milewski's Journal

The impala (Aepyceros melampus) has normal proportions for an ungulate, but its body movements are odd compared to antelopes and deer of similar size and shape.

Many types of antelopes and deer stot, but the kick-stotting of the impala is surprisingly different from gazelles or the kob (Kobus kob), which ecologically replaces the impala North of the equator. As it runs, the impala flings both its hind legs high in unison - in some cases so high that it seems to risk somersaulting - while waving its tail high as well. Few naturalists have observed kick-stotting in response to the approach of predators, possibly because this gait is reserved for the African hunting dog (Lycaon pictus). When charged by most types of predator, the impala neither stots nor raises its tail as it flees, which means that by far the most instances of kick-stotting photographed so far have been in social play, which is rehearsal rather than the real, life -or-death purpose of this gait.

The impala is surprisingly reluctant to trot, which is a normal gait taken for granted in most unguligrade and digitigrade mammals, and in some of them exaggerated into a form of stotting called 'style-trotting'. I have noticed that one of the few times when the impala trots is in slowing down to a halt after a bout of kick-stotting.

The impala is one of the few species of bovids or deer that is inept at swimming. This was first noticed during the rescue operations of wildlife stranded on islands in the rising waters of Kariba Dam. It seems odd that even desert gazelles, which spend their whole lives without encountering a river, can swim more competently than the impala, which often lives along river banks where it must risk being chased into the water by predators.

The impala is capable of standing and even (in the case if the courting male) walking bipedally. However, it seems unwilling to rise on its hind legs to forage, even in drought when the only remaining food is high on branches.

The impala is renowned for its bounding. However, this gait is not as distinctive as it may seem, for the ecological counterpart in India. the blackbuck (Antilope cervicapra), is similarly accomplished in bounding high and far.

Finally: even in the case of lying down to chew the cud, the impala seems odd. Whereas most other antelopes and deer - and even giraffes - are easy enough to spot lying down by day, the adult impala tends to remain standing during its midday rest, reserving its recumbency for the secrecy of night, which it tends to spend in certain open places away from vegetation. Perhaps this explains why photographers so seldom capture the impala in a lying position?

Antilopin bovids range widely in dry climates in Africa and Asia, but their greatest concentration of genera and species occurs in arid to semi-arid northeastern Africa (Somalia, Somaliland, Djibouti, Eritrea and parts of Sudan, Ethiopia and Kenya). In an area the size of France plus Spain located just north of the equator, seven species of gazelles, the beira (Dorcatragus megalotis), and nine species/subspecies of dikdiks (Madoqua) occur. This is more than all the species of bovids, antilocaprids and deer in the whole of the United States of America.

One clue to the reasons for such diversity is the poverty of antilopins in the similarly extensive arid to semi-arid climate in southern Africa, just south of the tropic of Capricorn, where only two species of antilopins occur in the relevant parts of Namibia, Botswana and South Africa. These are the springbok (Antidorcas marsupialis, which is a type of gazelle) and the steenbok (Raphicerus campestris, which is intermediate in size between dikdiks and the beira).

That is a dozen species versus a couple , in ostensibly similar environments on the same continent.

The antilopins of arid to semi-arid northeastern Africa include some of the most peculiar and specialised of ruminants. The gerenuk (Litocranius) has an exceptionally small face, an unrivalled ability to free-stand upright on its hind hooves, and such extreme economy of water that it refuses to drink even when raised in zoos. The smallest species of dikdik is the smallest of all ungulates, worldwide, that live in semi-desert. And Speke's gazelle (Gazella spekei) has an oddly inflatable nose even compared to its closest relatives within the same genus.

Both parts of Africa have similar mean annual rainfall, but a crucial difference is this. Whereas the dry parts of southern Africa have a single rainy season each year, those of northeastern Africa tend to have two rainy seasons each year owing to the East African Monsoon. This means that plant growth tends to be more reliable in the northeast than in the southwest of the continent, allowing greater specialisation of body sizes and shapes, diets, and foraging heights, and thus ecological niches. Whereas in general arid climates mean that 'beggars cannot be choosers' and survival depends on versatility, the antilopins of northeastern Africa include some of the choosiest species of ungulates known with respect to type of terrain, vegetation and diet.

