Voyage of the Kinkajous

For much of the time since the extinction of the dinosaurs, South America was an island continent. Developing in isolation, the mammals living there developed a number of unusual forms not seen elsewhere. Around 2.8 million years ago, however, South America became joined to its northern counterpart via the Isthmus of Panama. Northern animals flooded south, and relatively few headed in the opposite direction. As a result, unlike Australia, which remains an island continent today, the mammalian fauna of South America includes many animals at least broadly familiar from elsewhere.

Among the first mammal families to make the crossing was that of the raccoons. Although this had first appeared in Asia, living species are now found only in the Americas (ignoring some recent man-made introductions) with most of them found in tropical habitats. In fact, they reached South America almost ridiculously early, something we know because we have a fossil example that's around 7 million years old... at least 4 million years before the land crossing we'd expect them to have used had formed.

The raccoon family has only fourteen currently recognised species, and, of these, only three are actually "raccoons" in the sense of it being visibly obvious that that's what they are. As a result, it's one of those cases where using the technical term may be preferable to trying to find a common English one. So, fourteen species of "procyonid", then.

The first genetic analysis of the procyonid family tree was published in 2007, and it provided scientists with two surprises. The first was that the shape of the family tree was quite different to what we'd have guessed based purely on physical features - there had been apparently been more parallel evolution going on here than we'd thought. The second was that, by comparing the amount of divergence with inferred rates of genetic change, it appears that the different sub-groups within the raccoon family had already split apart from one another before the Isthmus of Panama formed.

To put it another way, some of the divergences between the different groups seen in South America today probably happened in North America... where most of them no longer live.

A good place to start if we want to find out more about what was going on here is with the oldest branch among the living members of the family. It turns out, according to the genetic evidence, that this belongs to the kinkajou (Potos flavus), a strange tree-dwelling animal that eats mostly fruit and is found from southern Mexico to central Bolivia and Brazil.

Unfortunately, kinkajous have no definitive fossil history. No fossils are known from South America, and the closest thing we have is the fossil Parapotos, which lived in Texas around 15 million years ago. The problem with that is, that while, as the name indicates, it was originally assumed to be closely related to kinkajous, it's equally plausible that it's actually an early olingo, a type of animal that we now know belongs to a different branch of the procyonid family tree.

Which leaves us with the genetic evidence. But there's a problem here, too, in that we can't compare the different species of kinkajou to see when they diverged because is there is only one species to look at. Indeed, that 2007 study, and two others that have broadly confirmed its findings since, used the DNA from a single individual to construct their family tree. Which is fine for what they were doing, but not if you want to untangle the history of kinkajous specifically.

Instead, what we need to do is sample the DNA from multiple different individuals, preferably from radically different parts of their range, compare them to see how long ago they last shared a common ancestor and to determine what patterns there may be among them. If nothing else, we should be able to prove that the seven recognised subspecies are exactly what we've always assumed them to be, and that nobody has ever got confused while trying to distinguish them.

Yeah, like that's going to happen.

Perhaps surprisingly, given that kinkajous are relatively obscure, if interesting, animals, there are no less than three such studies currently published. The most recent of these is also the largest, looking at DNA from 129 individuals and using five different genetic markers for each, rather than just one as is more commonly done. (This, presumably, is because the relevant technology is getting a lot cheaper these days...)

The results of this new, larger, analysis identified no less than eight clusters of related individuals within the species. Four of these correspond more or less directly to known subspecies described on the basis of subtle physical differences and a discrete geographic homeland. A fifth subspecies seems to consist of at least two different genetic groups, neither of which lives in exactly the right area, and a sixth couldn't be identified in the data at all. Since they couldn't get a DNA sample from the supposed seventh subspecies, which lives only in one relatively small area in Venezuela, we have no idea how "real" that might be.

Which leaves two genetic groupings that don't match up with anything we previously knew about. It's too early to name them as new subspecies, since that involves a certain degree of judgment and subjectivity about whether they are distinct enough to earn that honour. But it's certainly plausible to describe them as "evolutionarily significant units", a term that, broadly speaking, means that they are worthy of conservation if we want to maintain genetic diversity.

In fact, the authors of the study would go a good deal further than that. They point out that the genetic differences between some of the groups are much wider than we'd expect. So wide, they say, that some of the animals whose DNA they analysed must belong to entirely new and undescribed species. This really depends on how you're going to define the term "species", a notoriously tricky term, and not everyone is going to agree with them. But, while they don't go as far as naming these new discoveries, conceding that more evidence is needed, there's at least a hint that there may be no less than five species of kinkajou, not just the one.

But, to get back to our original point, what does this tell us about the evolutionary history of the kinkajous? A significant finding is that the oldest of the eight groups the researchers found consisted of kinkajous living in Guatemala, Honduras, and southern Mexico, with the South American groups all having diverged from one another later. This suggests that kinkajous did, indeed, first evolve on the northern side of what is now the Isthmus of Panama.

This, they calculate, would have happened around 11 million years ago, a time of global cooling that would have altered the nature of the forests in which their ancestors would have been living. The next major step would have occurred around 5 million years ago, most likely along the Pacific coast of what is now Colombia and northern Peru. Another burst of global cooling, with subsequent expansion of grasslands at the expense of forests, would have seen the initial splits among the South American kinkajou population, some of which headed east around the northern end of the Andes, and into the Amazon Basin.

Further splitting occurs during the late Pliocene, and even more so during the Ice Ages of the Pleistocene. Obviously, the Amazon jungle was never covered in ice sheets, but the world had cooled enough that the forest was split in two by an expanse of relatively open country... just the sort of event that could isolate different populations from one another.

The dating of this does, however, confirm the earlier suspicions that kinkajous must have arrived in South America long before we think the Isthmus of Panama formed (as we know at least some fossil procyonids did). That might be because the Isthmus is older than we suspect, although, if so, it's notable that not much else seems to have made the same trip south. But, even if that's not so, the Isthmus didn't appear out of nowhere.

The usual theory is that, before it rose entirely above the waves, separating the Pacific and Atlantic Oceans, a chain of volcanic islands ran along roughly the course of the modern Isthmus. The land rose, the islands joined, and eventually, Panama formed. It still wouldn't be an easy trip, requiring some animals to be washed out to sea, possibly on bits of tree blown down during a hurricane, and then clamber out, bedraggled and hungry, on a new shoreline a few further miles down the chain.

But, when you have a million or more years to make the attempt... you get a lot of tries, and you only need to succeed once.

[Photo by "Ichtusvet", from Wikimedia Commons. Cladogram adapted from Koepfli et al. 2007.]