Recent confirmation of the ability of the African savanna buffalo to swim

@ludwig_muller @beartracker @davidbygott @tonyrebelo @matthewinabinett @tandala @oviscanadensis_connerties @paradoxornithidae @dejong @jeremygilmore @capracornelius @jandutoit @kristenmartyn @johuisamen @j-j @nyoni-pete @niallp @jmugleston @pd_pratt @nmoorhatch @mario_mairal @earley_bird @oudejans @ricky_taylor @simontonge @andrewdeacon @ryanvanhuyssteen @jorgejuanrueda @gareth_preiss @hermanviviers

Most mammals are able instinctively to swim ( and and,body%20axis%20and%20tail%20(ref. and and and

However, certain species are poor at swimming (e.g.

One of the differences between the ungulate faunas of Africa and tropical to subtropical Asia is the ineptitude of several African ungulates when it comes to swimming.

These include

(This is true, notwithstanding the fact that the African ungulate fauna contains species - particularly the hippopotamus ( but also e.g. a chevrotain ( - that are more specialised for living in water than is seen in any relative in southern/southeast Asia.)

Of the above forms, impalas are the most revealing, because there is no gross structural reason for their poor ability to swim (e.g. and and

Impalas seem to have partly lost the ability to swim that is normal for ungulates, based on changes at the level of neurological 'software', rather than the 'hardware' of their limbs. This means a 'psychological' rather than a morphological limitation.

Furthermore, it may be owing to the absence of hippos in Asia that rhinoceroses, bovines, and suids are attracted to water for foraging and shelter.

The hippopotamus not only fills an 'amphibious' niche in Africa, but is irascible and dangerous to other animals entering the water, including other ungulates ( and and and and and and and and and and and

Examples of Asian ungulates with a noticeable attraction to water include

Given that rhinoceroses differ so greatly (, between Africa and Asia, in their ability to swim:
Is the same true for bovines?

During the filling of Lake Kariba ( in 1958-1964, the African savanna buffalo (Syncerus caffer, was found to swim competently ( and see comment below).

However, this report seems to tread water, as it were, in the more general  pool of evidence.

There are thousands of photos of the African savanna buffalo in the Web and in print. Given this intensive coverage, immersion in water seems surprisingly scarce, even during wallowing ( and and and and and

Several photos show male individuals immersed up to the shoulders while foraging in marshes ( and

However, any wetting tends to be only partial, as if the African savanna buffalo lacks the trust of water seen in its Asian counterparts.

I suspect that this bovine avoids immersion, ultimately because of the occupancy of an ‘amphibious’ niche in Africa by that specialised and dangerous species, the hippopotamus.

This would be in line with waterbucks (Kobus ellipsiprymnus and Kobus defassa, - which are named after their affinity for water, but seldom enter water compared to their counterparts among Asian deer (e.g. and and and
Given all of the above, I found it noteworthy that several photos taken in Chobe National Park ( have fortuitously confirmed swimming in the African savanna buffalo ( and and
and and and

These photos give the impression that this bovine is out of its element while swimming. I do not know if this awkwardness is owing to a psychological limitation, or particularly heavy bosses on the head.

The African savanna buffalo, like impalas, qualifies as being able to swim. However, in both cases, swimming seems to be undertaken only reluctantly and apprehensively.

Publicado el junio 30, 2022 10:16 TARDE por milewski milewski


The following are notes from Child G (1968) Behaviour of large mammals during the formation of Lake Kariba. Kariba Studies, published by Trustees of the National Museums of Rhodesia.

Syncerus caffer, like Sylvicapra grimmia, is stated to be a strong swimmer.

Aepyceros melampus is a weak swimmer compared to other antelopes, such as Tragelaphus sylvaticus. It cannot swim as far as can the far smaller-bodied species, S. grimmia.

Raphicerus sharpei is a strong swimmer, considering its small body size.

By contrast, Oreotragus is reluctant to enter the water, swimming more slowly than any other bovid in the Zambezi Valley. When in the water, it attempts to use the same bounding gait that it uses on land.

Crocuta crocuta, Panthera leo, and Panthera pardus are strong swimmers, and take to water readily.