There have been many historical attempts to domesticate both sheep (Ovis) and gazelles (Gazella). What many naturalists may not realise is that several types of ostensibly wild gazelles seem to have originated, at least partly, by selective breeding in captivity.

The species/subspecies bilkis, dareshurii, erlangeri, farasani, hamishi and muscatensis seem never to have been found in wild populations, although three of these now occur free-range on small islands in the Red Sea or Persian Gulf. I suspect that all are anthropogenically modified variants of the mountain gazelle (Gazella gazella), transported by humans to their locations and at least partly bred in captivity in the past. Even in its original range (in the Levant), the mountain gazelle has long lived somewhat commensally with tolerant farmers because no truly wild situations have remained in the Biblical Lands for hundreds of years.

The main effects of quasi-domestication in the above gazelles seem to be darkened colouration and a reduction in the size of the brain.

Four North African species (cuvieri, dorcas, pelzelni and leptoceros) seem to have been kept in captivity, for example in oases in Tunisia and elsewhere in the Maghreb, for thousands of years. Probably in most cases the animals were hand-reared after being caught as infants, kept as pets, and not selectively bred. However, an odd aspect of Cuvier's gazelle, in addition to its overall darkness, is that in some individuals there are irregular whitish markings on the face, reminiscent of the asymmetrical colouration so often produced inadvertently by domestication.

Certain traditional populations in India have long cared for the chinkara (Gazella bennettii), which lives somewhat commensally as well as often being raised as a pet. However, no aspects of colouration suggest that this species has been modified by this relationship.

Why did domestication of gazelles prove to be so unsuccessful that that most naturalists assume gazelles to be purely wild animals? Possibly because all gazelles, unlike all wild sheep, have a territorial social system, which limits their amenability to herding. None of the twelve species of domestic hoofstock originating in Eurasia originate from territorial ancestors.

Which leaves us with an odd thought. Had things turned out differently and Gazella domestica arisen in place of sheep, would the Bible have spoken of gazherds rather than shepherds?

In my last post I mentioned the puzzle of apparent protective mimicry of the impala by the gerenuk. Here I summarise the similarities and differences in colouration.

Both species have dark fawn on the dorsal surface of the torso, giving way to paler fawn on the flanks and then white in the belly, the borders between the zones being oddly crisply defined. The main difference is that in the gerenuk the back-flank border is so emphasised that it appears like a pale horizontal stripe in its own right. Only the impala possesses a dark spot of bare skin at the stifle-fold, but this is mere punctuation rather than a 'capital letter' in the typeface the animal presents to predators.

The faces have patterns so similar in detail that it seems unlikely that such convergence would have evolved merely by chance. The back of the head differs, in that only the impala possesses dark posterior ear-tips and a sheeny-haired crown which switches from fawn to whitish in some lights. However, these differences would not be visible to any predator scanning the stationary figure as the alert antelope faces the intruder.

The legs have similar colouration except, for example, that the dark gland-tuft is on the foreleg of the gerenuk, vs the hindleg of the impala. It is true that only the impala has white pasterns; but this is invisible unless the animal is standing on bare ground or the most closely-cropped of lawns.

As for the hindquarters, the patterns are different in detail but give a similar overall impression of inconspicuous vertical bars of whitish near the tail. Only the impala possesses dark pygal stripes, but this presents as a bar-code rather than a flag.

Both the gerenuk and the impala have tails surprisingly unlike those of other antelopes (including other gazelles in the case of the gerenuk) or deer. They differ in that the terminal, insect-swishing tassel is white in the impala vs blackish in the gerenuk. However, this is hard to see when the tails of apprehensive animals are held still and, in the case of the impala, tucked between the legs.

The bottom line is this: there are many detailed differences in colouration between gerenuk and impala, but none large-scale enough to distinguish the two species in an obvious way unless viewed at close quarters. Any human observer with any experience can easily tell them apart even at distance, but this is because the neck of the gerenuk is so long, and its face small to the point of near-absurdity.

So how do wild predators perceive them, and of what adaptive value are the convergences in colouration?

