Skip to main content

The remarkable life appearance of the Woolly rhino

One of the Pleistocene mammals depicted without fail in popular books – encyclopedias of prehistoric life and the like – is the Woolly rhinoceros Coelodonta antiquitatis (the species name is written antiquus in many sources).

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


One of the Pleistocene mammals depicted without fail in popular books – encyclopedias of prehistoric life and the like – is the Woolly rhinoceros Coelodonta antiquitatis (the species name is written antiquus in many sources). Originally named in 1807 (but known for some time prior), this cold-adapted, shaggy-coated rhinocerotid rhino occurred from the Atlantic fringes of Europe all the way east to Beringia, and as far south as the southern Caucasus and south-east China. Why it never moved into North America is unknown – it should have.

As is the case for various ‘Ice Age’ megamammals, the Woolly rhino wasn’t necessarily an inhabitant of freezing cold places with blizzards and thick snow on the ground, or even of tundra-dominated habitats. Spanish specimens come from dry, temperate habitats dominated by grasses and broadleaved trees. Fossils of other Coelodonta species show that this group originated in the Tibetan region during the Pliocene, their evolution perhaps driven by the uplift of the Qinghai-Tibet Plateau (Deng 2002, Deng et al. 2011). C. thibetana from Tibet is currently the oldest known member of the group. Previously, the oldest Coelodonta species was the animal usually called C. nihowanensis. It’s been argued that this name is not available as it’s a nomen nudum (a name lacking a type specimen). A Middle Pleistocene rhino, C. tologoijensis, is known from Transbaikalia, Mongolia and perhaps south-western Siberia. These older species are smaller than C. antiquitatis and have more slender limb bones (Kahlke & Lacombat 2008). [Image below by Atirador.]

Plant fragments stuck in Woolly rhino teeth (most typically inside the infundibula – the crescent-shaped recesses present in the middles of the molars) show that they were grazers, 96% or so of their diet being made up of grasses, with mosses and forbs forming the remainder (Guthrie 1990). However, preserved stomach contents show that Woolly rhinos also ate dwarf willows and birches.


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


The form of the Woolly rhino’s skull and teeth are in agreement with this grazing lifestyle: the mouth and lips are broad, and the head dipped downwards towards the ground, even when the animal was in its normal, relaxed posture. The skull is unusual in having both an extensively ossified nasal septum, and a down-turned anterior region on the premaxilla that contacts the edge of the upper jaw. As a consequence, it has ‘reverted’ to the possession of enclosed bony nostrils – a condition present in mammal ancestors but not, ordinarily, in mammals themselves. The snout region is also unusual in that incisors are wholly absent in both the upper and lower jaws. This is presumably an extreme specialisation for grazing and raises the question as to whether Coelodonta had keratinous pads, or some other, similar structures, in these parts of the jaws.

What does Coelodonta look like in popular depictions? Artists have most usually shown it looking just like a modern Black or White rhino dressed in a brown furry coat. I’ve never been satisfied with these sorts of reconstructions since they ignore the many nuances of Woolly rhino anatomy, some of which have been obvious since the 1870s at least.

As a child of 10 years old or so, I recall one of the most memorable specimens at London’s Natural History Museum being their cast of the Polish Starunia mummy, preserved in a petrochemical seep and discovered in 1929 (Nowak et al. 1930). The cast at London used to be part of a fossil mammal display, so close that you could touch it. It’s still on display at the NHM today, but it’s tucked away in an overhead recess near the museum’s side entrance. You can’t get close to it and I doubt that anyone who isn’t a hard-core palaeozoology uber-nerd would even know what it is.

What ancient people depicted

Woolly rhinos are not depicted all that frequently in Palaeolithic cave art. They’re massively out-numbered by illustrations of mammoths, bison and horses (Guthrie 2005). In fact, until recently, only about 20 Woolly rhino images were known (a lesser number of portable depictions of Coelodonta are known as well) (Bahn & Vertut 1997). However, this number has more than doubled thanks to the discovery of the remarkable Chauvet Cave in France, where there are about 60 rhino images.

