Essentials of Animal Behaviour

Essentials of Animal Behaviour is an introduction to the study of animal behaviour and

is primarily intended for first or second year undergraduates attending short courses
in the subject. The book concentrates on putting across the basic principles as
briefly and lucidly as possible with the aid of carefully selected examples from both
the recent and classic literature, together with numerous illustrations. It will enthuse
readers with this active and exciting area of research, and will lay a solid foundation
on which further study may be based. Its simple and readable style, helped by an
extensive glossary, will also make it useful to senior level school students, their
teachers and those with a general interest in the subject. It will be particularly
rewarding for all those needing the basics in animal behaviour, behavioural ecology
and comparative psychology.
The Institute of Biology aims to advance both the science and practice of biology.
Besides providing the general editors for this series, the Institute publishes two
journals Biologist and the Journal of Biological Education, conducts examinations,
arranges national and local meetings and represents the views of its members to
government and other bodies. The emphasis of the Studies in Biology will be on
subjects covering major parts of first-year undergraduate courses. We will be
publishing new editions of the ‘bestsellers’ as well as publishing additional new titles.

Titles available in this series

An Introduction to Genetic Engineering, D. S. T. Nicholl
Introductory Photosynthesis, 5th edition, D. O. Hall and K. K. Rao
Introductory Microbiology, J. Heritage, E. G. V. Evans and R. A. Killington
Biotechnology, 3rd edition, J. E. Smith
An Introduction to Parasitology, Bernard E. Matthews
Essentials of Animal Behaviour

P. J. B. Slater
Kennedy Professor of Natural History
University of St Andrews
          
The Pitt Building, Trumpington Street, Cambridge CB2 1RP, United Kingdom

   

The Edinburgh Building, Cambridge CB2 2RU, UK http://www.cup.cam.ac.uk
40 West 20th Street, New York, NY 10011-4211, USA http://www.cup.org
10 Stamford Road, Oakleigh, Melbourne 3166, Australia

© Cambridge University Press 1999

This book is in copyright. Subject to statutory exception

and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without
the written permission of Cambridge University Press.

First published 1999

Printed in the United Kingdom at the University Press, Cambridge

Typeset in Monotype Garamond 11/13 in QuarkXPress™ []

A catalogue record for this book is available from the British Library

Library of Congress Cataloguing in Publication data

Slater, P. J. B. (Peter James Bramwell), 1942–
Essentials of animal behaviour / P. J. B. Slater.
p. cm. – (Studies in biology)
Includes bibliographical references and index.
ISBN 0 521 62004 X (hb). – ISBN 0 521 62996 9 (pbk.)
1. Animal behavior. I. Title. II. Series.
QL751.S616 1999
591.5–dc21 98–34540 CIP

ISBN 0 521 62004 X hardback

ISBN 0 521 62996 9 paperback
Contents

Preface page ix

1 Asking questions about behaviour 1

2 Motor patterns 20

3 Sensory systems 39

4 Motivation 57

5 Development 84

6 Evolution 112

7 Function 139

8 Communication 161

9 Social organisation 186

Suggested further reading 208

Glossary 211

List of species names 223

Index 227
Preface

The aim of this book is to provide a brief, but reasonably comprehensive,

introduction to the study of animal behaviour. It has grown out of my earlier
book, An Introduction to Ethology, which was published 12 years ago and has
been out of print for some time. Twelve years is a long time in an active field
of science, and a lot of interesting things have happened during that period,
so the book needed a good going over and sprucing up. But fashions change
in science, as in all else, so some parts of the book, describing fields that have
seen little recent work, are much as they were before, while I have written new
sections and modified the text extensively in those areas in which there has
been a lot of recent interest. One obvious change is in the title: the word
‘ethology’ to describe animal behaviour, and ‘ethologist’ for someone who
studies it, while admirably concise, have rather fallen from use. Unfortunately
they have tended to be tied in people’s minds to the particular school of study
and theories that emanated from Europe in the middle years of this century.
Many of these theories have not been supported by later work, and the word
ethology has tended to sink with them. It probably sank somewhere in the
middle of the Atlantic, as it never really made it to the far side anyway!
The theories of the ethologists were clear and simple and provided a
magnificent body of hypotheses on which subsequent work could be based.
For this reason, the approach I shall take here is a somewhat historical one,
looking at the ideas of the early ethologists, notably Konrad Lorenz and Niko
Tinbergen, and then surveying developments over the past few decades to
identify subsequent work which has been influential in leading to what we now
think about animal behaviour. I have tried to choose some of the best exam-
ples, from among both classic studies in the field and the great range of recent
x Preface

