When the Hippos Changed

The hippopotamus family contains just two living species, including, of course, the well-known common hippo (Hippopotamus amphibius). As is so often the case, though, the family has a long fossil history including a number of other species and genera, although, while the fossil species were more widespread than the living forms are today, it's fair to say that it was never a huge or diverse group in the way that, say, antelopes are.

Among the fossil species, several are very closely related to the living common hippo, including both the stalk-eyed hippo (H. gorgops), which probably weighed over 3 tonnes, and the pig-sized Maltese hippo (H. melitensis). Quite where the living pygmy hippo (Choeropsis liberiensis) fits in the fossil family tree is much less clear, but there are a number of species that aren't particularly close to either of the surviving forms.

Taking a broad view of the fossil history of the family, then, palaeontologists have tended to group the hippos into two subfamilies. One are the "hippopotamines", a group of broadly "modern" hippos that includes both of the living species. At least two other genera are also considered to belong to this group, one of which, Hexaprotodon, lived everywhere from Madagascar to Spain and Indonesia, taking in much of northern Africa and southern Asia on the way. Most of these lived during the Pleistocene and Pliocene epochs, which is to say, the Ice Ages and the epoch immediately preceding them. The earliest forms, including the other genus, Archaeopotamus of Kenya and Arabia, lived during the late Miocene, first appearing somewhere around 8 million years ago.

The second subfamily, now long gone, were the "kenyapotamines", named for the mid to late Miocene Kenyapotamus, but now thought to also include genera such as Morotochoerus, which date back to the early Miocene, 22 million years ago. This group, which were generally more primitive than the hippotamines, died out around 8 million years ago.

You've probably noticed that I've just said "around 8 million years ago" twice. That is, the last of the early hippos died out about the same time that the first of the modern hippos appeared. Which actually isn't the way that these normally work. Quite often, we find that supposedly primitive forms survive for ages after the ones that are more recognisable to us first appear. For example, sabretooth cats lived alongside lions and tigers for millions of years before the last of them died out. And, after all, why shouldn't they? These older groups have often survived for a very long time, and presumably were very good at whatever it is that they did, so there's no reason that they should be immediately wiped out as soon as something that looks more modern turns up.

But, in the case of hippos, that doesn't appear to be the case. It's not just that kenyapotamines evolved into hippopotamines, although they quite probably did, it's that they were then very rapidly replaced by them, effecting a sudden change in the large mammal fauna of African wetlands. Something dramatic happened around 8 million years ago, and the best way to find out what it was is to take a close look at African fossil deposits of the right age. Which is a bit of an issue, because there's an almost complete gap in the African fossil record between... you guessed it, 9 and 7 million years ago.

The operative word, fortunately, being "almost". There is one site on the continent that does hold a relatively good sample of fossils from the right age, located near the village of Chorora in Ethiopia. In the south of the Afar Triangle, this is now a desolate region of desert scrubland, largely uninhabited by humans, let alone by anything that likes wallowing in wide rivers. But, back in the late Miocene, it wasn't so, and, since the site was first discovered in the 1970s, there have been a number of studies of the animals it holds - the most significant of recent years being the discovery of a primitive gorilla that pushed back the origin date for such animals, and therefore also of the human/chimp lineage.

But, until recently, there has not been a study of the fossil hippos found at the site. This may in part be because none of them are particularly complete, although there quite a few of them. A few months ago, however, a new species was identified, and named as Chororatherium roobii ("roobii" apparently means "hippopotamus" in the local Oromo language). It's based, as is often the case, on only a few teeth and bits of jaw, but these happen to include the key features used to distinguish the two fossil subfamilies. On this basis, while the animal shows a mix of "primitive" and modern features, the researchers describing it conclude that it is over the line and into the living, hippopotamine, group.

Nonetheless, it is the oldest, and most "primitive" (or, more accurately, least modern-looking) of the hippopotamines. The deposits in which it was found have been dated pretty tightly to almost exactly 8 million years ago. There's also a shift in the number of fossils of different kinds as we move from older to younger deposits within the Chorora Formation, with hippos becoming steadily more common as time wears on. Given that it's just one locality, this could, of course, just be due to some change in the nature of the environment. But it does fit with the broader pattern across Africa that kenyapotamines were relatively rare, even in the right habitats, before 9 million years ago, and hippopotamines quite common in river valleys after 7 million years ago.

This allows the authors of the paper describing the new species to propose an outline for what appears to have happened during that crucial time in hippo evolution. Roughly between 8.5 and 8.0 million years ago, the kenyapotamines are rapidly replaced by the earliest hippopotamines - that is, their new species - barely living alongside one another at all. Over the next 500,000 years or so, the new sorts of hippo suddenly become much more common, pushing out the other large river-dwelling herbivores of the day. Not long after this, the teeth of hippos reached their modern form, as found (more or less) in the four previously known genera from the late Pliocene to the present day.

So the change that sparked all this not only caused the hippotamines to appear in the first place, it also gave the resulting new species a significant boost, allowing them to become more common than their ancestors had been. Given that we know this from the way that their teeth changed, something to do with their diet seems a likely culprit. Analysis of the oxygen isotopes in the fossils suggest that the hippos of the day were already semi-aquatic (unlike the very first hippos, over 10 million years before), so it's not just that they suddenly started entering the rivers.

A more likely reason is the rapid expansion of grasses across this part of Africa at the time. This, in turn, is due to the climactic changes of the day, but the benefit to the hippos would have been indirect, with other animals (such as horses and antelope) grazing on the grasses, leaving other plants more abundant in the places that grasses had failed to colonise. While the kenyapotamines had fed on at least a moderate amount of grass, Pliocene hippopotamines fed mainly on softer water plants, and the tooth structure of these recently discovered fossils suggests that they did the same. It was a switch in diet, freeing them from competition with grazing animals - so long as they stayed in the rivers.

In this respect, early hippopotamines would have most resembled the living pygmy hippo, which eats mainly ferns and forest plants. The common hippo has, at some point, switched its diet back again, since it now eats almost nothing but grass. Yet, they may owe their very existence to the fact that, at one point, their ancestors did the exact opposite.

[Photo by Charlesjsharp, from Wikimedia Commons.]