Prehistoric artwork frequently shows the rhino’s head dipping down towards the ground as suggested by the anatomy. The majority of depictions show the rhinos as two-horned: the posterior (or frontal) horn is sometimes shown as being shorter or very much shorter than the anterior (or nasal) horn, but sometimes the two are about the same length, and sometimes (as in a painting from the Cave of Les Combarelles, Dordogne) the posterior horn is shown as being slightly longer than the anterior one. As discussed below, preserved horns show that the rhinos really were this variable in horn proportions. Of course, the possibility exists that some of the cave art was inaccurate – we know that ancient artists were often brilliant and faithfully accurate illustrators of the animals they saw, but they were virtually always doing things from memory and did make mistakes.

All Woolly rhino images consistently show a massively deep, hugely convex shoulder hump that extends all the way forward to the back of the head. The hump is so massive that, when the animal is depicted in a grazing pose, the anterior edge of the hump is sometimes shown overhanging virtually the whole of the face. Limb length and how close the body is to the ground varies quite a bit. Some illustrations show the limbs as being proportionally tiny, the belly being almost in contact with the ground, and a long, shaggy fringe of fur along the ventral surface of the belly. The majority, however, do not show this, and instead the belly is well up off the ground and a ventral hair fringe is missing. Perhaps the coat was variable according to season. What about colour? We’ll come to that in a minute.

More on horns: short and stout vs long and flattened

As just noted, we know quite a bit about horn morphology in this animal: both cave art and some remarkable permafrost specimens from Siberia show that Coelodonta was twin-horned, with a gigantic, gently curved anterior/nasal horn and an often shorter or slimmer posterior/frontal one (Fortelius 1983, Shidlovskiy et al. 2011). Only two mummified specimens have both horns in position, but a large number of isolated horns are also known. Only seven or so horns had been reported prior to the 1990s but we now know of over 40 horns that are variable in shape and size. The horns were originally thought to be the detached claws of gigantic birds, in particular those of the Siberian mythical super-bird Pyne who battled a sea monster and a giant fish (Fortelius 1983).

The posterior horn is often short, squat and sub-circular in cross section, with a strongly pointed apex. The anterior horn is usually far longer (about four times longer), but it isn’t a subconical structure, rounded in cross-section, like the anterior horns of living rhinos. Rather, it’s laterally compressed, lenticular in cross-section and with a triangular cross-sectional shape to its anterior border. This cross-section was seemingly produced as wear facets were formed along both sides of the horn’s leading edge (Fortelius 1983).

The horn is in fact so flattened from side-to-side that it has sometimes been described as plank-like: during the 1760s, naturalist and explorer Peter Simon Pallas proposed that this flattened form was artificial and caused by people cutting material away from the sides, but this is not correct. The erroneous idea that the flattened form results from post-mortemdistortion has also been mooted. Some authors have also suggested that the anterior horn must have been too fragile to be used in fighting (while this might seem intuitively reasonable, intuition is often wrong on such matters), and the inevitable idea that it was used for scraping snow aside has been fairly popular.

Regularly spaced, dark bands along the horn (closely spaced at the base but well separated towards the tip) were presumably visible in life. These have sometimes been suggested to be annual growth bands, in which case they show that Woolly rhinos lived on occasion to over 30 years, a lifespan about similar to that of modern rhino species. However, Shidlovskiy et al. (2011) noted that dark bands of this sort might be more to do with the rate of melanin deposition and may not be annual at all.

The anterior horns that have been found so far vary somewhat in proportion and detail. Some are longer and more slender than others, and have a blunter tip. It has been inferred, on the basis of comparison with living rhinos, that these longer, slimmer horns were those of females (Guthrie 1990). However, Fortelius (1983) thought it more likely that Woolly rhino reproductive strategies would have been based around short, intense rutting seasons and that this might have driven the evolution of larger body size – and greater horn length – in males. Modern rhino horns are used in intraspecific combat and in fighting predators, but also in foraging and especially in breaking branches. Indeed, it has been suggested that horn elaboration in rhino evolution was driven by their use in foraging, an intriguing hypothesis that needs more study.