Loxodonta africana (even when only four months old), Kobus ellipsiprymnus (including newborns, with the umbilicus still present), Tragelaphus sylvaticus, and Strepsiceros strepsiceros are, likewise, strong swimmers, and take to water readily when threatened. The horns of males of K. ellipsiprymnus and S. strepsiceros are somewhat of a handicap while swimming.

Papio ursinus is apparently a strong swimmer.

Diceros bicornis is a weak swimmer, apparently finding it difficult to hold its head above water. In contrast to all Asian spp. of rhinoceroses, it dislikes deep water. It refuses to cross rapid-flowing streams deeper than 38 cm. However, one individual hid in water, near the shore of the lake as it filled.

Diceros bicornis has small home-ranges (e.g. 4-6 square miles), partly because it needs to drink. "Black rhino are probably susceptible to abnormal floods where these affect the greater part of their home range, even if the water is only a foot or so deep. A female and calf drowned in this much water within a few days of an island submerging, indicating it is essential for rhino to lie down, as they often do, when sleeping."

Equus quagga chapmani is a weak swimmer, second only to D. bicornis. I find this remarkable, because it is well-known that Equus quagga boehmi is a competent swimmer across the Mara River in the Serengeti ecosystem (

Potamochoerus larvatus swims readily, second only to K. ellipsiprymnus. It swims strongly despite its large, heavy head.

Phacochoerus africanus is stated by Child (1968) to be a strong swimmer, which may possibly be contradicted by other authors.

Publicado por milewski hace más de 1 año
Publicado por milewski hace más de 1 año
Publicado por milewski hace más de 1 año
Publicado por milewski hace más de 1 año

I have an impression that the nostrils are differently-placed in the African savanna buffalo vs the river buffalo.

In Syncerus caffer, the rhinarium is rounded, and does not seem flexible in a dorsal direction:

In Bubalus bubalis, the rhinarium seems flexible, so that lifting of the nostrils keeps them above the surface of the water:

Syncerus caffer:

Bubalus bubalis:

Publicado por milewski hace más de 1 año

The impala (Aepyceros elampus) seems to differ from its Asian counterparts in their ability to swim.

The chital (Axis axis) is a strong swimmer ( and and

The blackbuck (Antilope cervicapra,, which tends to live far from rivers, does not seem to have been recorded swimming.

However, most other gazelles swim competently, suggesting that the blackbuck swims more strongly than does the impala.

Publicado por milewski hace más de 1 año

Unusually clear illustration of the relative body sizes of Homo sapiens and Ceratotherium simum simum:

Publicado por milewski hace más de 1 año


Camelus dromedarius ( is able to swim. This behaviour has been recorded in both Oman and India, and is associated with foraging on mangroves:,vid:j3DpKZA2378

The alpaca (Vicugna pacos, has an affinity for shallow water, and may possibly be able to swim:,vid:KoeoZ9a3g7s

The llama (Lama glama, seems to be a weak swimmer:

Publicado por milewski hace más de 1 año
Publicado por milewski hace más de 1 año

@milewski The following link,, is an image of ""swimming"" Camelus bactrianus. Who knew they could 'swim so well' when the related natural form (Camelus ferus) was (& still is) confined to the deserts of Dzungaria & the Tarim Basin, & with feral populations of the Camelus bactrianus present in the Transoxiana/Sogdiana & Kirghiz Steppe. Those aforementioned places are highly arid & very mountainous (when it comes to the Pamirs & Badakhshan), but I suppose rivers are present so they would have no trouble experiencing & traversing water. It's definitely worth noting (like you have pointed out) that the Llama (Lama glama) is a weak swimmer, perhaps being confined to montane areas & being domestic may have impaired it's swimming capabilities.

Publicado por paradoxornithidae hace más de 1 año

@milewski Inspired by your comment on Camelids swimming, I decided to find media showing the Lama vicugna (vicuña) swimming &/or traversing water, & apparently there's no media of that.

Publicado por paradoxornithidae hace más de 1 año


Many thanks for your comments, which add so richly to iNaturalist.

Camelidae are indeed remarkable w.r.t. swimming ability, and our knowledge is still in its infancy. However, I note the following.

Camelus bacterianus and Camelus dromedarius interbreed freely, and might legitimately be regarded as a single, ecotypically plastic, species. Thus, given that we know that C. dromedarius swims well, it seems quite plausible that C. bactrianus also swims well, despite having so little opportunity/need to do so in its habitat that most populations and generations never actually experience the act of swimming.