work, to illustrate my themes as clearly as possible for those with no previous

knowledge of the subject. These have the excitement of approaching this fas-
cinating subject for the first time; if I can share some of the thrill I felt when
I did so, this book will have succeeded.
I remain grateful to those who read sections of my earlier book. Three ref-
erees made very useful comments and suggestions about what to change and
what to leave as it was, and the ideas of various colleagues have helped me
with parts of the book, while Jeff Graves read the glossary and Vincent Janik
made helpful suggestions on the whole text. It is a particular pleasure to thank
Nigel Mann for his delightful illustrations, as well as Jan Parr for some that
have been carried over from my previous book, and Tracey Sanderson and
Jane Bulleid for seeing the book speedily through the presses.

P. J. B. S. St Andrews, 1998
1
Asking questions about
behaviour

People have been fascinated by the behaviour of animals for a long time. Their
interest was caught both by the eye-catching activities that they could see
around them in the natural world and by the need to understand and control
the behaviour of their own domestic animals. Questions naturally arose in
their minds. Aristotle, for example, wondered where swallows went in the
winter and, seeing them gathering in reedbeds, he speculated (as he did about
a lot of things!) that they hibernated in the mud at the bottom of ponds.
But the scientific study of behaviour is a recent phenomenon and, as with
so much else in biology, it received its most important boost from the writ-
ings of Charles Darwin. Darwin included a chapter in The Origin of Species on
‘Instinct’, a term used in his time to refer to the natural behaviour of animals.
He also wrote a book specifically about behaviour called The Expression of the
Emotions in Man and in Animals. Despite this, in the half century after Darwin
there was little work on the behaviour of animals, while zoologists grappled
with trying to understand the fundamental principles of systematics, physiol-
ogy and developmental biology. A few scientists from that time, like Julian
Huxley in Britain, Oscar Heinroth in Germany and Charles Otis Whitman in
America, stand out for their contribution to behaviour, but they were a small
band. It was only in the 1930s that a comprehensive theory of animal behav-
iour began to emerge through the writings of Konrad Lorenz and, later, of
Niko Tinbergen. These men founded the European school of ethology, ‘the
biological study of behaviour’ as Tinbergen defined it. In 1972, this field really
came of age as a science when Lorenz and Tinbergen received the Nobel Prize
for physiology. They shared it with Karl von Frisch, who discovered the
remarkable dance of the honey bee, which enables foragers to tell others in
2 Asking questions about behaviour

their hive the location of good food sources. This prize was recognition
indeed that these three men, whom many saw simply as naturalists, had made
a fundamental and lasting contribution to science (Figure 1.1).
It is, in fact, excusable to think of the study of animal behaviour as being a
branch of natural history, for the diversity of nature has always been a source
of interest and wonderment to those who, like ethologists, studied the natural
behaviour of animals. Observing and describing exactly what animals do is
fascinating in its own right, and it is also an essential prelude to a more
scientific analysis of their behaviour, as was stressed by the early ethologists.
It pays to know your animal! Thus spending long hours patiently watching
animals (Figure 1.2) can, in itself, be quite revealing even if it is not extended
to forming hypotheses and carrying out experiments.
As a result of this preliminary period of thorough and careful description
one can make an inventory, or ethogram, of the behaviour patterns of the
species being studied. To the casual observer it might look as though different
species of birds or of fishes behave much the same as each other. It might also
seem that the behaviour of each animal is a highly varied business, not easy to
split up into particular categories. Fortunately, for most animals, these impres-
sions are not totally true. Each species tends to have an array of stereotyped
behaviour patterns, some of which may be shared with related species but
others of which are unique to itself. Describing them and recognising them
each time they appear is not as difficult as it might at first seem.

1.1 A case history

Let us illustrate this point with an old favourite of behavioural studies, the
three-spined stickleback, an especially easy species to study as it behaves more
or less normally in an aquarium tank. Male sticklebacks come into breeding
condition in the spring when daylength increases and the streams in which
they live become warmer. They change colour, becoming bright red on the
underside and iridescent blue on the face, and their behaviour also alters. They
gather weed and collect it at a particular spot on the bottom of their pond,
gluing it together to form a nest with a special movement that extrudes a sticky
secretion from their cloaca. If another male approaches, the territory owner
will chase him away or, if he persists, threaten him by adopting a head-down
posture which shows off the red belly in all its brilliance (Figure 1.3a). Signals
such as this are known as displays. If a ripe female stickleback appears, her
belly swollen with eggs, our territory owner behaves quite differently, showing
another display known as the zig-zag dance (Figure 1.3b). He darts alternately
 A case history 3

