Volume: 17.2



The Common Zebra is an Uncommon Mammal

The Common Zebra is an uncommon mammal Richard D Estes, PhD, explores the evolutionary rationale behind this most intriguing paradox.

If you have ever tried to sex and age the Common or Plains Zebra, you will know it is hard work. Eagle-eyed driver-guides will tell you that the stallions have brighter, more contrasting stripes, but I find that method unreliable. Wildlife researchers and managers tasked with telling the sexes apart can also pick out mature stallions from their more robust build, especially as reflected in their necks, which are thicker than those of mares. And it helps to know that herd stallions usually bring up the rear of a family unit. But the most reliable way to ID the sex of both adults and young is from a full rear view (Figure1.)


The female’s vulva forms a broad band of black skin between the buttocks. Males also have bare black skin extending down from the anus, but this extends only a short way and is of a more triangular shape. The difference is not always clear, however – you really need an under-tail good look. But why is this scrutiny necessary? Why not just check out the male’s external genitalia? Because these are hard to see, that’s why. Seen from the rear, the testes blend in with the broad black stripe running along the belly midline, and the hind legs obscure the penile sheath except in a three-quarter view looking from front to back.

The difficulty of telling males and females apart begs the question: why is the Common Zebra the hardest of all plains game to sex and to age (with the possible exception of the oryx)? Such species contradict Darwin’s theory of sexual selection (Darwin 1871), according to which zebra stallions should look very different from mares. The theory holds that, in polygynous mating systems, in which some males can monopolize mating with more than one female at the expense of other males, male sexual competition (coupled with female preference) will lead to increased size and conspicuous display organs – that is to say, pronounced sexual dimorphism.

The Plains Zebra has an unusual (in mammals) harem social/mating system, shared by the Mountain Zebra and the horse, where stallions own harems numbering as many as five or six mares, captured one by one in fierce competition. Mare-less stallions associate in all-male (bachelor) groups.

Clearly then, the Common Zebra is an exception to Darwin’s theory. How come? I was confronted with the same paradox in my studies of antelopes. The wildebeest, subject of my PhD dissertation and later research, also displays minimal sexual dimorphism despite being highly polygynous. Females even have a faux penile tuft that makes it harder to determine gender. When I am recording the sex and age of a long line of passing gnus, I rely on differences in the horns, which are thicker and wider in the males. The same applies to the related topi and blesbok. In the oryx, however, females have horns as long as or longer than the males’ horns, and sport identical, highly conspicuous colouring and markings. Even the penile sheath is inconspicuous.

It took me decades to come up with a plausible explanation. The first clue emerged when I made a survey of the social and mating systems of the 72 species of African Bovidae, comprising antelopes, sheep and goats, and the buffalo, for a symposium on the behaviour of hoofed mammals (Estes 1974). It turned out that nearly all the herd forming (i.e. sociable) species with minimal sexual dimorphism live for at least part of the year in aggregations that include adults of both sexes. There had to be some kind of counter-selection, I reasoned, against development of conspicuous male secondary characters in species that live in sexually integrated groups.

Conversely, species with pronounced sexual dimorphism (such as impala, sable, Grant’s gazelle and kob) tend to remain segregated according to sex and age except for breeding. In territorial species, only males with exclusive territories succeed in breeding. The small über-class of breeding males enforces separation of bachelor males from females and young. Non-territorial bovids – notably kudu, nyala, eland, cattle, goats and sheep – are still more sexually dimorphic, and are also the most segregated. Male sexual competition is more rigorous in such dominance-hierarchy breeding systems, causing males to keep growing long after females mature, while at the same time developing the most extreme gender differences in size and display organs.

What is the source of selection that counters male peer competition for conspicuous secondary characters? Broadening the scope of the inquiry to other classes of vertebrates, we find that the sexes are virtually indistinguishable in schools of many fish species and in many birds that live in flocks. Pigeons, geese, and swans come to mind – “Birds of a feather flock together,” as the saying goes. It is clear that in such societies natural selection would weed out individuals that fail to fit the mold; ones that are different are likely to be peripheral to the school or flock and are most likely to be picked off by predators. An outstanding characteristic of these schools and flocks is the wonderful coordination of all the members; they wheel and turn simultaneously as though responding to an unseen conductor.

