Sunday 16 April 2023

The Life and Habits of the Father of Cats

Today, all large land-dwelling carnivorous placental mammals belong to a single group named, appropriately enough, the Carnivora. This includes all the cats, dogs, bears, badgers, and many others besides, and it has long been clear that they are a genuine biological group, all more closely related to each other than to anything else. Indeed, the similarities between these animals are sufficient that, even though the current name only dates from 1821, it's essentially identical to the "Ferae" order first described back in the 18th century.

Even if we look at the fossil record, we find that for millions of years, the great majority of such animals were also members of the Carnivora even if, in some cases, their specific family no longer exists. The great majority, but not all. In 1875, American palaeontologist Edward Drinker Cope coined the term "creodont" for a collection of early carnivorous mammals that lacked the key defining features of the carnivorans. His original definition was relatively narrow, but over the next few decades it was expanded until, in 1909, it came to include essentially all of the large land-dwelling carnivorous placentals that weren't carnivorans.

That didn't quite stick, because by the 1960s it had become clear that some particularly fierce-looking fossil animals belonged to an entirely different branch of the mammalian family tree. Nonetheless, the term "creodont" remained in use for a significant number of fossil animals which, superficially, looked like carnivorans but which lacked the particular features that define the group. Taking that definition, we can say that the creodonts were the dominant land-dwelling carnivorous mammals up until around 35 million years ago, when the carnivorans that we now know began to supplant them. Even so, the last ones died out, in Africa, just 9 million years ago during the Late Miocene.

This is no longer considered valid.

That is because, in the 1990s it started to become clear that the "creodonts" were not themselves a single group of mammals - that is, they did not share a single common ancestor that they didn't also share with something else. This has, in effect, brought back the old "Ferae" grouping, albeit with a wider definition than it had when originally proposed. Although not all scientists necessarily agree, the consensus view now is that what we previously thought were "creodonts" are actually two different groups of mammal.

One of these, probably the closest relatives of the carnivorans, are the hyaeonodonts. It is this group that lived alongside, and was eventually supplanted by, the carnivorans. Hyaeonodon itself is the most famous example, but I've discussed others on this blog in the past. The second group is rather less well-known.

These are the oxyaenids, and part of the reason we don't know much about them is that they lived a very long time ago. While they first appeared at the same time as the first hyaenodonts, they only lasted until around 40 million years ago, dying out well before the carnivorans started to become dominant. From a modern perspective, this makes them look like an evolutionary failure, but they still lived for over 20 million years - which is longer than great apes have so far managed.

The current Wikipedia article on oxyaenids states (without attribution) that they "walked on flat feet" and "were capable of climbing trees", suggesting something rather like a pine marten. But is this really true of all members of the group?

While many oxyaenids were roughly badger-sized, one of the larger examples was Patriofelis, with an estimated nose-to-rump length of about 120 cm (4 feet) - the size of a small leopard. It was first described by Joseph Leidy in 1870, and given a name that translates as "father of cats". All he had to go on at the time was a jawbone and, indeed, this looks vaguely cat-like, being relatively short with powerful meat-tearing teeth - but, of course, it isn't really a cat, or even the ancestor of one. 

With only a jaw to look at, all Leidy could conclude was that the animal was a carnivore, but more complete skeletons soon surfaced. In 1894, Jacob Wortman published a detailed description, concluding that the animal walked on the soles of its feet and had a long, powerful tail. Based on which, together with some other features of the feet, he concluded that it was probably semi-aquatic, like a sort of gigantic otter. Indeed, at the time, he thought it might have been an ancestor of seals.

There has been argument about this pretty much ever since, although the debate has generally shifted away from the semi-aquatic hypothesis to the narrower question of whether or not it regularly climbed trees. However, the last time anyone gave a detailed description of the animal's skeleton was in 1938, so it has been more a matter of arguing over past interpretations than examining new evidence about what are, after all, relatively obscure animals.

Nonetheless, one fossil dug up in Wyoming in 1953 was unusually complete, including parts of both left legs, the hips, ribs, and much of the backbone... but at the time, only the attached skull was described. Last year, researchers published the first description of the rest of the skeleton, with a view to settling the question of how it lived.

This will probably not be the final word on the matter; if palaeontologists have been debating this since the 19th century, there's no particular reason to suppose they're about to stop now. Even so, we do have what appears to be a new piece of information, which is the first description of the animal's backbone. This turns out to have been relatively inflexible, quite different from those of modern carnivorans and possibly a holdover from the creature's own ancestors. 

A similarly stiff backbone is, however, seen in many fast-running hoofed mammals, providing a brace for the limb muscles. The problem here is that Patriofelis had short, if powerfully muscular, limbs with flat bear-like feet. Not the sort of thing that would be good for prolonged running, although it wouldn't rule out the short bursts of speed that a bear is capable of. They would also be rubbish for swimming and an inflexible back would make that even worse, so the 'semi-aquatic' idea is pretty much ruled out.

More significantly, that backbone would also make it difficult to climb trees. Furthermore, the claws are not hooked as they might be for digging into bark, and the fingers and toes aren't long enough to grip branches. Plus, Patriofelis is just really big. This doesn't necessarily mean that the animal couldn't climb at all, but, in the authors' opinions, it's unlikely to have spent much time in the branches. At best, they suggest, it could have behaved like a leopard, climbing when it had to, but not exactly what you'd call an arboreal animal. And even that may be a stretch.

So what are we left with? Well, it does have attachment points for large, strong, muscles, especially on the limbs and neck. And that jaw is a little cat-like, with powerful sharp teeth that mean it can only have been an aggressive predator. At the time it was alive, 48 million years ago, Wyoming was covered in subtropical forest dominated by palm trees and sequoia, with a thick undergrowth of ferns. (As I said, this was a long time ago...) This, the authors suggest, implies that the animal was an ambush predator, lying in wait in dense vegetation and then suddenly leaping out on passing prey animals such as early brontotheres or the tapir-like Hyrachyus

The short legs, flat paws, and inflexible spine mean that it would not have been as effective at leaping as a modern cat, but otherwise, the suggested lifestyle is similar to that of some big cats today. The powerfully muscular limbs would have helped it wrestle its victims to the ground and the inflexible spine may even have protected against the risk of it being twisted or broken during a fight. 

So far as we can tell from other fossils in the same deposits, Patriofelis would have been the largest predator in the locality at the time. It may not have been great at climbing trees, but it sounds as if it would have been pretty fearsome.

[Photo by "Gally242" from Wikimedia Commons.]


  1. As oxyaenids, hyaenodonts, and carnivorans derive their carnassials from different teeth, their common ancestor would presumably be a (dentally) unspecialized omnivore?

    1. I believe the most primitive (putative) relatives are thought to be insectivores, similar to shrews. So, dentally unspecialised, certainly.