Patterns and pigmentation

What do we know about pigmentation in the Woolly rhino? Thanks to cave art, we have excellent, detailed and apparently very accurate information on the life appearance of quite a few Pleistocene animals. Sometimes, this information confirms what we already suspected but, on other occasions, ancient art surprises us. Prehistoric depictions of the giant deer Megaloceros and of some ancient reindeer, horses and leopards, for example, show that these animals did not look like the plain versions shown in the majority of reconstructions. European cave art depicting Coelodonta shows it as dark, but with a massively thick, near-black band completely encircling the animal’s middle. The band is depicted with care, “sometimes engraved at the edges and filled with flat wash” (Bahn & Vertut 1997, p. 153).

The breadth of this band is variable. In some images, it’s narrow and belt-like and restricted to the middle part of the body or to the region just anterior to the pelvis. In others, it is huge, extending all the way across the body from the hips to the shoulders.

Suggested explanations for this band include that it might represent vertical folds in the skin, and a tongue-in-cheek suggestion that it might represent a saddle has also been made. While it’s perhaps plausible that the band could be a symbolic feature of some sort, it seems most reasonable to conclude that it really was a genuine anatomical feature of this animal: that is, that these rhinos had a very dark, almost black, band of pigmentation wrapping around their middles. The fact that the bands are only present in the cave art of western Europe might, Guthrie (2005) suggested, show that it was unique to the population (subspecies?) of Woolly rhinos present here.

Subspecies of C. antiquitatis have, incidentally, been named: a Middle Pleistocene specimen from France was regarded as the type of the new taxon C. a. praecursor. Like the Asian C. tologoijensis and other more archaic Coelodonta taxa, it has a proportionally broader skull, more anteriorly located orbits, and more curved zygomatic arches than the classic Woolly rhinos of the Late Pleistocene. However, Kahlke & Lacombat (2008) removed C. a. praecursor from C. antiquitatis and synonymised it with C. tologoijensis.

Dark transverse bands do have a precedent in the living world but they’re rare and I can’t think of any examples that quite resemble the condition reconstructed for Coelodonta. For this reason, a bit of scepticism is warranted. Guthrie (2005) pointed to several living animals that have such structures (including cattle and guinea-pigs) but most are domestic and hence not subjected to natural selection pressures. He also pointed to the Giant panda Ailuropoda melanoleuca and Tanuki Nyctereutes procyonoides as examples: however, their bands are either incomplete (Tanuki), or restricted to the shoulder region, not the middle part of the body.

What else do we know?

What else do we know about the life appearance of this rhino? The pelt of the Woolly rhino was long, consisting of both light, fine strands as well as coarser, darker guard hairs. The skin (preserved on the Starunia carcass) is covered with small, abundant protuberances (Nowak et al. 1930). The hair follicles vary in size according to the hairs they supported, with those on the back of the neck being largest: a fact consistent with the presence of a mane. Three or four low humps formed from skin and fat – not all of which correspond with the underlying neural spines – are present on the middle part of the neck and shoulder, as they are in living white rhinos.

The lips are broad, vaguely rectangular (with rounded anterolateral ‘corners’), and thus, again, something like those of living white rhinos. The upper lip seems to extend over the lower one in some specimens and it appears to have been highly mobile, though has been some contention as to how far the upper lip extended anteriorly: some authors have argued that it overlapped the lower one entirely and was unusually prominent, perhaps recalling that of the Moose Alces alces (there a 1924 article on this, I believe by Hilzheimer, but I’ve been unable to find the full citation). Others have contested this, stating instead that the lips simply fitted tightly together (Nowak et al. 1930).

Mummified specimens reveal that the tail is proportionally short, as is predicted for cold-adapted mammals. The tail is also flattened, naked on the ventral surface of the distal half or so, and with long hairs around its edges and at its tip. The ears are far narrower than those of living rhinos – they’ve even been described as lanceolate in form. The idea mentioned above that Woolly rhinos couldn’t fight with their horns is disproven by cave art which clearly shows the animals fighting. Chernova & Kirillova (2010) looked at the horn’s internal microanatomy and concluded that the horns are especially resistant to fracturing. The unusual ossification of the nasal region might also be explained by vigorous use of the anterior horn in combat (Shidlovskiy et al. 2011).

