Mammals are animals that have hair and make milk. That’s an oversimplification, but it works. However, some people also add warm-blooded to the list. They’re not wrong, but you should know that the earliest mammal-like creatures were actually cold-blooded. But when exactly did this iconic transition take place?

In a new study, researchers at the University of Lisbon, the Natural History Museum in London, and the Field Museum of Natural History in Chicago, have found evidence that the first warm-blooded mammal-like ancestors may have appeared some 233 million years ago, during what’s known as the Late Triassic period. The rest, as they say, is history and to this day mammals and, interestingly enough, birds are the only warm-blooded groups of animals.

The ears of a warm mammal

Regardless of how hot or cold it may be outside, your body always maintains a constant internal temperature, with a bit of wiggle room for slight fluctuations. Just like a furnace, our body generates heat by burning the food we eat. This makes humans, and all mammals (with the notable exception of the naked mole rat and a weird squirrel named Xenospermophilus), endothermic, or warm-blooded animals. Meanwhile, ectotherms, such as fish, amphibians, reptiles, and invertebrates, get their body heat from external factors like the sun, meaning they’re cold-blooded animals.

Mammals evolved from reptiles, so they were naturally cold-blooded at first and stayed that way for tens of millions of years until they converted to an endothermic metabolism.

You can tell that early mammal ancestors, known as mammaliamorphs, eventually switched to endothermy at some point in geological history because we start seeing anatomical features linked to warm-bloodedness, as opposed to characteristics you normally see in crocodiles, snakes, and other cold-blooded reptiles. However, these features do not fossilize easily, so it’s difficult to pinpoint the exact time when the transition occurred.

But researchers now believe they’ve found a reliable proxy for the timeline of this transition to endothermy in the form of semicircular ducts in the inner ear. These ducts contain a fluid called endolymph, which helps us balance and experiences viscosity changes depending on the temperature of the animal. To accommodate the optimal flow of this fluid, mammalian inner ears have evolved different shapes, so by examining structural changes in these canals over time, it is possible to indirectly tell when the jump to warm-bloodedness was made.

“The origin of warm-bloodedness in our own mammalian lineage is interesting in itself, but a puzzling problem. We cannot stick thermometers in the armpit of your pet Dimetrodon, right? So, how on Earth would we be able to estimate body temperature in a creature 300 million years old? The origin of endothermy was based on a few key observations. First, the biomechanical equations governing the inner ear function contain a temperature term, so by reworking the equations you would be able to estimate the body temperature of any animal. Second, when any liquid heats up, its viscosity decreases. Think about the oil you put in a pan: initially it is very viscous at room temperature, but eventually it becomes more and more fluid as it heats up to fry some delicious French fries. Thirdly, the semicircular canal system of the inner ear should be viewed as a sensor, and as such, it should be optimized for the range of information it will detect,” Ricardo Araújo, a junior researcher at Instituto de Plasmas e Fusão Nuclear at the University of Lisbon in Portugal and one of the paper’s lead authors, told ZME Science.

“Therefore, given that the equations governing the semicircular canal system depend on temperature, when body temperature changes (increases for endothermy), then there must be changes in the system itself to keep it functioning correctly. In the case of the semicircular canal system in the mammalian lineage, there were changes in size and form of the semicircular canal system.”

Endothermy in mammals may have appeared in a flash

The researchers carefully examined the inner ears of 56 synapsid species, extinct ancestors to today’s mammals, and compared them to 243 living species. They found that the necessary changes required for endothermy, such as narrower ducts, appeared relatively abruptly during the Late Triassic, about 230 million years ago — some 20 million years later than scientists had previously thought mammalian warm-bloodedness evolved. This endothermic acquisition took less than a million years, rather than very gradually over tens of millions as believed before — and this was a nice surprise for the researchers when they first looked at the results.

“This was already the culmination of months, if not years, of research unveiling in front of our screens. We didn’t know what to think of it, but we also had no pre-expectation as the method we are employing was completely new. It was a mix of curiosity and bewilderment. It was like: ok, we got this. What to make of it now? I remember I was relieved that the results were very consistent, there was no real weirdness with the data. The signal was clear,” Araújo wrote in an email.

These early warm-blooded mammaliamorphs looked not all that different from your typical run-of-the-mill mammal. They had fur and had many of the body features you’d recognize in a house mouse. It is in slight changes in skull anatomy, such as different grooves in the jaw and the development of inner ear bones, as well as the rest of the skeleton that we start seeing important differences between mammals and mammaliamorphs (mammal-like ancestors but not quite mammals yet).

Before this important transition took place, cold-blooded synapsids had to hibernate and burrow in shelters in order to survive in the cold temperatures of higher latitudes. But once they transitioned to warm-bloodedness, they could roam through increasingly colder climates, all the way to the poles. They could also move faster and stay active for longer, enhancing their predation, but also their ability to evade predators of their own. But mammaliamorphs weren’t alone.

In a 2020 study, University of Bristol paleontologist Professor Mike Benton suggests the Permian-Triassic mass extinction — the most devastating mass extinction event in history, which killed as much as 95% of life — triggered an ecological arms race of sorts between mammaliamorphs and dinosaurs, and this competition brought about endothermy in both major groups. Benton’s assertions are based on hundreds of ancient fossilized footprints showing a posture shift in both dinosaurs and mammalian ancestors that happened almost instantly from a geological standpoint, sometime between 250 and 200 million years ago.

