The origin of mammalian endothermy

[¡Léalo en español!] [Przeczytaj to po polsku!] [pdf file v. 30.09.2016] [doi: 10.13140/RG.2.2.35081.44641]

Probable hair in a therapsid coprolite from the Permian of Russia (photo by K. Owocki; see Bajdek et al., 2016)

Ectothermic or endothermic

An obvious difference between modern reptiles and mammals is that the former are ectotherms and the latter endotherms: whereas the body temperature of reptiles is dependent on the prevailing environmental temperatures, it is regulated to remain nearly constant in mammals. We can also note that: (1) mammals have fur and reptiles do not, and (2) reptiles are characterized by a much longer digestion (slower metabolism) than mammals of a comparable body mass. As a consequence, many modern reptiles have a “lower-energy” lifestyle than mammals.

It becomes less obvious when we go back in time… Mammals evolved in the Triassic Period, and questions paleontologists ask themselves are: when did the endothermy in their evolutionary lineage appear? Were already the late Palaeozoic and early Mesozoic ancestors of mammals (called therapsids or mammal-like reptiles) endothermic?

In the 1970s, the American paleontologist Robert T. Bakker published innovative ideas on the physiology of mammal-like reptiles (and dinosaurs). He hypothesized that therapsids had fur and provided several lines of evidence that they were endotherms: (1) Therapsid bones lacked growth rings and had closely packed blood vessels and Haversian canals. (2) Some of them were distributed in cold temperate zones. (3) Short, stocky body proportions in many therapsids might have been a device to conserve heat. (4) Their predator–prey ratios, which depend on the energetic requirements of the predators, were lower than those of ectotherms.

R.T. Bakker’s ideas on the physiology of therapsids (and dinosaurs) have been accepted by most paleontologists and are well-known to the public. Nevertheless, because many of them are rather indirect lines of evidence, we now seek for new clues. Coprolites, i.e. fossil feces, being metabolistic byproducts provide such new and valuable data on the metabolism of their producers. In fact, the study of coprolites has during the last couple of years shed new light on the physiology of mammalian ancestors and their relatives. For example, it was argued in the article on coprolites of a giant dicynodont from the Triassic of Poland, that these herbivorous and toothless therapsids had a rather slow metabolism.

Furthermore, in 2016, another piece of the puzzle of mammalian endothermy was added when our team composed of seven researchers from Poland, Sweden, and Russia (P. Bajdek, M. Qvarnström, K. Owocki, T. Sulej, A.G. Sennikov, V.K. Golubev, and G. Niedźwiedzki) published a new paper on coprolites. The coprolites we studied were produced by Late Permian carnivorous therapsids, over 252 million years ago, and excavated during a Polish–Russian expedition to the Vyazniki site, European Russia.

Undigested bones and the fast metabolism

The Vyazniki site has yielded various morphotypes of coprolites (which will soon be discussed again on the blog). Our paper from 2016 focuses on only two big coprolite morphotypes: A and B. Whereas undigested bone fragments are present in the type-A coprolites, they are quite rare and highly degraded in the type-B coprolites. As said above, reptiles are characterized by a much longer digestion than mammals, and, for example, crocodiles practically completely digest the bones they ingest. On the contrary, undigested bones are commonly found in the feces of mammals. Following these arguments, the bone-rich coprolites of type A would have interestingly been produced by some kind of animals of a fast metabolism.

As such, the team of Krzysztof Owocki and Grzegorz Niedźwiedzki ascribed, in 2012, the bone-rich type-A coprolites to therapsid carnivores and the bone-barren type-B coprolites to archosauromorphs or other non-therapsid carnivores. Both therapsids and archosauromorphs are known from the fossil record of Vyazniki, but therapsids would have been more expected to have a fast metabolism than early archosauromorphs (ancestors of modern crocodiles and birds). This interpretation is supported by finds from the Upper Permian of South Africa. Already in 2011, the paleobiological context allowed Roger M.H. Smith and Jennifer Botha-Brink to link several bone-rich coprolite morphotypes from the Upper Permian of South Africa to carnivorous therapsids.

The oldest hairs

The researchers from South Africa have found more than just bones in the Permian coprolites. Some coprolites contain enigmatic elongated structures, that are on average 14 μm in diameter and reaching up to 5 mm in length. Roger M.H. Smith and Jennifer Botha-Brink suggested that these structures were remains of plants, fungi or, perhaps, hairs. It was exciting to our team to find comparable structures in a therapsid coprolite from the Upper Permian of Vyazniki, Russia. By the use of light and scanning electron microscopes we studied them in great detail, including their geochemistry. The structures we described from Russia are ten time larger in diameter than those from South Africa, and the largest one is over 5 mm long. They are interpreted as molds of hair-like elements; some even appear to show bifurcated hair roots! Hairs are well-resistant to digestion and often found in feces of modern carnivores.

If this interpretation is correct, these hairs are two times older than the previously earliest record of known hairs from Jurassic-Cretaceous mammals and imply that some therapsids had acquired insulation by the latest Paleozoic, prior to the rise of mammals. Hairs would probably have had a thermoregulatory function, as an insulation. Some researchers have also suggested that hairs could be tactile in origin. In 1968, G.H. Findlay hypothesized that perforations present in a skull of the Late Permian therapsid Olivera parringtoni reveal the presence of tactile hairs. Such hairs could have been of great use especially if mammals are descended from nocturnal reptiles. Hairs would make up for poor vision and moreover allow to conserve heat at night.

