A hyperconspicuous horse, hiding in plain sight


Much has been written about the narrow escape from extinction by the takhi (Equus przewalskii, https://en.wikipedia.org/wiki/Przewalski%27s_horse), the last form of wild horse on Earth. However, the adaptive colouration of this species has not been explained.

Just as the striping of zebras is puzzling, so the hyperconspicuousness of the takhi is puzzling. The difference is that the former puzzle is popular, whereas the latter puzzle is unacknowledged.

The colouration of the takhi makes it conspicuous at all ages and in all seasons. It gleams out from the treeless grasslands during the sunny time of year (https://www.inaturalist.org/observations/32959824 and https://www.inaturalist.org/observations/108162555). Why would a large ungulate advertise itself to this degree to its predators, such as wolves?

Particularly puzzling is the fact that infants of the takhi are both extremely pale for Equus, and far paler than adults of the species.

Please note that, in infants,

In this complex pattern of depigmentation and sheen in infants, the takhi is extreme among all the species and subspecies of Equus.

Now, it is one thing for large animals, living in open environments, to have colouration making the adults even more obvious to predators than they already are. It is another for infants – which are precocial in all equids including the takhi - to adopt this showiness from the start. And it is something else again for the conspicuous colouration of infants to be additionally configured, as opposed to being merely precocial.

Inasmuch as it self-advertises at several ontological levels, the takhi can perhaps be characterised as a hyperconspicuous ungulate.

In order to understand the adaptive value of the particular conspicuousness of the takhi, let us first review the colouration of the infants of comparable wild ungulates.

In Equus there are three patterns in the colouration of infants relative to adults.

Firstly, the colouration is similar to that of adults right from birth in zebras and African asses – none of which is as self-advertising in its colouration as the takhi. In these forms, infants show nearly adult colouration, partly because the mane is fully developed from the start (https://www.alamy.com/zebra-mother-and-baby-zebra-running-side-by-side-on-grass-in-masai-mara-kenya-image362235495.html).

Secondly, in Asiatic wild asses the infants resemble adults, but are not quite as self-advertising, because the mane – conspicuously dark even if short in adulthood – remains poorly developed at birth.

In the following species, infants are almost as pale as those of the takhi. The difference is that adults lack the dark contrasts conferred on the takhi by the lower legs, substantial mane, and broad tail-tassel:

And thirdly, we have the takhi with its characteristic pallour, as described above.

Among large, gregarious ruminants with habitats resembling those of Equus, the general pattern is for infants to have disruptive (camouflage) or cryptic (plain) colouration, regardless of the adult colouration. Even in habitats with vegetation too low or sparse to conceal adults, the infants hide in the environment, usually by lying flat, apart from the herd for most of the day and night, for several weeks.

An example is the wapiti (Cervus canadensis), a species of large deer, partly sympatric with the takhi in Mongolia. In this species, the colouration of infants is spotted for camouflage (https://www.gettyimages.co.uk/detail/photo/wild-baby-elk-or-wapiti-in-green-grass-royalty-free-image/657484904 and https://www.leesonphoto.com/image/I000081bIE2S5edQ).

In most large African antelopes, the colouration of infants is medium fawn, plain, and not particularly pale. This confers inconspicuousness, provided the infants lie stationary.

Even hartebeests (Alcelaphus, https://videohive.net/item/red-hartebeest-and-suckling-calf/21285845 and https://www.dreamstime.com/stock-image-baby-red-hartebeest-antelope-image13753251 and https://www.dreamstime.com/stock-image-red-hartebeest-image779081), which are exceptionally fast and enduring runners, retain the usual pattern in which infants hide rather than joining the group.

No form of horse, ass or zebra conforms to the typical pattern of ruminants. This is because the infants of Equus are so precocial that they are active immediately after birth, remain with the herd from the start, and are as apparent as their parents. But why has the takhi evolved to be conspicuous?

One approach to this puzzle may be to focus on the prime examples of infantile conspicuousness among ruminants, namely certain wildebeests (Connochaetes) of Africa.

Wildebeests are

  • the only ruminants that emulate Equus, in having newborns so precocial that they can keep up from the start with adults fleeing at maximum speed from predators,
  • the most specialised of ruminants for long-range seasonal migration, and
  • more synchronous in their breeding than expected, considering their year-round territoriality in tropical climates.

