Skip to main content

How To Take a Dinosaur's Temperature

Jurassic sauropod
A Jurassic sauropod. (Image credit: Illustrated by Russell Hawley, Tate Geological Museum)

This Behind the Scenes article was provided to LiveScience in partnership with the National Science Foundation.

For the longest time, most scientists believed it wasn't possible to accurately measure the body temperature of dinosaurs. They could only make educated guesses by, for example, calculating how fast the creatures ran based on the spacing of their tracks, or from measuring the growth rates of their bones.

How warm (or cold) these long extinct creatures were remained an enduring mystery — until now.

Using a new approach, a team of researchers led by the California Institute of Technology (also known as Caltech) figured out how to take the body temperatures of dinosaurs by analyzing the concentration of certain isotopes preserved in the mineral bioapatite, found in teeth.

Elemental teeth

A dinosaur tooth is drilled in preparation for an analysis of the concentrations of isotopes in the tooth, which reveals the body temperature of the extinct creatures. (Image credit: Caltech / Lance Hayashida)

The researchers studied two stable, but slightly heavier, isotopes of carbon and oxygen — carbon-13 and oxygen-18. The isotopes tend to bond with each other, or "clump," at lower temperatures, and the lower the temperature, the more carbon-13 and oxygen-18 will clump. By analyzing the clumping of those isotopes, the researchers were able to determine fairly precise temperature values — to within a range of 2 to 4 degrees Fahrenheit (1 to 2 degrees Celsius).

"The foundation of this measurement was an idea I had eight years ago to commission a mass spectrometer capable of analyzing simple gas molecules, including carbon dioxide, containing two or more rare isotopes," said lead study researcher John Eiler, a professor of geology and geochemistry at Caltech. "There were various excuses for doing so, but the real reason was that no one had ever seriously studied such things in natural samples ... so it was done purely out of desire to 'go' somewhere no one had gone in the study of naturally occurring isotopes."

The researchers relied on a decades-old process, but researchers never adjusted the process to try to quantify dinosaur body temperatures.

"Once this measurement basically worked, and revealed interesting results for atmospheric CO2 [carbon dioxide] I was looking around for another direction to go with the idea," Eiler said. "Studying CO2 released from carbonate minerals is a common and important experiment in stable isotope labs, and was an obvious thing to try. It was clear, to me at least, that it should yield a thermometer if you could get the carbon-oxygen bonds out of the solids without modification. But I was not sanguine about the prospects."

Isolating atoms

"I thought it most likely that the acid digestion reaction used to produce CO2 would ruin whatever information might be present in the minerals," Eiler said. "So, I dithered and put it off until I was really bored with the work on atmospheric gases."

"But, the experiment was equally obvious to a couple of my colleagues, and they wouldn't stop hectoring me to try, so in 2005, I had a couple of postdoctoral scholars in my lab start the relevant experiments. Surprisingly, it worked as a thermometer in a relatively straightforward way," Eiler said. "It was obvious we were on to something big."

One of the researchers, Prosenjit Ghosh,experimented on a human tooth — his son's — and aTyrannosaurus rex tooth, and, "both measurements seemed to more or less work," Eiler said, although Ghosh later moved on to a position outside of the country and did not continue the work.

Delayed dentistry

Caltech geochemists Rob Eagle (left) and John Eiler display a dinosaur tooth that was part of research to determine body temperatures for several of the animals. (Image credit: Caltech / Lance Hayashida)

At the same time, other scientists working on teeth and bones contacted him, asking whether the method would work on their samples. He invited one of them to his lab to try it. "The study was a mess, as we couldn't get the extraction chemistry to work consistently with the complex, carbonate-poor minerals in teeth," Eiler said. "But I was convinced it just needed a bit more elbow grease to work our way to a reproducible method for these materials."

Lack of funding to hire another researcher further delayed the work which was, "at that point, demonstrably bad," Eiler said. "Fortunately, one of my colleagues at Caltech offered to fund the position with campus funds once he heard what I wanted to do. At that stage I hired Rob Eagle, from Cambridge, and put him on the task. He quickly had it up and running."

Eagle, a postdoctoral scholar at Caltech, likens the work to "being able to stick a thermometer in an animal that has been extinct for 150 million years."

Taking dinosaur temperatures

Theanimals, as it turned out, were relatively warm — although not necessarily warm-blooded. They could have been cold-blooded (based on the definition for that metabolic design), but with warm body temperatures because of their large size — a phenomenon known as gigantothermy.

"What our study resolved is that at least some dinosaurs were physically 'warm', and so — in the simplest sense — were warm-blooded," Eiler said. "But we need more information to tell how they were warm-blooded — whether through carefully controlled endothermy, like mammals and modern birds, or through some other physiological strategy, like the 'gigantothermy' used by some large ectothermic animals. We are actively working on the measurements of smaller dinosaurs that will resolve this issue."

The researchers analyzed 11 teeth, dug up in Tanzania, Wyoming and Oklahoma, that belonged to Brachiosaurus brancai and Camarasaurus. They found that the Brachiosaurus had a temperature of about 38.2 degrees Celsius, or 100.8 degrees Fahrenheit, while the Camarasaurus was about 35.7 degrees Celsius, or 96.3 degrees Fahrenheit, which is warmer than modern and extinct crocodiles and alligators, but cooler than birds.

Prying teeth

Camarasaurus tooth from the Jurassic Morrison Formation of North America that researchers analyzed in this study. (Image credit: Thomas Tütken (Bonn University))

One challenge the researchers faced was persuading curators to part with tooth samples, since the process leaves damage on the teeth. "We grind the enamel from the tooth surface, producing a powder, and a great deal of material is needed — about 100 milligrams or more — so I'm afraid you can easily tell a tooth that has been through the mill," Eiler said.

"Some of the most interesting samples are sufficiently rare and no one wants to give up their personal favorite tooth," he adds. "It is understandable, but we have to push anyway. One of the experts … was surprised we worked on adult sauropods (large herbivores with a long neck and small head), which he considered to be relatively uninteresting as targets for a body temperature measurement [but] adult sauropod teeth are about the only thing you can get in any abundance you want. So there was little choice."

The results were published online June 23, 2011in the journal Science, and as is often the case with scientific advances once thought impossible, the study will be closely scrutinized in the research community.

"Our approach is a new one, so faced with this, the scientific community tends to split between excitement and skepticism, so my interactions with the senior paleontologists over our work is very much split along those lines," Eagle said. "Of course, it's our job to convince the skeptics that our approach is as exciting a new method as we think it is."

Eiler, has been fascinated with dinosaurs since childhood — when at age six he wrote his first paper on them, entitled "Dinosaurs Ror!" — and he believes skepticism was inevitable, "but I'm an optimist," he said.

"Once you understand the physical chemistry behind the isotope effects we use in this thermometer, you can't help but see it working all around you. So, from that perspective, it doesn't take much imagination to see a tool like this working. Actually, there is a large set of related ideas we are currently cooking up involving the application of similar approaches to the study of organic tissues, so we can measure temperatures of wood, wax, skin, etc." Eiler said. "But that is a story for another day."

To learn more, watch this video about Eiler and Eagle's research.

Editor's Note: The researchers depicted in Behind the Scenes articles have been supported by the National Science Foundation, the federal agency charged with funding basic research and education across all fields of science and engineering. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation. See the Behind the Scenes Archive. You can contact the author, Marlene Cimons at