The Mammalian Radiations: An Analysis of Trends in Evolution, Adaptation and Behaviour 
by John Eisenberg.
Athlone, 610 pp., £32, December 1981, 0 485 30008 7
Show More
Show More

‘Natural history’ is seen by some professional biologists as hardly deserving to be regarded as a part of science. Compared to the experimental sophistication of molecular biology, and the apparent generality of its conclusions, natural history is no more than a collection of particular facts of little theoretical or practical import. There are two reasons why one should dissent from this judgment. The first is that the task of biology is to explain the living world, and that world is irreduciby complex. If I am interested in molecular biology – and I am – it is because it helps to explain how that complexity arose. A biologist who was not interested in the diversity of living things would be like a historian who did not care what had actually happened in history, and confined his attention to the study of experimental psychology.

The second reason for taking natural history seriously is that the central theoretical idea in biology is that of evolution by natural selection, and this idea was formulated by men who were naturalists first and evolutionists second. Charles Darwin collected beetles before he embarked in the Beagle, and it was only in thinking about the experiences of that voyage that he became an evolutionist. Alfred Wallace differed in that he was an evolutionist (although he had not conceived the mechanism of natural selection) before he undertook his journeys to South America and Malaysia, but it was his experience as a naturalist in this country which made him an evolutionist. A knowledge and love of natural history were not sufficient to bring Darwin and Wallace to the theory of natural selection, but they were necessary.

Although Darwin, in The Descent of Man and Selection in Relation to Sex, and in The Expression of the Emotions in Man and Animals, made a decisive start on the study of the evolution of behaviour little further progress was made during the next fifty years, and it is only in the last fifty that the subject has come of age. That it has done so we owe in part to the ethologist, in particular Lorenz and Tinbergen, who taught us to look at what animals actually do and to ask questions about how and why they do it, and in part to men like Lack and MacArthur, who related the observed behaviour of particular species to their ecology, and to the selective forces acting on them. It is a striking fact that all these four men, and many of their immediate followers, worked primarily on birds (for the sake of my theme, I shall forget Tinbergen’s classic work on sticklebacks). Birds are conspicuous diurnal animals, and they exist in a rich variety of species in Europe and North America.

In contrast, mammals – about which we might be expected to have a more immediate curiosity – are much more difficult to study in the field, particularly if one happens to live in Europe. Most mammals are nocturnal. Often they rely on means of communication (scents; high-frequency sounds) which we find harder to perceive than the songs and plumage displays of birds. Many of them live underground, or in the forest canopy. The number of species living in Europe is relatively small. Yet despite all these difficulties, knowledge of mammalian behaviour has grown at an astonishing rate, particularly during the last ten to fifteen years. This has in part depended on field studies in the tropics: we know more about the behaviour of lions than we do of badgers. It has also been helped by technical advances – for example, the development of miniature radio transmitters which can be attached to animals as small as bats and mice without serious inconvenience. But in the main it has been achieved by the determination, ingenuity and endurance of a band of field naturalists. As a largely chairbound biologist, I am repeatedly amazed at the quality of the information gathered by colleagues.

Yet despite this recent and exciting body of work, it is still true that most biologists, when thinking about the evolution of behaviour, are likely to think of birds rather than mammals – of robins and great tits if they are English and of jays and red-winged blackbirds if they are Americans. We have been waiting for a book which would summarise the recent work on mammals and help us to think realistically about it. John Eisenberg has now written that book; it is hard to imagine anyone else who could have done so. He has studied a range of mammalian species in the wild. As assistant director of the Washington zoo, he has a wide familiarity with mammals in captivity. He has an astonishing knowledge of the literature on mammals, including many relatively obscure natural history journals.

His book was written for professional biologists, and will become a standard work of reference. How far is it a book for the amateur? The answer must depend on what it is that the amateur loves. This is not a book for a reader who wants only general ideas. To enjoy it, you must belong to that select class of people who like to know things about animals. For example, did you know that the sloth, that most arboreal of mammals, descends to the foot of a favoured feeding tree to defaecate, carefully burying its faeces? Did you know that there are hairless mole rats in southern Africa which live communally underground, and which form digging chains, passing the earth from one to another? To be an evolutionary biologist, you would need to find these facts fascinating in themselves, but you would also need to ask questions about them. Why does it benefit a sloth to take the risks involved in coming down from the trees to defaecate? There is a suggestion in the literature that it pays a sloth to fertilise its favourite tree, but I find that hard to accept. How did mole rats come to be so co-operative? Are the members of the digging chain relatives? Do they take it in turns to do the hard work of excavation?