In addition to their extreme precociality, certain species of wildebeests qualify as conspicuous in the following ways (see https://www.inaturalist.org/posts/54094-adaptive-colouration-in-wildebeests-part-1-large-scale-conspicuous-features#):

  • adults have particularly bold patterns, by virtue of gross combinations of dark and pale (although the configurations of contrast differ from that seen in the takhi),
  • infants are self-advertising, by virtue of their overall paleness, and
  • the fawn colour of infants differs from a cryptic plainness, because it contrasts with the precocially dark face.

In both the takhi and wildebeests, then, there is a combination of

  • infantile paleness,
  • adult conspicuousness owing to dark-pale contrasts,
  • sheen/antisheen at all ages,
  • immediate integration of infants into the group, and
  • extremely seasonal breeding.

In ruminants, conspicuous colouration is often associated with self-advertisement to predators. This is based on the principle that

  • gregarious species living in the open tend to be conspicuous even if plain-coloured, and
  • those individuals proving their fitness by 'stotting' are unlikely to be targeted for pursuit by predators scanning the group.

‘Stotting’ in wildebeests is subtle and complex. Adults react to the approach of potential predators by cavorting or style-trotting (https://www.shutterstock.com/nb/video/clip-5778965-cavorting-animal-mad and https://whyevolutionistrue.com/wp-content/uploads/2021/06/download-1-1.png), as if amused and exhilarated by the menace. Infants of wildebeests run in particularly bouncy ways on such occasions, as if playing.

It seems to make sense that if the best way to avoid being targeted is to show off to potential predators, then self-advertising colouration enhances this strategy. But, returning to equids: the trouble is that nobody has ever observed the takhi ‘stotting’, even in infancy.

The question of why zebras are striped has produced several potential explanations, which continue to be tested; the puzzle has fully gripped the imagination of biologists. The takhi poses a parallel puzzle, because its appearance is equally counterintuitive in its own way. However, in its case the question has, until now, been overlooked.


The following shows the distinctive pallour of the newborn takhi, a pattern transcending that in any species of ass or zebra. Despite being at home in well-watered lands, the takhi starts life paler than any other species in the genus Equus and as pale as the most arid-adapted large antelope, namely the addax (Addax nasomaculatus) of the Sahara.
Equus przewalskii:

The following shows that the pallour of infants of the takhi results partly from the initially poor development of the dark markings on the legs (note the trace of striping on the fore leg of this individual infant). The mane, although showing the first signs of dark bristles, is also poorly developed relative to the newborns of other forms of Equus.
Equus przewalskii:

The following shows that sheen contributes to the conspicuousness of infants of the takhi. Several features make adults of the takhi obvious to the human eye even at distances of more than a kilometre. Firstly, the pale ventral surface of the torso is extended so high on the flank that it catches the light; this pale area exhibits sheen in the summer coat. Secondly, the muzzle is conspicuously pale. Thirdly, the lower legs, mane and tail-tassel are conspicuously dark. The colouration is similar between seasons except that the sheen seems to be restricted to the summer coat.
Equus przewalskii:

The following shows that adults of the takhi are conspicuous even in snow by virtue of the darkness of the mane, tail and lower legs. The pallour of infants might hypothetically hide them in such backgrounds, but the seasonal demands of reproduction tend to preclude births in winter.
Equus przewalskii:

The following shows that in the western white-bearded wildebeest as in the takhi, infants are pale in contrast to the large-scale contrasts between pale (mainly owing to sheen) and dark in adults. Also note the precocial locomotion.
Connochaetes mearnsi:

The following shows the sheen on the legs, and a precocially dark face, in infants of the western white-bearded wildebeest; these are consistent with the fact that the infants of wildebeests lack hiding behaviour by day or, as far as we know, by night.
Connochaetes mearnsi:

The following shows that the western white-bearded wildebeest is born as pale as the ‘white’ beard of adults, consistent with the adaptive conspicuousness of infants.
Connochaetes mearnsi:

The following shows that infants of the Indian wild ass are almost as pale as those of the takhi. However, this paleness is not anomalous relative to adults, the dark mane of which is shorter and thus less noticeable than in the takhi. Asiatic wild asses, although conspicuous, do not qualify as hyperconspicuous because the colouration of infants is precocial rather than ontogenetically distinctive.
Equus khur:

The following shows that the dark mane and tail-tassel have yet to develop in infants of the Tibetan wild ass. However, infants of this species are not as pale as those of the other forms of wild asses.
Equus kiang:

The following shows that the Somali wild ass differs from the takhi in that its infants are not particularly pale, and differs from Asiatic wild asses in that its mane is so precocial that it has developed by the time of birth.
Equus somaliensis:

The following shows that in the plains zebra infants can be as intensely marked as adults.
Equus quagga boehmi:

The following shows that mountain zebras resemble the plains zebra in the relationship of infants to adults in colouration.
Equus zebra:

The following shows that Grevy’s zebra, although having the narrowest striping among zebras, likewise has infants striped much like adults.
Equus grevyi:

Publicado por milewski milewski, 28 de mayo de 2022


Another very interesting post. Thank you.