And in 2017, researchers led by Kevin Rey, now a postdoc at Vrije Universiteit Brussels, found mammal-like creatures could have first developed a warm-blooded metabolism as early as 252 million years ago, based on an analysis of the ratios of oxygen isotopes of fossils belonging to cynodontia (mammal ancestors).

However, the new findings suggest that mammalian endothermy happened much more recently, around the same time proto-mammals started to evolve whiskers, fur, and specialized backbones. In other words, warm-bloodedness appeared around the same time that mammal ancestors started to look and behave like the mammals we know today, with the important distinction that endothermy likely predated the appearance of the mammalian order. Whether or not the shape of the semicircular canals is a more reliable proxy for the origin of endothermy in mammals than other lines of evidence is debatable, but the authors of the new study make a convincing case supporting their findings.

In an email to ZME Science, Romain David, a postdoctoral researcher at the Natural History Museum and one of the study’s lead authors, says that like Rey and colleagues, they also found an increase in body temperature among cynognatians. However, David adds that the new study additionally provides values and ranges for increases in body temperature, showing that they probably fell below the known lower threshold for extant endothermy (about 31 degrees Celsius). “In addition, Rey et al. did not test Probainognathians for which we found particularly low body temperatures, suggesting that if we posit that Cynognathians were endotherms, this had likely nothing to do with the origin or mammalian endothermy and was a side experiment on their branch of the phylogenetic tree,” David says.

“Concerning things like posture changes, they are one of a variety of potential proxies for endothermy that have been discussed over the years. The problem with many of them is that they often give fairly vague or contradictory results, and the degree to which they are potentially reflective of endothermy versus other aspects of biology and ecology isn’t always clear. Our method contrasts with that in a couple of important ways. First, it is based on physical first principles of how the inner ear functions, so there is a strong reason to expect semicircular canal form to be related to body temperature. Second, our approach has been validated using a large sample of modern animals with known body temperatures. The success of our approach in estimating temperatures for those species helps to give us confidence in our results when applied to fossils. Finally, our result gives specific numeric predictions about body temperature, instead of a more qualitative endotherm/ectotherm classification,” added Kenneth Angielczyk, a paleobiologist at the Field Museum of Natural History in Chicago and lead author of the new study.

Who were the first warm-blooded creatures: dinos or mammals?

This isn’t the final nail in the coffin though. Many questions and lingering mysteries remain surrounding the origin of endothermy in mammalian creatures. For instance, we still don’t know if dinosaurs or mammalian ancestors were the first to evolve warm-bloodedness.

“Currently, the timing of the origin of endotherms in mammals and birds is much discussed. The range of suggestions is from Carboniferous to Late Jurassic, and yet this paper lines up the point of origin in mammals with the origin of mammals pretty much, as had long been suggested. However, the mammal line showed many prior adaptive shifts towards endothermic-like characters (such as lumbar ribs and presumed diaphragm, dorso-ventrally flexing vertebral column, upright posture) in the Late Permian, and strong evidence for hair in the Early Triassic. So, the origin of endothermy was probably stepwise with several important prior steps before this shift in inner ear characteristics,” commented Professor Benton for ZME Science, who reviewed the new study that appeared in Nature.

“This research sounds great. Trying to figure out the origin of mammalian endothermy has been of a subject of study for a while. This new method is interesting and proposes a new age for the origin of mammalian endothermy which seems to be very restrictive and abrupt in time. As this is a new promising proxy, future studies, and with a better knowledge of the different parameters estimated and calculated might be able to answer the remaining uncertain taxa and possibly re-estimate previous ones. In my opinion, they managed to set up an upper limit of the origin of mammalian endothermy, when it was fully formed and really similar to what we have today. The divergence of time with other studies might be due to the development of, at least, partial endothermy which took time to get printed in the morphology,” Rey told ZME Science.

“We found out that the origin of endothermy in the mammalian lineage happened more recently than the Permo-Triassic boundary, namely at around 233 million years ago. Now, the million-dollar question is on the dinosaur side: did dinosaurs or their ancestors become endothermic before, at the same time, or after mammaliamorphs did? Once this question is finally settled we can truly evaluate competition models such as between endothermic mammal ancestors and dinosaurs,” Araújo said.

All in all, although there are many unknowns and unsolved mysteries concerning the evolution of warm-bloodedness, this new study adds great depth to the debate, inching us closer to answering one of the greatest riddles of paleontology. And it’s all thanks to the institution of science, which always builds upon previous work, brick by brick, as Araújo kindly reminds us:

“I will never forget after the first round of reviews when we were working for months to no end, from cold nights to torrid summer afternoons, collecting data from literally hundreds of papers to have a solid comparative database of current climates, body mass, body temperature, stride frequency, aerobic speed, and so on. And I remember thinking: this fundamental research on the tailbeat frequency of a deep-sea fish is actually useful to help to solve one of the longest riddles yet of modern physiology and paleontology. It was a great feeling to realize that we were a tiny part of an entire effort of generations of scientists that are attempting to understand our world. From the scientist that put on a scale in a gelid night a Greenland shark, to the scientist that implanted a thermometer under the skin of a Madagascan tenrec, we couldn’t be more grateful that such data exists. This realization underscores the serendipitous importance of fundamental research.”