Eisenberg is not interested only in facts: he wants to make sense of those facts. To do so, he makes use of the ‘comparative method’ in two rather different ways. He compares the mammalian radiations in different geographical regions, and he compares the behavioural systems of different taxonomic groups.

We are fortunate that the wandering of the continents has ensured that the evolution and radiation of the mammals have taken place independently, or largely so, in a number of different regions. Very crudely, the picture is as follows. The northern temperate regions (North America, Europe and Asia) have been in intermittent but effective communication throughout. As a result, we often find closely similar species in North America and Eurasia (the caribou and the reindeer; the American and European bison). Many groups (horses, elephants) which are today confined to the Old World were until recently represented in the New. The tropical regions of Africa, India and Malaysia have been intermittently connected to the northern land mass, but Africa has been connected to the other tropical regions only via the north. Three regions – Australia, Madagascar and South America – were separated from the other continents for most of the period during which mammals were evolving.

Australia is the most distinct. Until 25,000 years ago, when men first reached Australia, bringing with them dogs which later ran wild (dingos), the only ‘placental’ mammals in Australia were the bats, and some mouse-like rodents. The ‘marsupial’ mammals which radiated in Australia failed to solve the central problem of a prolonged intra-uterine life: how to prevent the mother from treating the foetus as a foreign body and reacting against it immunologically, as we react against an organ transplant. In consequence, marsupial babies are born in an immature state, and must make their way to the mother’s teats, to which they become permanently attached for many weeks or even months. In Australia, the marsupials gave rise to herbivorous and carnivorous, and to burrowing, tree-living and plains-living forms. Madagascar, also, was separated from mainland Africa throughout the mammalian radiations. Excluding the numerous recent introductions by man, the mammals of Madagascar are descended mainly from three immigrant groups: the tenrecs (a group of primitive insectivores), the lemurs (primitive primates) and the viverrids (carnivores related to the civets and genets). South America has a more complex, and for a zoologist a more frustrating, history. Earlier, during the age of reptiles, it formed part of Gondwanaland, along with Africa, Antarctica and Australia. After it separated from Africa, there evolved a unique mammalian fauna. The carnivorous elements of the fauna were marsupials, the herbivores placentals. Some ten million years ago – geologically, a mere yesterday – South and North America came together. The resultant migrations were followed by massive extinctions. A few of the southern forms – for example, the marsupial opossum and the placental armadillo – successfully invaded the north, but the successful invasions were predominantly in the other direction, resulting in the extinction of a vast range of mammalian species, including a marsupial analogue of the sabre-tooth cats and a single-toed herbivore resembling the horses.

The existence of these independent radiations in Australia, Africa, South America and Madagascar enables us to ask how far similar climatic conditions cause the evolution of similar mammalian species. To compare these faunas is analogous to comparing the civilisations of the Mayans, the Egyptians and the Sumerians. Similarities suggest necessity: differences suggest the operation of chance.

Certainly there are striking similarities. Those who learnt zoology when I did will remember the difficulty of distinguishing the skull of a dog from that of the marsupial ‘Tasmanian wolf’ (it was essential to remember to count the incisors). More significantly, Eisenberg emphasises that each of the major faunas evolved the same ecological types (aboreal forms eating leaves, and others eating fruits, and still others eating insects). However, my own strong impression is of the differences. Is it not strange that the major plains-living herbivores of Australia (kangaroos) progress in a series of leaps, whereas those of Africa (antelopes) gallop? Why should the monkeys of South America have prehensile tails and those of Africa not? It is true that the ant-eating armadillos of South America show a remarkable convergence with the pangolins of the Old World. But there are no arboreal armadillos. I cannot resist adding the curious fact that some armadillos regularly produce four genetically identical quadruplets at a birth; this has been a considerable irritation to theoreticians, because armadillo siblings ought to be particularly nice to one another, but there is no evidence that they are.

It may be largely a matter of taste and temperament whether one is struck by the similarities or the differences between the mammalian radiations. I remember my friend Dick Lewontin giving a seminar on ‘causal and historical explanations in science’. To illustrate his point, he raised the question: why are there no penguins in the Arctic? It might be (causal explanation) that some features of the Arctic make the penguin way of life ill-adapted. Alternatively (historical explanation), it might be that penguins evolved in the Antarctic, and have never been able to cross the tropics to invade the Arctic. After discussing the question for some time, he asserted that the wise biologist would answer that there are penguins in the Arctic, only we call them auks (e.g. puffins, guillemots). In other words, we should not bother with the fine distinctions between auks and penguins, or between armadillos and pangolins, but should content ourselves with understanding the features common to different ecosystems.