For some strange reason, the gerenuk (Litocranius walleri) and the impala (Aepyceros melampus) have similar colouration. This is true despite the fact that the two species are unrelated phylogenetically and differ ecologically, seldom occurring together.
.
The gerenuk is specialised to forage with upright bipedality, can forgo drinking for years on end, and tends to be solitary. By contrast, the impala has not been observed even to prop its forelegs on a plant stem, must drink nearly daily, and is gregarious. In the narrow zone where the two species share the same landscapes in Kenya, the gerenuk prefers thorn scrub while the impala prefers grassland.

Perhaps the gerenuk has come to mimic the colouration of the impala for protection against predators. The impala tends to be common where it occurs, while the gerenuk is everywhere scarce. Based on the likelihood that the gerenuk is not as enduring a runner as the impala, a naive predator might be misled to turn down hunting opportunities after spotting the gerenuk.

However, there are obvious problems with this explanation. The gerenuk is so much lankier than the impala that the colouration could hardly disguise its identity. Most of the distribution of the gerenuk, on the Horn of Africa, is hundreds or thousands of kilometres from the nearest impala. And there is scant convergence between the two species in their behaviour in alarm. For example, the gerenuk habitually trots while the impala is remarkable in how reluctant it is to trot; and the stotting behaviours, in display to scanning predators, are extremely different.

Here is a frontier of understanding in Biology, where new hypotheses are needed. I can help by describing in detail the subtle similarities and differences in colouration, which will be the topic of my next post...

One would not expect the gerenuk (Litocranius walleri) to be particularly frequently photographed. It lives in remote areas, its populations are sparse, it is shy, and its appearance is rather dull apart from a graceful lankiness . In contrast to other gazelles, it is seldom successfully kept in zoos.

In the sixties and seventies, few photos of the gerenuk were available. Pierre Dandelot and Helmut Diller, painting the species for the best field guidebooks of the seventies and eighties, erred considerably in their depictions, presumably because they had little material to examine.

I lived in Kenya for about six years in the eighties and nineties, and toured that country fairly widely. Yet I spotted the gerenuk only a few times.

So imagine my surprise when I recently undertook a search for photos on the Web, and found them to be innumerable. The gerenuk is surpassed only by the blackbuck (Antilope cervicapra) as the most frequently-photographed gazelle.

In the case of the blackbuck, popularity is understandable. This gregarious and spectacular species still occurs widely in India, where the rising tech-savvy Middle Class, with a typically Hindu love of animals modernised by colonial English influence, has produced more amateur photographers than expected in an underdeveloped country. The blackbuck is one of the most successful of the gazelles in zoos worldwide. And introductions to the USA and Argentina, where the species is commercially hunted, have so added to the opportunities for photography that there are more photos on the Web of the blackbuck in Texas alone there are of most species of gazelles in any situation.

None of the above explanations apply to the gerenuk. Possibly this species has become commoner in the last few decades in a few parts of its range, such as the private game ranches of the Laikipia region of Kenya. But it remains manyfold less likely to be encountered by photographers than the extremely photogenic Thomson's gazelle, which is abundant in the Serengeti and can be found right on the outskirts of Nairobi itself. So I find it noteworthy that, when I recently scoured the Web for photos of the nominate subspecies of Thomson's gazelle (which is the one occurring in Kenya except for the Mara), I found so few photos that I could not complete my research. To this day I remain unsure of subspecific differences in Thomson's gazelle.

One obvious appeal of the gerenuk is the inbuilt 'anthropomorphism' of its extreme bipedal ability. People may love penguins and the gerenuk for similar reasons. But is this enough to explain the proliferation of photos of the gerenuk?

Most gazelles (genera Gazella, Eudorcas, Nanger, Litocranius, Antilope, Ammodorcas and Antidorcas) are only moderately sexually dimorphic. The adult male is not strikingly larger than the adult female but possesses horns, or at least larger horns.

However, the blackbuck (Antilope cervicapra) presents an intriguing comparison with the gerenuk (Litocranius walleri). In both, the adult female is hornless and weighs about 30 kg, and the female colouration is the least conspicuous among gazelles. The dark flank-band typical of gazelles has been lost; there is a pale horizontal stripe along the upper flank; there is pale around the eye but the pale facial stripe and dark malar stipe, both typical of gazelles, are minimal; and on the hindquarters the bold effect of most gazelles is lost because the pygal band is minimal, the white on posterior haunch is restricted, and the tail-tassel is small/nondescript.