The wild ancestor of C. dromedarius seems to have lived in a small area of the eastern coast of the Arabian Peninsula, where its natural habitat combined an arid climate with the availability of mangroves as forage. So, it seems to make sense that it evolved an ability to swim as part of its ancestral dependence on commuting among patches of mangal vegetation. I.e. it is not necessarily as incongruous as it first may seem, that this arid-adapted camelid swims well.

One might well explain the poor ability of South American domestic camelids to swim in terms of the lack of opportunity/need to swim in the Andes. However, Lama guanicoe is known to be a competent swimmer.

In the case of Vicugna the situation is more complex and interesting than at first it might seem. This is because, until ten thousand years ago, Vicugna occurred in the pampas of Buenos Aires province (see page 42 in, where its habitat was presumably marshes. To this day, Vicugna in the Andes likes to have its feet wet as it forages in high-altitude marshes (

This proclivity is noteworthy because a member of the same guild, in the pampas, was Hydrochoerus (which survives there to this day,

What I infer is that vicugna and capybara shared the same grasses and sedges in the pampas, but differed in that the former relied on running for safety, whereas the latter relied on swimming. This raises the possibility of two forms of similar body mass, ecologically similar in many ways in being semi-ungulates, but diverging in anti-predator strategy to the degree that one was incapable of swimming while the other is more aquatic than any comparable ungulate other than Hippopotamidae.

Alternatively, it remains possible that Vicugna actually can swim as well as has been shown in L. guanicoe.

Publicado por milewski hace más de 1 año
Publicado por milewski hace más de 1 año

I have considered what you have written in relation to Camelus being a monotypic taxa, or perhaps a species complex, since you have said the Camelus bactrianus and Camelus dromedarius possibly represent a 'single, ecotypically plastic, species'. If it's the case that they are a single species, would the Camelus ferus (wild bactrian camel) be included as apart of it? According to such a scheme, would all three species of Camelus consequentially become ssp. instead of spp.?

Publicado por paradoxornithidae hace más de 1 año


Many thanks for questioning the taxonomy of Camelus.

By the way, a similar taxonomic problem applies, even more, to the American camelids. To this day it remains unknown whether a) Lama glama is derived from Lama guanicoe or from some other, now extinct, wild ancestor, b) Lama pacos is derived from L. guanicoe or some other wild ancestor, and c) L. pacos is derived from intergeneric hybridisation between Lama and Vicugna (please see Bonavia 2008,

What this means is that the taxonomic status of all four domestic spp. of camelids on Earth remains unresolved.

In the case of Camelus, it is indeed possible that we are dealing with four subspecies of a single, ecologically plastic species, viz. a) the wild form of two-humped camel, b) the domestic form of two-humped camel, c) the extinct wild form of one-humped camel, and d) the domestic form of one-humped camel.

The main reasons to view Camelus this way are a) two- and one-humped forms interbreed easily, b) the two-humped forms are adapted to cold, and not merely different in geographical distribution, c) there has been no geographical disjunction between the two domestic forms.

If it is true that the original range of the wild (ancestral) one-humped form was limited to the eastern coast of Arabia, and ecologically peculiar in its combination of desert and mangal vegetation, then this is ambivalent. It seems arbitrary as to whether this supports viewing it as a subspecies, or a species (including its domestic derivative, viz. dromedarius) in its own right.

Furthermore, it remains uncertain that ferus is a truly wild form. It may be merely feral, which would mean that all truly wild, ancestral forms of Camelus are extinct (and unavailable even in the fossil record).

The main reasons to view Camelus as consisting of two species, one with two humps and the other with one hump, are a) despite the broad zone of intergradation in e.g. Iran, the two types remain polarised between the southwest (Arabia and northern Africa) and the northeast (the Asian interior), b) at their extremes, two-humped camels look quite different from one-humped camels, and c) most other domestic mammals (including the genera Equus, Sus, Bos, Bubalus, Ovis, Capra, and Canis) are also hybrids among several species of wild ancestors.

Indeed, it is possible that a fundamental technique, in mammalian domestication, has been interspecific hybridisation, hand-in-hand with selective breeding.