Publicado por doug263 hace 2 meses (Marca)

Style-trotting, a form of stotting, in subadult male Cervus canadensis: https://whyevolutionistrue.com/wp-content/uploads/2019/10/5l8a8366.jpg.

Publicado por milewski hace 2 meses (Marca)

@doug263 You're most welcome.

Publicado por milewski hace 2 meses (Marca)

The Tibetan wild ass (Equus kiang) is another species possessing conspicuous colouration.



The pale panels on the Tibetan wild ass extend far too high on neck, scapula and haunches, and contrast far too much with the dark part of the flank, to be explained as counter-shading. The effect is to catch the sunlight like a flag rather than to hide the animals. Like ruminants of open habitats (e.g. springbok, Antidorcas marsupialis), this equid is coloured conspicuously, making it easier for the herd to monitor members’ positions and communicate any alarm instantly. Furthermore, the dark-pale contrast on the flanks may make it hard for the main predator of the kiang, namely the wolf, to distinguish heavily pregnant, vulnerable individuals.

Publicado por milewski hace alrededor de 2 meses (Marca)

The following footage https://www.youtube.com/watch?v=5CaQcaNq8DA nicely shows how conspicuous the takhi is in its natural environment, even at a distance.

Publicado por milewski hace alrededor de 2 meses (Marca)

The takhi can hardly hide in its open environment. The following show how the colouration of the takhi, at all seasons, boost its conspicuousness.
The following shows that in winter the pale flanks of the takhi do blend with a snowy background, but there is no chance for the animals to hide by virtue of their overall colouration for the simple reason that their manes, tails and feet are far too large and dark to allow any kind of crypsis or camouflage. It is no accident that these parts of the body are the ones with dark colouration because they are also likely to be the ones moving even when the animals are standing. Most ungulates and Carnivora can easily detect the slightest movement, and so the combination of dark-pale contrast and movement ensure that the takhi is conspicious in this kind of setting.

The following are in an intermediate setting, neither clearly wintery nor clearly summery. Here both the paleness of the flanks and the darkness of the mane, tail and legs ensure that the animals are conspicuous as opposed to cryptic or camouflaged.

The following shows a summery scene in which the takhi appeals overall too pale to blend into the green of the pastures. Again, the dark parts accentuate the animals, but the difference from winter is that it is the overall paleness (which in this season, with the pelage short, involves sheen as well as a lack of pigmentation on the fur) relative to the background that spoils any chance of crypsis. So the combination of pale flanks and various dark and mobile body-parts ensures that the takhi remains conspicuous at all seasons.

Publicado por milewski hace alrededor de 2 meses (Marca)

An underplayed aspect of the colouration of equids is the various rump patterns in the various species.
The photos in http://www.equinestudies.org/mammalian_species_2008/mammalian_species_equus_caballus_pdf1.pdf illustrate the rump pattern in the takhi (Equus przewalskii).
Please note the following features.
Firstly, there is only a trace of the midline dark stripe which is so much more prominent in e.g. Equus grevyi, Equus hemionus, and Equus kiang, and present also in Equus zebra. This stripe is essentially vestigial in the takhi, in the sense that it makes negligible difference to the animal’s appearance (https://www.dreamstime.com/przewalski-s-horse-dzungarian-horse-przewalski-s-horse-dzungarian-horse-rare-endangered-subspecies-wild-horse-image175746959 and https://www.dreamstime.com/stock-photo-przewalski-s-horse-wild-mongolian-horses-eating-hay-latin-name-equus-przewalskii-image70687864).
Secondly, the main dark/pale contrast of the hindquarters, visible at great distance in the field, is the contrast between the large blackish tail tassel, which extends up as high as the root of the tail at the level of the anus, and the adjacent extensive panels of extremely pale pelage on the posterior haunches (https://www.mimercentre.org/index.php/blog/wild-horses-in-mongolia and https://en.wikipedia.org/wiki/Return_of_the_Wild_Horses#/media/File:N%C3%A1vrat_divok%C3%BDch_kon%C3%AD_2013_5.jpg and https://www.flickr.com/photos/pokerbrit/3651762957 and https://audreydradam.files.wordpress.com/2014/09/takhi.jpg and https://www.activewild.com/wp-content/uploads/2018/09/wild-przewalskis-horse-in-mongolia.jpg).