Figure 1.1. Ethology’s three Nobel

prize winners: Konrad Lorenz, Niko
Tinbergen and Karl von Frisch.
4 Asking questions about behaviour

Figure 1.2. Though much of the study of animal behaviour today takes place in
the controlled environment of the laboratory, the essence of the subject is to
understand behaviour as it occurs in nature and, for this reason, much ethological
research is still conducted in the wild (photograph © K.J. Stewart).

towards and away from the female in a very striking manner and, if she follows
him, he draws her slowly towards his nest. Once there she may creep through
the nest and spawn, and he will then follow, fertilising the eggs she has pro-
duced as he does so. Her part is then over; indeed he is likely to chase her away,
for care of the eggs, and of the young after they hatch, is carried out by the
male alone. When he has eggs he stays close by the nest and repairs any
damage that it may suffer, as well as showing fanning, a movement that serves
to drive a stream of water over the eggs and so keep them supplied with
oxygen (Figure 1.3c).
This description allows the identification of certain behaviour patterns
which are common to all male three-spined sticklebacks in breeding condi-
tion: ‘gluing’, ‘head down posture’, ‘zig-zag dance’, ‘creeping through’,
‘fanning’. All these would appear in an ethogram of this species. But the
description also raises a great many questions, and it is here that the scientific
aspect of ethology begins. Niko Tinbergen recognised that the sorts of ques-
tions one could ask about behaviour fell into four different categories: ques-
tions about development, about causes, about functions and about evolution.
Interest in development might lead one to ask how a male comes to behave in
 A case history 5

(a) (c)

(d)

(b) (e)

Figure 1.3. Some of the fixed action patterns shown by a male three-spined
stickleback in breeding condition. (a) The zig-zag dance with which the male leads
the female to the nest, at which he shows her the entrance (b) and she creeps
through and spawns (c); (d) shows the male fanning at the nest and (e) the head-
down threat display he shows when another male approaches.

the way that he does during the course of his lifetime. For example, does his
skill at nest building improve with practice? Does he court a female the very
first time that he sees one or must he learn that this is an appropriate way to
behave towards her? On the subject of causes, one asks about the mechanisms
underlying behaviour, and this concerns both internal states and external
stimuli. What is it that signals the male to come into breeding condition in the
spring, and how does this affect his physiology so that he is ready to fight and
to court? He behaves differently towards females and towards other males:
what difference between them leads him to do so? Functional questions
concern the advantages of behaving in a particular way. Why does the male
show the zig-zag dance rather than simply swimming to the nest? Does his
head down posture actually deter other males from approaching, as it should
do if its function is to act as a threat signal? Finally, we can ask about the evolu-
6 Asking questions about behaviour

tion of behaviour. Comparison between different species of sticklebacks can

give us clues about how ancient or recent are particular forms of behaviour.
Comparison between displays and other behaviour patterns may suggest what
actions formed the originals from which displays have been derived.
This account of the questions with which the study of behaviour is con-
cerned makes it sound as though the subject is a straightforward one attack-
ing well-defined problems and without any great controversy. Such an
impression would be totally wrong, however. Scientists interested in behav-
iour have had to contend with gales blowing from various different directions
during their short voyage, as well as with awkward questions from some of
their own number determined to rock the boat. This has been no bad thing.
The early ethologists, and particularly Konrad Lorenz, put forward sweeping
general theories based more on careful observation and brilliant intuition than
on thorough experimental evidence. These formed a marvellous source of
hypotheses for later research, and as a result of this work more detailed
knowledge has accumulated and many of the broad and simple theories have
had to be replaced. There has also been a growing reluctance to generalise as
it has become clear that different animal species vary a great deal in their
behaviour so that all-embracing theories are not likely to be very helpful. Rats
are not just large mice, far less small people!