Could any of this information help explain the case of the zebra? I think it can:


    A. Minimal sexual dimorphism correlates with living in mixed groups. The closer the resemblance between the sexes, the more integrated and coordinated are group movements.


      B. Obvious sexual dimorphism correlates with sexual segregation. Males and females live in largely separate societies except for breeding males that, intolerant of potential rivals, are the enforcers of segregation.

How might these findings apply to Plains- Zebra society? In zebra aggregations, you have herds of bachelor stallions interspersed with families of females and young, each guarded by a stallion. Herd stallions interact with one another and with bachelors, yet rarely fight; most encounters take the form of the greeting ceremony. This is really a test of fitness, whereby a herd stallion serves notice that he is the owner of a group of females and young. An illustration of the greeting ceremony (Figure 2) adorns the cover of my book, The Safari Companion (Estes 1999).


The respect of a stallion’s right of ownership is remarkable; it is comparable with the respect of the property rights of territorial males. In both systems, males have made a huge effort to win mating rights, and are prepared to defend those rights against all comers.

Probably the harem system of the Plains and Mountain Zebra and the horse evolved from a territorial system, which persists in other extant members of the family: the wild asses and Grevy’s Zebra. My colleague, Hans Klingel, who carried out his seminal study of Plains Zebra socio-ecology at the same time as I was studying the Ngorongoro wildebeest (1967; in press), believes that early equids such as the ‘dawn horse’ Eohippus defended resource territories like solitary antelopes (e.g. did-dik, oribi, blue duiker).

By exchanging ownership of real estate for ownership of females, the Plains Zebra and the horse gained the freedom to lead a nomadic existence, whether in separate herds or in aggregations of many units. This very unusual social and mating system largely explains why the Plains Zebra is – and why the undomesticated horse was – among the most numerous and successful of hoofed mammals.

As a consequence of the gentleman’s agreement about ownership of mares, males compete mainly over fillies. Now we’re talking about serious fights. Beginning in their second year (indeed, as early as 18 months), females come into heat for as many as five days every month. They advertise their condition by frequent urination and a conspicuous posture with hind legs straddled and tail slightly raised. Reacting to this come-on, stallions converge from far and wide, each hell bent on adding a filly to his own harem. They fight with the father, who behaves as though his mares were up for grabs, and with one another. Battles lasting for hours and whole days test the mettle and endurance of the contestants (Figure 3). Eventually a stallion succeeds in abducting the filly, which – after a trial period perhaps lasting weeks – is accepted into the company of mares already in the seraglio.

Adult mares display the estrous posture only immediately before copulation, ensuring that other stallions are not attracted and the family is left in peace. When approached by a stallion, and during copulation, young and adult mares assume a facial expression that Germans call Rossigskeitgesicht – a splendid term with the mundane English translation “mare-in-heat face” (Figure 4). This display also signals submission.

Unfortunately for the stallion that abducted her, however, the filly comes into estrus again a month later. The battle royal begins all over again and he may well lose her to another stallion. And she keeps coming into heat without becoming pregnant for up to a year. When she is finally in foal, the filly settles down for good in the sire’s harem.

The resemblance between male and female Plains Zebras is close enough that, in a band containing family and bachelor groups, it is behaviour, rather than looks, that makes the stallions stand out. The herding (Fig. 5) and greeting behaviour of herd stallions and the horsing around often seen in bachelor herds are a give-away. Call me crazy, but I see a parallel between zebras and pigeons (which I kept as a child). Like zebras, male pigeons are a bit more robust than females, but you cannot pick them out in a flock until the males start courting and fighting.

Alert readers may at this stage be saying to themselves: “This is all very well, maybe even interesting, but the real question – how did this masking of male secondary characters came about – remains unanswered.” Point taken… My attempt to account for this looks at the zebra’s presumed ancestral social organization and draws on my explanation of minimal sexual dimorphism in wildebeest, oryx, and the like. Both may be considered highly speculative. Indeed, the theory I advanced, in Estes (1991), to account for the condition in bovids is not widely known and has even been discounted – at least by some colleagues of mine who have read the paper.