This is essentially the same pattern as seen in, for example, roan antelope (Hippotragus equinus), but is well-developed in the takhi.
Thirdly, the pale panels of the posterior haunches do not extend above the level of the base of the tail. In this way the takhi resembles e.g. hartebeests (Alcelaphus), the difference being that in the latter the boundary of the pale panels is far sharper, and the dark tail far smaller.
Fourthly, in the takhi viewed directly from behind, there is also a noteworthy pattern of blackish on the posterior surfaces of hock, fetlock and pastern (https://www.dreamstime.com/stock-image-przewalski-horse-image25512241 and https://www.agefotostock.com/age/en/details-photo/close-up-of-the-backside-from-two-przewalski-s-horses-equus-ferus-przewalskii-standing-on-a-meadow-in-the-bavarian-forest-national-park-germany/ZV4-2094452/1). This means that the lower legs are dark enough, in part, to stand out as part of the conspicuous posterior aspect of the animal, even at a distance. And, as it happens, the black tail tassel is so long that effectively a single blackish panel results, consisting of the root of the tail, the tail tassel, the hocks, the fetlocks, and the pasterns, all collectively accentuating the appearance of the animal in collective contrast with the effectively whitish buttocks.
By comparison with this vivid display, the rump pattern of the extinct quagga (Equus quagga quagga), although never described or depicted as such in the literature, is nondescript. It does nominally consist of a pale tail-stalk and a dark tassel, but the tail tassel of the extinct quagga is far smaller than that of the takhi, and not as dark. So the dark/pale contrasts are limited. Furthermore, the rump and posterior haunches/buttocks of the extinct quagga are pale brown rather than extremely pale fawn. I am unsure about how much paler the base of the tail is than the posterior haunches, because few photographs of the museum specimens show this perspective.
The takhi thus has one of the most vivid displays, whereas the extinct quagga has one of the least vivid displays.

Publicado por milewski hace alrededor de 2 meses (Marca)

The following is further evidence that the spotted hyena (Crocuta crocuta) in Eurasia depended partly or largely on a wild horse, with the predator >100 kg body mass and the prey >400 kg body mass, i.e. fourfold larger than the predator.

See http://www.fedoa.unina.it/1935/1/Meloro_Scienze_Terra.pdf

In accordance with Bergmann's rule, the spotted hyena of Pleistocene Europe was larger-bodied than that found in Africa today despite being the same species.

Meloro (2007), on the basis of fossil found e.g. in Kent’s Cave in England, assumes a mean adult body mass of 102 kg. The species in the Serengeti today weighs an average of about 50 kg, the males slightly less and the females slightly more. In Kruger Park the body mass is more like 70 kg, with an average adult body mass certainly >60 kg.

So, the spotted hyena in Pleistocene Europe seems to have been about double the body mass of that recorded recently in the Serengeti.
The plains zebra (Equus quagga boehmi) in the Serengeti has a body mass ca 250 kg. Extant Equus przewalskii weighs about 300 kg. Even if we assume that the ancestral horse in Italy weighed about 350 kg (similar to the extinct quagga), what we have is a prey species of 350 kg and a predator species of 100 kg. In terms of body size, the horse would have been vulnerable to the spotted hyena, bearing in mind that the modern plains zebra, although five-fold the body mass of the modern spotted hyena, certainly is vulnerable to this predator.

If one follows the fivefold yardstick, the spotted hyena in Italy would have been able effectively to predate any wild equid up to about 500 kg, which is about the body mass of a thoroughbred. Indeed, as Meloro (2007) mentions, another important prey of the spotted hyena in Europe was a species of rhinoceros, as well as Bison priscus, Bos primigenius, andCervus elaphus.

A prey species of the large form of the spotted hyena in the treeless grassland of Eurasia in the Pleistocene was Equus dalianensis (https://www.sciencedirect.com/science/article/abs/pii/S0277379120306533).