However that may be, Eisenberg’s book is a wonderful source of data for the comparison of independently evolved faunas. For myself, I shall continue to find it interesting that auks can fly and penguins can’t, and that horses gallop, kangaroos hop, and men and ostriches run.

Eisenberg’s second approach is to compare particular habits – feeding, sleeping, breeding etc – in different families and orders of mammals. He discusses how mammals clean themselves, fight, forage for food and curl up, or do not curl up, to go to sleep. He summarises a vast mass of data on pairing behaviour, copulation, gestation, lactation and parental care. However, the most topical problem he attacks in this way is the nature and significance of social behaviour. The problem is topical partly for theoretical reasons – evolutionary biologists are much concerned with how apparently co-operative behaviour could arise from natural selection favouring individual reproductive success – and partly because such a large part of what we know about the social behaviour of mammals has been discovered in the last fifteen years.

Although social behaviour is found in many mammalian orders – for example, bats, rodents and ungulates – I shall discuss only the carnivores and the primates. Many carnivores are solitary, except at the time of mating and for the longer-lasting bond between a female and her offspring, but larger social groups are found in several species which co-operate in hunting large prey. In the Cape Hunting Dog. Lycaon pictus, several adult males and females form a stable group, hunting co-operatively. However, only one male and one female breed; all the adults help to feed the female and her young, enabling her to raise a larger litter than would otherwise be possible. Eisenberg sees this form of society (which is also found in wolves) as an adaptation for hunting large prey which has evolved from the monogamous pattern typical of canids, in which the male helps to feed the female and her young.

Lions also hunt co-operatively, but the permanent group consists of several females and their young. All the females are reproductively active, and may suckle each other’s cubs. At any one time such a group of females will be associated with a group of two or more males, often but not always related to one another. Any one of the males may mate with an oestrus female. From time to time the males holding a group of females will be violently displaced by another set of allied males; at such times, existing infants are usually killed. The female offspring remain in the group, whereas the males leave when adolescent. This matrilincal structure is typical of mammals. The fact that offspring of one sex leave the group before breeding has, reasonably, been explained as having evolved because the offspring of matings between close relatives are usually of low fitness. It is still a matter of debate, however, why, in mammals, it is usually the males which leave the group, whereas in birds it is usually the females.

Eisenberg sees lions and hunting dogs as the two extremes of carnivore social structure. Spotted hyaenas, which, contrary to popular belief, often hunt and kill large prey, live in a society in which females rather than males are dominant. As in lions, there is a group of reproductive females, but the females are larger than the males and dominant to them. Strangely, females have external genitalia, used in greeting ceremonies, which are visually indistinguishable from those of males.

These three examples might suggest that, in carnivores, societies arise in species which need to co-operate in hunting. Unhappily, this is not the whole truth. Dwarf mongeese live and forage in groups, yet feed on insects and small mammals which are caught individually. As in wolves and hunting dogs, only one pair is reproductively active, but other group members help in defence against predators. It is far from obvious what reproductive advantage these helpers gain by group membership; the matter is made still more puzzling by the fact that the helpers are often recent recruits from outside the group, unrelated to the breeding pair.

The breeding systems of primates can be classified into three main types: monogamous, single-male and multi-male. In monogamous species, such as gibbons and marmosets, a mated pair stay together, and co-operate in defending a territory and raising the young. In other species, of which the gorilla and the langurs are examples, a single adult male lives with several adult females, and has sole sexual access to those females. Finally, in chimpanzees, macaques and baboons, the group consists of several reproductively active males and females. In such species, highly complex social relationships can arise between the members of a group, but it is in general not true that permanent bonds are formed between a particular male and female.

It is natural to ask what was the social structure typical of our own ancestors. We cannot tell, although there are tenuous indications from comparative anatomy that the harem-holding structure found in gorillas and langurs may be nearest the mark. One difficulty is that we do not know why different primate species display different social structures. Eisenberg suggests that social organisation must be related to the distribution of food resources, and I agree with him. I fear, however, that he may know too much about mammals to come up with simple explanations, if such there be. But whether or not he has thought of all the right answers, the study of mammals will not be the same again.

Send Letters To:

The Editor
London Review of Books,
28 Little Russell Street
London, WC1A 2HN

Please include name, address, and a telephone number.

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences