Fossil Martens... or Not?

When talking about fossil animals on this blog I often mention the earliest known example of a particular group. But this often hides a degree of uncertainty, or even controversy, because such the exact identity of such fossils can be difficult to pinpoint. That's partly because, being, by definition, older than other fossil examples of the group, they are also the most likely to be incomplete or poorly preserved. Often, since we're talking about mammals here, the "oldest known fossil" may consist of little more than a distinctive tooth. 

The second, and perhaps even bigger, problem is that the further you go back to the origin of a group the more it blurs into whatever preceded it. Even if we had perfect remains, or if we could travel back in time and see the animals in life, or take blood samples from them for genetic analysis, there would always be a question of what exactly we were looking at. Where do you draw the line when, in reality, one group will have slowly and perhaps imperceptibly, evolved into a newer one?

The mustelids, or weasel family, date back a long way, probably somewhere around 30 million years to the Late Oligocene. For a large chunk of the time since, the fossils that we can place in the family don't necessarily belong to any of the living subfamilies, which seem to pop up around the Middle Miocene. So, when we look at fossil mustelids of about that age, there's naturally a question as to what exactly we're looking at and where it might fit in the relevant family tree.

Through much of the 20th century, most small to medium mustelid fossils from the Middle Miocene have been placed, if somewhat tentatively, in the genus Martes. This is the genus to which the living pine and beech martens belong, among other animals, making it the longest surviving of all mustelid genera, allegedly dating back 18 million years. Recently, however, this placement has begun to be questioned.

The problem is that martens, in many respects, are not highly specialised mustelids, even to the extent that their closest relatives, the wolverines, are. So it could be that the features we see in these often fragmentary fossils that remind us of martens are simply ancestral traits that had not yet been lost in lines leading to other animals - or in lines that simply died out. What would have, if this is right, is what's termed a "wastebasket taxon" - a category used to place fossils in when we're not really sure what they are, but we have to call them something.

So what is the oldest fossil marten? At least three possible candidates have been proposed over the last few decades. In 1994, in a book on marten conservation, the chapter on evolution proposed that the oldest fossil belonged to the species Martes laevidens, which lived in Germany 20 million years ago, during the Early Miocene. However, more recent studies have shown that it probably isn't, and in any event, it predates even the most generous estimates from genetic analysis as to when the group can plausibly have arisen. 

That leaves us with two obvious candidates (and there may be others). Martes mellibula lived in Spain about 9 million years ago, while Martes wenzensis dates from just 4 million years ago in Poland. The latter is indisputably later than the split between martens and wolverines and it looks remarkably like a pine marten, differing only in being slightly bigger. So there's not really any doubt that it, at least, really was a marten; the question is as to whether some of the older examples might be ones, too.

A recently published paper concludes that Martes mellibula probably was a marten, while acknowledging that the question has yet to be fully resolved. But that still leaves open the issue of all those other supposed members of the genus that have been identified over the years. Do they count?

Significantly, the paper describes a new species of fossil mustelid and tries to use it to put these other early possibilities into context. They named the new species Araganonictis araid, and it lived in Spain around 12 million years ago. This places it as living at around the time that molecular estimates tell us that the marten/wolverine lineage diverged from the "musteline" lineage that includes weasels, stoats, polecats, and mink. Nonetheless, and bearing in mind that they only have a lower jaw and its teeth to go on, they believe that it belonged to the former, but not to any of the living genera. (In other words, it's much an "early wolverine" as it is an "early marten" being related to the common ancestor of both of those animals, and of the South American tayra).

By comparing this new fossil with others placed in the genus, the authors have at least made a start at clearing out the wastebasket, assigning some early fossils to other genera, and changing some of the nomenclature. Aragonictis itself, for example, seems to be a close relative of "Martes" laevidens, the other proposed early marten, which the authors tentatively assign to the genus Circamustela, first identified from 10 million year old deposits in Spain and more recently noted as also living in Germany alongside the possibly bipedal ape Danuvius.

Arranging these various fossils into a time frame, it appears that the marten/wolverine group underwent two rapid bursts of evolutionary change in Miocene Europe. The first of these would be the one that heralded their origin, during the very warmest spell of the Miocene, although it's hard to pin down a more precise dating. The second corresponds with the appearance of Aragonictis and of the many other early marten-like animals that have so often been mixed up together with their modern relatives.

Aragonictis was smaller than living martens, but larger than typical weasels. It probably looked more like the former, and its sharp teeth indicate a shift towards a more heavily meat-based diet than that of its ancestors.  At least in Spain, the arrival of these new species may be connected with increasing rainfall at the time and the consequent expansion of the local forests. The fossil was discovered in what would originally have been a wooded area in a river delta; the site has also revealed fossil turtles and beavers among many other animals. This would have been a good habitat for small carnivores with marten-like habits, as animals like the living pine martens and beech martens show, which may explain why so many suddenly appeared. 

Over the next few million years they increased in size, with some eventually splitting off to produce the first true wolverines. But the complexities of evolution and the lack of hard dividing lines means that, among mustelids as amongst other animals, there are still more nuances and details to be uncovered.

[Photo by Surrey John, from Wikimedia Commons.]