This was larger-bodied, with withers height of about 142-152 cm, than its close relative, the still-extant Equus przewalskii. Grevy’s zebra, the largest extant zebra, has withers height of about 145 cm and body mass ca 390 kg (max. 450 kg). I infer that E. dalianensis was nearly double the body mass of Equus quagga boehmi, and if anything larger-bodied than Grevy’s zebra. With a body mass of 450 kg or more, it approached the size of a thoroughbred.
The treeless grassland of Eurasia was colder than Italy, hence the spotted hyena was probably larger. I assume that the subspecies that preyed on the large wild horse E. dalianensis was about 110 kg, double the body mass of the form that preys on the plains zebra in the Serengeti today. This makes sense given that the spotted hyena can only just manage to run down and kill the plains zebra in the Serengeti, this ungulate being too large and self-defensive to be killed easily by this predator.

Also see https://www.horsetalk.co.nz/2019/05/18/extinct-horse-species-originally/.

Publicado por milewski hace alrededor de 2 meses (Marca)

The following shows the sheen in the colouration of the takhi: https://www.youtube.com/watch?v=4RKRU2xQoTA.

Publicado por milewski hace alrededor de 2 meses (Marca)

All forms of zebras have precocial manes, but the extreme example is perhaps Equus quagga crawshayi, illustrated below. The subsequent photos show Equus przewalskii, in which the mane is certainly present at birth, but not nearly so well-developed relative to the adult.
Do not be confused by the fact that the mane is smaller in adult E. przewalskii than in adult E. q. crawshayi. The point in the following photos is that the mane is actually better-developed in the infant of E. q. crawshayi than in its mother, whereas the mane is more poorly developed in the infant of E. przewalskii than in its mother.
The following all show Equus quagga crawshayi:




The following all show Equus przewalskii:









Publicado por milewski hace alrededor de 2 meses (Marca)

The following shows the conspicuous paleness of infants in Equus przewalskii: https://www.alamy.com/stock-photo-przewalskis-horse-equus-przewalskii-new-born-foal-running-in-a-field-17755358.html.

Publicado por milewski hace alrededor de 2 meses (Marca)

Another equid with conspicuous colouration is Equus kiang (https://www.natureplprints.com/2020-may-highlights/kiang-equus-kiang-herd-grazing-steppe-evening-19989577.html#openModal and https://www.researchgate.net/figure/An-adult-male-Equus-kiang-subspecies-kiang-in-summer-coat-from-eastern-Ladakh-India_fig1_262137189 and https://fineartamerica.com/featured/kiang-wild-ass-anthony-mercieca.html?product=puzzle&puzzleType=puzzle-18-24).

Unlike E. przewalskii, Equus kiang sometimes displays when fleeing, by lifting the muzzle above the horizontal: https://www.reddit.com/r/HardcoreNature/comments/tb1lk3/a_pack_of_wolves_chasing_a_kiang/ and https://roundglasssustain.com/photo-stories/tibetan-wild-ass.

This display has also been photographed in infants of the kiang: http://images.fineartamerica.com/images-medium-large-5/kiang-foal-margaret-s-sweeny.jpg and http://c8.alamy.com/comp/DDCFWB/kiang-foal-DDCFWB.jpg.

Please scroll in https://www.ju-lehadventure.com/ladakh-information/wildlife-of-ladakh for another clear photo of the 'high-muzzle' display in Equus kiang.

The following shows both mother and juvenile Equus kiang displaying: https://www.mindenpictures.com/stock-photo-kiang-equus-kiang-mother-and-foal-running-native-to-asia-naturephotography-image00539327.html and http://media.gettyimages.com/photos/kiang-mother-and-foal-running-native-to-asia-picture-id537009377?s=612x612.

Here is another photo showing the 'high-muzzle' display in an Asian wild ass: https://i.pinimg.com/236x/4c/cb/c3/4ccbc38a3333dc7a122df1aa479e41d0--whippet-silk-road.jpg.

Presumably the muzzle is held high to indicate energetic fitness. Possibly the feet are lifted higher, during trotting, than they would be if the animals were merely running as efficiently as possible.

This 'high-muzzle' display is never seen in the plains zebra or the takhi. However, a similar display occurs in Rangifer tarandus and Cervus canadensis.

The following shows the display of the muzzle in combination with style-trotting/proud-trotting (a form of stotting), in Rangifer tarandus: https://www.naturalworldsafaris.com/~/media/images/destinations/arctic-canada/nws-st-arctic-canada-wildlife-caribou.ashx. Not only is the gait exaggerated, but the caudal flag is fully activated.

Publicado por milewski hace alrededor de 2 meses (Marca)

Agregar un comentario

Acceder o Crear una cuenta para agregar comentarios.