaver we find 
a rodent which apparently limits its numbers mainly by disease 
in the form of local outbreaks, and not by widespread epi- 
demics ; and which is unaflFected by cycles of climate with a 
short period, so that the numbers of the population as a whole 
— apart from the effects of trapping by man — remain the same 
from year to year. 

15. Fluctuations in the numbers of any one species in- 
evitably cause changes in those of others associated with it, 
especially of its immediate enemies. This is well shown in 
rodent fluctuations. In Canada the arctic fox fluctuates with 
the lemmings, the red fox with mice and rabbits, the fisher 
with mice and rabbits and also fish, the lynx with rabbits, 
and so on. All these carnivorous mammals, which prey 
largely or partly on rodents, have periodic fluctuations of 
great regularity. This is true of weasels, ermine, martens, 
foxes, wolverene, skunks, mink, lynx, and others. The sable 
of Siberia depends to a great extent upon squirrels, and 
fluctuates accordingly. In all cases there is a lag in the increase 
of the carnivores owing to their larger size. Many birds of 
prey show similar fluctuations due to the variations in their 
food-supply, but here the question of numbers is complicated 
by migration, and it is rather hard to get at the real facts about 
changes in the whole population. But it can at any rate be 
said that the numbers of hawks and owls often vary regularly 
in any one place. One of the best descriptions of the enormous 
changes wrought in the abundance and habits of carnivorous 
birds and mammals by rodent cycles is that given by Cabot 11 ^ 
at the end of his book on Labrador. 

There is little doubt that insectivorous mammals and 
birds, such as shrews, moles, and warblers, are subject to 
considerable fluctuations dependent upon the abundance of 



140 ANIMAL ECOLOGY 

their food, or upon other factors in their surroundings. Such 
variations have a regular periodicity in the case of the shrews ^Sa 
at any rate. 

1 6. Rodent fluctuations have always been recognised, 
although the universal existence of the phenomenon has not 
always been realised. It is less well known that the larger 
herbivorous mammals (ungulates) are also subject to periodic 
epidemics in nature, which are separated by much greater 
intervals of years than those of rodents, owing to the slower 
rate of increase of the former consequent upon their larger 
size. A rough idea of the rates of increase of the mammals 
of various sizes may be gained by the following figures. If 
one pair were allowed to increase unchecked, or at any rate 
were not subject to very severe checks (as by epidemics), a 
dense population would be produced by mice or lemmings 
in three or four years, by squirrels in five years, by hares or 
rabbits in about ten years, by sheep in about twenty years, 
by buffaloes in about thirty years, and by elephants in about 
fifty or more years. These figures are of course rough approxi- 
mations, but they enable one to see why mammals of different 
sizes have different intervals between their epidemics and 
between their maxima. 

17. Good evidence on the subject of fluctuations in wild 
ungulates is rather scattered and hard to obtain, since during 
the last few thousand years there have been many epidemics 
among man or his domestic animals which have become 
communicated to wild species and have thus interfered arti- 
ficially with the natural balance of numbers of the latter. For 
instance, cholera and liver-rot seem to have been serious 
factors affecting the numbers of wild boars and deer on the 
continent of Europe during early times. ^^^^ But there are 
several bits of information which give us a glimpse of the way 
in which large ungulates used to regulate their numbers when 
living in an undisturbed state of nature. Brooks ^ says that 
on Vancouver Island the wild deer are subject to cycles of 
abundance and scarcity. These are followed closely by the 
cougar or mountain lion, which preys exclusively on deer, 
just as the lynx does upon rabbits. We do not know in this 



VARIATIONS IN THE NUMBERS OF ANIMALS 141 

case whether disease is the cause of this scarcity, but it seems 
very probable. Again, Fleming,^i« speaking of the wild deer 
in England in 1834, refers to an epidemic *' which from time 
immemorial had broken out at irregular intervals and swept 
off thousands of these animals." Percival i^k says of the 
African zebra that ** the free animal is carried off in considerable 
numbers by periodic outbreaks of disease, which has been 
traced to a lungworm." 

18. It should be sufficiently clear by now that the numbers 
of many animals are subject to great fluctuations from year to 
year, and that, in the majority of cases which have been in- 
vestigated, these fluctuations can be traced ultimately to pulsa- 
tions or changes in the environmental conditions affecting the 
animals. If we follow up the implications of these facts it 
becomes possible to see why so many animals appear at intervals 
in vast plagues, often of great economic importance. Every 
herbivorous animal is adapted to increase at such a rate that 
enough extra individuals are produced to satisfy the require- 
ments of its carnivorous enemies ; while the latter, being 
usually larger, cannot increase so fast as their prey, and are 
accordingly adapted to a certain rate of increase which will not 
cause them to over-eat their food-supply. There are therefore 
small herbivores increasing fast, and larger carnivores increasing 
more slowly. The larger they are the more slowly they in- 
crease ; this is a result of size- differences, but at the same time 
acts as an adaptation, in that it makes possible the existence 
of the food-cycle at all. Now, suppose one of these small 
herbivores — a mouse or an aphid — is suddenly able to accele- 
rate its rate of increase, either as a population or as an individual. 
The change might be caused by a favourable winter which 
would enable the population to start in spring with a larger 
capital of numbers than usual. Again, it might be brought 
about by increased numbers of young being produced in each 
brood, as in the case of the varying hare or the short-eared 
owl. If the herbivore accelerates its rate of increase in this 
way, the carnivore, being larger and with slower powers of 
increase, is entirely unable to control the numbers of its prey 
any longer. And if the latter continues to increase at this 



142 ANIMAL ECOLOGY 

rate, its enemies will also do so, but will lag far behind and 
be too late to have any predominant effect on the numbers of 
the prey. 

19. It seems, therefore, that the food relationships of animals 
result in the numbers being controlled in the majority of cases 
by carnivorous enemies, but that when disturbances- in the 
environment cause a sudden acceleration in the rate of increase 
of some smaller species, its enemies no longer act as an efficient 
control. Actually the numbers of an animal are ultimately 
very often controlled by organisms smaller than itself, i.e. by 
parasites which produce epidemics. The following parallel 
may perhaps make this argument clearer. In ordinary 
weather the fire-brigades in a town are sufficiently numerous 
to keep down any outbreaks which may occur, but if there 
occurs a spell of very dry weather for several months there 
will be a sudden increase in the number and spreading 
powers of fires ; they will flare up more quickly and be more 
difficult to extinguish. In such an emergency the number of 
fire-brigades will be inadequate for dealing with the situation, 
and there will be a long delay before new firemen can be 
properly trained and equipped with helmets and hoses. Mean- 
while the fires will go on spreading and setting light to other 
houses, and in bad cases the whole town may be burnt down 
and have to be rebuilt again, as in the great fire of London in 
1666. In just the same way, once a mouse population has 
*' broken out " and escaped from the control of its enemies, it 
will give rise to a number of other mice, which in turn will 
increase and spread Hke the fire. The firemen (represented 
here by owls and foxes) cannot bring up and train more of their 
species in time to be able to stop the outbreak. 

zo. In the case of a bad fire the situation can usually be 
saved by a migration of fire-engines from neighbouring towns, 
but if it is a dry summer everywhere the fire-brigades will be 
busy putting out their own fires, and so unable to help any one 
else. Similarly, in a mouse plague there is often a huge 
migration of owls to the spot, which increase the chances of 
reducing the numbers. But here again the importance of 
climatic factors is seen, since the latter may cause plagues over 



VARIATIONS IN THE NUMBERS OF ANIMALS 143 

large areas, so that migration of enemies hardly takes place 
at all. 

Although we have been taking mammals and birds as 
examples, since more is known about them and they are more 
familiar to most people, the principle outlined above applies 
universally in animal communities and is probably the ex- 
planation of most sudden plagues of animals. It results from 
the variable nature of the environment on the one hand, and 
the food- cycle mechanism of animal communities on the other. 

Having shown the fact of fluctuations in numbers among 
many animals, and having discussed the causes of such fluctua- 
tions, we may next turn to a consideration of some of the effects 
which these fluctuations have upon animals in nature. 

21. There are a number of curious and interesting conse- 
quences, of which we can only describe a few ; but it will 
become clear that variations in numbers play a very big part 
in the ecology of animals. 

For instance, the food habits of many species depend not 
only upon the quality but on the quantity of their prey — a 
fact which is often lost sight of in experiments upon food 
preferences. As was pointed out earlier, the lower limit to 
the size of an animal's food is partly determined by the ease 
with which the latter can be caught ; and this in turn is tre- 
mendously affected by its abundance ; for if the prey is too 
scarce it takes too long to collect enough to satisfy the animal's 
needs. Time is a vital factor in the lives of most animals, 
and especially of carnivores. The animal must secure a certain 
amount of food every day or month or year, in order to survive 
and breed successfully. The seriousness of this factor varies 
in different species and at different times and places, but it is 
nearly always present during some part of the animal's life. 
It is said that the sheep in Tibet have to feed at a run, since the 
blades of grass are so scattered that only by being very active 
and energetic can the sheep get enough in a given time to sup- 
port life.^^^ With temperature-regulated animals the problem 
is of course more urgent, since they cannot usually allow the 
body-fuel to run too low, while with cold-blooded animals it 
is more a question of accumulating reserves of food big enough 



144 AXLMAL ECOLOGY 

to carry- through their breeding activities successfully. But in 
anv case the problem remains. 

22. The consequence of this is that as soon as a food animal 
sinks below a certain degree of abundance its enemies either 
stan-e or turn their attention to some other source of food. 
Fluctuations in numbers have therefore a potent influence 
upon the food-habits of animals. In fact, if several important 
kev-industr}- species become suddenly ver\' abundant or very 
scarce, the whole food-cycle may undergo considerable changes, 
if only temporarily. The various automatic balanced systems 
which exist ^^ill tend to bring the numbers, and therefore the 
food-habits, back in the long run to their original state. x\nother 
way of putting it is that the favourite food of an animal is not 
usually the most abundant one. Many animals have a definite 
scale of food preferences, depending upon qualit}', but if a 
favourite one becomes suddenly common the animal will 
abandon its previous food ; and conversely, if a common food 
becomes suddenly scarce it may still continue to seek it up 
to a certain point, beyond which taste must give way to 
necessity'. These variations in food-habits may lead to corre- 
sponding variations in the habitat to which the animal resorts. 
This is shown by the follo\^ing example of the habits of the 
common rook {Corviis frugilegus) , which is meant only to 
illustrate the idea, and not to prove anything final about the 
rook itself. During a particularly dull railway journey the 
writer noted dowii the number of pasture fields and of ploughed 
fields recently sown, and at the same time counted the number 
of fields of each kind which contained flocks of rooks feeding. 
The figures obtained were as shown in the table : 





No. of fields. 


No. with rooks. 


Per cent, 
with rooks. 


Pasture 

Sown ploughed . . 


.. 144 
II 


19 

4 


\l 



It will be seen that although there were comparatively few 
ploughed fields they appeared to be much more sought after 
than the pasture land (assuming for the sake of argument that 
the figures are sufficiently large to prove this — which they are 
not). But the absolute number of rook-flocks on pasture 



VARIATIONS IN THE NUMBERS OF ANIMALS 145 

land was much greater than on ploughed land. If rooks had 
been much rarer birds we might have found that they were 
exclusively attached to ploughed fields at that time of the year. 
23. Dugmore's description-*^^ of the African buffalo (Bos 
coffer) in the Sudan shows in a very interesting way how the 
habits of an animal may depend upon its numbers. The 
wild buffalo which inhabits a large part of Africa used to be 
one of the most abundant large animals in that country, but in 
1890 a frightful epidemic of rinderpest swept the continent, 
killing off, amongst other species, enormous numbers of 
buffaloes, and almost exterminating them in many parts of 
the country-. Before the great epidemic the buffalo used to 
live in herds out in the open grassland and feed by day, but 
" for many years after [1890] the few remaining animals fed at 
night and retired to forests and dense swamps during the 
day." This was still the condition of affairs in 1910, but 
within recent years the buffalo have begun to increase again 
greatly and appear to have gone back to their former habits. 
This fact is confirmed by the obsers'ations of Percival ^-^^ and 
Chapman *^^" in Africa. It will be noticed that the African 
buffalo took about thirty-five years to recover its numbers. 
It is interesting, therefore, to find that the greater kudu 
{Strepsiceros bed) recovered in a much shorter time (ten to 
twenty years) from the rinderpest epidemic, although at the 
time it was almost ^^^ped out. This quicker recovery is 
associated ^^ith the fact that it is a much rarer animal than 
the buffalo, and so requires less time to regain its maximum 
density of numbers. ^-^^ 



CHAPTER X 

DISPERSAL 

The study of animal dispersal involves (i) a large number of subjects 
besides biology, and we can only deal here with the general ecological 
aspects. (2) Most animal dispersal is directed towards the ordinary 
feeding, breeding, and other requirements of animals, not directly 
towards spreading the species, as shown (3) by the example of the 
capercaillie in Scotland. (4) The terms used in discussing the subject 
need to be carefully defined : the " spread " of any species involves 
" dispersal," " establishment of the individual," and " establishment of 
the species." These three phases sum up to control the " distribution " 
of a species at any one time. (5) Ecological succession plays a large 
part in slow dispersal of animals. (6) Movements of animals are often 
periodic and regular, but these do not necessarily extend the range 
of the species. (7) Besides active migration, many animals become 
dispersed by " accidental " means (which are often very definite and 
regular) ; they employ wind, water, logs, ships, seaweed, etc., and 
(8) other animals. Direct evidence on this subject is hard to obtain, 
and (9) is usually encountered by accident in the course of other ecological 
work. It is most important to publish such incomplete observations, as 
they may be unique. (10) Dispersal may take place as described above, 
through the ordinary activities of the animals, but where it is a definite 
large-scale phenomenon resulting in spread of the species, it may be 
either for the purpose of getting away from an overcrowded population 
or for getting towards some new place, or for both. (11) Much abortive 
colonisation is always taking place, partly owing to ordinary ecological 
factors being unsuitable and (12) partly to the difficulty of finding a mate 
upon arrival at a new place. (13, 14) Three methods of finding the 
right habitat are employed by animals : broadcasting huge numbers so 
that a few by chance reach the right habitat ; directive migration, by 
means of special instincts or tropisms ; and a combination of general 
broadcasting and local direction-finding. 

I. When we are studying any particular animal or com- 
munity of animals, we are brought up, sooner or later, against 
questions connected with dispersal : with the movements of 
animals in search of food, of shelter, or of their mates. This 
movement, on a large or small scale, is characteristic of animal 
communities, as compared with plant communities, and it 
forms a very important part of the lives of wild animals. Dis- 
persal is such a large subject, and has so many sides to it, that 

146 



DISPERSAL 147 

it will only be possible here to give a general outline of some 
of the ways in which it affects animal communities. If we 
studied the factors which cause the onset of migration in animals 
like birds or lemmings, we should soon find it necessary to go 
quite deeply into psychology, in order to find out why animals 
come together into herds at all, why they sometimes become, 
as it w^ere, hypnotised or obsessed with the migration impulse, 
and why they migrate in one direction rather than another. 
If we studied at all closely the reactions of insects, which lead 
them to seek a particular food-plant, or to lay their eggs in a 
particular place, we should soon become involved in remote 
branches of organic chemistry. Dispersal is especially attrac- 
tive as a subject for study if only because it is so easy 
at any moment to escape from the purely biological side of the 
work, and seek a change in psychology, chemistry, meteorology, 
or oceanography. At the same time, we are chiefly concerned 
in this chapter with providing a general orientation towards 
the problem of dispersal, in so far as it forms a factor of the 
life of animals. It is a subject which requires clear thinking, 
since in the usual discussions about the present, and even 
more the past, distribution of animals, one notices a remarkable 
lack of comprehension of the extremely complicated processes 
involved in their spread, whether as individuals or species. 

2. Since most plants cannot move about of their own 
accord, they usually have some special adaptation for the 
spreading of seeds or other reproductive products ; whereas 
animals, possessing the power of active movement on a large 
or small scale, do not have such highly developed special 
means of dispersal. Furthermore, their powers of movement 
are not usually employed for the direct purpose of spreading 
the species over the widest possible area. Although an animal 
like a rabbit, or an earthworm, or an earwig, has considerable 
powers of dispersal, these are directed towards its immediate 
needs of obtaining food, finding water, or a mate, or avoiding 
enemies, rather than towards the occupation of new territory 
for its own sake. The latter often takes place automatically 
as a result of the former type of activity, and this casual un- 
directed movement may become elaborated into regular 



148 ANIMAL ECOLOGY 

migrations ; but these, again, are usually for the purpose of 
finding food, etc., and only result secondarily in the spread of 
the species. Of course this is not to say that some animals 
have not got very remarkable and specialised means of dis- 
persal, which exist merely for the purpose of spreading the 
species. Such adaptations exist in nearly all sessile or very 
sedentary animals such as marine hydroids on the one hand, 
or spiders on the other. But the spread of species in an 
ordinary animal community (excluding coral reef and other 
marine littoral or intertidal sessile animal communities) usually 
consists of a rather vague and erratic shifting of the animals 
from one place to another, which after a number of years may 
sum itself up as a change in distribution. The cases of large- 
scale migration in a definite direction are either periodic (as 
in birds), i.e. they consist in a movement backwards and for- 
wards between two or more places ; or else are rather excep- 
tional, as in the case of locust migrations or the outbursts of 
sand-grouse from Asia. The latter type of migration is very 
striking both from its size and also from its regular periodicity, 
but the number of species in which it takes place at all often is 
probably rather limited. 