1.2 Development and causation

The first real storm to hit animal behaviour blew across the Atlantic, and came
in a confrontation between the ethologists and the American school of
comparative psychology. The two groups shared an interest in the behaviour
of animals, but they approached it from very different viewpoints. The ethol-
ogists worked largely in continental Europe and, being zoologists, they had a
respect for evolution and were thus interested in a wide variety of species and
the different ways in which they behaved. Despite their name, the comparative
psychologists at that stage tended not to be concerned with such comparisons
and to study very few species, usually just rats and pigeons, their interest being
to look for general laws of behaviour that would hold regardless of the species
being studied, and preferably apply to humans as well. Their reputation was
for rigorous experimental work in carefully controlled laboratory conditions;
most ethologists on the other hand simply observed their subjects behaving
freely and they did so in the totally uncontrolled conditions of the animal’s
natural surroundings, those to which selection had adapted it (Figure 1.4).
That two such different approaches to very similar topics should lead to
 Development and causation 7

Figure 1.4. A traditional view of the distinction between ethology and psychology
was that the psychologist put his animal in a small enclosure and peered in to see
what it was doing, while the ethologist put himself in the box and looked out at
what the animals round about were up to. Both approaches have their advantages
and this distinction between psychology and ethology is now blurred.

confrontation is not surprising. The battle was fought over the subject of
behaviour development, a subject on which the views of the two schools were
especially starkly contrasted.
The different views on development of the ethologists and the comparative
psychologists stemmed, in essence, from the fact that one stressed nature and
the other nurture. To most psychologists the learning ability of animals and
the flexibility it gives their behaviour is the main interest in studying them, for
these aspects may shed light on the equivalent attributes of humans, hence
8 Asking questions about behaviour

their stress on nurture. To ethologists, on the other hand, the study of species-
typical behaviour was a prime concern: they therefore tended to concentrate
on patterns which were highly stereotyped and of similar form throughout a
species, and they often referred to such acts as ‘innate’ or ‘instinctive’. The
assumption here was that nature was all-important and nurture was of little
consequence. Indeed, Konrad Lorenz once remarked that the developmental
origins of behaviour was a subject of more interest to embryologists than to
ethologists.
This controversy was a bitter one, but it was also fruitful, for each side had
much to gain from the other, and its resolution brought them closer together.
Psychologists came to recognise that evolution had led animal species to be
different from each other and placed constraints on what each could learn. For
their part, ethologists came to reject the idea that any behaviour was fixed and
inflexible and to realise that the acts they studied, no matter how stereotyped,
may have been influenced by learning and by other environmental influences.
They also came to appreciate the merits of a carefully controlled experimental
approach. Thus today many ethologists work in the laboratory and some of
them even use the sorts of equipment developed by psychologists for the
study of learning, adapted to shed light on ethological questions. In the battle
over nature and nurture, the result has been a synthesis: both sides have gained
from the realisation that neither nature nor nurture can be ignored in the
development of any behaviour pattern. The borderline between ethology and
psychology, once hotly contested, has now broken down and those trained in
either field may be found working on a variety of topics of interest to both.
A good example of work that spans the interests of both fields is given in Box
1.1.

Box 1.1 Development: fear of snakes in rhesus monkeys

Not surprisingly, wild monkeys are normally frightened of snakes. They have special
alarm calls that are given when one is about, and these lead others to be very cau-
tious. They will approach and have a good look, but not get too close. Until the
snake has passed the behaviour of the whole troop is altered.
Susan Mineka and her colleagues have studied how young rhesus monkeys in
captivity come to recognise snakes and be so afraid of them. They have used a piece
of equipment (known, rather pretentiously, as the Wisconsin General Test
Apparatus), which is simply a plexiglas chamber into which an object is placed. A
hungry monkey is then put on one side of this chamber with food on a shelf on the
other, and its fear of the object is assessed by how willing it is to reach across. Wild
 Development and causation 9

caught monkeys will not do so when there is a snake in the box, but young ones
reared in captivity show no such fear, reaching casually across for the food. If,
however, they see another monkey being fearful in that situation, even if only on a
video screen, they too will become scared next time they themselves are tested. This
suggests that the fear is being passed from one animal to another through social
learning. But there is more to it than that, as there is something special about
snakes that facilitates the learning. If a videotape is edited, so that the tutor monkey
appears to be frightened of a bunch of flowers that has been cut-in in place of the
snake that was really there, the observer does not become frightened of flowers. So
we have an interplay here between learning from others and a predisposition to
learn about snakes.

A lot of behaviour development involves such interplays. In this case it is a very

neat arrangement. The spectrum of predators in one place may be different from
that in another within the range of a species of monkey, and some things that are
harmless look very like ones that are not. Learning from the experience of others,
but with a bias towards learning to be cautious about some general categories of
animal (long thin ones are not a bad start!), is a good developmental strategy, so
that the animal comes to avoid things that might harm it but to ignore ones that
will not.