Despite this uniformity of females and juveniles, the mature males could hardly differ more. In the gerenuk the male has modest horns and feminine colouration except on the crown of the head. In the blackbuck the male grows extremely long, corkscrew horns; and his colouration is so converted into a whole-body dark-and-pale 'beacon' that little remains of the pattern of gazelles. Even the face becomes so showily dark-and-pale that it looks categorically different from that of the gerenuk and most gazelles.

This correlates with the ecological and social differences. The blackbuck is a specialised grazer which drinks frequently and concentrates in large groups, whereas the gerenuk is a specialised browser which can forgo drinking for years and is often solitary. Both species are territorial, but in divergent ways.

In the blackbuck, territories are so small, crowded and hectically defended that the competing males show off to each other for most of the time. They alternate this with visual appeasement, because they can forage only by trespassing their way to nearby, untrampled pastures, excusing themselves gesturally along the way there and back. In the gerenuk, the territories are so large that males seldom even see each other. Not only do they not need to trespass, but they only ever patrol a limited central part of the territory - using smell rather than sight.

What this means is that - despite females being so similar - males have social modes so different that the male blackbuck makes no attempt to hide from predators, whereas in the gerenuk the male remains thoroughly secretive.

And the male genitalia differ too, despite the fact that in both species the courting male following the female by walking bipedally, unsupported by the forefeet. In the blackbuck, the penis extends far forward; in the gerenuk it is the scrotum that instead seems shifted forward. It is understandable that the gerenuk avoids pressing the testicles between the hind legs while foraging upright. But the penile aberration in the blackbuck is only partly explained by the fact that this species is unusually touch-averse among gazelles -even in sexual ardour.

Stripes, whether pale or dark, can make animals hard to tell from their surroundings. However, horizontal stripes along the torso (as opposed to the classical vertical pattern seen in e.g. the tiger, Panthera tigris) are rare in large mammals. Among ungulates, the gerenuk (Litocranius walleri) - the lankiest of all gazelles and indeed of all antelopes or deer - is a noteworthy exception.

The gerenuk, in both sexes and at all ages from infancy, has a pale horizontal stripe running from the dorsal base of the neck across the flank to the rump. This is the major feature of what is otherwise an unusually plain colouration among gazelles. Presumably the stripe breaks up the figure, helping the gerenuk to hide in the sparsely woody vegetation it inhabits.

The only other gazelles with a similar stripe are the blackbuck (female and juvenile colouration of Antilope cervicapra) and some populations of the goitred gazelle (Gazella subgutturosa), typically in Azerbaijan. In the blackbuck the stripe is outweighed, in subspecies rajputanae, by the conspicuous ventral white which catches the light on the shoulders and above the elbow. Furthermore, the stripe is converted to a series of vague spots as it approaches the rump. In the goitred gazelle, the stripe is merely a local variation of what, in most populations, is a pale band; and either way it is hardly noticeable relative to the bold dark-and-pale of hindquarters which are far more conspicuously marked than in either the blackbuck or the gerenuk.

Horizontal pale stripes occur also in several genera of deer (Dama, Cervus, Axis) and in several species of the antelope genus Tragelaphus. However, in these animals they tend to be mere details of complex patterns of spotting and striping.

This raises the question of what relationship there might be between the unusual striping of the gerenuk and its extreme ability to forage bipedally. Although various species of deer and antelopes can stand on the hindlegs to reach high on plants, only the gerenuk can remain free-standing for minutes at a time without the forefeet being used as props on branches. And only the gerenuk rises bipedally so frequently that this seems to be its main posture in foraging.

What is intriguing is the idea that, in the upright stance, the stripe becomes vertical and thus tends to align with the main stems of the tall shrubs and short trees typical of the habitat of the gerenuk. Perhaps the gerenuk is unusually striped because its torso is so often viewed in the vertical.

Perhaps supporting this explanation is the observation that a slightly less lanky gazelle, the dibatag (Ammodorcas clarkei), lacks any stripe on the torso despite also foraging to some extent bipedally. A difference is the typical habitat of the dibatag is dominated not by 'acacia' (Vachellia species such as tortilis) but instead by sundry tall shrubs in a distinctive vegetation type on sand, called 'gedguwa'. I hypothesise that this form of 'open thicket' tends to be more cluttered with foliage at about one metre above ground than is the case in 'acacia scrub', leading to a subtle difference in visibility and thus in adaptive colouration of the gazelles in question.