3. The idea with which we have to start is, therefore, that 
animal dispersal is on the whole a rather quiet, humdrum pro- 
cess, and that it is taking place all the time as a result of the 
normal life of the animals. A good example of this sort of 
dispersal is provided by the capercaillie (Tetrao urogallus), 
which became extinct in Scotland about 1770 but was reintro- 
duced in 1 837, and has spread gradually, so that it now occupies 
a very large part of the country again. The manner of its 
spread has been well described by Ritchie, ^^^ who gives a very 
interesting map showing the dates of its arrival at various 
places along the routes of its migration from each centre of 
introduction. The capercaillie is found almost exclusively 
in pine woods, and its migration was apparently caused by birds 
flying occasionally from one wood to the next, and so gradually 
occupying new territory. Parallel with this process went 
the increase in numbers, but it is almost certainly wrong to 
imagine that it was direct " pressure of numbers " that caused 



DISPERSAL 149 

it to migrate ; although, without accompanying increase in 
numbers, the species would not spread far. Ritchie says : 
" Dr. Harvie-Brown was of opinion that the capercaillie 
viewed prospective sites from its old establishments, and this 
very probable selection by sight, together with the fact that 
most of the woodland lay along the watercourses, would 
determine the capercaillie's dispersal along the valleys. 
Indeed, judging from the dates of the advent or establishment 
of birds in new areas, the valley systems ranked second only 
to the presence of fir woods in determining the course of the 
migrating capercaillies." The squirrel, which was also intro- 
duced into Scotland, behaved in a similar way, migrating along 
the valleys and frequenting the fir woods.-^^^ 

4. So far we have used the term " dispersal " in a rather 
loose way. It is necessary, however, before entering further 
upon the subject, to analyse the process of dispersal and define 
rather carefully certain terms which are usually used in a 
vague sense, or at any rate in different senses by different 
people. The following system of terms is used here, and 
whether or not it agrees with the definitions of other writers, 
it will at least be clear what is meant by them here. By 
" dispersal " is meant the actual migration or carriage of 
animals from one place to another : e.g. the floating of young 
spiders on streamers of gossamer, the flight of a migrating 
goose, the transport of water-mites on the legs of water- 
beetles, or the drifting along of jellyfish in the sea. When 
such an animal reaches its destination it may either die or 
survive. It usually dies, but if it does not, we say that it has 
" established itself as an individual." Thus, a large logger- 
head turtle {Thalassochelys caretta) reached the coast of 
Scotland (Skye) in December, 1923, quite alive, and full 
of eggs.^-^ This species is an inhabitant of tropical and 
subtropical seas. Again, in certain years large numbers of 
clouded yellow butterflies (Coltas edusa) arrive in England 
from the Continent. But in such cases the individuals are 
unable to breed successfully, or else the young are unable 
to survive. The next stage is, therefore, that the animal 
must *' establish itself as a species." It does this if it is 



ISO ANIMAL ECOLOGY 

successful in breeding and starting a permanent popula- 
tion of its kind. Frequently a species may reach some new 
place and breed, and may establish itself for a short time, but 
then is wiped out. This often happens because the animal is 
not adapted to some periodic factor which acts at fairly long 
intervals, e.g, a very bad winter, or an epidemic. Or the 
species may die out simply because its numbers and rate of 
increase are not suitably adjusted to the new environment in 
which it finds itself. For instance, after big invasions of 
sand-grouse or of crossbills, pairs of the birds have been known 
to breed in some localities for a year or two after their first 
appearance. ^^^ But they usually die out in the end, and no 
more are seen until the next invasion. Some botanists employ 
the word " ecesis " (from the Greek word which means 
dwelling at home) instead of " establishment." It is rather a 
useful word, owing to its clear meaning and shortness, but at 
the same time rather an ugly one. 

The combined results of dispersal and of the establish- 
ment of the individual and then the species in a new place we 
call the *' spreading " of the species. The area covered by 
it at any one time is its '' distribution." These terms may be 
summed up as follows : 

Dynamic . . . . Dispersal ^ 

Establishment as individual J- Spread. 
Establishment as species J 

Static . . . . Distribution. 

The loose and undefined use of words like dispersal, spread, 
and distribution, by faunists and palaeontologists has led to 
a good deal of confused thinking on the subject of past move- 
ments in the animal world, and it is well to be quite clear on the 
subject, and to realise that the spreading of a species involves 
at least three fairly clear-cut phases, and that successful spread- 
ing involves overcoming a series of obstacles at each of these 
phases, so that it may be absurd to attribute the final result 
to any simple factor or factors. 

5. One of the important ways in which slow dispersal takes 
place is through the migration or spreading of the environ- 
ment, i.e. by ecological succession. The patch of vegetation 



DISPERSAL 151 

forming any particular habitat hardly ever remains for a long 
time in the same spot, owing to the process of ecological succes- 
sion which is almost everywhere at work ; and since succession 
takes place fairly slowly, any habitat carries along with it a 
complete set of animals, which thus gradually extend or change 
their range. This extension of range takes place slowly and 
quietly, and the species may end up by living in the same 
habitat but in an entirely different locality, as it were, " without 
knowing it." A simple example of this sort of thing is the 
southward extension of arctic species of animals during an ice- 
age in Europe. The animals must in many cases have changed 
their range with almost imperceptible slowness, as the physical 
conditions or vegetation gradually shifted southwards ; often 
the process would be irregular, sometimes backwards and some- 
times forwards, and over and over again patches of vegetation 
would be completely destroyed and wiped out. But on the 
whole the tundra zone, and the forest belt, would move south, 
and with it the fauna, until the latter had penetrated hundreds 
of miles south of its original home. The point is that such 
spreading would not necessarily be accompanied by any 
violent special migrations, except in so far as the latter were 
normally found for various other reasons. 

Another example which illustrates the same thing on a 
smaller scale is the formation of '' ox-bow " ponds in river 
valleys. On the flood plain of a place like the Thames Valley 
there are a number of ponds which owe their origin to the un- 
easy movements of the river in its bed. Arms, loops, or back- 
waters of the river become cut off as the latter shifts its course, 
and in the ponds so formed there are usually a number of 
molluscs or fish which would not have been able to reach the 
pond if the whole environment had not shifted. The examples of 
ice-ages and ox-bow ponds are only special cases of a universal 
and very important way in which species spread over the 
surface of the earth. The process consists of a combination 
of the migration of the environment and of the local move- 
ments of the animals in pursuit of their normal activities. 
Any new extension of a habitat, such as the edge of a pine wood, 
is immediately filled up in an almost imperceptible way. 



152 ANIMAL ECOLOGY 

When palaeontologists speak of the migration of animals 
into a new country, they usually have a vague idea of some 
species like the caribou walking from its original home in a 
large crowd, and arriving in force at the destination. What 
probably happens is rather the process described above, which 
is a great deal less spectacular and lasts over a very much longer 
time. Big-scale migrations do exist and of course attract 
attention from their very size, but it cannot be too much 
emphasised that the spreading of species is, in ninety-nine 
cases out of a hundred, a slow process intimately bound up 
with the local habits and habitats of the animals, and with 
ecological succession. 

6. Among many animals the daily migrations in search of 
food and other necessities are more or less regular and well 
defined. Crozier^^* observed an amusing case of this sort 
among the molluscs of the coast of Bermuda. He found that 
Chiton tuberculatus (a species about nine centimetres in 
length) does not wander very far frond its *' roosting-place," 
but travels within about a radius of a metre from its home. 
One individual which was observed had a small Fissurella 
(length about 0*9 cm.) living on it, browsing upon the 
epiphytic growths which covered the valves of the Chiton, 
The Fissurella also wandered about all day over the Chiton^ 
but always returned in the evening to the third valve ! 

Seasonal migrations of a regular nature occur among many 
animals besides birds. (For an up-to-date account of bird 
migration the reader may be referred to A. Landsborough 
Thomson's Problems of Bird Migration,^^^ since the subject 
is far too large to be adequately dealt with here.) Instances 
which will immediately occur to the naturalist are the migra- 
tions of fish in search of their spawning grounds, or of herbi- 
vorous mammals in search of pasture. In Palestine a certain 
amount of excitement and discomfort was at one time caused 
by large numbers of migrating scorpions, which invaded the 
military camps that happened to have been planted in their 
path. In the same way it has been noted ^^^ that the common 
bandar or rhesus monkey and the Hanuman monkey of India 
trek in parties (formed of a mixture of the two species) from 



DISPERSAL 153 

the plains to the hills of Nepal in the hot season, and return 
in the cold season, carrying their young with them. 

Such seasonal migrations may be on any scale, ranging from 
the thousand-mile journeys of some birds, to the crested 
newt's spring migration to the nearest pond for the purpose 
of laying its eggs, or the even shorter journey made by aquatic 
worms and rotifers, when they repair to the hot surface- 
layers of conferva in ponds for the same purpose. 

7. Amongst the various means of dispersal we can dis- 
tinguish roughly between the more voluntary, or at any rate 
active, migration of the animal itself on the one hand, and the 
numerous means of *' accidental " dispersal on the other. 
An enormous number of the smaller species of animals get 
about from one place to another by special means of transport 
other than their own legs, wings, or cilia. In certain cases, as 
when the larva has a special instinct or tropism which leads 
it to hang on to a particular animal, the dispersal may be 
apparently almost solely for the purpose of disseminating the 
species. Thus, the glochidia larvae of the mussel have a special 
reaction to certain fish which causes them to shoot out the sticky 
threads by w^hich they hang on to their temporary host. But 
at the same time, many of these cases involve the extraction 
of food from the host while the animal is hanging on, and so 
the dispersal is to that extent a secondary consideration. 

Amongst the more curious means of dispersal is that 
adopted by the Emperor penguin, which breeds in mid- 
winter on the sea-ice on certain parts of Antarctica. Being 
a large bird, development takes rather a long time, and in spring 
the young birds are not always ready to take to the water by 
the time the sea-ice begins to break up and drift north. This 
difficulty was seen to be overcome by the birds taking their 
chicks to the edge of the sea-ice where it was beginning to 
break away, and waiting until the chunk upon which they 
were sitting freed itself and floated away northwards. The 
chicks were thus able to continue their development for some 
time longer before entering the water. In polar and tem- 
perate regions floating mats of seaweed carry large numbers of 
animals about attached to and living in them. Similarly, drift- 



154 ANIMAL ECOLOGY 

wood is an important agent of accidental dispersal. Beebe ^^^ 
states that on one log collected out at sea during the voyage 
of the Arcturus, he counted fifty-four species of marine 
animals, including numbers of crabs, fish, and worms. In 
modern times, ships act as the most gorgeous hollow floating 
logs and have been used by animals on a large scale. 

8. Animals form the means of transport of a great many 
other species. Apart from parasites, permanent or temporary, 
there are a number of cases of accidental carriage which have 
been recorded. Thus, Kaufmann^^^ noticed that the fresh- 
water ostracod Cyclocypris Icevis^ which he was keeping in 
an aquarium, had the habit of occasionally hanging on to 
the legs of certain water-beetles, so that it is possible 
that they may get about in this way from pond to pond. 
Water-mites have the similar habit of hanging on to all manner 
of insects (beetles, bugs, flies, etc.) in the larval state, and in this 
way getting carried to new ponds. There are a number of 
other examples which might be quoted, but to do so would 
obscure the fact that in an enormous majority of cases we have 
not the slightest idea in what way animals do commonly 
get about. Take the case of Polyzoa living in ponds. It is 
easy enough to conjecture that they may get about on birds' 
feet or some such manner, but very difficult indeed to find an 
actual case of their doing so. Only occasionally are we lucky 
enough to catch them in the act, as when Garbini^^^ found the 
statoblast or winter egg of Plumatella on the beak of a heron. 
Often it is possible to obtain circumstantial evidence as to the 
methods of dispersal, evidence which usually takes the form 
of saying : " There are only two ways these animals could 
have got to this island (or pond, or wood), and this one seems 
the most likely." Thus Leege^^^ studied the fresh-water 
Crustacea in a pool of water on an island off the Friesian coast, 
an island which had only risen out of the sea during the pre- 
vious ten years, and which must have obtained its fresh- water 
fauna from elsewhere by accidental dispersal. In 1907 there 
was no pool ; in 1908 it was formed, and in 1909 von Leege 
found a large number of specimens of four species of common 
water-fleas {Daphnia pulex^ Simocephalus exspinosus^ Pleuroxiis 



DISPERSAL 155 

adufictis, and Chydoriis sphcericus). The island was frequented 
by enormous numbers of birds, and since these species of 
water-fleas are all littoral ones and furthermore form winter 
eggs which easily become entangled with birds* feathers, he 
concluded that the birds had been the transporting agent. 
This is usually the sort of evidence which one gets about 
accidental animal dispersal ; the final proof is seldom forth- 
coming, owing to the difficulty of catching the animals in the 
act, especially when the dispersal may be a rather rare event 
in any case. 

9. One is struck by the way in which the true and often 
quite important facts about animal dispersal, and especially 
accidental dispersal, are usually met with when one is work- 
ing on something else in the field. In fact, it is almost impos- 
sible to spend any time profitably in a deliberate study of 
animal dispersal as such, except in certain well-defined direc- 
tions, as in seasonal migrations of birds and fish, or parasites, 
or species whose larvae regularly hang on to a definite host. 
Thus the writer has found a frog carrying a fresh-water bivalve 
(Sphcerium) attached to its hind leg, while the real work on hand 
was the collecting of neotenous newts. And while studying 
the habits of rabbits in Herefordshire, he has suddenly noticed 
that hordes of young toads were migrating across the country 
in the direction of the local lake, and was surprised to find 
that they were making for a piece of water which they were 
unable to see, since it was hidden by a high wall. It seemed 
clear that they must be responding to the humidity of the air ; 
but even that seems a little hard to believe, when one considers 
that they were more than a hundred yards from the water. 
It is on such facts as these that our knowledge of animal dis- 
persal will always have to be built up, facts which are usually 
encountered by accident, in a casual way, and often in circum- 
stances which make it impossible to follow up the problem 
any further. It is therefore especially desirable to publish 
such notes on dispersal, however fragmentary they may be, 
since there is not usually much chance of getting better ones 
for some years. And it may be noted here that there is a 
certain reluctance among zoologists to publish incomplete 

LU ' L ! B R A R Y 



156 ANIMAL ECOLOGY 

observations, which is quite justifiable in the case ot definite 
experiments, or of the descriptions of dead structures which 
will wait for you to observe them completely, and which can 
be checked by other observers if they are sufficiently in- 
terested. But in accumulating the life-history records of wild 
animals it is essential to publish any tiny fact about them which 
seems unlikely to be encountered in the near future, or which 
helps to provide another piece in the complete jig-saw puzzle 
which ecologists spend their time putting together. A good 
example in this direction has been set by C. B. Williams, ^^^ 
who has clearly demonstrated in his studies on the migration 
of the painted-lady butterfly the value of collecting together 
small and apparently isolated pieces of information, both from 
his own notes and from the observations of others, which, 
when carefully collated, throw a great deal of light on the 
process of dispersal. 

10. We have said earlier that when an animal has accom- 
plished the business of dispersal from its starting-point, the 
next process consists in the establishment of itself as an 
individual and then as a species, in its new home. There is, 
however, this exception to be noted. It seems highly probable, 
although difficult in the present state of our knowledge to prove 
conclusively, that many animals migrate on a large scale in 
order to get away from a particular place rather than to go 
towards anywhere in particular. That is to day, there are often 
very cogent reasons why a large section of the population should 
migrate somewhere else, the most common one being over- 
population. We see such pressure of numbers acting in the 
case of lemmings, locusts, sand-grouse, and aphids, when they 
suddenly depart in huge swarms for new lands. The point 
here is that the immediate reason for migration may be to 
relieve the situation in their normal habitat, and although it 
is all to the good for each species to establish itself from one 
of these migrating swarms in another locality (as happens in 
the case of the Caucasian locust migrations), yet the main 
effect of the migration is to remove a surplus population 
from the area usually inhabited by the species, and not to 
colonise other areas. It should, therefore, be borne in mind 



DISPERSAL 



157 



that any large-scale migration of this sort may have two reasons, 
either to get away from the centre of distribution in order 
to prevent disaster through overcrowding, or to reach some- 
where at the circumference and so extend the range of the 
species. 

II. It is plain that an enormous wastage must occur 
while the establishment after dispersal is taking place, and 
that only a tiny fraction of the original emigrants will ever 
succeed in establishing itself even temporarily. We cannot 
do better than quote here the words of Wood-Jones,^°^^ who 
had a peculiarly good opportunity of appreciating the factors in 
dispersal and establishment of species, in connection with the 
arrival of new animals and plants on the coral islands of Cocos- 
Keeling. He says : *' Those creatures that are settled and 
established are the elect, and they are appointed out of a 
countless host of competitors, all of whom have had equal 
adventure but have gone under in the struggle, through no 
fault of their own. They are the actual colonists, the survivors 
of a vast army of immigrants, every one of which was a potential 
colonist." One particularly striking example was noted by 
him ^^''^ : occasionally huge flights of dragon-flies would arrive 
(belonging to the species Pantula flavescens^ Tramea rosen- 
bergiij and Anax guttatus)^ and would live for some time 
and feed ; but, owing to the absence of any permanent open 
fresh water on the islands, they never succeeded in establish- 
ing themselves permanently, although they actually laid eggs 
in temporary pools, which were not suitable for their breed- 
ing purposes. Another example of abortive colonisation on 
a huge scale was encountered by the sledging parties of the 
Oxford University Arctic Expedition whilst crossing the ice- 
cap of North-East Land in the summer of 1924.^^ One day 
in August, all three parties in different parts of the country 
encountered vast swarms of aphids {Dilachnus picece), normally 
found on the spruce {Picea ohovata) of Northern Europe, 
together with large numbers of hover-flies of the species 
Syrphiis ribesii. These insects had travelled on a strong gale 
of wind for a distance of over eight hundred miles, and had 
been blown in a broad belt across the island of North-East 



158 ANIMAL ECOLOGY 

Land (which is about the size of Wales). Since the surface 
is entirely covered with ice and snow, or else consists of very 
barren rocks with a high-arctic flora, there was not the remotest 
chance of either the aphids or the hover-flies surviving. As a 
matter of fact, the majority of them were wiped out by a 
blizzard which occurred three days later. 

