Camarasaurus seems to have gone on yearly migrations, travelling 600 kilometres in a round trip. Credit: S. Cunningham/Alamy

A collection of teeth from dinosaurs in the western United States has produced the first solid evidence that these ancient animals undertook seasonal migrations.

It is often hard to determine what dinosaurs looked like from fossils — and trying to infer behaviour adds a whole different level of difficulty. Nevertheless, it has often been assumed that dinosaurs did migrate.

Henry Fricke, a geochemist at Colorado College in Colorado Springs, and his colleagues now seem to have hard evidence that sauropod dinosaurs moved hundreds of kilometres every year. They publish their findings today in Nature1.

"In a theoretical sense, it's not hugely surprising. They are huge — they would probably have eaten themselves out of house and home if they stayed in one place," says Fricke. "Now we have evidence that demonstrates that, and a method to move forward and study other dinosaurs."

The evidence collected by Fricke and his colleagues takes the form of fossil enamel chipped from 32 teeth that belonged to sauropods of the genus Camarasaurus. The teeth, which date from the late Jurassic period (160 million to 145 million years ago), were collected in the Morrison basin in Wyoming and Utah.

The researchers measured the ratios of two isotopes, oxygen-16 and oxygen-18, in this enamel, then compared them with the ratios in the sedimentary rocks found in the area. In vertebrates, the oxygen ratio relates to the ratio in the water they were drinking when their teeth were growing; sedimentary rocks record local ratios.

Thus, if the oxygen ratios in a tooth do not match those in the rocks near where the tooth was found, the animal must have been somewhere else when growing that tooth.

That is exactly what the team found. Some teeth matched the basin, but others had a much lower proportion of 18O, indicating that their former owners had probably spent time at higher elevations. Levels of 18O are low over high ground because the heavy isotope is rained out of clouds first as they rise and cool. The Morrison basin Camarasaurus must have migrated at least 300 kilometres between the basin and highlands to the west, says Fricke.

Travel over time

The researchers then took samples from a set of teeth from one animal, starting at the bottom of the tooth and working up. The youngest enamel, from the base of the teeth, indicated a highlands location. The older enamel nearer the tip was formed in the basin environment.

This indicated that the dinosaur was moving out of the basin into the highlands when the teeth were forming. But because the animal was found in the basin, it must have moved back there at some point, probably in a seasonal migration.

"This study was able to establish movement across a landscape in one individual. That pretty much nails it," says Eric Snively, a palaeobiologist at the University of Alberta in Edmonton, Canada, who has worked on dinosaur migration.

He adds that much previous work on dinosaur migrations has looked at north–south movements, especially with 'polar dinosaurs' found in the far north2. The latest paper, with its east–west migration, "almost turns our perspective by 90 degrees".

Next on Fricke's to-do list is a similar study on some of the many teeth from other dinosaurs that are also found in the region.

"The ones we're really interested in right now are the associated carnivores," he says. "The question is whether the carnivores stayed in one place and waited for the sauropods, [or whether] they followed them on their migrations."

For now, the results are only preliminary. But Fricke is "pretty confident" that he will be able to tell whether other dinosaurs stayed at home or joined the sauropods on their migration.