We have talked about taxonomy a couple times on this site. For those who don’t know, taxonomy is the science of how living things are related to one another. In the modern day, we have the extremely advanced version of taxonomy known as cladistics, based on in-detail examination of as many features as possible to determine these relationships. However, the further back we go, the more primitive our understanding of the tree of life got. One unusual theory to emerge in the days before more refined cladistics was the Haematotheria hypothesis, and it returned more recently than you would think.
The realization that all living things could be classified in a complex tree predates any understanding of evolution. While Carolus Linnaeus made the first version of our modern system, some of the biases that caused the emergence of Haematothermia were apparent as early as classical Greece, about 300 BC. Aristotle’s work with biology was quite impressive for its time, and he knew that among animals, there were those that could regulate their body temperature on their own (warm-blooded) and those who could not (cold-blooded), so this formed one of his major divisions among vertebrates. Aristotle, however, did not have a systematic understanding of animal relationships, so it was only with taxonomy that this strange grouping was born.
He also suggested that animals could be split into those who lay eggs and those who give live birth, but some snakes fall into the later group, and they’re just one example of boundary-breakers in the animal kingdom.
Richard Owens is sometimes recognized as the originator of the idea, but as far as I can tell, he explicitly rejected it. The closest he came was saying that crocodiles were an intermediary between birds and mammals. In his series “On the Anatomy of Vertebrates”, he expresses multiple times that while birds and mammals share similar adaptations for thermoregulation, birds are more closely related to the egg-laying reptiles. The term “haematothermia” was coined as an unnatural grouping for convenience, not a real taxonomic one. Owens instead thought that the ancestors of birds were the pterosaurs, citing the presence of wings and air sacks in both groups. This wasn’t true, but he wasn’t the only person to believe it, and this belief led to the earliest connection between birds and mammals I can find.
Pterosaurs have been known for longer than dinosaurs, and while the general view has always been that they were reptiles, there were always some dissenters who instead thought that they were mammals. Samuel von Soemmerring, a scientist whose work on the human nervous system still holds up, suggested that Pterodactylus was a bat, but a bat that was also intermediate between mammals and birds. As such, he gave it the generic name “Ornithocephalus“, effectively “bird head”. However, it should be noted that he didn’t mean intermediate in the sense that bats evolved directly into birds through pterosaurs. His understanding was more of an archetypical one, similar to the idea of Platonic ideals. To him, a pterosaur was metaphysically both a bird and a mammal.
Edward Newman’s reconstruction of mammalian pterosaurs.
Mammalian pterosaurs were still occasionally seen in palaeoart up through the 1850s, but they were long discredited by the time the modern haematotheria hypothesis emerged. This would end up being 1982, when recently passed British paleontologist Brian G. Gardiner first published his new view of tetrapod classification. The mainline view was (and still is) that mammals split off early from other amniotes (reptiles, birds, and mammals) and birds came from dinosaurs. Gardiner agreed that birds were dinosaurs, but suggested that crocodiles and mammals shared a common ancestry with them that excluded all other living animals. In other words, they were all thecodonts.
Thecodontia was a popular taxonomic grouping at the time, a group of Permian and Triassic reptiles defined by having their teeth rooted in sockets. This distinguished them from most modern reptiles, which have teeth either effectively “glued” to the top of the jaw or nested along the inside of it. However, mammals also have socketed teeth. The mainstream view is that this is a convergently evolved feature, but Gardiner saw it as one of the main points of evidence that mammals were thecodonts. Birds don’t have teeth, but their dinosaur ancestors did, and they were indeed socketed.
Euparkeria by Nobu Tamura. The ancestors of the archosaurs (theocodonts) likely looked a lot like this. If the haematotheria hypothesis was correct, they would also have a coating of feathers.
Gardiner also suggested that feathers and hair are homologous, which means their ancestor would have had a similar coating. This isn’t just interesting because of the images it has resulted in illustrating that potential ancestor; it also suggests that warm-bloodedness is their ancestral condition, with the fur serving to retain body heat. In this view, full endothermy, where a creature entirely regulates its own body temperature, only evolved once and was lost in crocodiles.
Mammals are classically considered as the last living members of a larger group called the synapsids. Other members of the lineage include what are casually called “mammal-like reptiles”, most famously animals like Dimetrodon. Gardiner argued that the pelycosaurs (the family of synapsids that includes Dimetrodon) were a group of reptiles unrelated to mammals, while the therapsids (whose best-known non-mammalian members are probably the gorgonopsids) were thecodonts and the ancestors of mammals. On a surface level, this is reasonable enough, with the two groups having very different appearances and apparent relations to mammals.
Inostrancevia alexandri (a therapsid) hunting a Scutosaurus (not a synapsid). Inostrancevia was relatively close to mammals and might have even had hair and mammalian snouts, like a dog’s wet nose. Mauricio Anton.
However, it was well-evidenced then and even more now that the therapsids evolved from within the pelycosaurs, developing more mammalian features along the way. A more recent study was able to show the gradual cumulation of mammal traits between the pelycosaurs and later synapsids, with even gaps in the fossil record matching the same rate of evolution shown in complete lineages. This separates mammals from birds about as clearly as possible. Modern genetic evidence also supported the traditional view that mammals were the first group of modern amniotes to branch off, with even turtles being closer to birds than us.
Haematotheria was never a popular hypothesis, and it has been rendered completely irrelevant in recent years, so why did I talk about it at all? Mostly for the really wild creatures it implies, like furry crocodile ancestors and the hypothetical common ancestor of all these very different animals, but also to explore how our understanding of taxonomy has changed over the centuries.
Want more paleontology? Click the links below!