12. Abortive colonisation is happening everywhere in 
nature, on a smaller scale, and the two examples quoted above 
are given in order to drive home the fact that dispersal by 
itself may have absolutely no effect upon the distribution of 
a species, unless it is accompanied by eifective establishment 
at the other end. The factors controlling the survival of the 
species have been dealt with in a general way earlier in this 
book, and it is therefore unnecessary to say more about 
them here. We may note, however, that one of the chief 
difficulties facing an animal upon its arrival is that of finding 
a mate, with which to co-operate in perpetuating its race. The 
chances of one individual copepod reaching a new piece of 
water by accidental dispersal may be small, and that of two 
individuals of opposite sexes doing so smaller still. But the 
chance of these two meeting and mating and bringing up 
young would often seem to be extremely remote indeed. We 
find, therefore, that animals which have to reach new places 
by definite migration or other special means of dispersal are 
either parthenogenetic during part of their life-cycle, or else 
migrate in large parties, like locusts. In the case of locusts 
there are well known to be special tropisms which cause the 
animals to stay together in swarms, and to follow the move- 
ments and flight of their neighbours, so that there is a good 
chance of a swarm of both sexes arriving at the other end.^^ 
In the case of most water-fleas the egg which is transported 
by accident hatches into a female, which is able by par- 
thenogenesis to produce a huge population, amongst which 
males occur at a later stage. Tliis is also a common method of 
colonisation used by parasites, whose Hfe-history is often so 
risky that the chance of two animals of opposite sexes arriving 
together in one host is negligible. 

13. With regard to the methods employed by animals to 



DISPERSAL 



159 



find their proper habitat, when they are migrating or otherwise 
becoming dispersed, there are three important devices to 
be noticed. The first class of animals employs the method of 
broadcasting enormous numbers of young ones or even of eggs, 
with the result that some of them fall on stony ground and 
perish, while others reach the right place, where they have at 
least a very thin chance of establishing themselves. This 
method is employed by all those spiders which float away on 
gossamer threads when young ; by locusts which start on migra- 
tion with full air-sacs and loaded fat-bodies, and fly along until 
the air and the fat are used up, when they have to descend and 
can then only undertake small local movements ^^ ; and it is 
also used by an enormous number of sessile and sedentary 
marine animals, which produce free-floating larvae that have 
only small powers of directive movement. This broadcasting 
involves a huge wastage of life, and is usually confined to young 
animals, except in cases which, like the locusts, are partly 
concerned with the reUef of pressure in the home population. 

The second method is to have some special reaction which 
enables the animals to find their suitable habitat ; this is a 
very widespread method, and is much less wasteful than the 
broadcasting one. Pettersson ^^^ has studied the fluctuations in 
the Baltic herring fisheries and found that the herring pro- 
bably only enter the Baltic when water of a certain salinity 
penetrates there. They follow the salt water, and refuse to 
go in water with a salinity of less than 32 or 34 per mille. 
At the present time the lower layers of the Baltic are 
hardly ever more than 28 per mille ; but this condition 
is influenced by the tidal effects of the moon, so that it 
appears that in the past there have been at certain times 
invasions of the Baltic by comparatively salt water, such as 
still occurs in the Skagerak and Kattegat. His work on this 
problem makes it practically certain that the existence of a 
definite salinity-preference on the part of the herring, combined 
with peculiar tidal phenomena, has caused in the past regular 
fluctuations in the prosperity of the Baltic and neighbouring 
fisheries, whose period is about i8| years, with a superimposed 
longer period of iii years. 



i6o ANIMAL ECOLOGY 

14. On the other hand, some marine fish react to other 
factors than salinity. Thus, the mackerel in the Black Sea 
are said to migrate in response to a change in temperature. 
Whether this is the actual factor atwork or not, it appears certain 
that they do not react to salinity. ^^* With birds, the precise 
way in which they find their right habitat is not known, but 
in many cases they appear to know simply by a rather elaborate 
process of experience and memory, in others by some more 
mysterious sense of direction. Many insects have definite 
chemotropic reactions which lead them to choose the right 
habitat either for feeding or for egg-laying. For instance, 
Howlett ^^^ made Sarcophaga oviposit in a bottle with scatol in 
it, this being a decomposition product of albuminous sub- 
stances ; while Richardson ^^^ made house-flies oviposit in re- 
sponse to ammonia together with butyric and valerianic acids. 
Barrows ^^'^ found that the positive reaction of Drosophila to 
fermenting fruit was due largely to amyl, and especially ethyl, 
alcohols, acetic and lactic acids, and acetic ether. 

But although a considerable amount of work has been done 
in this direction, and many more examples could be quoted, 
the fact remains that we are hardly able in any case to say how 
a particular insect does manage to find the right habitat to 
live in. That is to say, we can see that it inhabits certain 
vegetation and physical conditions, and that these are best 
suited to its physiological endowments, but it is hard to find 
out, and usually unknown, by what indication it is enabled 
to find these optimum conditions. This point has already 
been touched upon in the chapter on environmental factors 
(p. 40). 

The third method of finding habitats when dispersal is 
in progress is by general broadcasting combined with local 
directive movement. This appears to be practised by some 
birds (homing pigeons amongst others) and by a number of 
insects. Thus, a butterfly may undertake a huge migration 
whose direction is only determined by the particular winds 
blowing at the time ; but if it arrives at any place at all like its 
original home it will then be able to find the right food-plant 
for its larvae, by chemotropic or other means. It is possible 



DISPERSAL i6i 

to perceive here the natural place of the study of animal be- 
haviour and of tropisms in biology. Experimental studies in 
this field are almost always conducted from the point of view 
of physiology or psychology purely and simply ; but there is 
a large field for studying the natural significance of such 
reactions with reference to the normal life of the animals. 



M 



CHAPTER XI 

METHODS 

There are several points about methods which (i) are of general importance 
in ecological work, e.g. (2) the recording of facts with an eye to the use 
to which they will be put in the future, and (3) the correct identification 
of species, which latter depends both (4, 5) upon a pleasant and com- 
prehending attitude of systematists towards ecological work and (6) 
upon the collection of good systematic material by ecologists, who alone 
can provide the right data with the specimens. (7) The usual mistake 
among beginners is to under-estimate the number of animals of each 
kind. (8) Information from other people can be of great value if backed 
up by specimens of the animals concerned. (9) The carrying out of a 
biological survey involves various things : first, the listing of the main 
habitats, then (10) the collecting of the animals, together with careful 
habitat- and other notes, and finally (11) the construction of food-cycle 
diagrams, which (12) necessitates exhaustive study of the food habits 
of animals, a study which can be made in at least ten ways. (13) The 
community-relations of animals can be worked out in two ways, either 
separately or combined together. (14) The numbers of animals require 
special methods for their recording : one may use censuses in a given 
area or (15) in a given time, while (16) for recording variations in numbers 
it is advisable not to refer to " the usual " as a standard, but (17) to the 
numbers in the previous year or month, etc. (18) Finally, in publishing 
the results of ecological surveys it is desirable to include an index of 
species or genera, and (19) to employ certain special methods for 
recording the facts about food, etc. 

I . Although the whole of this book is really concerned with 
methods of tackling ecological problems, rather than with 
an inexorable tabulation of all the important facts which are 
known about ecology, it is advantageous to draw together all 
the various lines of thought into one chapter, and to mention 
a few general ideas which may be of use as a background to 
ecological work. One of the most striking things about 
natural history facts is the haphazard way in which they are 
usually recorded. We are not referring so much to the fact 
that our knowledge of so many life-histories of animals has to 
be built up by piecing together fragmentary observations of 
different people, since it is impossible for any one person to be 

162 



METHODS 163 

lucky enough to work out a complete picture of the life and 
habits of any one animal in all its aspects and phases. The 
thing which strikes one is rather the way in which the observa- 
tions are recorded, there being in many cases no principle 
followed. After all, it is impossible to describe an occurrence 
in the most useful way, without having some idea of how the 
information is going to be used Adams 1* has emphasised the 
importance of this in a very helpful chapter on ecological 
methods, and quotes Van Hise, who said : "I have heard a 
man say : * I observe the facts as I find them, unprejudiced 
by any theory.' I regard this statement as not only condemning 
the work of the man, but the position as an impossible one. 
No man has ever stated more than a small part of the facts 
with reference to any area. The geologist must select the 
facts which he regards of sufficient note to record and describe. 
But such selection implies theories of their importance and 
significance. In a given case the problem is therefore reduced 
to selecting the facts for record, with a broad and deep com- 
prehension of the principles involved, and a definite under- 
standing of the rules of the game, an appreciation of what is 
probable and what is not probable." 

2. The first point of importance is therefore to have a 
very clear idea of the use to which your observations will 
probably be put afterwards, whether by yourself or others. 
At the same time, of course, facts are constantly assuming an 
unforeseen importance in the light of later discoveries ; we 
are merely pointing out that it pays to try to look ahead and 
make records in such a way that they will be as intelligible and 
valuable as possible. These remarks may sound commonplace 
and superfluous, but an example will show the great importance 
of the point raised. When an ornithologist records the food 
of a particular species of bird, he very seldom troubles to find 
out the exact species of food- animals concerned. For instance, 
many food records merely refer to " mayflies " or " worms " 
or " Helix. "" Conversely, when an entomologist records the 
enemies of some caterpillar he will often enough refer to them 
as " small warblers," or if a worm-lover were to speak about 
the enemies of worms upon mud-flats he would probably 



1 64 ANIMAL ECOLOGY 

talk about " wading birds." Although it is possible to find 
out a great deal about the food-cycles in animal communities 
by working in terms of wider groups of animals than species, 
yet it is essential for a complete understanding of the problem 
to know the species of eater and eaten — a thing which we very 
seldom do know. If the ornithologist or entomologist took 
the small extra trouble of getting the foods accurately identified 
down to species^ their observations would be increased about 
a hundredfold in value. 

3. Although the number of observations about the food 
and enemies of various animals is prodigious, yet the majority 
of these data are just too vague to be of much value in making 
a co-ordinated scheme of the interrelations of animals. In 
other words, when one animal is seen eating another, it is very 
desirable to record the exact names of both parties to the 
transaction. The record of " green woodpecker eating flies " 
is of some use, as is the record of " Woodpecker eating Borborus 
equinus/* but the ideal observation is " green woodpecker 
eating Borborus equinus.^^ This point leads on to another 
important one, namely, the necessity for cultivating a proper 
** species sense." The extent to which progress in ecology 
depends upon accurate identification, and upon the existence 
of a sound systematic groundwork for all groups of animals, 
cannot be too much impressed upon the beginner in ecology. 
This is the essential basis of the whole thing ; without it the 
ecologist is helpless, and the whole of his work may be rendered 
useless, or at any rate of far less use than it might otherwise 
have been, by errors such as including several species as one, 
or using the wrong names for animals. The result of such 
errors is endless misinterpretation of work, especially by 
people in other countries. It is possibly to this danger that 
we must attribute a certain lack of sympathy for ecological 
work, politely veiled or otherwise, which is sometimes met 
with among systematists. They realise that ecological observa- 
tions are dependent upon correct nomenclature, and are 
therefore to some extent ephemeral, in cases where the latter 
is not yet finally settled. Added to this is the feeling that 
ecologists are rather parasitic in their habits and are to 



METHODS 165 

some extent using other people (systematists) to do their 
work ! 

4. This feeling is natural enough, and arises from the fact 
that a systematist has two distinct functions : one is to describe, 
classify, and name all the species that exist (or have existed) ; 
the other is to identify specimens for other people, especially 
when elaborate technique and considerable skilled knowledge 
are required in the process. Now, it is only recently that the 
animal kingdom has begun to be completely explored. The 
systematist is still busy putting his own house in order, or, 
what is also often the case, putting in order the houses of other 
people who have died leaving them in a considerable mess. 
But in a great many groups of animals, we are really in sight 
of the time when there will be comparatively little purely 
descriptive work and classifying of species ; although, to the 
study of the exact limits of species and varieties, and what 
they are, there will never be an end. The point is that the 
system of classification is rapidly becoming standardised, and 
we shall, in the near future, be able at least to reach agreement 
(often arbitrary enough) as to what is meant by any specific 
name. It follows from what has been said, that the task of 
the systematist will become more and more that of the man 
who identifies specimens for other people, and less and less 
that of the describer of new species. 

5. One of the biggest tasks confronting any one engaged 
upon ecological survey work is that of getting all the animals 
identified. Indeed, it is usually impossible to get all groups 
identified down to species, owing to lag in the systematic 
study of some of them (e.g. Planarian worms). The material 
collected may either be worked out by the ecologist himself 
or he may get the specimens identified by experts. The latter 
plan is the better of the two, since it is much more sensible to 
get animals identified properly by a man who knows them well, 
than to attain a fallacious sense of independence by working 
them out oneself — wrong. Also, in the majority of cases, 
there is simply not time for the ecologist to work out all 
the material himself, and it seems certain that nearly all 
primary survey work will in the future have to be carried on by 



i66 ANIMAL ECOLOGY 

co-operation on a large scale between ecologists in the field and 
experts in museums. At the same time it is useful to know 
how to name the more obvious species of animals, and to know 
also where to find out general information about any particular 
group or species. A list of the works dealing with a number 
of groups of British animals is given in the bibliography at the 
end of this book. The Hst is necessarily incomplete, since in 
some cases {e.g. fresh-water planarian worms) no comprehensive 
work has been published ; while in many others, although the 
systematic work has been thoroughly done, the results are 
scattered in a number of periodicals, or are in relatively in- 
accessible foreign works, or else remain locked up inside the 
heads of experts who have not yet had time or opportunity 
to write the necessary monographs on their groups. Further- 
more, in the list of works quoted, no complete treatment is 
attempted of protozoa, parasites, or marine animals. 

6. The vital importance of good systematic work and the 
desirability of making it as far as possible available in a simplified 
form to working ecologists has been pointed out. We may 
now turn for a moment to the other side of this matter. 
It is very important that ecologists should, during the course 
of whatever work they are doing, pay attention to the collec- 
tion of material which can be used for systematic studies. 
Ecologists often have unique opportunities for collecting large 
series of animals from one place at different times, and such 
series are often invaluable in helping to decide the limits of 
variation of different species. It is becoming more and more 
clearly realised that the habits and habitats of animals may 
form systematic characters quite as important as structural 
features, and that unless information of this type is accumulated 
in the form of good specimens with full data about habitats, 
etc., attached, there is no proof that one " species " does not 
contain a number of species, differing in such ways, but not 
in obvious structures. A striking example of this kind of 
thing is Daphnia pulex, whose life- cycle in Europe includes 
the formation of fertilised winter eggs which enable the species 
to tide over the winter until the following spring. In Spits- 
bergen there is also a Daphnia which is identical in structure 



METHODS 167 

and habitat with the European form, except that it has the 
unusual power of forming winter eggs parthenogenetically, 
without the necessity of fertilisation by a male. There are in 
consequence no male Daphnia in Spitsbergen at all.^-^ Another 
good example of the importance of field observations for dis- 
tinguishing species is that of the British warblers, which are in 
some cases much more easily distinguished by their songs 
and nesting habits than by their appearance. 

Now, the systematist is not usually a trained field naturalist, 
or, if he is, he lacks the knowledge of plant and animal associa- 
tions which is required in order to define accurately the habitat 
of the specimens he is collecting. The ideal procedure would 
seem, therefore, to be that as full data as possible should be 
entered upon labels and handed over to the systematist with 
the specimens, and that a more detailed account of the environ- 
ment, and in particular of the animal environment, should be 
published by the ecologist himself, who can employ, if neces- 
sary, some means of referring to the actual specimens collected. 

7. We have dwelt at some length on the necessity of getting 
absolutely reliable identification wherever it is possible, because 
mistakes in this matter are one of the most fruitful causes of 
misunderstanding, while vagueness in description of an animal 
may render the most brilliant observations upon its ecology 
more or less valueless. The usual mistake made by beginners 
is in under-estimating the number of species in a genus and so 
becoming careless about checking all specimens obtained in 
order to get exact identification in all cases. Thus, suppose I 
record '' Daphnia pulex eaten by sticklebacks " ; there are 
two quite different species of sticklebacks, the ten-spined 
(Gasterostetis pungitius) and the three- spined (G. aculeatus)^ 
and since I had not distinguished them, you might begin to 
wonder whether I was aware of the existence of different 
species of Daphnia also. This element of uncertainty makes the 
value of the observation very small. In practice it often re- 
quires only a very small extra amount of trouble to collect a 
few specimens of the animal on which the observations have 
been made, or in the case of animals like birds and fishes, to 
look up in a book to see how many species there are and which 



i68 ANIMAL ECOLOGY 

it was. It frequently happens that the person who chances 
to notice some fact of vital interest to the ecologist working 
on some problem is not the ecologist himself, but some other 
kind of biologist, or perhaps some one who is not a scientist at 
all. If the observations made by such people could be backed 
up by specimens of the animals, it would be possible to collect 
a vast amount of very valuable data. 

8. It is worth while bearing in mind that the ecologist can 
frequently get, in this way, facts which he would otherwise 
never come across at all by himself, and he should make every 
effort to enlist the help of other people to co-operate. In this 
connection it is worth while quoting from the rules which were 
made out by Dr. Levick for the use of non-biological members 
of the Northern Party of Scott's Antarctic Expedition.121 
They ran as follows : " Members are invited to write in this 
book notes on anything of interest seen by them relating to 
birds, seals, whales, etc., appending their initials and bearing 
in mind the following observations : 

" (i) Never write down anything as a fact unless you are 
absolutely certain. If you are not quite sure, say ' I think I 
saw ' instead of ' I saw,' or * I think it was ' instead of ' It 

was ' ; but make it clear whether you are a little doubtful or 

very doubtful. 

" (2) In observing animals disturb them as little as 

possible. . . . 

" (3) Notes on the most trivial incidents are often of great 

value, but only when written with scrupulous regard to 

accuracy." 

These rules are useful for zoologists also. 

9. Having given these general suggestions about ecological 

work, we will now consider the best methods of carrying out 

a general primary ecological survey of animal communities. 

Many valuable hints are contained in Tansley's book,^^ since 

to a large extent the methods of primary survey are essentially 

the same for plants and animals. The process of making such 

a survey is as follows : 

First of all, have a general look round the area to be studied, 

and get an idea of the main habitats that exist, and in 



METHODS 169 

particular of the main plant associations. Don't bother about 
details yet, but simply try and get a grasp of the big habitats 
and habitat gradients. When you have made a list of the 
important habitats, come down more to details and subdivide 
these into smaller areas or zones, in the manner indicated in 
Chapter II. Thus, at this point your notes would be in the 
following form : 

" The country can be divided roughly into the lower- 
lying parts which are cultivated, and the upper hilly parts 
which are not. The uncultivated area can be divided into 
three very distinct main zones : 
'' I. Grassland. 

''2. Bracken, with scattered trees, forming a sort of 
bracken savannah. 
" 3. Woodland. 

" I and 2 are more or less abruptly separated, but 2 and 3 
grade into one another at their margins owing to the com- 
plicated distribution of shrubs, such as bramble, and small 
trees, such as hawthorn." 

Supposing we then took the woodland, the notes would go 
on something Hke this : 

" There are several fairly distinct types of woodland. 
" I. Ash, with some sycamore. 

"2. Oak woods (which species ?), with hazel undergrowth. 
"3. Oak and sweet chestnut woods. 

" These again vary much in undergrov^h owing to the 
effects of fires, and felling, and age. N.B. This summer is 
dry enough to have caused grass fires, but the woods have not 
caught seriously." 

If, then, we considered the oak-hazel wood, we might 
write : 

" The oak wood can be divided into vertical strata : 
" I. Tree-tops — 
" (a) Leaves. 

" (b) Twigs and branches. 
*' (c) Under bark and rotten wood of branches. 
" 2. Trunks— 

*' (a) Upper part with lichens (drier). 



170 ANIMAL ECOLOGY 

" (b) Lower part with mosses and liverworts (damper 
than last owing to run-off from the trunk, and 
height often only a foot, but varying according 
to the aspect). 

" 3. Hazel undergrowth, with some other shrubs. 

"4. Herbaceous undergrowth. 

"5. Litter of dead leaves, etc. ; or 6. Moss carpet. 

" 7. Soil, underground." 

10. This listing of habitats does not take very long to 
carry out, and is absolutely essential. Wherever possible the 
co-operation of a plant ecologist should be enlisted, in order 
that the plant associations may be accurately determined. 
But often it is sufficient to make lists of plants yourself 
from each of the habitat divisions (perhaps with the aid of 
some field botanist who knows the species well). Druce's 
Botanisfs Pocket-hook^^ is extremely useful for accurate 
identification of British plants. When this has been done, the 
next thing is to start collecting the animals from these different 
habitats, and in doing this there are several points to be borne 
in mind. First, it is vitally important to make as full notes as 
possible on the animals, and to record full details of the exact 
habitat, e.g. the species of plant on which they are found, 
whether they were on the upper or on the lower sides of the 
leaves, and any other observations made at the time, such as 
the reaction to light or rain, or food-habits, or numbers. The 
last is especially important. The data can either be written 
on a label with the specimen or, what is usually more con- 
venient, a number can be placed in the tube or box containing 
the specimen and a corresponding number entered against the 
notes upon it in your notebook. A rather convenient method 
of making notes is to carry a few record cards such as are used 
for a card index, instead of the usual notebook. On the other 
hand, if a notebook is used, it is possible to take a carbon copy 
at the time, which may often be useful. Usually, however, 
it is impossible to make anything but very brief and rough 
notes in the field, and they have to be written up carefully at 
home afterwards. It is customary to warn students that they 
must make notes on the spoty and not afterwards. A trained 



METHODS 171 

ecologist can, however, quite safely carry a lot of the details 
in his head and put them on paper at the end of the day. This 
is a matter of practice, and is a habit worth cultivating, as it 
saves much time and also makes it possible to do better work 
in wet weather, when note-taking is awkward. The chief 
time when it is best to take notes on the spot is when one is 
trying to prove something definite, since at such times it is 
very easy to forget the facts that do not agree with one's theories. 
It usually happens that a certain number of animals are found 
in odd places which do not fit in with any of the habitats 
originally listed, and these will necessitate some revision of 
the habitat-divisions you started with. Again, a number of 
animals are always occurring accidentally in the wrong habitat, 
and although they should be recorded carefully, the amount 
of detail as to their habitat need not be great. Discretion has 
to be used in this matter. 

II. When a general idea of the distribution of the fauna 
has been gained, it is advisable to attempt the construction of 
a rough food-cycle diagram showing the relationships of the 
species. To do this accurately it is necessary to get the 
specimens identified, but a rough preliminary idea can be 
formed without knowing the exact species, although it is useless 
to publish such a diagram unless backed up by lists of the 
actual species concerned. It will be found necessary to organise 
a sort of ring of consulting systematists who are willing to 
work out material from the various groups of animals. It is a 
good plan, when sending large numbers of specimens (most of 
which will probably be quite common ones) to include, if 
possible, some which seem unusually interesting or rare, 
since in this way the expert who is working out the material 
for you will find it more interesting, and will be the more 
willing to help in the future. ^ For details of methods of 
collecting and preserving animals the reader may be referred 
to the British Museum Handbook for Collectors, 1^2 which 
covers a number of animals, to Ward and Whipple, ^'^ who 
give excellent directions for most fresh- water animals, and to 
the various books given in the bibliography of special groups 
of animals. 



172 ANIMAL ECOLOGY 

It is very important that specimens should be killed and 
preserved in the appropriate way, as otherwise they may be 
useless for purposes of identification, or, at any rate, cause a 
lot of unnecessary trouble. 

12. For constructing food-cycles there is only one method — 
the patient collecting of all kinds of information about the 
food and enemies of the species that are being studied. Direct 
observation in the particular place in v^^hich you are working 
is the best, since the food habits of animals are often very 
variable at different times and in different places. It may be 
convenient to summarise the various ways in which evidence 
about animals' food may be obtained. 

(i) Watch the animal eating and, if necessary, take a speci- 
men of its food (and of the animal itself). This is the type of 
evidence that is most difficult to obtain. 

(2) Examine the contents of the crop or stomach or in- 
testine. This may give good positive evidence but is useless 
for proving a negative, e.g. remains of butterflies disappear 
very quickly under the influence of the digestive juices of 
birds. 

(3) Finding stores of food, etc. 

(4) From a study of the animals and plants associated with 
it, deduce the animal's probable food. This enables the field 
of observation to be narrowed down. For instance, the 
writer saw a ptarmigan rise from a hillside, and on going to the 
spot where it had been, found that a number of seeds and 
flowers had been eaten from various plants. Since this bird 
was the only large herbivorous animal in the region, it was 
certain that the bird had eaten them itself. 

(5) Experiments may be made to confirm such proba- 
biHties. 

(6) Examine excretory products, e.g. castings of owls 
containing remains of small mammals, or droppings of terns 
containing limbs of Crustacea. 

(7) The structure of the animal will help to narrow down 
the field to a particular size of food. 

(8) Note any food preferences, with reference both to 
quality and quantity. 



METHODS 173 

(9) The amounts eaten per day are of great interest, e.g. 
counts of the number of animals brought to its nest by a bird 
in a given time. 

(10) Finally, the numbers of two species will often give 
a clue to the fact that one is feeding on the other, e.g. 
birds attracted by an unusual abundance of caterpillars on 
oaks. 

13. There are two ways of tackling the problem of food- 
cycles and community-organisation of animals. One way is 
to start with one particular species and radiate outwards along 
its various connections with other animals and follow the 
train of associations wherever it leads. This was the method 
described in Chapter V. It is a very fascinating form of 
ecological work, owing to the variety of interesting facts and 
ideas which are encountered, and it also has the advantage that 
it can be carried out without any very elaborate previous survey 
or listing of all the species of animal in the district. On the 
other hand, one may list all the animals and then subdivide 
them according to their place in the community — herbivores 
and carnivores, key-industries, terminal species, large and 
small, and so on. The separate food-chains can then be worked 
out, and in this way one gets a better perspective of the whole 
community. Perhaps a combination of the two methods 
would be the best procedure. 

14. Another important subject about which something may 
be said here is that of numbers. The study of numbers is a 
very new subject, and perfect methods of recording the numbers 
and changes in the numbers of animals have yet to be evolved. 
In practice, we have to deal with two main aspects of this matter. 
The first question is as to the best way of taking censuses of 
the animal population at any one time, and the second is the 
question of recording changes in the numbers from one period 
to another. With regard to the first, a certain amount of 
work has been devoted to the methods of estimating the absolute 
numbers of various animals. Quantitative work on plankton 
has reached a very high degree of efficiency ; the usual method 
consists in doing counts of small samples from the whole 
collection and then multiplying by a factor to get the total 



174 ANilVEAL ECOLOGY 

numbers present. The method of weighing material is also 
used. These methods are fully dealt with by Whipple,^^ and 
by Birge and Juday.92 

One of the more important recent inventions, not described 
in these books, is the apparatus designed by Hardy ,1^3 which 
enables marine plankton to be collected continuously on a 
band of silk as the ship moves along at sea. This apparatus 
has already shown good results on the Discovery whaling 
expedition, and will provide information of great ecological 
interest. For by means of it, a belt-transect of the plankton 
can be made along any desired line, and variations in the 
fauna which are clearly shown can be used for correlation 
with the physical and chemical gradients in the sea, or with 
changes in the distribution of larger animals such as whales 
and fish. 

Similar methods can be employed for soil-animals, and 
in fact for any animals which are sufficiently small and 
numerous to be susceptible to mechanical sampling and 
counting. The problem becomes much more difficult in the 
case of the higher animals like birds and mammals, which are 
more mobile, are constantly shifting their place of abode, and 
are, relatively speaking, so scarce as to make it impracticable 
to kill large samples and count them. However, it is com- 
paratively easy to make accurate censuses of nests during the 
breeding season, and a good deal of work along these lines has 
been done in the United States. The reader may be referred 
to a recent book by Nicholson, ^^'^ who gives an account of 
the methods of bird census successfully employed by him 
in England. 

15. Grinnell and Storer^o have successfully employed a 
different method of recording the numbers of birds. They 
say : " Instead of using a unit of area, we used a unit of time. 
Birds were listed, as to species and individuals, per hour of 
observation. In a general way this record involved area too. 
Our censuses were practically all made on foot, and the distance 
to the right or left at which the observer could see or hear 
birds did not differ, materially, in different regions. The rate 
of the observer's travel did, of course, vary some . . . also, in 



METHODS 175 

some places, the greater density of the vegetational cover 
acted to limit the range of sight. But for each of these adverse 
features of the method there were certain compensations." 
One of the advantages of this method is that it gives a good 
idea of the relative numbers of animals in any association, and 
this is one of the most important types of fact about which 
we require information. It seems probable that the method 
will give information of great value, so long as a sufficient 
number of censuses are obtained by different people, in order 
to eliminate the effects of factors like the weather, time of day, 
rate of travel, etc. 

16. In addition to censuses giving the average numbers 
of animals in different habitats, we require methods of recording 
the changes in numbers from month to month or year to year. 
Of course, a series of censuses of the kind described will provide 
this information, but in many cases there is not the time, staff, 
or opportunity for carrying out censuses of sufficient accuracy, 
in which the methods will remain the same as time goes 
on, so that the results are comparable. The chief difficulty 
of recording changes in the numbers of any animal which 
undergoes violent fluctuations in numbers is in finding a 
standard to which the abundance in different years can be 
referred. Such statements as " wasps are more abundant 
than usual " cannot be safely used, for two reasons. The 
first is that we do not know what '' usual " means ; the second 
is that its meaning varies from year to year, and in the minds 
of different people. The latter is due to the fact that most 
people do not remember with any accuracy for more than 
about five years ; and also that more significance is attached to 
recent years than to earlier ones. The result of all this is 
that the word " usual " when applied to numbers may mean 
practically anything, according to the particular emotions, 
powers of observation, and strength of memory of the observer. 
The records of butterfly abundance in England given in the 
Phenological Reports of the Royal Meteorological Society, 
prove conclusively and surprisingly that butterflies are " scarcer 
than usual " in about one year out of every five ! It comes to 
this, that records referring to the '* usual " are only of value 



176 ANIMAL ECOLOGY 

when they refer to years of very great scarcity or very great 
abundance. In intermediate years they are almost, if not 
quite, valueless. 

17. The best method of recording the relative changes in 
numbers of fluctuating animals appears to be as follows : the 
numbers in any one year are referred to the abundance of the 
previous year. Thus we might say " small tortoiseshell 
butterflies more abundant this year than last year." If a 
continuous series of such records be made, we can then get 
a very clear idea of the relative abundance from year to year, 
and if there is any regular periodicity in the numbers, the 
maxima and minima will be quite easily distinguished. The 
advantage of this method is that it avoids the errors which arise 
when a fictitious average ('' usual ") is used as a standard. 
Furthermore, most people can remember pretty clearly what 
the numbers were in the previous year, and so there is no 
danger of introducing a great error in this way. It is advisable 
to keep at least two separate records, one referring to the 
breeding season, and the other to the non-breeding season. 
Then the numbers in the breeding season of one year can be 
compared with those both of the breeding and non-breeding 
seasons of the previous years. The method is, of course, 
equally applicable to monthly or other variations in numbers ; 
its limitation is that it can only be used on fairly conspicuous 
animals. This method of recording changes in the numbers 
of animals requires if possible to be backed up by actual 
census figures in some years at least. In this way it would 
be possible to give the curve of fluctuating numbers an 
absolute value. 

18. In conclusion, it is desirable to say something about pub- 
lication of the results of ecological work and the best methods 
of presenting the facts so as to be of the greatest use to other 
people. We have already dealt with some of the more im- 
portant errors into which it is possible to fall — insufficient 
description of the habitat and inadequate or inaccurate identifi- 
cation of species. There are one or two other points which 
are worth mentioning also. The first is that primary survey 
work and other ecological work dealing with large numbers of 



METHODS 



177 



animals belonging to different taxonomic groups has ultimately 
as its main use the elucidation of particular problems about 
individual species, by providing a picture of the biological 
surroundings of the animals. The result of this is that writers 
of ecological papers should aim at making their results as 
accessible as possible to the man who is working on one group 
or one species {e.g. some animal of economic importance). 
Now, it is usually impossible for such a man to pick out what 
he wants from amongst the great mass of facts contained in an 
ecological survey paper, with its huge lists of species. If, 
however, a short index to species or genera, or even families, 
is included at the end of the paper, it immediately increases its 
practical value to other biologists about a thousand-fold. 
Wherever possible, therefore, an index giving page references 
should be included, thus enabling the information about any 
one animal to be picked out with the greatest ease and saving 
of time. 

19. Another method of presenting the results of ecological 
surveys, which has advantages, is that used by Richards, ^^ and 
consists in tabulating the lists of species in the following way : 



Animal Community 


OF Typical Callunetum. 


Common name. Latin name. 


Food habits. 


Habitat. 


Moth 

Hover- fly 


Scythris variella 

Stph. 
Sphcerophoria 

scrip ta L. 


Larva on Cal- 

luna and Erica 
Larva on aphides 


Adult hops about 
on bare ground. 
Adult on flowers. 



The best way of describing and recording food-cycles is 
another important problem to be faced in the publication of 
ecological work. Simple diagrams like those on pp. 58, 66, 
can be employed ; these are all right for showing general 
results, but when we wish to include a large number of species 
something more is required. Perhaps the most effective 
method would be to put in on the general diagram the group 
names, e.g. " aphids," together with a number referring to a 
list of the actual species in question — a list which would be too 

N 



178 



ANIMAL ECOLOGY 



long to include on the diagram. It is also sometimes useful 
to include the relative sizes of the different animals, but here 















/ 

/ 


Mice 
100 


"™"^ 


Red Fox 
600 




/ 
/ 
/ 

/ 






/ 






PLANTS K^-^ 


Varying Hare 
430 


• 

.1 — > 


Lynx 

900 




\ 














Deer 
1500 


. .. . ^ u . -..- ^ 


Wolves 
1000 



















Fig. 13. — The diagram shows part of an animal community in Canada, 
and illustrates the method of including food-chains and the size of the 
animals in the same diagram. (The figures are lengths of the animals in 
millimetres — average for both sexes, tip of nose to base of tail.) This 
diagram should be compared with that in Fig. 12, giving the length of 
the period of fluctuation of the animals. (From Elton. ^*) 

one is usually hampered by lack of data or by the difficulty of 
finding a standard to which animals of different shapes can 
be referred. 



CHAPTER XII 



ECOLOGY AND EVOLUTION 



(i, 2) Although the ordinary theory of natural selection appears, at first 
sight, to explain almost all the phenomena produced by evolution, the 
two greatest arguments in its favour being (3) the existence of so many 
perfect adaptations in animals and the difficulty of imagining how 
any but useful characters could spread in a population ; yet (4) there 
are certain cases of colour dimorphism among animals which cannot 
be explained on the hypothesis of natural selection. Of these one of the 
most striking is the arctic fox, with its blue and white phases ; (5) 
another example is the white-eared cob of the Sudan. In fact (6) it 
seems very likely that most so-called adaptive colours in mammals are 
not actually adaptive at all. (7) Furthermore, Richards and Robson 
have shown that it is highly probable that very closely allied species 
hardly ever differ in characters which are adaptive, although less closely 
allied species may do so. (8) These lines of evidence (from field 
observation on the ecology of the animals, and from systematics) make 
it very probable that there must be some process in nature which 
allows of the spread of non-adaptive characters in the population of a 
species. (9) The nature of this process will probably be revealed by 
ecological work on the numbers of wild animals, and (10) it is suggested 
that one factor in the spread of non-adaptive mutations is the expansion 
in numbers of a species after each periodic minimum in numbers, at 
which times the struggle for existence tends to cease or to become 
reduced. (11) Whatever may be any one's particular views on evolu- 
tion, there is no doubt that ecological work is absolutely essential for a 
solution of certain aspects of the problem. 

