Sunday 18 October 2015

Fossorial Fossils

Skeleton of a European mole
(Note the odd-looking humerus and shoulder girdle)
The mole family includes over 40 species, found across the Northern Hemisphere. They are, of course, most noted for living underground, burrowing their way through the soil in search of worms and similar food. But, in fact, this is by no means true of all members of the family.

Indeed, we can divide the lifestyles of these animals into three broad categories. First, there are the truly fossorial moles, meaning those that spend virtually their entire lives underground. This is the case for about three quarters of the living species, including the familiar ones that so annoy gardeners. But there are also the semi-fossorial "shrew-moles", that do quite a lot of digging, but spend much of their life above ground, retreating to their burrows primarily to sleep, and perhaps to supplement their surface diet. And, finally, there are a few species in the family that are semi-aquatic, spending their time swimming in rivers, even if they, too, have burrows to rest in.

It's generally thought that it's the shrew-moles that most resemble the earliest prehistoric members of the family. That, in other words, while moles may always have dug holes, the specialised features that allow them to spend their whole lives below the soil took some time to evolve. Which is probably what you'd expect anyway, especially since the closest relatives of moles alive today are the shrews and hedgehogs.

But how and when did this change to a fully underground life begin? One of the interesting things about this is that all three general lifestyles are found in species on both sides of the Atlantic. The European mole (Talpa europaea) that we find in places like Britain is a different species from, say, the eastern mole (Scalopus aquaticus) found throughout the eastern US, but the two live very similar lifestyles, producing lots of mole-hills, and so on. But, given that they live on different continents, and given that there are also semi-fossorial shrew-moles on both of those continents (in the Far East and the Pacific Northwest, respectively), are they actually that closely related?

Granted, yes, they are both moles, and that's not in dispute. So they aren't distantly related to one another in the way that they are to the mole-like animals of the Southern Hemisphere. The question is whether or not their subterranean lifestyle evolved just the once, with fossorial moles crossing over the Bering land bridge (in whichever direction), or whether it evolved twice, with shrew-mole-like creatures retreating underground separately on each continent.

Fossil moles, it is fair to say, have received less attention than dinosaurs or sabre-tooth cats. Even leaving aside the fact that they're obviously less dramatic, the bones of small mammals tend to be fragile, and don't always preserve well in the fossil record. As with so many other creatures, a lot of what we know about fossil moles comes purely from teeth, not from whole skeletons. Teeth - at least the molars and premolars - have the advantage not only of preserving well, but also of having precise shapes that tend to be distinctive between different groups of animal. If you find an isolated tooth, you can often work what it belonged to, but an isolated rib or leg-bone may look much the same regardless of what it comes from.

But teeth, great though they are for elucidating where an extinct animal might fit in its evolutionary tree, tell you pretty much nothing about how much digging it did. Fortunately, in the case of moles there is another option: the humerus. This, of course, is the bone in the upper forelimb (or upper arm, in humans), and, because of all the digging they have to do with their forelimbs, it's unusually thick and sturdy in moles. In fact, it's almost ridiculously sturdy, having a shape that really doesn't look like that of any other bone in any other animal. Crucially, it doesn't even look like the same bone in shrew-moles, or in the semi-aquatic species, so it's both more likely to survive than such bones usually are, and unusually helpful to us when it does so.

The oldest moles we know of date back to the late Eocene, around 36 million years ago, with specimens of roughly the same age being known from both Europe and North America. But these are, as one might expect from the above, only teeth. While we can't know for sure, we'd also assume that these, being so early, were still primarily surface-dwelling creatures. The oldest known mole that we can confidently assert was good at burrowing is Geotrypus minor, a specimen from the early Oligocene of Germany, roughly 30 million years ago. From the shape of its humerus, it was not yet as good at digging as modern species, but it seems to have been well on the way, and it's at least plausible that it was already throwing up mole-hills.