Development is just one area that is of interest to both biologists and psy-
chologists studying behaviour. Another is in the field of causation, the study
of those outside influences and internal states that lead animals to behave in
the way that they do. The senses of animals keep them informed about
changes in the external world so that they can react appropriately to many
10 Asking questions about behaviour

different sorts of stimuli, escaping from those that are potentially dangerous,
attempting to capture and eat those that look like food and approaching and
courting those that may be prospective mates. Understanding the sensory pro-
cesses involved is important to shed light on how behaviour is caused, and
behavioural studies here border on the interests of the sensory physiologist
and the perceptual psychologist. All have the common aim of understanding
how events in the outside world are translated into nervous signals and hence
into behaviour.
Both physiological psychologists and biologists interested in behaviour may
also be interested in how behaviour is influenced by internal events, such as
low blood glucose or high levels of a hormone. Just as it is possible to present
an animal with a loudspeaker playing a courtship call or a model showing a
threat display, so its internal state can be changed: for example, it can be
deprived of food so that its blood glucose is lowered or it can be treated with
a sex hormone to see whether this affects its behaviour. A full understanding
of the causation of behaviour requires knowledge of how both external and
internal events affect the nervous system to produce behaviour. Box 1.2 gives
an example of a study looking at how these factors interact. To understand
causation at all levels also means looking into the exact neural mechanisms
involved, the centres and pathways which intervene between senses and
movement. This is the realm of the neurobiologist but, from the point of view
of understanding the causes of behaviour, it can also be fruitful to treat the
animal as a ‘black box’ rather than probing inside it. For example, one can
study the events in the outside world that lead to a behaviour pattern being
shown, or how an animal decides which, of the many behaviour patterns in
its repertoire, it will carry out at a particular moment. The fact that animals are
more willing at some times than at others to show particular responses, like
eating, drinking or mating, has led to many different theories of how internal
and external factors combine to affect behaviour, collectively known as the-
ories of ‘motivation’. Much of the attention of the early ethologists, such as
Lorenz and Tinbergen, was devoted to these theories, but recently they have
rather fallen from vogue. This is partly because it has become more fashion-
able to explain the mechanisms underlying behaviour in terms of the animal’s
neural machinery, as many neurobiologists are trying to do. But it is also
because motivation is a very complex matter which requires the taking into
account of many different factors for each system of behaviour and cannot
be summed up by a simple overall model of the sort that ethologists originally
put forward.
 Development and causation 11

Box 1.2 Causation: the incubation of barbary doves

The reproductive behaviour of barbary doves, or ring doves as they are called in the
United States, has been studied extensively, particularly initially by Danny Lehrman.
If a pair of doves is put together and given a nest and eggs they will not incubate
straight away, so the right external stimulus is not all that they need to get the
behaviour going. But if a pair has been together for several days, particularly if they
have had pine needles with which to build their own nest, they will start to sit when
given a nest and eggs even though the female has not laid. Clearly something has
happened to change their behaviour.

When the pair are first put together the male starts courting the female, using a
display called the ‘bowing-coo’, with which he struts up and down in front of her
cooing (like a dove!). To start with she responds rather little, but after a few days
copulation first occurs. At this stage too the male has started carrying nest material
to their nest bowl and calling there with another display, the nest-coo. The female
follows him to the nest, but leaves when he goes in search of more material. After a
day or two, however, she remains behind when he leaves and starts to shape the
nest; a further day or two thereafter, when the nest is complete, the first egg
appears and they are all set to incubate.
These changes will take place even if the two birds are separated from each other
12 Asking questions about behaviour

by a wire mesh, so mating itself is not required. But, as far as the female is con-
cerned, the displays of the male are an important factor. Indeed, female doves can
be stimulated to lay by the shadow of a male projected onto frosted glass.
Interestingly, what these displays appear to do is to encourage the female herself to
nest-coo, and this in turn may influence her by self-stimulation. But what leads to
incubation? At one level the answer is the stimulation received from displaying and
nest building. But what these do is to lead to secretion of hormones. Some of these
lead to the production of eggs – obviously also essential for incubation – but the
one important for incubation itself is progesterone. If female doves are injected with
this hormone before pairing, and given a nest and eggs as soon as they are placed
with a male, they will sit at once.
The breeding behaviour of doves is a fine example of how internal factors, such
as hormones, and external ones, such as a displaying partner, nest material and
eggs, combine to give changes in behaviour. Not only that, but the two birds are
nicely synchronised with each other and the changes in behaviour take place at just
the right time. Because of their earlier experience, which changed their internal
state, the birds are ready to sit as soon as their eggs appear.

