Life of an Almost-Mammal

Scaloposaurus, a therocephalian.
This artist has chosen to show it hairy, which is plausible,
but uncertain.

When I was at school, I was taught that mammals evolved from reptiles. However, at least under the modern definition of 'reptile', that isn't really true. In fact, the mammals belong to their own, distinct, evolutionary lineage, one that has existed alongside the reptiles ever since the latter first came into being. This is a staggeringly long period of time - by definition, it's from long before there were any dinosaurs, for example.

However, this evolutionary group didn't just include the mammals themselves. Mammals may not have evolved from reptiles, but they haven't been around anything like as long in the fossil record. So, if we look back to before the first mammals evolved, we find that there were many other kinds of creature belonging to this group that are no longer around today. Because evolution isn't really a ladder, some of these animals survived alongside the early mammals. In fact, the last ones didn't die out until well after the split between placentals and marsupials.

Studying fossils of these creatures can tell us a lot about where mammals came from, and how they became so different from the reptiles.

This evolutionary group, taken as a whole, is called the Synapsida. Within it, those animals that most resemble mammals, without actually being them, are called the Therapsida. Strictly speaking, that's not a sound definition, since any formal description would necessarily have to concede that mammals are just a special kind of therapsid (and therefore have to be included). But, for our purposes, it will do as a general term for these 'not-quite-mammals'.

The therapsids survived for a very long time, and there were lots of different kinds, some of them more mammal-like than others. The group that probably includes the immediate ancestors of the mammals are called the cynodonts, but, apart from them, the next most mammal-like group are the therocephalians.

In all probability, therocephalians are not the ancestors of the mammals, or of the cynodonts in general. Instead, they are generally thought to be the sisters of the cynodonts; their closest relatives on the evolutionary tree. They lived between the Permian and Triassic periods, dying out not long before the first dinosaurs appeared. What's particularly interesting is that they survived the mass extinction at the end of the Permian - a vastly destructive event that far outweighs the relatively puny K-T extinction that would later wipe out the dinosaurs. Sure, the forms living after the event were smaller and more lightly built than those that preceded it, but the fact that they survived at all is quite impressive.

One of these survivors was Tetracynodon darti, a slim animal, about the size of a small rat, that lived in what is now South Africa around 250 million years ago. In comparison, the first dinosaurs didn't appear until about 230 million years ago, so we are really talking 'old' here. It was first named and described by Denise Sigogneau in 1963, and relatively little work has been done on it since. She regarded it as a very close relative of the earlier fossil Tetracynodon tenuis, which, amazingly, lived on the other side of the Permian-Triassic boundary. The fossils do look remarkably similar, but is it really the case that these animals survived virtually unchanged for millions of years, and across the greatest extinction event in Earth's history?

Maybe not, since, while there doesn't seem to be any doubt about Sigogneau's work, there is some debate about the other species, which is known only from a single fossil. It's part of a general debate about small therocephalian species: namely, how many of them are really small species, and how many are just fossils of infants? In the case of T. tenuis, it may well be a juvenile of the closely related - but much larger - Lycideops. And maybe it just looks similar to the other species because the young animals looked more alike than the adults. If so, the name Tetracynodon isn't valid, and T. darti is going to have to be given a new one. That hasn't happened yet, though, so I'm going to have to stick with the only name I've got.

Sigogneau had just one fossil to work from, but since then, we've found plenty more belonging to the same species. A recent study by Trond Sigurdsen and co-workers reviewed over a dozen of these fossils, as well as making use of technology developed since 1963 to build up a far more complete picture of what these animals were like.

A look at its skull shows a curious blend of mammalian and "reptilian" features. One of the defining features of a mammal, at least in fossils, is that the lower jaw is composed of a single bone on each side, and there are three bones in the middle ear. A cursory look at the skull of T. darti appears to show that it's jaw, too, has only a single bone on each side. However, if you turn it over and take a look at the inner surface of the jaw (that is, on the side towards towards the tongue) it becomes apparent that there are several other bones there, crucially including one of the ones that's part of the middle ear in mammals. As with other therapsids, then, the jaw is neither truly mammalian, nor reptilian, but somewhere in between.

Another distinction between reptiles and mammals lies in the teeth. Although the pattern is highly modified in various animals, the basic layout is that mammalian teeth include incisors, canines, premolars, and molars, while reptiles have no such distinction. In this creature, the teeth look somewhat reptilian, mostly consisting of stabbing pointed teeth, arranged much as they would be in, say, a lizard. Again, though, there's more to it than that, because these animals did have a pair of large canine teeth, something today found only in mammals. The presence of an additional pair of un-erupted canines close to the visible ones may suggest that it was capable of replacing its canines at least once in its life (more, if what we're seeing aren't "milk teeth").