I proposed that male sexual competition comes in two forms with opposite effects: the familiar peer competition that Darwin recognized, and a much less familiar kind called despotic competition that he didn’t. Peer competition, involving males of the same age and development stage, promotes sexual dimorphism. Despotic competition involves aggression by older and bigger males against younger, weaker males. Competition of the latter kind leads to the eviction of adolescent males from female herds when they develop secondary characters that reveal their gender and trigger the aggression of breeding males. Following eviction, the sex ratio, which remains roughly equal as long as both sexes stay in their natal herd and home range, becomes skewed due to higher male mortality rates.

Accordingly, natural selection should favour a suppression of male secondary characters until the stage is reached when the benefits of leaving (to join bachelor herds and engage in peer competition) outweigh the costs of skulking in female herds. Considering that the fittest males pass on their genes to far more offspring than females can (large antelopes produce only one calf a year for say eight to ten years), mothers should do whatever they can to promote survival of their sons. The same holds true for zebras and horses, which have a 12- month gestation and can produce only one foal a year.

In my obscure 1991 paper, I proposed that female bovids get around the problem of keeping their sons with them, while at the same time allowing them to develop the weapons essential for peer competition, by themselves growing horns and copying other male secondary characters.

When horns of similar size and shape are present in both sexes, they cease to be badges of gender. Ditto markings, beards, manes, and other traits evolved by males to advertise their gender. While I know that females with horns use their horns as weapons, I maintain that natural and sexual selection for keeping males with females until they are prepared to join male society is stronger than selection for horns as weapons in females. There is not much evidence, after all, to suggest that horned females are subject to lower predation rates than hornless ones.

My theory would explain how it happened that oryx males and females are more alike than, say, wildebeest or topi. These desert species roam widely and exist at very low population densities. Evicted subadult males would be unlikely to find a separate bachelor herd they could join. Consequently, the costs of leaving are greatly outweighed by the benefits of staying. So oryx herds include bachelor males intermixed with females. The alpha male treats them all the same as long as they behave the same: that is, submissively. Female-mimicry of male horns and other markings originally reserved for males to the adult stage might explain how this unusual transformation of an originally territorial mating system came about.

Enough said already about bovids, however. This article is supposed to be about a hoofed mammal from an altogether different order: an odd-toed, not an even-toed ungulate. What gave rise to the selection pressure that would account for look-alike male and female Plains Zebras?

I cannot even use the argument that the risk of being evicted from the herd is the source of counter-selection, because male offspring are not evicted. In the harem system, selection favours stallions that take a paternal interest in their offspring by allowing them to stay on in the natal family.


Eventually, at between two and three years of age, they leave of their own accord and join a bachelor herd, where they remain until mature and ready to start their own harems at five years or so.

If the zebra’s ancestors were territorial, then that might explain how it all started. The benefits of staying with the mother in a female herd would select for sons that avoid eviction by territorial stallions through continuing to look like females. Regardless of how the harem system might have come about, minimal sexual dimorphism would facilitate association in aggregations with minimal strife and become what socio-biologists call an evolutionarily stable strategy (ESS).

I want to make just one final point: Grevy’s Zebra was once widely distributed in Africa. It is now an endangered relict species that the Plains Zebra has replaced everywhere except in the most arid parts of its former range. The same may be said of the wild asses. This is proof (I would argue) that the nomadic harem system of the Plains Zebra and the horse is superior to the territorial society of the Grevy’s Zebra and the wild asses. Although the explanation I have offered for the Plains Zebra’s lack of sexual dimorphism is pure speculation and is unprovable, I hope that I have made the case that toning down gender differences by reducing male aggressiveness helps Plains Zebras to live together in relative peace.

References: Darwin, C R 1871. The Descent of Man and Selection in Relation to Sex. Appletonn New York.

Estes, R D 1974. Social organization of the African Bovidae. Proceedings of an International Symposium on the Behavior of Ungulates and its Relation to Management. IUCN Special Publication (New Series) No. 24, Morges, Switzerland, pp. 166- 205.

Estes, R D 1991, The significance of horns and other male secondary sexual characters in female bovids. Applied Animal Behavior Science 29: 403-51.

Estes, R D 1999, The Safari Companion. Chelsea Green, White River Junction, VT.

Klingel, H 1967,. Soziale Organisation und Verhalten freilbender Steppenzebras (Equus quagga). Zeitschrift für Tierpsychologie 27: pp. 580-624

Klingel, H (in press). Equus quagga (in) Kingdon, J S and Hoffmann, M: The Mammals of Africa Vol. 5: Equids