1.3 Evolution and function

As mentioned earlier, investigating the evolution of behaviour and its func-

tion, or adaptive significance, are two other fields of interest, particularly from
a biological perspective. In these areas animal behaviour borders on genetics,
evolutionary biology and ecology, rather than on topics of interest to psy-
chologists and neurobiologists. The study of behavioural evolution is not an
easy task, for behaviour leaves no fossils, so the best that can be done is to
reconstruct the course that it must have taken from a comparison of the
behaviour of species alive today. Selection experiments and the study of
mutants that behave differently from normal animals can also help us to
understand the changes in behaviour that must have taken place during the
course of evolution. Rarely, as in the case outlined in Box 1.3, evolutionary
change can actually be observed taking place.
The major thrust of recent ethological research has, however, been in the
field of function: studies aimed at understanding the adaptive significance of
behaviour. The last few decades have seen a revolution in our understanding
of evolution theory, with the emergence of many stimulating new ideas. Some
of these concepts, such as ‘inclusive fitness’ and ‘evolutionarily stable strate-
 Evolution and function 13

gies’, have particular relevance to behaviour and have led to intensive study,
especially of the social behaviour of animals in the wild, in an effort to dis-
cover just how natural selection has led their behaviour to be as it is. This field
of study, on the border between animal behaviour, ecology and evolution
theory, is usually now referred to as behavioural ecology, and has generated
some beautiful work. Some aspects of it, especially the subject of sociobiol-
ogy, which concentrates on putting social behaviour in an evolutionary frame-
work, have also led to a great deal of controversy, much of it rather fruitless.
One area of argument has been over the relevance of these ideas to humans,
as some sociobiologists (often now calling themselves evolutionary psychol-
ogists, just to confuse!) are enthusiastic about the application of evolutionary
ideas to humans, believing that our behaviour can be best understood if
viewed in the context of our evolutionary heritage.

BOX 1.3 Evolution: the behaviour of guppies in Trinidad

Guppies are small fish that occur in the streams and rivers of northeastern South
America and a few Caribbean islands, and they are also common aquarium fish as
they are easily kept and bred in captivity. The males have attractive iridescent
patches of various colours with which they display to the females. They have been
studied particularly in Trinidad by a succession of people, most recently Anne Houde
and John Endler, and Anne Magurran and Ben Seghers together with various collab-
orators. In Trinidad guppies occur in numerous rivers, though obstacles such as
waterfalls have sometimes halted their spread. Their morphology and behaviour
varies between the different rivers and this has caused particular interest: it provides
a rare opportunity to study evolution in action.
A critical factor turns out to be the distribution of predators between the rivers.
Some sites are ‘high risk’ with a range of predatory fish, such as pike-cichlids, while
14 Asking questions about behaviour

others are ‘low risk’ with rather few of them. At the high risk sites males are less
brightly coloured and there is some evidence that the females in these areas may
prefer them that way, unlike their usual preference for the brightest. In places where
a predatory shrimp is common, females prefer males with orange colouration;
orange is a colour the shrimp cannot see.
When it comes to behaviour, fish in the high risk sites spend much more of their
time in schools, and they are more tightly grouped within these. They are also more
cautious when a predator is present, keeping a greater distance from it than fish
from low risk areas. That such differences have a genetic basis has been confirmed
by the fact that the differences in schooling behaviour persist after several genera-
tions of breeding in captivity. Further evidence comes from an experiment where
200 guppies were moved from a high risk area to a virtually predator-free one,
above a waterfall where there were previously no guppies. Captive breeding from
this population 34 years later showed that their offspring schooled less and were
not as wary of potential predators as their ancestors at the high risk sites. Schooling
has costs, such as increased competition for food, so that selection might be
expected to lead to a decrease in the absence of predators. It is striking that such a
decrease could be found after only 100 generations or so.
Guppies breed rapidly and have a short generation time. Not many other species
give us opportunities to observe evolution actually happening in the wild. However,
as we will see in Chapter 6, comparisons between species can often allow us to
reconstruct the sorts of changes that must have taken place during their evolution.

Another, related, bone of contention has been whether or not the claim that
behaviour is adaptive means that it is unmodifiable and under tight genetic
control: some sociobiologists have written as if it did mean this, and have hence
been charged with ‘genetic determinism’, an especially heinous crime if they
are also writing about humans! But, in truth, all that is required for behaviour
to be acted on by natural selection is that it has some genetic basis, no matter
how slight. There is no reason why its development should be in any sense
tightly controlled, so adaptation does not presuppose genetic determinism.
And, in the case of humans, behaviour may come to be adaptive for a great
many reasons other than through natural selection, notably through our
remarkable ability to learn by our own experience and from the experience of
others. Indeed, for many of us the environment we occupy is so odd compared
with that in which we evolved that it would be hard to argue that our adaptation
to it had much to do with selection! Yet we have both modified our environ-
ment to suit ourselves and adapted ourselves to it in a fine degree of detail.
These controversies have generated a good deal of heat, and there are still
 Evolution and function 15

those who are fiercely committed to one side or the other without a hint of
compromise. But reason, as so often, takes a middle course, and neither
extreme position is very plausible. Both evolutionary heritage and the genes
which are its legacy from one generation to the next clearly influence all the
behaviour of all animals, ourselves included, but to argue that either is all-
important is as mistaken as to deny that either has any importance. A more
serious difficulty with functional reasoning, as with much evolutionary
thought, is the ease with which hypotheses can be generated but the difficulty
of subjecting them to any rigorous test. Writing popular books, packed with
brilliant hypotheses dressed up as facts about the relevance of animal behav-
iour to humans, is quite a money spinner. Many of the ‘just so stories’ with
which these are filled are so compelling that one feels they just must be true –
until someone else comes up with an even neater explanation!
The problem is that functional questions are not often easy to test experi-
mentally. To carry out an experiment what one needs is two groups of animals,
an experimental and a control, which differ only in that aspect of their behav-
iour one wants to study. It is not easy to achieve such a very specific change,
leaving all other aspects of behaviour the same. Furthermore, adaptation is
only relevant to the particular environment in which the species evolved, so
the study is most likely to be useful if it is carried out in nature, which further
limits the sorts of experiments that can be done. Faced with such problems,
many behavioural ecologists have rejected experimentation and concentrated
on observation and correlation instead. The results of this can be persuasive,
particularly, as in the example shown in Box 1.4, where they are geared to
testing specific hypotheses, in this case that bumblebees forage in the most
efficient possible manner.
But a correlational approach has problems of its own. The size of monkey
troops may correlate with the size of their home ranges, but this fact does not,
in itself, provide an explanation. A larger group may require a larger range to
feed on, a larger range may require a larger group to defend it, or both fea-
tures may be caused by a third factor and not directly related to each other at
all. For instance, a food supply which is briefly abundant in a small area, then
dies out and ‘blossoms’ once more some distance away, may lead to both. It
may encourage a large home range so that there is always at least one source
of food present in it. Furthermore, as each source in turn is abundant, group
size can be large without disadvantage to the members of the group. Thus
large group and large range will both arise without one causing the other. The
moral is that one must be cautious about interpreting correlations and that, in
function as in other aspects of behaviour, greater certainty can be reached if
it is possible to carry out experiments. This is not easy but, with ingenuity,
some clever tests of functional hypotheses have been devised.
16 Asking questions about behaviour

BOX 1.4 Function: the foraging behaviour of bumblebees

The relationship between flowers and the insects that pollinate them is a close and
fascinating one. Insects fly from flower to flower feeding on the nectar that is pro-
duced for them as a reward for carrying pollen from one flower to the next. But this
is only of benefit to the flower if the insect takes the pollen to one of its own
species, for pollen is the plant’s equivalent of sperm and is wasted unless it fertilises
an egg of the same species. Insects, however, do tend to specialise on the flowers of
one species because skills are usually necessary to reach the nectar and, having
developed these, they are better equipped to do so. Indeed flower species probably
differ from each other in shape very substantially for this very reason: only insects
that have developed the appropriate skills can feed on them, and this forces the
individual insect to stick to the same species and thus carry pollen from one plant to
another of the same sort.

Bumblebees are no exception to this rule, and the adaptiveness of their behaviour
has been shown beautifully by Bernd Heinrich. Each bee specialises on one particu-
lar flower, referred to as its ‘major’, while taking occasional nectar meals from other
species, its ‘minors’. Worker bumblebees are short lived and tend to stick to the
 Keeping things simple 17

same major, but queens live longer and may switch from one to another. Having
minors probably enables them to check up on whether another flower has become
more profitable than their current favourite. As flowering seasons are short, the
abundance of different sorts of flower changes through the year and thus bees are
well advised to check that their major still gives the best returns.
Whether it is worth a bee feeding on a particular flower species depends on
several factors. Some flowers are much more common than others, so there is little
flying time between one meal and the next. Some flowers also produce much more
nectar than others so that the meals they provide are larger. Finally, the profitability
of a type of flower depends on how many bees are feeding on it. If lots are doing
so, then a bee looking for a food source will find that flower rather unprofitable, for
there will be little nectar available on each visit. As a result, it is only common
flowers with a high rate of nectar production that have large numbers of bees
feeding on them. Each bee arrives shortly after another one and so receives little
reward, but then it does not have to fly far to the next meal. On the other hand,
few bees specialise on rare flowers with a low nectar yield and so each of them gets
a large meal at each visit, though it has to fly a long way between one flower and
the next to get this reward. Taking all these points into account, Heinrich argued
that every bee does about as well as every other in terms of nectar intake per unit
time. It follows that it would not pay any of them to switch to a different flower
from the one they are majoring on except if these change in frequency. Their behav-
iour is thus beautifully adaptive to themselves, as well as to the flowers.

1.4 Keeping things simple

One nice thing about animal behaviour as a field is that it is relatively free of
jargon compared with many other branches of biology. A glance at the glos-
sary at the end of this book will yield few words that are totally unfamiliar. But
there is a hidden problem here, for many of these words are used in a special
sense that is not the same as their everyday meaning, and is usually more
precise and more restricted. The word function is a good example here. While
it is in the vocabulary of most people, rather few mean selective advantage
when they say it! There has also been a recent tendency to adopt words from
everyday language, which are not entirely appropriate when applied to animal
behaviour. Male animals may often mate with unwilling females, and for a
while the word rape was used to describe this. But that word carries with it a
good many overtones, for example of culpability, which one would not want
18 Asking questions about behaviour

to apply in animal examples. To refer to ‘enforced copulation’ may be more

cumbersome, but it is also safer. The same argument could be applied to
words like ‘lying’ and ‘cheating’, though these are now firmly in the behav-
ioural vocabulary. It is important to realise that these words do not imply any
malice or evil intention on the part of animals, or indeed that they are think-
ing anything out for themselves at all. A bird may avoid eating a brightly
coloured butterfly that tastes nice because its colour pattern mimics that of
another species that tastes nasty. The bird is certainly deceived, but to label
such a beautiful adaptation as ‘lying’ on the part of the butterfly is really going
too far!
Problems in the use of words like these are part of the more general issue
of ‘anthropomorphism’, looking at animals as if they were people. The way
in which they behave often looks extraordinarily clever and sophisticated, as
if they had thought about the problem and decided what the best thing to do
was. But beware! Millions of years of evolution can come up with an answer,
often in very simple animals with small nervous systems, based on a few rules,
which looks not dissimilar to the solution to the same problem we might work
out for ourselves. Animals may indeed think things out, have intentions, and
behave in all sorts of other ways like we do, but these are very difficult ideas
to test scientifically. One of the most important rules in science is that known
as ‘Occam’s razor’ which states, in essence, that one should test the simplest
hypothesis first and only if it is found wanting move on to more complex ones
which are less easy to disprove. Fortunately, as we shall see, some delightfully
simple ideas have proved adequate to account for many aspects of the behav-
iour of animals.
Related to the assumption that animals have intentions is to write as if
natural selection had. This is the problem of ‘teleology’: the implication that
there is purpose or design in nature. Natural selection does not act ‘in order
to’ achieve something, but works in retrospect, successful traits persisting and
unsuccessful ones dying out. It is a common shorthand among biologists, nev-
ertheless, to write as if selection had purpose; as with the actions of animals,
it is best not to for fear of misunderstanding.
Having come up with these strictures, I had better try to abide by them
myself without resorting to contorted sentences and leaden prose!

1.5 Book plan

The study of animal behaviour concentrates on four different types of ques-

tion: causation, development, evolution and function. These topics will form
 Book plan 19

the core of this book and are dealt with, each in turn, in Chapters 4–7. But, as
the best studies have always started with a period of description, we will follow
this tradition in the next two chapters by describing the motor patterns of
animals, their form and what is known of their control, then the senses which
give the animal its outlook onto its environment, and how stimuli influence
these to produce behaviour. Much of the book will be concerned with indi-
vidual animals, but we will often refer to their interactions with one another,
as when they fight or they mate. In the last two chapters of the book we will
come to the social aspects of behaviour more specifically, to discuss
communication and the social organisations which are built up by the com-
municative interactions between individual animals. At the end of the book
there is a glossary. This deals with words and phrases in common use among
those studying behaviour. Most of these will be found elsewhere in the book,
and referring to the glossary may help to remind the reader of their meaning,
but I have included others as well in the hope that this dictionary will help the
reader who takes the subject on through wider reading. The literature list at
the end of the book will, I hope, also help in this. It is not intended to be
exhaustive, but gives a list of books in which the reader will be able to take the
topics of each chapter to greater depth, and journals where it is worth looking
for the latest findings.