I . It may at first sight seem out of place to devote one chapter 
of a book on ecology to the subject of evolution. The reason 
for doing it is that the ecologist working in the field is con- 
tinually being brought up with a sharp bump against the 
species problem. There are, in- fact, certain aspects of the 
problem of the origin of species which can only be successfully 
tackled along ecological lines, and it is with these aspects that 
we shall deal in the present chapter, although it is here only 
possible to touch on some of the most important points. In 
order to make quite clear what part of the evolution problem 
is affected by ecological work, we must give a brief summary 

179 



i8o ANIMAL ECOLOGY 

of the present position of the subject. Every biologist accepts 
the fact that evolution has taken place. The problem which 
has not yet been really solved is the exact manner in which it 
has happened. The existence of vast numbers of undoubted 
and complicated adaptations in physiological, psychological, 
and structural characters makes it reasonably certain that 
Darwin's theory of natural selection must be essentially true, 
however we may disagree about certain parts of it. We 
start, therefore, by assuming that natural selection is a very 
important factor in encouraging the spread and perpetuation 
in the population of some of the genotypic variations which 
are constantly arising, and the cause of which is at present 
obscure. As we shall have occasion to point out that natural 
selection entirely fails to explain a number of phenomena in 
nature, it is well to be absolutely clear about the matter right 
at the start. The writer assumes that natural selection is an 
important factor in evolution, while at the same time holding 
that there are other agencies also at work, the nature of which 
will be best discovered by field ecological work on animals (just 
as Darwin and Wallace both discovered the existence of natural 
selection after an extensive experience of field work on animals). 
It should be further stated that the writer does not believe 
that there is as yet any conclusive evidence in favour of " the 
inheritance of acquired characters." 

2. The ordinary hypothesis of evolution by natural selection 
may be summed up conveniently as follows : 



^e natural rate Checks limiting 


of increase increase 


/ 




^ yr 




Struggle for existence 


Genotypic 


among individuals 


variations 


\ 


/ 


Natural selection of individuals 


«= Survival of the fittest 


\ 




Natural selection 


of the race 


\ 




Evolution and formation of adaptations 



ECOLOGY AND EVOLUTION i8i 

Looked at from another angle, the process of formation of 
a new species can be divided up into three phases : 
(i) Occurrence of a genotypic (heritable) variation. 

(2) Spread of this variation in the population. 

(3) Isolation of this new stock so as to form, ultimately, a 

new species. 
The process of isolation does not necessarily come in in 
all cases, however, since a new variety might simply spread 
through the whole population of that species, and automatically 
change the w^hole stock. 

3. With the origin of genotypic variations we are not here 
concerned. Ecology has, however, a definite contribution to 
make towards the study of the second and third phases. The 
usual Darwinian assumes that a variation which crops up 
singly, or at any rate rarely, has absolutely no chance of spread- 
ing in the population unless it is favoured by possessing some 
advantage over its fellows. This argument appears at first 
sight irrefutable. If a cod has a million eggs, and one of these 
eggs contains a new hereditary factor, what chance has this 
particular egg of growing up to the one of the two successful 
cods out of that million ? And if it did reach the breeding 
stage, and had a million young itself, only one of these, if any, 
would survive in the next generation. The Darwinian 
assumes that the deadly chances against any new variation 
spreading to any extent in the population can only be wiped 
out by the favourable influence of natural selection. If this 
is so, then all the characters possessed by animals — at any rate 
those which separate closely allied species — must either be of 
some direct use to the species (or to one sex in the species), or 
else they must owe their existence to the fact that they are 
intimately bound up in development with some other character 
which is useful, e.g. both might be products of the same 
hereditary factor in the egg. We see, then, that one of the 
great arguments in favour of the natural selection theory is the 
difficulty of any other hypothesis about the spread of variations, 
once they have arisen ; while another argument is that all 
animals are simply masses of adaptations. 

4. So far, we have been arguing from one step to the next, 



i82 ANIMAL ECOLOGY 

until we are led by an apparently unassailable chain of reasoning 
to the existence of adaptations everywhere in nature. We 
must now leave arguments for a moment, and start at the other 
end by reviewing a few of the facts. In the front court of the 
British Museum of Natural History there are two cases which 
illustrate the beautiful colour adaptations of arctic animals to 
their surroundings, and will also serve to illustrate what we 
wish to point out here. There is in one case a group showing an 
arctic fox (Vulpes lagopus), some ptarmigan, and some ermine, 
in their summer dress of browns and greys, which match the 
surrounding vegetation with great exactness. In the other 
case the animals are shown in their winter dress of pure white, 
which makes them invisible against the snow. So far, so good. 
But further study of what is known about the field natural 
history of the arctic fox begins to reveal awkward facts, which 
do not fit in easily with this scheme of protective coloration, 
and in fact reveal a number of creaking joints in its harness. 
All over the arctic regions the arctic fox possesses two colour 
phases, one of which is brown in summer and white in winter, 
while the other is grey or black in summer, and " blue " — 
often quite black — ^in winter. The writer has seen a " blue " fox 
in summer which was the colour of a black cat, and startlingly 
visible against rocks and vegetation at a distance of a quarter 
of a mile. The blue and white phases occur equally in males 
and females, and interbreed freely, and in different parts of 
the arctic regions are found in various proportions in the 
population. In Iceland only the blue phase is found, while 
in Labrador it is rare. In Greenland, Alaska, and Spitsbergen 
both are common.ios if the whiteness in winter is an adapta- 
tion, the blackness of the other phase cannot also be advantage- 
ous. If the black colour is not adaptive, how did it evolve ? 
If the white colour is adaptive, how does the black survive? 
We have in addition to reckon with the fact that in many parts 
of the arctic, the fox can have no possible use for its colour 
in winter, because it subsists at that season upon carrion left 
by bears, out on the frozen sea-ice, or if it is on land, it depends 
almost entirely on caches of animals collected and stored up 
in the autumn. 



ECOLOGY AND EVOLUTION 183 

5. There are many similar cases of dimorphic forms which 
must have arisen by the spreading of colour varieties in the 
population, but which apparently cannot have been encouraged 
either by natural or sexual selection. Another case similar to 
that of the arctic fox, is the antelope called the white-eared 
cob {Adenota leucotis)^ which inhabits the steppe country of 
the upper Sudan.^^^ In this antelope there are two colour 
phases, one of which is light or tawny in colour and more or 
less matches its surroundings, while the other is dark or almost 
black. The interesting thing is that this colour dimorphism 
exists only in the male, the females being all light- coloured. 
The light individuals match their surroundings, the black do 
not. Taking the whole range of the species, there is an area 
in the middle of the range with black and light phases living 
together, and an outer zone with the light phase only. 
Examination of the horns showed that the differences in colour 
in the males were not due to age differences, as is often the 
case with such animals. 

Here again, what at first sight seems to be an admirable 
adaptation in colour, turns out to be no better off than its com- 
panion phase which does not match its surroundings at all. 
Even if we assume that the colours are correlated with some 
other adaptation, the difficulty remains. There are, of course, 
a vast number of cases in which effective colour adaptation 
almost certainly exists {e.g. in the ptarmigan and in many 
insects), but in these cases there are never important dimorphic 
phases. But the fact that the adaptation exists in a number of 
cases does not in any way affect the fact that in certain other 
cases it does not exist at all. 

6. It is rather interesting to find how emphatically nearly 
all naturalists who have had wide experience of wild mammals 
reject the idea of colour adaptation in these animals. Dug- 
more ^^^ says : " The whole theory of protective coloration 
in the larger animals may be open to argument, but from my 
own observations in the field I am firmly convinced that 
practically speaking there is no such thing," while Roosevelt 86b 
says : "In South America concealing coloration plays no 
more part in the lives of the adult deer, the tamandua, the 



i84 ANIMAL ECOLOGY 

tapir, the peccary, the jaguar, and the puma, than it plays in 
Africa in the lives of such animals as the zebra, the sable 
antelope, the wildebeeste, the lion, and the hunting dog." 
Chapman ^oe gives evidence in favour of the same views. 

It can always be argued about any of these animals that 
even if the colours are not directly adaptive they may be 
correlated in development with some character (perhaps 
physiological) which is adaptive. But such arguments cannot 
apply to species which are dimorphic, like the arctic fox or 
the white-eared cob. Similar colour dimorphism is found 
also in the Tibetan wolf,^^ the African lion,i08 and the 
American grey squirrel,^'^ but is comparatively uncommon in 
mammals. In birds, however, it is often found. In the 
Galapagos Islands there is a hawk {Buteo galapagoensis) 
which has two phases (independent of age or sex), one of 
which is dark, while the other is pale buff,^^^ and a species 
of gannet {Sula piscatrix wehsteri) on the same islands which 
has two phases, brown and white.^^^ Many more examples 
could be given ; a good deal of the very considerable evidence 
about birds has been summed up by Stresemann.^i^ Exactly 
comparable colour dimorphism occurs in certain American 
dragonflies on the genus Mshna.^^ 

7. There is another important Hne of evidence on the subject 
of adaptation which has recently been investigated very care- 
fully by Richards and Robson and reviewed in a paper .29 The 
gist of their conclusions is that very closely allied species 
practically never differ in characters which can by any stretch 
of the imagination be called adaptive. If natural selection 
exercises any important influence upon the divergence of 
species, w^e should expect to find that the characters separating 
species would in many cases be of obvious survival value. But 
the odd thing is that although the characters which distinguish 
genera or distantly allied species from one another are often 
obviously adaptive, those separating closely allied species are 
nearly always quite trivial and apparently meaningless. These 
two authors say, after reviewing the whole subject : *' It thus 
seems that the direct utility of specific characters has rarely 
been proved and is at any rate unlikely to be common. Further- 



ECOLOGY AND EVOLUTION 185 

more, since the correlation of structure, etc., with other 
characters shown to be useful does not at present rest on many 
well-proved examples, it cannot yet be assumed that most 
specific characters are indirectly useful. Thus the role of 
Natural Selection in the production of closely allied species, 
so far as it is known at present, seems to be limited. This 
statement is not to be taken as a wholesale denial of the power 
of Natural Selection. The latter is not in question when 
structural differences of a size likely to effect survival are 
involved. It is only the capacity of selection to use on a large 
scale the small differences between closely allied species that 
is unproved." 

8. It seems probable that the process of evolution may 
take place along these Hues : genotypic variations arise in 
one or a few individuals in the population of any species and 
spread by some means that is not natural selection ; this 
process, combined with various factors which lead to the 
isolation of different sections of the population from one 
another, results in the establishment of varieties and species 
which differ in comparatively trivial and unimportant 
characters. Later on, natural selection is ultimately effective, 
probably acting rather on populations than on individuals. 
Some such hypothesis seems absolutely necessary to account 
for the facts (driven home by ecological work in the field, and 
by careful systematic work at home) that, on the one hand, 
remarkable adaptations exist in all animals, while on the other 
hand the differences between closely allied species are not 
adaptive. This view is opposed to most of the current teach- 
ing about evolution, which tends either to exalt unduly or deny 
completely the power of natural selection, but it has the 
advantage of fitting the facts, which is after all not a bad 
recommendation. The most obvious question which arises 
is how a variation can spread in any population unless it is 
in some way favoured by natural selection. The process 
which, as we have pointed out, must happen and be happening, 
must be a mechanical one which allows of the spread of all 
characters indiscriminately. Any really harmful one would 
be wiped out soon enough by natural selection, and any really 



i86 ANIMAL ECOLOGY 

useful one would be encouraged by natural selection. It is 
the indifferent characters with which we are concerned. 

9. There is little doubt that it will be through ecological 
work upon the numbers of animals that this problem will be 
finally solved, and what we know already about the subject 
enables us to make certain suggestions. It has been shown 
in Chapter VIII. that nearly all animals fluctuate considerably 
in numbers, some of these fluctuations being very violent and 
often very regular in their periodicity. For our present 
purpose the important thing to bear in mind is the fact that 
at frequent intervals (frequent compared to the time which it 
would take for a species to change appreciably) the population 
of many animals is reduced to a very low ebb, and that this is 
followed by a more or less rapid expansion in numbers until 
the former state of abundance is reached once more. After 
a lemming year, with its inevitable epidemic killing off of all 
but a few of the animals, the arctic tundra is almost empty 
of lemmings. The same thing can be said of the snowshoe 
rabbit. One year the country is pullulating with rabbits, the 
following year you may hunt for a whole summer and only 
see one. There is usually a rather rapid expansion after this 
minimum of numbers. In a stream near Liverpool studied 
by the writer, the whole fauna over a stretch of three miles 
was wiped out during the summer of 192 1, by a severe drought. 
Recolonisation took place from some deep ponds connected 
with the upper part of the stream, and after three or four years 
the population of molluscs, insects, Crustacea, fish, etc., had 
regained its " normal " density. A similar destruction of the 
fauna took place in 1921 in a small branch of the Thames 
near Oxford, but by 1925 the animals had reached '* normal " 
numbers again (through immigration and natural increase). 
The Gammarus pulex were very scarce in 1922, but had reached 
great abundance by 1925, when they were again practically 
wiped out, this time by an epidemic. 

10. Now, if you turn back to the diagram on p. 180 you 
will notice that the argument contains a certain fallacy. The 
original theory says that all animals tend to increase, and at a 
very liigh rate, but are prevented from doing so by checks. 



ECOLOGY AND EVOLUTION 187 

What has been said about fluctuations in numbers shows that 
such is not always the case. Many animals periodically 
undergo rapid increase with practically no checks at all. In 
fact, the struggle for existence sometimes tends to disappear 
almost entirely. During the expansion in numbers from a 
minimum, almost every animal survives, or at any rate a very 
high proportion of them do so, and an immeasurably larger 
number survives than when the population remains constant. 
If therefore a heritable variation were to occur in the small 
nucleus of animals left at a minimum of numbers, it would 
spread very quickly and automatically, so that a very large 
proportion of numbers of individuals would possess it when 
the species had regained its normal numbers. In this way 
it would be possible for non-adaptive (indifferent) characters 
to spread in the population, and we should have a partial 
explanation of the puzzling facts about closely allied species, 
and of the existence of so many apparently non-adaptive 
characters in animals.^^ 

II. There are many objections to this hypothesis, which is 
chiefly mentioned here, not merely because it affords a possible 
solution of the problem of the origin of species, but because 
it illustrates the fact that ecological studies upon animal 
numbers from a dynamic standpoint are a necessary basis for 
evolution theories. Another important result of the periodic 
fluctuations which occur in the numbers of animals is that the 
nature and degree of severity of natural selection are periodic 
and constantly varying. For instance, at a minimum of 
numbers, rabbits will undergo selection for resistance to bad 
climate or ability of males to find females, while at times of 
maximum there will be different types of selection, e.g. for 
resistance to disease, and ability of males to secure females 
in competition with other males. All these points about 
adaptation, numbers, and selection, prove that ecological work 
has a very important contribution to make to the study of the 
evolution problem. 



CONCLUSION 

There is no getting away from the fact that good ecological 
work cannot be done in an atmosphere of cloistered calm, of 
smooth concentrated focussing upon clean, rounded, and 
elegant problems. Any ecological problem which is really 
worth working upon at all, is constantly leading th-e worker 
on to neighbouring subjects, and is constantly enlarging his 
view of the extent and variety of animal life, and of the numerous 
ways in which one problem in the field interacts with another. 
In the course of field work one should have a rather uncom- 
fortable feeling that one is not covering the whole ground, 
that the problem is too big to tackle single-handed, and that it 
would be worth while finding out whether So-and-so (a 
botanist) would not be able to co-operate with benefit to both, 
and that it might be worth while getting to know a little about 
geology or the movements of the moon or of a dog's tail, or the 
psychology of starlings, or any of those apparently specialised 
or remote subjects which are always turning out to be at the 
basis of ecological problems encountered in the field. There 
is hardly any doubt at all that this feeling of discomfort or 
conscience, or whatever you choose to call it, required in all 
scientific work, if anything more than routine results are to 
be produced, is most urgently required in ecology, which is a 
new science. Its methods require a wholesale overhauling, 
in order that the rich harvest of isolated facts that has been 
gathered during the last thousand years may be welded into 
working theories which will enable us to understand something 
about the general mechanism of animal life in nature, and in 
particular to obtain some insight into the means by which 
anima} numbers are controlled. For it is failures in regula- 
tion of numbers of various animals which form by far the 



CONCLUSION 189 

biggest part of present-day economic problems in the field, 
and one of the aims of this book has been to indicate the lines 
along which the numbers of animals may be studied. The 
order of the chapters in the present book represents roughly 
the order in which it is necessary to study the ecology of animals. 
First, there must be a preliminary survey to find out the general 
distribution and composition of animal communities. Then 
attention is usually concentrated on some particular species, 
with the object of discovering what factors Hmit it in its range 
and numbers. As far as physical and chemical factors are 
concerned this can usually be done without reference to other 
species of animals, but one is practically always brought face 
to face also with biotic factors in the form of plants and 
animals. Ecological succession in plants, involving gradual 
migration of animal communities, has to be studied in this 
connection. When one starts to trace out the dependence of 
one animal upon another, one soon realises that it is necessary 
to study the whole community living in one habitat, since the 
interrelations of animals ramify so far. The study of an 
animal community is difficult, since so little work has been 
done on it ; but there are certain principles which seem to 
apply to nearly all communities, and which enable the inter- 
relations of different species to be understood fairly clearly. 
It is only when the limiting factors, biotic and otherwise, have 
been appreciated that it is possible even to begin to study the 
numbers of animals and the ways in which they are regulated. 
One of the main facts which emerges from this study is that 
numbers do not usually remain constant for any length of 
time, but usually vary cyclically, sometimes with extraordinary 
regularity. Bound up with various aspects of animal com- 
munities is the question of dispersal of animals, which can be 
shown to be partly connected .with the ordinary activities 
of animals, and partly with changes in numbers. Finally, 
what little we know about the regulation of numbers in 
animals, enables us to say that the problem of the origin of 
species can only be successfully solved by the aid of work on 
numbers. 

