Sunday 30 October 2022

Horns v. Geography - Relationships Among Rhino Species

White rhino
Rhinoceroses are amongst the largest land animals alive today, exceeded in size only by the elephants. As one might imagine, given their distinctive appearance, the group has a long evolutionary history. What is perhaps less obvious is that the family was once much larger than it is today, with many species living side-by-side. In total, we have so far named something like 100 species of fossil rhinoceros and, while some of those will probably not survive more detailed analysis, it's also true that there must be several we haven't found yet. Either way, it's quite a lot.

Most of these lived during the Miocene epoch (although the family is older than this) with the number of species thinning out during the following, Pliocene, epoch between around 5 and 2 million years ago. Even so, we know of nine species that lived during the later Ice Ages. Four of these, however, did not survive their end, leaving us with the five that - in some cases only just - survive today.

Perhaps the most visible and distinctive feature of rhinos is the presence of the large 'horn' on their nose. This is not a horn in the anatomical sense of that word used to describe goats or antelopes because it has no bony core; it is partially calcified keratin all the way through, structured slightly differently from the regular keratin of a true horn sheath, but nonetheless having no blood vessels, nerves, or other cellular components. Significantly, some rhino species have two horns, while others just have the one. Moreover, the one-horned species both live in Asia, but the two-horned species are spread between Asia and Africa.

This has led to two competing theories about how the five species of living rhinoceros are related to one another. One theory is based on the physical appearance of the animals, assuming, not unreasonably, that those which look the most similar are probably the most closely related. Specifically, this means that the two-horned rhinos must form one subgroup within the family while the one-horned rhinos form the other. Since most fossil rhinos had (so far as we can tell) only one horn, the idea here is that, at some point, a rhino evolved that had a second horn, and that the two-horned rhinos we have now all descended from that animal.

But, other scientists said, hang on a moment: one species of two-horned rhino doesn't live anywhere near the other two. This led to the second theory, that it is the rhinos that live closest to each other that are each other's closest relatives and the number of horns isn't all that important. Under this scheme, we would still have two evolutionary groups within the living rhinos, but these would consist of the three Asian species on the one hand, and the two African species on the other. The presence of a second horn would therefore have to be something that evolved twice... but it's a lot easier to see how the rhinos ended up where they are today.

Of course, this was all before modern genetic analysis came onto the scene and you'd think that that would have cleared everything up fairly quickly. But no, not really.

An early study in 2001 looked at changes in a couple of mitochondrial genes, including cytochrome b, a gene found in all mammals (and, indeed, many other animals) and often used as a means of constructing evolutionary trees. It showed that the geographical hypothesis was correct; the three Asian species are more closely related to each other than to the African rhinos. But as technology improved we were able to examine more genes at once, and get a better picture. Ten years later, in 2011, came a study adding nine nuclear genes to the analysis (including BRCA-1, a gene with considerable medical significance in humans). This new study showed that the horn hypothesis was the correct one; the three two-horned species are more closely related to each other than to the one-horned species.

Neither of these studies are alone, both being backed up by further findings using different methods. In fact, if we move on another decade or so to 2020, there's even a genetic study showing that both theories are wrong. The best way to resolve this is to throw more genetic data into the analysis... which brings me to phylogenomics.

The standard way of using genetics to establish evolutionary relationships is to analyse one or more genes, and see how the fine details of their code vary between individuals. We can analyse many more genes at the same time than we used to be able to, but it's still a sample and, as the case of the rhinos shows, what answer you get can sometimes depend on which genes you happen to look at. Phylogenomics, on the other hand, uses a much larger sample, comparing large chunks of the genomes of the animals concerned so that (in theory) any stray signals from individual genes that point in the wrong direction are swamped by the overwhelming majority that don't. It's not perfect, because there is more than one way of doing the analysis but, if done properly, it should nonetheless be helpful.