In front of the canines, however, there are seven incisors on each side of the upper jaw and five in the lower - more than in any living mammal. The upper jaw also has an additional two teeth on each side between the canines and the incisors proper; again, something we don't see in mammals. There are a number of teeth behind the canines, too, which are much simpler in shape than we'd find in most true premolars and molars.

In mammals, the chest is protected by a ribcage. Reptiles also have a ribcage, of course, but, unlike us, they often also have ribs connected to the vertebrae in the neck and abdomen. These are all "floating" or "spare" ribs, not joined to the breast bone, but, especially in the abdomen, they can be quite large. Unlike mammals, T. darti did have small ribs attached to the some of the bones in the neck. None of the fossils are complete enough to show us what's going on further back, although what we know from other therocephalians suggest that there should be rather small ribs attached to the lumbar vertebrae as well. It's been suggested that the fact that these ribs are so small suggests a mammal-like anatomy, with a muscular diaphragm separating the chest and abdomen - something reptiles don't have.

We can't say whether therocephalians - still less this particular species - really had a diaphragm or not. But that doesn't mean that we can't say anything about their soft anatomy. With the aid of modern CT scanning, used in medicine to examine the insides of a patient's body, we can look at the inner surface of the skull without having to break open the fossil. That allows us to gain some idea of the shape of the brain.

It turns out that the brain of T. darti is also intermediate between that of reptiles and mammals. For instance, it has the overall tubular shape that we see in reptile brains, but not as extreme. The outer surface of the forehead is also solid, without the small hole that is found in earlier therapsids, and in many reptiles (including some living ones). That hole is there to house a strange, light-sensitive organ, sometimes called the "third eye", which these creatures, like mammals, evidently didn't have.

Incidentally, if reading that last sentence leaves you thinking "wait, are saying that some therapsids did have three eyes? WTF?", then, yes, you read it correctly. I mean, it wasn't, strictly speaking, an actual eye, and it was pretty tiny and probably completely covered by skin... but, yeah. Three "eyes". I kid you not.

Anyway, as I mentioned above, a long-standing question about these animals is whether they're really just juvenile representatives of some other species, possibly one we already know about. To check that out, the researchers examined the bones of one of the fossils under the microscope, to see how mature they looked. They conclude that the animal wasn't a juvenile, although there are plenty of other features in the skeleton that suggest it hadn't finished growing yet, either. So probably a "sub-adult", maybe a year or so old, and not far off the adult size. You might expect that to be the case, given that the earliest mammals were quite small, but it's worth noting that many other early therapsids were actually fairly large, and so the small size might be somehow related to its survival through the devastating end-Permian extinction.

A particularly interesting question, though, is whether these things were warm-blooded, and therefore, whether they were likely to have hair. It's hard to know that for sure, for obvious reasons, but it's been suggested that one clue might lie in the turbinate bones. These are thin, delicate, pieces of bone inside the nasal cavity, that act, in part, as heat-exchange baffles, keeping the air inside the respiratory system warm and humid. They are also found in reptiles and birds, partly because they also house the cells responsible for the sense of smell, but they are particularly well-developed in mammals.

The evidence from the CT scans seems to show that the turbinates in T. darti were not very extensive. They were certainly there, but that's hardly surprising, and they generally seem to be closer in structure to those of earlier therapsids than to those of mammals. So probably not warm-blooded, then. On the other hand, they do have a partially developed hard palate, something that's sometimes also thought to be associated with a higher metabolic rate (crocodiles have something similar, too, but that's probably to stop their nose filling with water when they open their mouth under the surface). The diaphragm, if they had one, would be another pointer in that direction, so, just possibly, they were slightly more warm-blooded than living reptiles are.

In summary, what we have here is an animal that clearly is not a mammal, and yet that has a number of features that are quite mammal-like. Since mammals are not descended from therocephalians, much of this may be parallel evolution, building on the features found in the common ancestor of these two groups. The slender snout, pointed teeth, and agile-looking limbs of this animal suggest a fast-moving carnivore, snapping at large insects, or maybe even especially small reptiles. A vicious little predator, a rat-sized, half-reptile-half-shrew, roaming South Africa 20 million years before the first of the dinosaurs.

[Image by Dmitri Bogdanov, from Wikimedia Commons]