Guthrie (2005) noted that few cave art image of rhinos show spears or arrows embedded in the body (a handful do). This might indicate that rhinos were rarely hunted. With numerous horse, deer and bison around, we might predict that people rarely hunted these giant, formidable, thick-skinned animals. A rhino at Chauvet seems to be shown with blood gushing from its mouth and nose while one at Colomière seems to have several arrows projecting from its belly. A few other French examples look like they have spears stuck in the side of the body (Guthie 2005).

All in all, we clearly know a lot of detail about the appearance of this animal and artists who choose to depict it should clearly incorporate these numerous details. All too often, they have not: you can see inaccurate horn and lip shapes in numerous reconstructions, and many don’t incorporate the apparently genuine details of pigmentation that we know about, the dark body band in particular. And I’m becoming increasingly frustrated with palaeontologists who are paid to be consultants for palaeontological artists but who don’t know, or – better – don’t care about getting things right. Here’s a Woolly rhino cheat-sheet, compiled by combining everything we know...

I really should write about the life appearance of other Pleistocene megafauna at some stage. And one day I will! For previous Tet Zoo article on rhinos, see…

Note that many of the papers listed below are available at the amazing Rhino Resource Centre. Remember also to donate to the RRC if you can, and to support rhino charities and to spread the word about the obscene destruction of rhino populations currently fueled by the absurd interest some members of our species have in obtaining rhino-horn products.

Refs - -

Bahn, P. G. & Vertut, J. 1997. Journey Through the Ice Age. Weidenfeld & Nicolson, London.

Chernova, O. F. & Kirillova, I. V. 2010. New data on horn morphology of the woolly rhinoceros (Coelodonta antiquitatis Blumenbach, 1799). Proceedings of the Zoological Institute 314, 333-342.

Deng, T. 2002. The earliest known wooly [sic] rhino discovered in the Linxia basin, Gansu Province, China. Geological Bulletin of China 21, 604-608.

Deng, T., Wang, X., Fortelius, M., Li, Q., Wang, Y., Tseng, Z. J., Takeuchi, G. T., Saylor, J. E., Säilä, L. K. & Xie, G. 2011. Out of Tibet: Pliocene Woolly rhino suggests Highhplateau origin of Ice Age megaherbivores. Science 6047, 1285-1288.

Fortelius, M. 1983. The morphology and paleobiological significance of the horns of Coelodonta antiquitatis (Mammalia: Rhinoceratidae). Journal of Vertebrate Paleontology 3, 125-135

Guthrie, R. D. 1990. Frozen Fauna of the Mammoth Steppe: the Story of Blue Babe. The University of Chicago Press, Chicago and London.

- . 2005. The Nature of Paleolithic Art. The University of Chicago Press, Chicago.

Kahlke, R.-D. & Lacombat, F. 2008. The earliest immigration of woolly rhinoceros (Coelodonta tologoijensis, Rhinocerotidae, Mammalia) into Europe and its adaptive evolution in Palaearctic cold stage mammal faunas. Quaternary Science Reviews 27, 1951-1961.

Nowak, J., Panow, E., Tokarski, J., Szafer, W. & Stach, J. 1930. The second woolly rhinoceros (Coelodonta antiquitatis Blum.) from Starunia, Poland (Geology, Mineralogy, Flora and Fauna). Classe des Sciences Mathématiques et Naturelles, Série B: Sciences Naturelles, Supplément 1-47.

Shidlovskiy, F. K., Kirillova, I. V. & Wood, J. 2011. Horns of the woolly rhinoceros Coelodonta antiquitatis (Blumenbach, 1799) in the Ice Age Museum collection (Moscow, Russia). Quaternary International 255, 125-129.

Darren Naish is a science writer, technical editor and palaeozoologist (affiliated with the University of Southampton, UK). He mostly works on Cretaceous dinosaurs and pterosaurs but has an avid interest in all things tetrapod. His publications can be downloaded at darrennaish.wordpress.com. He has been blogging at Tetrapod Zoology since 2006. Check out the Tet Zoo podcast at tetzoo.com!

More by Darren Naish