A study published last year not only looked at the genomes of all five living species of rhino, but was also able to add in the partial genomes of three of the four additional species that went extinct towards the end of the Last Ice Age. I'm not going to claim that this is going to be the last word on the matter, because it probably isn't, but it does seem to have had at least some acceptance as a likely explanation for what's happening and provides support for what seems to be more popular of the two theories.

It showed that the rhino family, as it exists today, can best be divided into three subgroups. One group consists of the two-horned African species, the black (Diceros bicornis) and white (Ceratotherium simum) rhinos, while a second group contains the one-horned species, the Indian (Rhinoceros unicornis) and Javan (Rhinoceros sondaicus) rhinos. That much all the theories had agreed on - the question was where the remaining species, the Sumatran rhino (Dicerorhinus sumatrensis), went. This has two horns, like the African rhinos, but lives in Asia, like the Indian and Javan rhinos.

The phylogenomic study clearly places it as closer to its Asian relatives, although not by much. According to this analysis, its ancestors split off from those of the one-horned rhinos around 14.8 million years ago, just 600,000 years after the split with the African species. This was the hottest time of the Miocene and, probably not coincidentally, not long after Africa collided with Asia and first allowed rhinos to move from one continent to the other. From the exact pattern of the genetic discrepancies, the authors of the study conclude that "African" rhinos first evolved in Asia, and to some extent interbred with their early neighbours, confusing the genetic signal, before they headed south to their new home.

As for the extinct species, both the woolly rhinoceros (Coelodonta antiquitatis) and Merck's rhino (Stephanorhinus kirchbergensis) resolve as close relatives of the Sumatran rhino, which fits with earlier studies, although they lived further north than the Sumatran rhino does today, with fossil sites known from Spain to Korea. They were not able to analyse the DNA of one of the other late-surviving species, the narrow-nosed rhinoceros (Stephanorhinus hemitoechus) of Europe, but it's thought to be a very close relative of Merck's species, so it likely belongs in the same group.

The remaining fossil species, the "Siberian unicorn" (Elasmotherium sibiricum) does not belong to any of these three groups, but it would have been very surprising had it been otherwise. It was an exceptionally large animal, living in western Siberia and Central Asia, and, for what it's worth, only had a single horn. Significant differences had long resulted in it being placed in a different subfamily from living rhinos, and the phylogenomic study estimates that the last common ancestor of it and modern species probably lived around 35 million years ago, over twice as far back as the earliest split between the surviving groups.

The same study also appeared to show that the genomes of rhinoceros species indicate that a relative lack of diversity and high levels of inbreeding have long been a natural condition for them, something that they're actually adapted to. With three out of the five living species of rhino critically endangered, it's obvious that this can't be the whole story, but if it is true, it does offer a glimmer of hope. It may not be easy, but if we can save these species and stop them going extinct, it could be that they will recover rather more quickly than we might otherwise expect.

[Photo by "Coralie" from Wikimedia Commons. Cladogram adapted from Liu et al. 2021.]


  1. So the probable scenario is that the LCA of extant rhinos had two horns, and the Indian-Javan clade lost one?

    Are horned rhinos a clade, btw, to the exclusion of indricotheres and whatnot?

    1. With regard to the first question, it's difficult to tell, since many fossil species aren't sufficiently complete for us to know how many horns they had. But I'd agree it's the logical inference. But as to your second point, that's a pretty clear 'yes'. While some early members of the true rhinoceros family probably lacked horns, indricotheres and all the rest are members of different, but related clades. (That is, they are rhinoceratoids, but not rhinoceratids).

    2. Oh, and when I said "something like 100 named [fossil] species"... I wasn't even counting the indricotheres, etc. Just the rhinoceros family itself (although, as noted in the post, this does include the elasmotheres).

  2. Two typo :Stephanorhinus* , Rhinoceros unicornis*

    1. Thanks - fixed! (I bet the second one was my spellchecker being overeager and me not noticing...)