But that, of course, doesn't prove that the subterranean lifestyle evolved only once, or that, if it did, it was in Europe. There were already moles of some sort in North America by that point, and they had had about 6 million years to cross over, one way or the other. Given the scarcity of their fossils - and, perhaps, the small number of palaeontologists interested in looking for moles specifically - there's no reason to suppose that we haven't just failed to find a crucial piece of the puzzle.

Mole family according to Schwermann & Thompson
Black - fossorial
Brown - semi-fossorial
Blue - semi-aquatic
So, instead, we have to make inferences, based on what we know of the living species, and of the fossils that we do have. A re-analysis of previously known fossils using micro-CT scans of the rock in which they are embedded, and comparison with other fossils gives us a considerable amount of new information to throw into the mix. But the resulting picture is, as so often, far from simple and unambiguous.

The fossils examined in this study were previously thought to belong to three different groups - a relative of the living Eurasian moles, a relative of the living North Americans moles, and a semi-aquatic desman. The study confirms that this is so, which is significant, because all the fossils were dug up in southern Germany. Hardly surprising for the "Eurasian-like" species or the desman, but arguably somewhat more so for the one apparently related to the North American moles.

In fact, it's not quite as surprising as it sounds, since it has long been known that a living species from central China, the Gansu mole (Scapanulus oweni) is, in fact, a relative of the North American, not the Eurasian, forms. Apparently, "North American" moles once lived over a much larger region of the Earth, which means that we can be confident that they must have crossed over the Bering Straits at least once. And, if once, why not more than once?

The finding that is surprising, though, is that this analysis suggests that the "North American" and Eurasian moles share a single common ancestor, to the exclusion of the other members of the family. This suggests that the subterranean lifestyle evolved once, probably somewhere in Asia, and that the various shrew-moles and semi-aquatic species are descendants of earlier forms that never took up digging to that extent.

The reason that this is surprising is that it isn't what analysis of the genetics and molecular biology of the mole family tells us. The most detailed evidence we have from that source suggests that, as a minimum, American shrew-moles (Neurotrichus gibbsi) are also descended from the common ancestor of the two kinds of subterranean mole. Which means that either fossoriality evolved twice, or that American shrew-moles were originally subterranean, and then partially gave up on that lifestyle in favour of spending at least some of their time running around in the leaf litter instead.

However, even the molecular studies are not wholly consistent with one another. One study from 2004, for instance, suggests that desmans are also descended from this common ancestor, and are particularly closely related to the Eurasian moles (which makes sense, as desmans only live in Europe). However, a more recent analysis, from 2013, places the desmans separately, which might support one (admittedly controversial) theory that early moles were, like desmans, semi-aquatic, and that their spade-like forefeet were originally paddles for swimming.

So we have a contradiction, and, as they so often say in science "more study is needed". However, the fossil CT scans might provide more support for the model of separate evolution of fossoriality than at first appears to be the case. That's because they also showed that one fossil species, Geotypus montisasini, is a primitive member of the Eurasian mole group. Why does that matter? Because, while it certainly would have done a lot of digging, the forelimb of Geotypus is not as well adapted to that as more modern forms are. It's as if it's half way between the shrew-mole lifestyle and that of the likes of the European mole.

But if it's identifiably a member of the Eurasian group, then it must have appeared after the split between those animals and their "North American" kin. Which would be a bit odd if the common ancestor of those two groups was already a highly specialised fossorial animal. So, yes, perhaps the two groups do share a single common ancestor, and presumably it dug tunnels or burrows of some kind. but perhaps the truly subterranean lifestyle came later - once separately, in each group.

And the fact that the two groups look so similar now is because there's only so many ways for a small insectivore to get really good at digging holes.

[Photo by Didier Descouens, from Wikimedia Commons. Cladogram adapted from Schwermann & Thompson 2015.]


  1. Why necessarily via Bering's Strait? Isn't there supposed still to've been a direct connection between N America and Europe in the Eocene?

    1. Yes, quite true, that's also possible. Depending on how early it happened, of course.