The order of study of animals in nature therefore falls 



190 ANIMAL ECOLOGY 

naturally into a series of ecological phases : preliminary survey, 
factors (biotic and otherwise), animal communities, numbers, 
regulation of numbers, dispersal, and (if so inclined) origin of 
species. 

Human ecology and animal ecology have developed in 
curious contrast to one another. Human ecology has been 
concerned almost entirely with biotic factors, with the effects 
of man upon man, disregarding often enough the other animals 
amongst which we live. Owing to the fact that most of the 
workers in this subject are themselves biotic factors, an undue 
prominence has been given in history and economics to these 
purely human influences. It is only recently, under the 
influence of men like Huntington ^^^ and Hill,^"^^ that the 
importance of physical and climatic factors in man's environ- 
ment has become recognised. In animal ecology it has been 
entirely the other way about. Attention has been concentrated 
on the physical and chemical factors affecting animals, and if the 
biotic factors of vegetation and other animals have been studied, 
they have played quite a minor part in ecological work, or have 
been studied from the point of view of evolution, either to 
prove or to disprove the powers of natural selection in producing 
adaptations. As a matter of fact, we are now in a position to 
see that animals live lives which are socially in many ways 
comparable with the community-life of mankind, and if these 
resemblances be only considered as analogies, there yet remains 
the important fact that animal communities are very compli- 
cated and subject to regular rules, and that it is impossible 
to treat any one species as if it were an isolated unit, when 
we are studying its distribution and numbers. 

I am ending this book with a diagram which attempts to 
illustrate in a rough way the relation of the various branches 
of ecology to each other, and to other branches of science. 
The diagram is based upon the order of the chapters in this 
book ; and it may serve as a reminder that ecology is quite 
a large subject. 



CONCLUSION 



191 



Biological Surveys. 

Climatology 
Systematics 



i 



Limiting Factors. Ecological Succession. Animal Communities. 



Physiology 

Psychology 

Chemistry 

Physics 

Geographical distribution 

Geology 



Geology 
Paleeontology 
Climatic changes 
Forestry and agriculture 
Plant succession 

I 

Time Factors; 

Meteorology 
Astronomy 



Regulation of Numbers 

Astronomy 
Economic zoology 
Evolution 
Human economics 



Dispersal. 

Psychology 

Chemistry 

Oceanography 



Social science 
Sex biology 
Psychology 
Physiology of food 
Adaptation 




LIST OF REFERENCES 

The list of references given below is divided into three sections. 
The first comprises books and papers which will be found stimu- 
lating to the imagination and productive of ecological ideas. The 
list is necessarily rather arbitrary, and does not pretend to be 
anything more than suggestive. The second section contains 
references to various works on special subjects mentioned in the 
text, and to the examples which have been used to illustrate ideas 
in this book. Some of these occur also in the general works listed 
in Section i. The third section is devoted to works on the sys- 
tematics and natural history of particular groups of animals, as 
explained on p. i66. 

Section i.— General Works 

1. Adams, C. C. (1913). Guide to the study of animal ecology. New 

York, la : p. 40. lb : p. 82. 

2. Buxton, P. A. (1923). Animal life in deserts. London. 2a : p. 87. 

2b : p. 115. 2c : p. 132. 

3. Carpenter, G. D. Hale (1920). A naturalist on Lake Victoria. 

London. 3a : p. 39. 3b : p. 53. 3c : p. 190. 

4. (191 3). Second report on the bionomics of Glossina fiiscipes 

(palpalis) of Uganda. Reports of the Sleeping Sickness Com- 
mission of the Royal Society No. 14. 4a : p. 14. 4b : p. 37. 

5. (19 1 9). Third report on the bionomics of Glossina palpalis 

on Lake Victoria. Reports (as above) No. 17. 5a : p. 3. 

6. Carr- Saunders, A. M. (1922). The population problem. Oxford. 

7. Darwin, C. (1845). The voyage of the Beagle. London. 

8. Hewitt, C. G. (1921). The conservation of the wild life of Canada. 

New York. 8a: p. 20. 8b: p. 210. 8c : p. 117. 8d : p. 232. 

9. Howard, H. Eliot (1920). Territory in bird life. London. 

10. Humboldt, A. von (1850). Views of nature or contemplations on 

the sublime phenomena of creation. (Translated.) London. 
10a : p. 199. 

11. Longstaff, T. G. (1926). Local changes in distribution. Ibis, 

London. 11a : p. 654. lib : p. 656. 

12. Percival, a. B. (1924). A game ranger's note-book. London. 

12a: pp. 158, 160. 12b: p. 209. 12e : p. 254. 12d : p. 302. 
12e : p. 332. 12f : p. 344. 12g : p. 345. 12h : p. 307. 12k : 
p. 303. 121 : p. 141. 

13. Ritchie, J. (1920). The influence of man on animal life in Scotland. 

Cambridge. 13a: p. 501. 13b: p. 508. 13c: p. 513. 13d: p. 272. 
13e: p. 290. 

Z92 



LIST OF REFERENCES 193 

14. Shelford, V. E. (1913). Animal communities in temperate America ; 

Chicago. 

15. Tansley, a. G. (1923). Practical plant ecology. London. 15a : 

p. 97. 

16. (191 1). Types of British vegetation. Cambridge. 

[7. Thomson, G. M. (1922). The naturalisation of animals and plants 
in New Zealand. Cambridge. 17a: p. 27. 17b: p. 81. 17c : 
p. 88. 17d: p. 154. 

18. Richards, O. W. (1926). Studies on the ecology of English heaths. 

Journ. Ecology, Vol. 14. 18a : p. 246. 18b : p. 264. 18c : 
p. 249. 

19. Farrow, E. P. (1925). Plant life on East Anglian heaths. Cambridge. 

8a : p. 38. 

Section 2. — Special References 

20. Shelford, V. E. (1907). Preliminary note on the distribution of the 

tiger beetles {Cicindela) and its relation to plant succession. Biol. 
Bull. Vol. 14, p. 9. 

21. Alaska, in the Encyclopeedia Britannica, nth edition. 1922. 

22. Elton, C. S. (1925). Thedispersalof insects to Spitsbergen. Trans. 

Ent. Soc. London. August 7th, p. 289. 

23. (1924). Periodic fluctuations in the numbers of animals : their 

causes and effects. British Journal of Experimental Biology, 
Vol. 2, p. 119. 

24. (1925). Plague and the regulation of numbers in wild mammals. 

Journ. Hygiene, Vol. 24, p. 138. 

25. Summerhayes, V. S., and Elton, C. S. (1923). Contributions to the 

ecology of Spitsbergen and Bear Island. Journ. Ecology, Vol. 11, 
p. 214. 25a : p. 265. 25b : p. 268. 

26. SoPER, J. D. (1921). Notes on the snowshoe rabbit. Journ. Mam- 

malogy, Vol. 2, pp. 102, 104. 

27. (1923). The mammals of Wellington and Waterloo Counties, 

Ontario. Journ. Mammalogy, Vol. 4, p. 244. 

28. Sanders, N. J., and Shelford, V. E. (1922). A quantitative and 
seasonal study of a pine dune animal community. Ecology, Vol. 3, 

p. 306. 

29. Richards, O. W., and Robson, G. C. (1926). The species problem 

and evolution. Nature, March 6th and 13th. 

30. Darw^in, C. (1874). On the structure and distribution of coral reefs. 

London. (2nd edition.) p. 20. 

31. Yapp, R. H. (1922). The concept of habitat. Journ. Ecology, 

Vol. 10, p. I. 

32. Levick, G. M. (1914). Antarctic penguins. London. 32a : p. 132. 

32b : p. 135. 

33. Collett, R. (1911-12). Norges Pattedyr. Christiania. 33a: p. 9. 

33b : p. 144. 33c : p. 223. 

34. Wilkins, G. H., quoted by Wild, F. (1923). Shackleton's last voyage. 

Appendix 2, p. 335. 

35. Mass ART, J. (1920 circ). La biologie des inondations de I'Yser. 

Brussels. 

36. Beebe, W. (1924). Galapagos : world's end. 36a : p. 72. 36b : 

p. 92. 36c : p. 122. 36d : p. 222. 36e : p. 321. 36f:p. 287. 

37. Austin, E. E. (1926). The house-fly. British Museum (Natural 

History), Economic Series No. ia. London, p. 41. 

38. UvAROV, B. P. (1923). Quelques problemes de la biologie des saute- 

relles. Ann. des Epiphyties, Vol. 9, p. 87. 

O 



194 ANIMAL ECOLOGY 

89. Clements, F. E. (191 6). Plant succession. Washington. 

40. Grinnell, J., and Storer, T. I. (1924). Animal life in the Yosemite. 

Berkeley, California, p. 22. 

41. Williams, C. B. (1924). Bioclimatic observations in the Egyptian 

Desert in March, 1923. Ministry of Agriculture, Egypt : Technical 
and Scientific Service Bulletin No. 37. Cairo. 

42. MacGregor, M. E. The influence of the hydrogen-ion concentration 

in the development of mosquito larvas. Parasitology, Vol. 13, p. 348. 

43. Rawling, C. G. (1905). The great plateau. London. 43a: p. 316. 

43b: p. 316. 

44. Russell, J, (1923). The micro-organisms of the soil. London. 

44a : Chapter 9, by A. D. Imms. 44b : Chapter 5, by D. W. 
Cutler. 

45. Dugmore, a. R. (1924). The vast Sudan. London. 45a : p. 274. 

45b : p. 281. 

46. Hadwen, S., and Palmer, L. J. (1922). Reindeer in Alaska. U.S. 

Dept. of Agric. Bull. No. 1089. 

47. HiNTON, M. A. C. (191 8). Rats and mice as enemies of mankind. 

British Museum (Natural History) Economic Series, No. 8, p. 45. 

48. HoLiNSHED, quoted by Maxwell, H. E. (1893). The plague of field 

voles in Scotland. The Zoologist, p. 121. 

49. Piper, S. E. (1908). Mouse plagues, their control and prevention. 

U.S. Yearbook of Agriculture, p. 301. 

50. Cronwright-Schreiner, S. C. (1925). The migratory springbucks 

of South Africa. London, p. 75. 

51. Harmer, S. F. (1913). The polyzoa of waterworks. Proc. Zool. 

Soc, p. 426. 

52. Church, A. H. (1919). The plankton-phase and plankton-rate. 

Journal of Botany, June. 

53. LoRTET, L. (1883). Poissons et reptiles du lac Tiberiade, etc. Arch. 

Mus. Hist. Naturelle de Lyon, Vol. 3, p. 106 (quoted in " Nature ")• 

54. Martini, E. (1925). Neues iiber Wanderungen und Wirtswechsel bei 

parasitischen Wurmen. Die Erde, Vol. 3, p. 24. 

55. Rothschild, N. C. (1915). Synopsis of the British Siphonaptera. 

Entom. Monthly Mag., March. 

56. Christy, C. (1924). Big game and pigmies. London, p. 231. 

57. Wallace, A. R. (1913). The Malay Archipelago. London, p. 89. 

58. Wright, W. H. The black bear. London, p. 73. 

59. Stewart, F. H. (1922). Parasitic worms in man. Nature, March 

23rd. 

60. Baty, R. R. DU. 15,000 miles in a ketch. London, p. 96. 

61. Pearson, T. G. (1924). Conservative conservation. The National 

Assoc, of Audubon Societies, Circular No. 9. 

62. Coward, T. A. (1920). The problem of the oak. Lancashire and 

Cheshire Naturalist, Vol. 13, p. 96. 

63. Allee, W. C. (1923). Studies in marine ecology. No. 4. Ecology, 

Vol. 4, p. 341- 

64. Rowan, W. (1925). On the effect of extreme cold on birds. British 

Birds, Vol. 18, p. 296. 

65. Baxter, E. V., and Rintoul, L. J. (1925). Fluctuations in breeding 

birds on the isle of May. Scottish Naturalist, Nov .-Dec, p. 175. 

66. KoFOiD, C. A. (1921). Report of the San Francisco Bay Marine 

Piling Survey ; the biological phase. San Francisco. 

67. Hansen, H. J., and Schiodte, L C. (1878-1892). Zoologica Danica, 

Vol. I. Pattedyr. Copenhagen, p. 88. 

68. Brehm,A. E. (1897). From North Pole to Equator. London, p. 253. 



LIST OF REFERENCES 195 

69. Troupe, R. S. (1921). The silviculture of Indian trees. Oxford. 

Vol. 3, p. 982. 

70. Powell, W. (1925). Rodents : description, habits, and methods of 

destruction. Union of South Africa, Dept. of Public Health. 
Bull. No. 321, p. 12. 

71. BuLSTRODE, H. T. (191 1). Report on human plague in East Suffolk. 

. . . Rept. of Local Govt. Board on Public Health, etc. New Series, 
No. 52. Part I. London. 

72. NiEDiECK, P. (1909). Cruises in the Behring Sea. London, p. 36. 

73. Perkins, R. C. L., and Sw^zy, O. (1924). The introduction into 

Hawaii of insects that attack Lantana. Bull. Exptl. Stn. of Hawaiian 
Sugar Planters' Assoc. Entomological Series, Bull. 16. 

74. Bailey, V. (1922). Beaver habits, beaver control, and possibilities in 

beaver farming. U.S. Dept. of Agric. Bull. 1078. 

75. Elliott, H. W. (1884). Report on the seal islands of Alaska. Wash- 

ington. 

76. Saunders, J. T. (1924). The effect of the hydrogen-ion concentration 

on the behaviour and occurrence of Spirostomum. Proc. Cambridge 
Phil. Soc. Biol. Sci., Vol. i, p. 189. 

77. Vallentin, R., and Boyson, V. F. (1924). The Falkland Islands. 

Oxford. 77a : p. 336. 77b : p. 309. 

78. MacFarlane, R. (1905). Notes on mammals collected and observed 

in the Northern Mackenzie River district. . . , Proc. U.S. Nat. 
Mus., Vol. 28, p. 743. 

79. MiALL, L. C. (1897). Thirty years of teaching. London, p. 105. 

80. Brooks, A. (1926). Past and present big game conditions in British 

Columbia and the predatory mammal question. Journ. Mam- 
malogy, Vol. 7, P- 37. 

81. Fleming, G. (1871). Animal plagues. London. 81a : Vol. i, 

pp. 66, 96, 117, 119, 140, 228, 26^. 81b: Vol. 2, p. 188. 81c: 
Vol. 2, p. 268. Sid : Vol. I, p. ii6~; Vol. 2, p. 180. 81e : Vol. 2, 
p. 243. 

82. Wii-SON, O. T. (1925). Some experimental observations of marine 

algal succession. Ecology, Vol. 6, p. 302. 

83. Hofman, J. V. (1920). The establishment of a Douglas Fir Forest. 

Ecology, Vol. I, p. I. 

84. Cooper, W. S. (1922). The ecological life-history of certain species 

of Ribes and its application to the control of the White Pine Blister 
Rust. Ecology, Vol. 3, p. 7. 

85. Chapman, A. (1921). Savage Sudan. London. 85a : p. 42. 

85b: p. 176. 85c; p. 284. 85d : p. 245. 85e : appendix. 85f : p. 190. 

86. Roosevelt, T. (1914). Through the Brazilian wilderness. London. 

86a : pp. 16-18. 86b : p. 94. 86c ; p. 88. 

87. Donaldson, H. H. (1924). The rat. Philadelphia, pp. 8, 9. 

88. Schweitzer, A. (1923)- On the edge of the primeval forest. London. 

p. 143. 

89. Wenyon, C. M. (1926). Protozoology. London. Vol. i. 89a: 

p. 385. 89b : p. 348. 89c : p. 358. 

90. Pearl, R., and Parker, S. L. (1923). American Journ. Hygiene, 

Vol. 3, p. 94- 

91. Robertson, T. B. Experimental stuides on cellular multiplication : 

I. The multiplication of isolated infusoria. Biochemical Journ. 
Vol. 25, p. 612. 

92. BiRGE, E. A., and Juday, C. (1922). The inland lakes of Wisconsin. 

Wisconsin Geol. and Nat. Hist. Survey, Bull. No. 64. Scient. Ser. 
No. 13, p. 4. 



196 ANIMAL ECOLOGY 

93. Mawson, D. (191 5). The home of the blizzard. London. Vol. z, 

p. 116. 

94. Howard, L. O., and Fisice, W. F. (1911). The importation into the 

United States of the parasites of the gypsy moth and the brown-tail 
moth. U.S. Dept. of Agric. Bureau of Entom. Bull. No. 91. 

95. Decoppet, M. (1920). Le hanneton. Geneva. 

96. Whipple, G. C. The microscopy of drinking water. 

97. Ward, H. B., and Whipple, G. C. (191 8). Fresh- water Biology. 
t psIewYork. 

98. Walker, E. M. (1912). The North American Dragonflies of the genus 

JEshna. Univ. of Toronto Studies, Biol. Ser. No. 11, p. 29. 

99. Druce, G. C. (1922). The Botanist's pocket-book. London. 

100. Seton, E. T. (1920). The arctic prairies. London, p. 109. 

101. (1920). Migrations of the grey squirrel (Sciurus carolinensis), 

Journ. Mammalogy, Vol. i, p. 53. 