The gerenuk (Litocranius walleri) is unusual in hiding more than other gazelles. Its secretive mature makes sense in view of its relatively densely vegetated habitat and its limited gregariousness.

The gerenuk has inconspicuous colouration because it has lost the extensive, eye-catching pale and dark on the flanks and hindquarters which are typical of gazelles. It has been frequently photographed despite its skulking habits and arid, remote habitat, because its extreme ability to forage bipedally has particular appeal in the human mind.

Oddly, the gerenuk retains the ability, typical of gazelles, to stot in display of its individual fitness. This takes the form of style-trotting and stiff-legged bouncing, complete with the flaring of the white fur on the posterior haunches (which has yet to be captured photographically). Two peculiarities are: 1) the tail, although proportionately long among gazelles, is hardly raised during stotting, and 2) emphasis is instead added by the folding of the legs at the height of each bounce.

Stotting is generally associated with prey species capable of particular speed and stamina, because it demonstrates to the scanning predator the likelihood that pursuit of the individual in question would be futile. The gerenuk has been assumed to lack speed and stamina, relative to most gazelles. This is based on its extreme lankiness, the clutter of tall shrubs and short trees in its typical habitat (which make for an obstacle-course rather than an open field of flight), and the inability or reluctance of the gerenuk to leap over plants in the way tragelaphins (e.g. kudus) do.

So, what is the adaptive advantage of stotting in the gerenuk?

One possible explanation is as follows. All large predators tend to be scarce and transient in the habitat of the gerenuk because of a lack of drinking water and the sparsity of other prey. This means that, often, the predator will be relatively naive with respect to the specific nature of the gerenuk, while nonetheless being dissuaded in a mentally 'hard-wired' way by any stotting displays observed.

What the gerenuk might be doing, in stotting, is to convert what is in other gazelles an eminently honest display of fitness into a deceptive display suggesting borrowed identity and false context. In short, confusing rather than informing.

Do readers have a better explanation?

I have pointed out, in a previous post, that the large animals of the Arabian Peninsula are consistently smaller than their closest relatives (usually of the same species or genus) on the mainland of Africa and/or Asia. Since the largest animals of islands worldwide tend to be small relative to comparable animals on relevant mainland areas, the Arabian Peninsula seems to act biogeographically like an island -which is surprising in view of its vast area and its broad connection to Asia.

Here I focus on what is perhaps the most complicated example of this pattern: the Arabian gazelle (Gazella arabica). This species, the adult female of which weighs only 10-15 kg, is restricted to the Arabian Peninsula and was formerly widespread on stony slopes with scattered thorny shrubs.

The other two species of Gazella of the Arabian Peninsula are easy to compare with mainland counterparts. Gazella marica is a diminutive (adult female 15-20 kg) version of Asian Gazella subgutturosa, and extinct Gazella saudiya is a slightly diminutive (adult female probably about 13 kg, making it the smallest of all gazelles) version of African Gazella dorcas.

What is odd about the Arabian gazelle is that it seems to be the diminutive counterpart of an antilopin belonging to a a genus restricted to Somalia and Ethiopia and containing only one species: the dibatag (Ammodorcas clarkei). The dibatag (adult female about 25 kg, with proportionately longer neck and legs than in any gazelle) is so different from the Arabian gazelle that their biogeographical relationship may not previously have been noticed. However, both the seldom-photographed dibatag and the frequently-photographed Arabian gazelle have the neck and legs so elongated that - particularly when they stand bipedally - they can reach higher than competing gazelles to eat from the scattered trees of the semi-desert.

The Horn of Africa, home to the dibatag, is like Arabia in projecting into the Indian Ocean. However, it does not qualify as a peninsula. Hence - so the hypothesis goes - the dibatag can afford to be specialised in body-form, as well as relatively large.

In the conventional view, the Arabian gazelle is merely a gracile-looking southern replacement for the mountain gazelle (Gazella gazella) of the Levant, with which it continues to be taxonomically confused. But another, perhaps more mentally stimulating, view of the Arabian gazelle is as a peninsular ecological counterpart for the dibatag of Africa.
.