The discoveries from South Africa and Russia suggest that Late Permian therapsid carnivores had developed (1) an insulation (fur) and (2) an accelerated metabolism. Taken together, these features make us suspect that the late Paleozoic ancestors of mammals were already endotherms.

Piotr Bajdek 1 and Martin Qvarnström 2
1 Częstochowa, Poland
2 Uppsala University, Sweden


Bajdek, P., Owocki, K., Niedźwiedzki, G., 2014. Putative dicynodont coprolites from the Upper Triassic of Poland. Palaeogeogr. Palaeoclimatol. Palaeoecol. 411, 1–17. doi: 10.1016/j.palaeo.2014.06.013

Bajdek, P., Qvarnström, M., Owocki, K., Sulej, T., Sennikov, A.G., Golubev, V.K., Niedźwiedzki, G., 2016. Microbiota and food residues including possible evidence of pre-mammalian hair in Upper Permian coprolites from Russia. Lethaia 49, 455–477. doi: 10.1111/let.12156

Bakker, R.T., 1971. Dinosaur physiology and the origin of mammals. Evolution 25, 636–658.

Bakker, R.T., 1975. Dinosaur renaissance. Scientific American 232, 58–78.

Findlay, G.H., 1968. On the scaloposaurid skull of Oliviera parringtoni, Brink with a note on the origin of hair. Palaeontologia Africana 11, 47–59.

Owocki, K., Niedźwiedzki, G., Sennikov, A.G., Golubev, V.K., Janiszewska, K., Sulej, T., 2012. Upper Permian vertebrate coprolites from Vyazniki and Gorokhovets, Vyatkian regional stage, Russian Platform. Palaios 27, 867–877. doi: palo.2012.p12-017r

Smith, R.M.H., Botha-Brink, J., 2011. Morphology and composition of bone-bearing coprolites from the Late Permian Beaufort Group, Karoo Basin, South Africa. Palaeogeogr. Palaeoclimatol. Palaeoecol. 312, 40–53. doi: 10.1016/j.palaeo.2011.09.006



  1. OON

    This is extremely exciting! I was excited by therapsids since my childhood when I had a wonderful book about prehistoric creatures, always wanting to know whether Inostrancevia and (probably surprise) Dvinia were really furry as they were shown there.
    Is it possible to pinpoint the lineage to which the animal that left those coprolites with hair belonged (though I wonder how it can be done)? If it was not a cynodont could it mean that hairs was a common feature of all theriodonts?


    • Piotr Bajdek

      Thank you for the comment! 🙂 Your question is very good, yet it may be hard to answer convincingly by now. All we have are bone-bearing carnivore coprolites, some of them containing possible hairs, and the list of fauna known from the sites.

      Carnivorous therapsids of Vyazniki include therocephalians, as Moschowhaitsia vjuschkovi, Megawhaitsia patrichae, and a new undescribed therocephalian. The big bone-rich coprolites of the morphotype A would have been produced most likely by a large therocephalian, as e.g. Moschowhaitsia vjuschkovi. Whereas, carnivorous therapsids of the Beaufort Group, Karoo Basin, South Africa, include gorgonopsians, therocephalians and non-mammaliaform cynodonts. The gorgonopsians were pointed out as probable producers of some bone-rich coprolites (Smith and Botha-Brink, 2011). Thus, I’m inclined to say that a relatively short digestion and a high metabolism rate were common features of all the theriodonts and to me personally, it seems quite conceivable that they all had fur. Possible hairs are known in coprolite materials from the Late Permian of both Russia and South Africa…

      On the other hand, note that the hairs quite likely derive from prey species, although animals sometimes swallow a little of their own hair. Although I would argue that dicynodonts were not fully-endothermic animals, quite interestingly, dicynodonts were among the most probable prey species in both the Beaufort Group, South Africa (as suggested by Smith and Botha-Brink, 2011), and the Vyazniki site, Russia… In fact, coprolites of the morphotype A from Russia contain well-vascularized bone fragments, most likely of therapsids and possibly of dicynodonts which were included in the Vyazniki fauna. Thus, we cannot rule out that not only theriodonts possessed fur but by now we don’t have enough evidence to really say.


      • OON

        Thanks for your reply and the papers you’ve sent (I actually wanted to check the first one before commenting but couldn’t do it right now) This all is very cool – a lot of info about those animals from the source I never thought of.
        If we suppose for a second that dicynodonts had hair after all I wonder whether they were somewhat endothermic in the primitive form and dropped their metabolic rates back when they became large. Or if their hair developed not for endothermy but as whiskers for sensation when they were small burrowers… Ok, as I’m not even sure that they were small burrowers, enough with diletant’s fantasies=))) Hope that you and other paleontologists will get more and more interesting finds!


  2. prak

    Amazing article, I have been looking for something like this for quite some time. I have heard that monitor lizards also have fairly high metabolic rates, is it possible for an ectotherm with a monitor-lizard like metabolism to have produced those coprolites?


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s