102. Hardy, A. C. (1924). The herring in relation to its animate environ- 

ment, part I. Ministry of Agriculture and Fisheries, Fishery 
Investigations, Series 2, Vol. 7, No. 3. 

103. (1926). The Discovery Expedition. Appendix 2 : A new 

method of plankton research. Nature, Oct. loth, 1926. 

104. Baker, J. R. (1925). A coral reef in the New Hebrides. Proc. Zool. 

Soc, p. 1007. 

105. Vershaffelt, E. (1910). Konink. Akad. von Wetenschappen te 

Amsterdam. Proc. Section of Sciences, Vol. 13, p. 536. 

106. CouEGNAS, J. (1920). L'aire de distribution geographique des 

ecrevisses de la region de Sussac (Haute- Vienne) et ses rapports 
avec les donnees geologiques. Arch. Zool. Exp6r. Notes et 
Revue, Vol. 59, No. 3, p. 11. 

107. Wood-Jones, F. (191 2). Corals and Atolls. London. 107a : p. 178. 

107b: p. 241. 107c : p. 205. 107d:p.3o7. 107e:p.3ii. 

108. Lydekker, R. (1903). Mostly mammals. London, p. 67. 

109. Haviland, M, D. (1926). Forest, Steppe and Tundra. Cambridge. 

110. Stresemann, E. (1925). Uber farbungs mutationen bei nichtdome- 

stizierten vogeln. Verhandl. Deutsch. Zool. Ges. Vol. 30, p. 159. 

111. Hesse, T. (1924). Tiergeographie auf okologischer grundlage. 

Jena. 

112. Handbook of Instructions for Collectors. Issued by the British 

Museum (Natural History). London. 1921 (4th edition). 

113. Berg, B. (1925). Mit den Zugvogeln nach Afrika. Berlin. 

114. WiTHERBY, H. F. (1920). A practical handbook of British birds. 

London. 

115. Gerard, C. (1871). Essai d'une faune historique des mammif^res 

sauvages de I'Alsace. Colmar. 

116. Hunter, W. D. (1912). Journ. Econ. Entom. Vol. 5, p. 123. 

117. Casserby, G. (1924). Where do wild elephants die ? Journ. Mam- 

malogy, Vol. 5, p. 113. 

118. Phenological report of the Royal Meteorological Society for 1879. 

In Quart. Journ. R. Meteor. Soc, 1880. 

119. Cabot, W. B. (1912). In Northern Labrador. London. Appendix. 

120. Summerhayes, V. S., and Williams, P. H. (1926). Studies on the 

ecology of English heaths. Part 2. Journ. Ecology, Vol. 16, 
p. 203. 

121. Priestley, R. E. (1914). Antarctic adventure. London, p. 146. 

122. Thorpe, W. H. (1925). Some ecological aspects of British ornithology. 

British Birds, Vol. 19, p. 107. 

123. Olofsson, O. (191 8). Studien iiber die Siisswasserfauna Spitz- 

bergens. Zool. Beitrage aus Upsala, Vol. 6, p. 422. 



LIST OF REFERENCES 197 

124. Crozier, W. J. (1921). " Homing " behaviour in Chiton. Amer. 

Naturalist, Vol. 55, p. 276. 

125. Ritchie, J. (1924). The loggerhead turtle in Scotland. Scottish 

Naturahst, July-August, p. 99. 

126. Thomson, A. L. (1926). Problems of bird migration. London. 

127. Anon. (1922). Trekking monkeys. The Field, February 2Sth, 

p. 248. 

128. BeebEjW. (1926). The Arcturus adventure. New York and London . 

p. 48. 

129. Kaufmann, a. (1900). Cypriden und Darwinuliden der Schweiz. 

Revue Suisse de Zool., Vol. 8, p. 209. 

130. Garbini, a. (1895). Diffusione passiva nella limnofauna. Memor. 

Accad. Verona, Vol. 71, Ser. 2, Fasc. i, p. 21. (Cited in Archiv. 
fiir Naturgeschichte, Vol. 61, part 2, no. 3, p. 181.) 

131. Leege, O. (1911). Die Entomostraken der Insel Memmert. 96th 

Jahresber. d. naturf. Gas. in Emden. (Cited by Wagler, E. (1923). 
Intern. Revue ges. Hydrobiol. und Hydrogr., Vol. 11, p. 45-) 

132. Williams, C. B. (1925). The migrations of the painted lady butterfly. 

Nature, April nth, p. 535. 

133. Pettersson, O. (1912). The connection between hydrographical and 

meteorological phenomena. Quart. Journal Royal Meteorological 
Soc, Vol. 38, p. 173. 

134. Galtsoff, p. S. (1924). Seasonal migrations of mackerel in the 

Black Sea. Ecology, Vol. 5, p. i. 

135. HowLETT, F. M. (1915). Bull. Entom. Research, Vol. 6, p. 297. 

136. Richardson, C. H. (1916). A chemotropic response of the house fly 

(Musca domestica L.). Science (New Series), Vol. 43, p. 613. 

137. Barrows, W. M. (1907). The reactions of the pomace fly (Droso- 

phila ampelophila Loew.) to odorous substances. Journ. Exp. Zool., 
Vol. 4, p. 515. 

138. Tansley, a. G., and Adamson, R. S. (1925). Studies of the vege- 

tation of the English Chalk. Part 3. Journ. Ecology, Vol. 13, 
p. 213. 

139. Manson-Bahr, P. H. (1925). Manson's tropical diseases. London. 

pp. 508, 540. 

140. Kirkman, F. B. (1910). The British bird book. London. Vol. i. 

141. Flattely, F. W., and Walton, C. L. (1922). The biology of the 

seashore. London, p. 309. 

142. MoRLEY, B. (May, 1914). A larva plague in Deffer Wood, Yorks. 

The Naturalist, p. 151. 

143. Wheeler, W. M. (1913). Ants. New York. Chapter 19. 

144. Longstaff, T. G. (1923). The assault on Mount Everest, p. 323. 

145. Hill, L., and Campbell, A. (1925). Health and Environment. 

London. 

146. Huntington, E. (1925). The pulse of Asia. New York and 

London. 
(1926). Civilization and climate. New Haven. 

147. Nicholson, E. M. (1927). How birds live. London. Appendix. 

148. Sumner, F. B. (1925). Some biological problems of our south- 

western deserts. Biology, Vol. 6, p. 361. 

149. Bell, W. G. (1924). The great plague of London in 1665. London. 

150. Scott, R. F., in Scott's last expedition. London, 1923, p. 74. 

151. Scott, R. F. (1907). The voyage of the Discovery. London. 

Vol. 2, p. 224. 

152. Williams, C. B. Memoir No. i of Dept. of Agriculture of Trinidad. 



198 ANIMAL ECOLOGY 

Section 3. — Works on particular groups of British Animals 

The system upon which the works Hsted below have been 
chosen is explained in Chapter IL When it is stated that no suit- 
able work exists on any particular group, this does not imply that 
no good systematic work has been published on that group, only 
that it is not to be found in EngHsh, or that it is inaccessible, or in 
a very scattered form, or else not sufficiently up to date to be very 
reliable. In almost all cases, except very recent publications, 
there has been published a good deal of additional work in various 
periodicals, and often much work has been done but has not yet 
been published by the systematists concerned. The list is neither 
logical nor complete, but in spite of this, should be found useful 
to working ecologists. 

Mammals 

Barrett-Hamilton, G. E. H., and Hinton, M. A. C. (1910-1921). A 
history of British Mammals. London. 

Birds 

WiTHERBY, H. F., and others (1920). A practical handbook of British 

Birds. London. 
Howard, H. E. (1907-1914). The British Warblers. London. 

Reptiles 

Leighton, G. R. (1901). The life-history of British Serpents. Edinburgh- 
(1903). The life-history of British Lizards. Edinburgh. 

Fish 
Maxwell, H. (1920 circ). British freshwater Fishes. London. 

Molluscs 

Ellis, A. E. (1926). British Snails. Oxford. 

This book covers the gasteropods of land and fresh water. There 
is no convenient and accurate work on the freshwater lamellibranchs, 
and the systematic position of a good many of the species is still in 
dispute. 

Crustacea 

Webb, W. M., and Sillem, C. (1906). The British Woodlice. London. 

There is no comprehensive work upon the freshwater Crustacea 
of the British Isles, but much information can be obtained from Ward 
and Whipple's " Freshwater Biology," ®' which although dealing with 
American forms, is most useful, since many of the genera, and in the 
case of smaller forms, species, are the same in Europe and America. 
The following other works may be mentioned : 

Tattersall, W. (1920). The occurrence of Asellus meridianus Rac. in 
Derbyshire. Lanes, and Cheshire Naturalist, May, p. 273 (contains 
key to species of Asellus). 

Bars, G. O. (1895). An account of the Crustacea of Norway, Christiania 
and Copenhagen. Vol. i. (Includes an account of the genus 
Gammarus^ 



LIST OF REFERENCES 199 

Watermites 

Soar, C. D., and Williamson, W. (1925). The British Hydracarina. 
London. (Ray Society Publications.) 
Only the first volume is out at present. 
There are no convenient works on land mites, spiders, or myriopods. 

Insects 

Imms, a. D. (1925). A general textbook of entomology. London. 
An excellent work for general information and reference. 

Coleoptera. 

Fowler, W. W. (1887-1914). The Coleoptera of the British Islands. 

London. 

Six volumes. There are a number of later additions in various 

periodicals. 

Hemiptera. 

Saunders, E. (1892). The Hemiptera-Heteroptera of the British Isles. 
London. 

Primarily a systematic work. 
Butler, E. A. (1923). A biology of the British Hemiptera-Heteroptera. 
London. 

This is biological rather than systematic, but brings up to date the 
systematic work which was done after the publication of Saunders' 
book. 
Edwards, J. (1896). The Hemiptera-Homoptera (Cicadina and Psyllina) 

of the British Isles. London. 
Davidson, J. (1925). A list of British Aphides. London. 

Gives all references to the food plants of British species. 
Chrystal, R. N., and Storey, F. (1922). The Douglas Fir Chermes 
(Chermes cooleyi). Forestry Commission Bulletin No. 4. London. 
Gives keys to all British species of Chermes. 
Newstead, R. (1901). A monograph of the Coccidae of the British Isles. 
London. (Ray Society Publications.) 

Hymenoptera. 

Donisthorpe, H. St. J. K. (1927). British Ants. London. 

Sladen, F. W. L. (1912). The humble bee and how to domesticate it. 

London. 
Saunders, E. (1896). The Hymenoptera-Aculeata of the British Isles. 

London. 
Stellwaag, F. (1921). Die Schmarotzer-wespen als Parasiten. Mongr. Z, 

Angew. Entomologie, No. 6. 
SwANTON, E. W. (1912). British Plant-Galls. London. 

This deals mostly with the Cynipida, but also includes galls formed 

by mites and flies. 
Morice, F. D. Help-notes towards the determination of the British 

Tenthredinids, etc. A series of papers on sawflies scattered through 

the Entomologist's Monthly Magazine, from 1903 onwards. 
Morley, C. (1903). The Ichneumons of Great Britain. Plymouth. 

For looking up biological facts about ichneumons : it is inadvisable 

to attempt identification of these insects, since it is a difficult and 

tricky process. For this reason, no references are given in this list 

to the other parasitic hymenoptera. 



200 ANIMAL ECOLOGY 

Diptera. 

Verrall, G. H. (190 1, 1909). British Flies. London. Two volumes. 

This book covers a number of groups, but does not deal with many 

others, which have not been conveniently summed up in accessible 

publications. The following other works may be mentioned: 
WiNGATE, W. J. (1906). A preliminary list of Durham Diptera with 

analytical tables. Trans. Nat. Hist. Soc. of Northumberland, Durham, 

and Newcastle-on-Tyne. New Series, Vol. 2, p. i. 

This contains valuable tables for identifying many flies. 
LiNDER, E. (edited by) (1924). Die Fliegen der palaearctischen Region. 

Stuttgart. Still coming out in parts. 
Lang, W. D. (1920). A handbook of British Mosquitoes. London. 
Baer, W. (1921). DieTachinen als Schmarotzer der schadlichen Insekten. 

Berlin. 
Edwards, F. W. (1926). On the British biting Midges (Diptera, Cerato- 

pogonidas). Trans. Ent. Soc. London. Vol. 74, p. 389. 

Orthoptera. 

Lucas, W. J. (1920). A monograph of the British Orthoptera. London. 
(Ray Society Publications.) 

Odonata. 

Lucas, W. J. (1900). British Dragonflies (Odonata). London. 
Lepidoptera. 

Meyrick, E. (1895). A handbook of British Lepidoptera. London. 
Trichoptera. 

Lachlan, R. (1874-1880). A monographic revision and synopsis of the 
Trichoptera of the European fauna. London. 

Ephemeroptera. 

Eaton, H. E. (1883). A revisional monograph of recent Ephemeridae or 
Mayflies. Trans. Linn. Soc. Ser. 2. Zool. Vol. 3. 

Neuroptera. 

WiTHYCOMBE, C. L. (1922). The biology of the British Neuroptera 
Planipennia. Trans. Ent. Soc. London. 

Larvce of aquatic insects. 

Rousseau, E. (1921). Les larvae et nymphes aquatiques des insectes 
d'Europe. Brussels. 

Volume I deals with Hemiptera, Odonata, Ephemoroptera, Ple- 
coptera, Neuroptera, and Trichoptera. The second volume, not yet 
pubhshed, will deal with Lepidoptera and Coleoptera. 

There are no suitable works of reference on British Collembola, 
Thysanura, Protura, Plecoptera, or Psocoptera. 

Leeches 
Harding, W. A. (1910). A revision of the British Leeches. Parasitology, 
Vol. 3, p. 130. 

Flatworms 
No good reference works except a paper on free-living land flatworms : 
Percival, E. (1925). Rhytichodemtis britannicus, n. sp., a new British triclad. 
Quart. Journ. Micr. Science, Vol. 69 (new series), p. 343. 



INDEX 



Niimhers in ordinary type refer to pages 
list of referetices. 



those in italic refer to the 



AcanthQmyops flavus, 50 

Accipiter atricapillus , 123 

Adams, 3, i 

Adamson, 138 

Adenota leticotis, 183 

JEshna, 184 

Agouti, 66 

Agromyzid fly, 55 

AIgs, calcareous, 14, 29 

Allee, 129, 6 j 

Amblyrhynchus, 67 

American Bureau of Biological 

Survey, 6 
Anacharis canadensis, 112 
Anas creccay 24 
Anax guttatus, 157 
Animals, influence of, on plants, 16, 

22,30, 31, 51 

Ann AND ALE, 6 

Anopheles btfurcatus, 46 

— maculipennis, 46 

— plumbeus, 46 
Ant, driver, 63 

Ants, 50, 51, 52, 125, 126 

Anurida maritima, 92 

Aphids, 63, 66, 96, 126, 128, 129, 

157, 158 
Apode7nus sylvaticus, 12, 13, 43 
Arbacia punctulata, 129 
Army worm, 54, 55 
Aru blackbird, 67 
Asio flammeus, 119 
Aspen, 48 
Austin, 47, 57 

Bacteria, 78 
Bacteriophage, 78 
Bailey, 74 
Baker, 15, 104 
BalanuSy 24, 29 



Bamboo, flowering, 132 
— rats, 132 
Bandar, 152 
Barnacles, 24 
Barrows, 160, jj 7 
Bats, 88 
Baty, du, 60 
Baxter, 131, 65 
Bear, black, 90 
— , Kamskatkan, 96 
— , polar, 74, 121 
Bear Island, 58 
Beaver, 47, 138, 139 
Beebe, 126, 154, 36, I2y 
Bee-eater, 65, 122 
Bell, i4g 
Benthos, 13 
Berg, 104, 113 
Berlenga Island, 114 
Bird migration, 152, 160 
Birds and cold winters, 131 
Birds as dispersal agents, 155 
Birge, 70, 174, 93 
Bison, American, 105, 106 
Blackberry, iii 
Boar, wild, 140 
Bombus lapidarins, 51 
Bos caffeTy 145 
BoYSON, yy 
Brehm, 6S 
.Brooks, 140, 80 
Brown, 19 

Buffalo, African, 67, 145 
— , Burmese, 46 
Bulstrode, yi 
Buteo galapagoensiSy 184 
Butorides sundevalli, 126 
Butterflies, 109, 128 
Buxton, 88, 2 
Buzzard, 124 



202 



INDEX 



Cabot, 139, iig 

Cacergates leucosticta, 121 

Cactus, giant, 48 

Calcium carbonate in water, 42, 45 

Calluna, 26, 30 

Capercaillie, 11, 148 

Cardiosoma, 67 

Car ex aretiaria, 30 

Carpenter, 60, 88, 95, 115, 3, 4, 5 

Carr- Saunders, 6 

Casserby, iiy 

Castor fiber, 47, 138 

Cat, wild, 46 

Cats, 114 

Censuses, 173 et seq. 

Centurus uropygialis, 48 

Ceratophyllus sciurorum, 77 

— walkeriy 77 
Cereus giganteus, 48 

Chapman, 65, 67, 103, 145, 184, 8s 

Characteristic species, 12 

Chemotropism, 39, 160 

Chiffchaff, 40 

Chionis alba, 81 

Chiton tuberculatus , 152 

Cholera, 140 

Christy, 56 

Chrysops dimidiatus, 88 

— silaria, 88 
Church, 52 

Chydorus sphcericus, 155 
Cicindela, 25, 26 
Citellus tereticaudus, 46 
Cladocera, no 
Claussen, 133 
Climatic cycles, 130, 137 
Climax association, 22 
Clouded yellow, 100, 149 
Cob, white-eared, 183 
Coccosphaere, no 
Cockchafers, 129 

Colaptes chrysoides mearmi, 48 

Colzas edusa, 149 

Collembola, 63, 107, no 

CoLLETT, 125, 33 

Colour dimorphism, 182 et seq. 