The bottom line: it seems arbitrary as to how to classify camels taxonomically. I do not envisage any objective resolution of this problem. And the most basic reason for this is that the species-concept has always been as much convention as reality, in the first place.

Your further thoughts?

Also please see

Publicado por milewski hace más de 1 año


Thanks for the excellent reply.

In relation to the factsheet for extant camels, the page mentions an "unclear evolutionary relationship between C. bactrianus, C. dromedarius, and C. ferus." This is quite interesting, as it could indicate they have readily introgressed since diverging, leading to an "unclear evolutionary relationship."

The factsheet also states, "Bactrian camels and dromedaries diverged about 8–10 million years ago (Cui et al., 2007; Franklin, 2011)." That's a significant difference in the phylogenetic sense, if they diverged that long ago... So if the given divergence date is correct, then they would qualify as being different genera, since that was the Miocene, and in that timeframe given by Cui et al. and Franklin, it would be in a time in which Australopithecus had not yet appeared and Sahelanthropus had just appeared.

I found a source that contradicts Cui (a bit strange how the author has my surname) in the wikipedia article for Camelus, and it's written, "However, the fossil record suggests a far more recent divergence between the Bactrian camel and the dromedary because, despite a moderately rich fossil record of camelids, no fossil that fits within this divergence is older than the middle Pleistocene (about 0.8 Ma)." This implies the separation is 1 mya or less. Therefore, it can definitely provide support to the idea of there being one "ecotypically plastic" species in Camelus. Also, there is a significant amount of contradictory information on that factsheet when compared to the most current revision of the wikipedia article ( regarding the Camelus genus.

Publicado por paradoxornithidae hace más de 1 año


Lama guanicoe swims well.

Bonavia (2008,, on page 26, states:

"They are also skilled swimmers (Arbocco Arce 1974:10; Gilmore 1950:450), as Darwin (1921:238, 1969:172) had already noted: 'The guanacos readily take to the water: several times at Port Valdes they were seen swimming from island to island.' "

This suggests to me that any difficulty experienced by Lama glama, Lama pacos and Vicugna vicugna, in swimming, may be owing to their wool, rather than to a basic inability.

'Port Valdes' presumably refers to the Valdes Peninsula (, on the eastern coast of southern Argentina.

Here is footage ( of L. guanicoe wading, and probably also swimming, at a different location, namely Torres del Paine.

Publicado por milewski hace más de 1 año


The following are excerpts from pages 45-48 of "The Zambezi: River of the Gods" by Jan and Fiona Teede (third edition, 1990), Ruscombe Books, Zimbabwe ( and

The topic is Operation Noah, Lake Kariba (

"At first the rescuers attempted to force the animals to swim by banging dustbin lids and hurling thunder-flashes among them. Antelope such as bushbuck ( escaped into the water and lay still, their noses just above the surface...Baboons ( were uncooperative. Under pressure they dog-paddled to the mainland, but returned to their islands as soon as the men had gone. Waterbuck ( were observed swimming with their young clinging like jockeys to their backs. The adults helped each other. When one was tired, it would rest its head on another's back until it felt strong enough to assist the first in a similar manner. Impala, the most numerous of the antelope, refused to swim... Despite the feline distaste for water, it was found that lions and leopards could swim well when no other option presented itself. Some swam from island to island, dispatching the easy pickings they found there. Elephant and buffalo could swim to the mainland, but rhino ( remained stubborn to the last. One rhino cow became known as Greta Garbo, because of her truculence and reclusive habits. Refusing to budge from her island, despite repeated harrassment by the rescuers, she would chase them into the lake...Three cow elephants with calves refused to move from their island, knowing that their young ones would be incapable of swimming to the mainland...The mothers killed their babies before departing, rather than leaving them to drown or starve. The battered remains of the calves were found by the water's edge...Monkeys ( were found to be natural swimmers. They would dive and swim breaststroke to depths of over five metres, their rescuers unable to pursue them. Porcupines ( were equipped with a life jacket in the form of quills, and antbears ( were almost as buoyant. Both swam slowly but confidently."

Publicado por milewski hace más de 1 año

most interesting and remarkable @milewski

Publicado por paradoxornithidae hace alrededor de 1 año

Agregar un comentario

Acceder o Crear una cuenta para agregar comentarios.