Colours of animals, 182 et seq. 

Columbella, 129 

Competition between animals, 27 

Cooper, 31, 84 

Copepods, 66 

Corals, 14, 29, 68, 93 

Corvus corax, 97 

— frugilegus, 144 
Cotton worm, 128 
Cougar, 140 
Couegnas, 106 



Coward, 128, 6 j 
Crab, scarlet rock, 67 
Crayfish, 41, 42 
Crithidia hyalomma, 78 
Crocodile-bird, 74 
Cronwright-Schreiner, 50 
Crossbill, 100, 150 
Crozier, 152, 124 
Curlew, black, 60 
Cuttlefish, 61 
Cyclones, 130, 132 
Cyclocypris Icsvis, 154 
Cyclops, 69, 97 

— albidus, 97 

— fuscus, 97 

— serratulus, 97 

— strenuus, 76, 97 

— viridis, 97 
Cynictis pencillata, 83 



Daphnia, 69 

— pulex, 154, 166 
Dartford warbler, 124 
Darwin, i, 3, 68, 180, 7, 30 
Decoppet, g5 

Deer, 67, 115, 140, 141 

— , barking, 46 

Depressions, 93 

Development of plant communities, 

20 
Diaptomiis gracilis, 37, 38, 76 
Dilachnus picece, 157 
Diphyllobothrium latum, 76 
Dipodomys deserti, 46 

— merriajni, 120 
Dog, hunting, 63 
Dogfish, 57 

Dominance among marine animals, 
13, 14, 15, 29 

plants, 9 

Donaldson, 8y 

Dor-beetle, 51 

Dormouse, 77 

Douglas fir, 30 

Dragonflies, 121, 122, 157 

Driftwood as dispersal agent, 154 

Drosophila, 116, 160 

Droughts in England, 99, 132, 186 

Druce, 170, gg 

Dugmore, 145, 183, 45 



Earthworms, 54, 67, 107 
Ecesis, 150 
Echinorhynchus, 77 
Ecological survey methods, 168 



INDEX 



203 



Ecology, neglect of, i, 2, 3, etc. 

— , size of, 190, 191 

Ectocarpiis, 29 

Egret, buff-backed, 67 

Elephant, 46, 67, 130 

Elodea, 112 

Elliott, y^ 

Elton, 58, 134, 22, 23, 24, 25 

Erignathus barbatus, 121 

Ermine, 139 

Euphorbia, 77 

Eurytemora lacinulata, 36, 37, 38, 

41,42 
— rabotiy 38, 39 
Evotomys glareolus, 1 2 
Exclusive species, 11 



Falco sparverius, 48 

Farrow, 30, 52, 19 

Filaria, 73, 87, 91 

— bancrofti, 87, 88 

Finlaya ge?iiciilata, 45, 46 

Fish, chromid, 74 

Fisher, 139 

FiSKE, 94 

Fissiirella, 152 

Flattely, 141 

Fleas, 52,73,77, 78, 125 

Fleming, 141, 5 j 

Flies, blood-sucking, 73 

Fly, horse-, 73 

— , house-, 47 

— , hover-, 157 

Flycatcher, 48 

Food, methods of investigating, 172 

Food-chains, 55, 56 

Food-cycle, 55, 56 

Formica rufa, 52 

Fox, 139, 75 

— , arctic, 65, 74, 134, 139, 182 

— , red, 123, 136, 139 

Francolinus vulgaris, 46 

Frog, 155 

— , tree, 88 

Froghopper, sugar-cane, 128 

Fur-seal, Alas'kan, 47, 106 



Galtsoff, 134 
Gammarus pulex, 40, 186 
Gannet, 184 
Garbini, 154, 130 
Geotrupes, 51 
Gerard, 115 
Gerbille, 83, 135 
Gipsy moth, iii 



Glochidia, 153 

Glossina palpalis , 60, 88, 95, 121 

Gonyaulax, no 

Goshawk, 123 

Gossamer, 159 

Grapsus grapsus, 1 26 

Grebe, crested, 74 

Grinnell, II, 174, 40 

Grouse, red, 11, 23, 96 

— disease, 126 

Guillemots, 65, 104 

Gull, black-headed. 24 



Habitats, sharp distinctions between, 

10 
Had WEN, 46 
Hcematopus qiioyi, 60 
Hamadryad, 67 
Hamster, 135 
Hansen, 6y 

Hanuman monkey, 152 
Hardy, 57, 58, 174, J02, 103 
Hares, 66, 135 
Hare, varying, 123, 133, i35 
Harmer, 57 
Hartebeeste, 67 
Harvie- Brown, 149 
Haviland, 17, log 
Hawk, Galapagos, 184 
Hawks, 139 

Hay infusions, succession in, 25 
Heat, 46, 47, 153, 160 
Henry, 105 
Heron, 126 
Herring, 57, 58, i59 
Hesse, 17, iii 
Hewitt, 136, 5 
Hill, 190, 145 
Hinton, 47 
Hippobosca equina, 73 
Hippopotamus, 88 
HoFMAN, 30, S3 
Holinshed, 108, 45 
Holothurians, 68 
Hookworm, 73 
Horse-fly, 73 
^ouse-fiy, 160 
Hover-fly, 157 
Howard, Eliot, 69, 124, 9 
Howard, L. O., 94 
Howlett, 160, 135 
Hudson, 7 
Hunter, 116 
Hunting dog, 63 
Humboldt, von, 91, 10 
Huntington, 190, 146 



204 



INDEX 



Hyzena, spotted, 65 
Hyalommay 78 
Hybernia defoliaria, 128 
Hydrogen-ion concentration, 14, 15, 

45 
Hydroids, 14, 148 
Hymenolepis , 76 
Hymenoptera, parasitic, 78, 79 
Hyrax, 66 



Ice, as dispersal agent, 153 

Ice-age, 151 

Identification of animals, i, 165, 171 

— of plants, 170 

Illiteracy, advantages of, 7 



"Java, II 

JuDAY, 70, 174, g2 

Junco hy emails, 131 



Kangeroo-rat, desert, 46 
Kaufmann, 154, i2g 
Key-industry animals, 57 
King-snake, 67 
KiRKMAN, 140 
Kites, 67 
Kittiwake, 62 
KoFOiD, 66 
Kudu, geater, 145 



Lacewings, 66 

Lacuna, 129 

Ladybirds, 63, 66 

Lagopus mutus, 1 1 

— scoticus, 1 1 , 23 

Lantana camara, 54, 55 

Larus ridibundus, 24 

Leege, 154, 131 

Lemmings, 123, 125, 132 ei^e^., 156, 

186 
Lenmius lemmus, 132 
Leptomonas, 78 
Leptyphantes sobrius, 44 
Lepus americanuSy 123, 133, 135, 136, 

137 
Levick, 168, 32 
Life zones, 11 
Light, 8, 9, 10, 40, 43 
Limax cinereo-niger, 94 
Limncea truncatida, 54 
Limpets, 60 

LlNN^US, I 

Lion, African, 29, 63, 65, 70, 125, 184 



Lion, mountain, 140 

Liver-rot, 53, 54, 140 

Loa loa, 87 

Locust, 87, 109, 148, 156, 158, 159 

Loggerhead turtle, 149 

LONGSTAFF, II, I44 

Lophortyx gambeli, 46 
LoRTET, 74, 53 
Louse, 73 
Lunar cycles, 159 
Lungfish, 60 
Lydekker, 108 
Lymantria dispar, 11 1 
Lynx, 136, 139 



MacFarlane, 139, y8 
MacGregor, 45, 42 
Mackerel, 160 
Malaria, 53, 54, no 
Man, eating all sizes of food, 61 
Manson-Bahr, i^g 
Marmot, 137 
— , ground, 97 
Marten, pine, 139 
Martini, 54 
Massart, 24, 25, 35 
Mawson, 70, 105, 9 J 
Megalestris, 70 

— MacCorrnicki, 97 
Megastigmus spermotropJms, 30 
Melittophagus, 122 
Melolontha hippocastani, 129 

— vulgaris, 1 29 
Merops, 122 
Mesophytic, 22 
MlALL, I, 79 

Mice, 108, 123, 135, 137, 139 
Micropallus whitneyi, 48 
Microtus, 108 

— agrestis, 1 2 
Migration of birds, 96 

— of fish, 152 
Mink, 139 
Mole, 46, 139 
Molge cristata, 115 

— vulgaris, 115 
Mongoose, yellow, 83 
Monkeys, 46 
Mosquitoes, 45, 54, 88 
Mottled umber, 128 
Mountains, zonation on, 9, 11 
Mouse, desert, 120 

— , grass, 12 

— , long-tailed wood, 12, 13 
Mouse-deer, 66 
Mouse-hare, 135 



INDEX 



205 



Musca domestica, 47 
Muskrat, 137 
Mussarama, 67 
Mussel, fresh- water, 153 
— , marine, 24 
Myiarchus cinerascens , 48 
Mynah, 54, 55 
Myrmica scabrinodis , 51 
Mytilus, 24, 29, 45 



Natural history societies, i, 2 
— selection, 41 et seq. 
Nematodes, 81 
Neuroterus lenticularis , 128 
New Hebrides, coral reef in, 15 
Newt, crested, 115, 153 
— , smooth, 115 
Niches, 63 
Nicholson, 174, 147 

NiEDIECK, 96, y2 



Oak moths, 119, 128 

Obelia, 29 

Olofsson, 123 

Opossum, 112, 113 

Optimum density of numbers, 113 

Ostrich, 65 

Otus asio gilmani, 48 

Owl, elf, 48 

— , screech, 48 

— , short-eared, 119 

— , snowy, 123 

Owls, 139 

Ox-bow, 151 

Oxford water works, 37 

Oxshott Common, 22, 66 



Paddy, 81 
Painted lady, 156 
Palcemonetes varians, 25 
Palmer, 46 
Panther, 46 
Pantula flavescens , 157 
Paramecium, 25, 39 
Parasitoid, 79 
Parker, go 

Parthenogenesis, 158, 167 
Partridge, black, 46 
— , common, 11 
Passenger pigeon, 105 
Patella cenea, 60 
Pearl, 90 
Pearson, 61 
Pempelia subornitella, 51 



Penguin, Adelie, 70, 105 

— , emperor, 123, 153 

— , gen too, 81 

Percival, 56, 67, 70, 105, 141, 14s, 

12 
Perdix perdix, 11 
Peridinian, no 
Perkins, ys 
Pettersson, 159, 133 
Phasianus colchiais, 1 1 
Pheasant, 11 
Phenology, 175 
Phylloscopus collybita, 40 

— trochilus, 40 

Physiology, great knowledge of un- 
necessary, 39 
Phytomonas davidi, 77 
Picea obovata, 157 
Pieris brassicce, 39, 115 

— rapce, 39, nS 
Pig, 46 

Pigeons, homing, 160 

Piper, 49 

Plague, bubonic, 52, S3> 78, 84 

Plagues of animals, 141 et seq. 

Plankton, 13, 58, 70, 90, 104 

Plants, influence of on animals, 8, 

9, 32, 47, 48 
Pleuroxus aduncus, 154 
Plumatella, 154 
Podiceps cristatus, 74 
Polychaetes, 129 
Polj^zoa, 154 

Pontosphcera Huxley i, no 
Populus tre^nuloides, 48 
Powell, yo 
Prawns, 25 
Priestley, 121 
Protozoa in soil, 129, 130 
Pseudotsuga taxifolta, 30 
Ptarmigan, 11 
Pygoscelis papua, 81 
Pyramid of numbers, 68 



Quail, desert, 46 
Quercus robur, 12 

Rabbit, 51, 52, 66, 75, 89, iii, 135, 

139 
— ,jack, 135 
— , snowshoe, 186 
Rain, 94 

Rats, 52, 53, III, 114 
Rattus norvegicus, 53 
— rattus, S3 



206 



INDEX 



Raven, 97 

Rawling, 43 

Reindeer, 106 

Rhesus monkey, 152 

Rhinoceros, 56 

Ribes, 31 

Richards, 13, 66, 79, 177, 184, iS, 

29 
Richardson, 160, 136 
Rinderpest, 145 
RiNTOUL, 131, 65 
Ritchie, 23, 27, 148, i3y 125 
Robertson, gi 
ROBSON, 184, 29 
Rock crab, scarlet, 126 
Rook, 144 
Roosevelt, 183, 56 
Rothamsted soil census, 107, 130 
Rothschild, 55 
Rowan, 131, 6-^ 
Russell, 44 
Rust, white pine blister, 31 



Sable, 139 

Salinity, 14, 15, 36, 38, 39, 159 

Sanders, 90, 23 

Sandgrouse, 100, 148, 150, 156 

Sand-martin, 65 

Saprolegnia, 46 

Sarcophaga, 160 

Saunders, 45, 76 

Schiodte, 67 

Schweitzer, 63, 88 

Scorpions, 152 

Scott, 130, 151 

Seal, Alaskan fur, 47, 106 

— , great bearded, 121 

— , Weddell, 97 

Seaweed as dispersal agent, 153 

Sere, 22 

Seton, 137, 100, lOI 

Sheep, 67, 97, III, 112 

— , Tibetan, 143 

ShELFORD, 25, 27, 90, 14, 20y 28 

Ships as dispersal agents, 154 

Ship worm, 132 

Shrew, 139, 140 

— , common, 115 

—, pygmy, 115 

Silpha quadripunctata, 43 

Simocephalus exspinosus, 154 

Size of animals, 178 

— of food, 59 

Skua, 123 

— , antarctic, 70, 97 

— , arctic, 62, 74, 100 



Skunk, 130 

Sleeping sickness, 88, 126 

Slugs, 90, 94 

Snakes eating snakes, 66, 67 

Soil, numbers of animals in, 107 

Spangle gall, 128 

Sparrow, house, 6 

Sparrow-hawk, 48 

SoPER, 136, 26, 2y 

Sorex araneiis, 115 

— minutus, 115 

Species sense, cultivation of, 164 

Sphcerium, 155 

Spiders, 148, 159 

Spirostomum ambiguum, 45 

Springbok, 66, 109 

Spruce, 157 

Squirrel, 77, 139, i49 

— , grey, 28, 135, 184 

— , red, 28, 135 

— , round-tailed ground, 46 

Starling, 67, iii 

Stenocephalus agilis, 77 

Stenus, 63 

Stewart, 59 

Stoat, 62 

Storer, 11, 174, 40 

Strepsiceros bea, 145 

Stresemann, 184, no 

Sula piscatrix zvebsteri, 184 

Summerhayes, 58, 22, 25, 120 

Sumner, 148 

Sunspot cycle, 130 

Suricat, 82 

Suslik, 97 

Swans, 112 

SWEZY, 7J 

Syrphid flies, 66 
Syrphus ribesii, 157 



Tabanid flies, 73, 88 
Tansley, 3, 23, 32, i5y 16, 138 
Tapeworm, 73, 75, 76, 125, 126 
Taterona lobengula, 83 
Teal, 24 

Teredo navalis, 132 
Territory, 69, 70, 124 
Tetragnatha laboriosa, 90 
Tetrao urogallus, 11, 148 
Thalassochelys caretta, 149 
Theridium spirale, 90 
Thomson, A. L., 152, 126 
Thomson, G. M., ly 
Thorpe, 12, 122 
Thyme, wild, 51 
Thymus serpylliim, 51 



INDEX 



207 



Tiberias, Lake, 74 
Tick-bird, African, 67 
Ticks, 68, 78 
Tiger, 46, 125 
Toads, 155 

Tortoise, Galapagos, 106 
Tortrix viridana, 119 
Tragelaph, 88 
Tramea rosenbergii, 157 
Trichosurus vulpeculuy 112 
Tristan da Cunha, 114 
Tropisms, proper place of 

biology, 161 
Troupe, 6g 

Tsetse fly, 60, 88, 95, 116, 121, 
Tubular ia, 129 
Turtle, loggerhead, 149 
Turtle-dove, Chinese, 54 
Typhochrestus spetshergensis ^ 44 



Ungulates, variations in numbers of. 



140 

UVAROV, 85, j5 



Vallentin, 60, yy 
Verschaffelt, 105 
Vespa germanica, 1 28 
— vulgaris, 128 
Viscacha, 66 
Vole, bank, 12 
Vulpes lagopuSy 182 



Wagtails, 67 

Walker, g8 
Wallaby, 66 
— , black-tailed, 112 
Wallace, ii, 180, 57 
Walrus, 106 



Walton, 141 

Warblers, 139 

Ward, 171, 97 

Wasps, 128 

Water-bloom, 112 

Water-fleas, no 

Water-mites, 154 

Water supply, 46 

Weasels, 139 

Wenyon, 8y 

Whale, 105 

— , killer, 61, 62 

— , whalebone, 59, 61 

Wheeler, 143 

Whipple, 171, 174, 96, gy 

White Nile, birds of, 103 

— butterflies, 39 

WiLKINS, 81 

Williams, C. B., 22, 88, 156, 41 

132, 152 
— , P. H., 120 
Willow wren, 40, 124 
Wilson, 82 

Winters, periodic bad, 131 
Witherby, 114 
Wolf, 136 

— , Tibetan, 62, 184 
Wolverene, 139 
Wood-Jones, 132, 157, 107 
Woodpecker, desert, 48 
— , green, 51, 52 
WoRMius, Olaus, 133 
Wright, 58 



Yapp, 23, 31 
Yosemite, fauna of, 1 1 
Yser, floods in, 24 



Zebra, 63, 65, 70, 103, 105, 141 
Ziegler, 133 



printeli in tSrcat ffiri'afn 

BY 

WILLIAM CLOWES & SONS, LIMITED. 
LONDON AND BECCLES.