Premium

A ‘kitten-otter-bear’? Untangling the identity of a very strange skeleton

This ancestor of bears was the size of a raccoon with the claws of a cat and really bad teeth. No wonder it took decades for scientists to figure out what it is.

An Eoarctos, a raccoon-like carnivore with orange and black fur, eats a mollusk in its shell.
The new species Eoarctos vorax is depicted in a river environment 32 million years ago in what is now North Dakota. The animal, similar to a raccoon, represents a key time in mammal evolution.
Illustration by Mark Hallett
ByRiley Black
June 02, 2023
5 min read

One of the earliest relatives of bears looked less like a grizzly and more like a raccoon with a fondness for crushing snail shells. Named Eoactos vorax by paleontologists, the fossil mammal helps reveal how the group containing skunks, raccoons, bears, and even seals got their start some 32 million years ago, as the age of mammals entered full swing.

It’s taken paleontologists decades to untangle the identity of Eoarctos. Field expeditions to the Fitterer Ranch fossil site in North Dakota started to turn up the busted-up, bizarre jaws of some kind of carnivorous mammal back in the 1940s. Every time paleontologists returned to the site, they hoped they would find more of the animal in the ancient rock. Then in 1982 a field crew led by paleontologist Robert Emry at long last found the elusive fossil—a nearly complete skeleton of the small carnivore.

At the time, Emry and paleontologist Richard Tedford, an expert on fossil carnivores, had planned to describe the skeleton. The animal appeared to be an arctoid, which is a broad group of mammals more closely related to dogs than cats.

“Tedford realized for decades that there was an important, undescribed species at Fitterer Ranch that could provide important insight into arctoid evolution if they could just get a good enough specimen out of the ground,” says North Dakota Geological Survey paleontologist Clint Boyd. Only, Tedford passed away in 2011 and the project was left unfinished. Now Emry, Boyd, and colleagues have completed what’s taken nearly 80 years to uncover, publishing their description of the animal in the Journal of Vertebrae Paleontology.

The nearly complete skeleton of a male Eoarctos, which included a baculum or penis bone to identify the animal’s sex, acted as a kind of paleontological puzzle box that allowed other fossils of the same species to be recognized. The paleontologists didn’t just have one stunning specimen, but jaws and skull fragments from multiple Eoarctos.

Eoarctos is certainly a new and very interesting taxon,” says Swedish Museum of Natural History paleontologist Lars Werdelin, who was not involved in the new study. Until now, he says, the earliest arctoids have been so poorly known that Eoarctos provides one of the first detailed looks at how this family of mammals moved and behaved in the ancient landscape.

Finally reassembled, Eoarctos isn’t quite like any mammal alive today. The earliest shell-eating mammal yet known, it had racoon-like proportions and cat-like claws that likely helped it climb up and down the trees of the ancient wetland it called home. “Our internal working name for it was kitten-otter-bear,” Boyd says, a stranger beast than paleontologists ever expected when those first fragments of jaw began to emerge from the rock.

A climbing snail-muncher

In the evolutionary tree of mammals, Boyd and colleagues confirmed, Eoarctos is an early arctoid. The group encompasses everything from seals to grizzly bears to spotted skunks. While Eoarctos is not the oldest member of its group, Boyd notes, most of those earliest arctoids are only known from bits of skull and jaw. “There were major questions about what early arctoids were doing in terms of locomotion,” he says. Whereas the first dogs quickly evolved to be runners, no one really knew what their arctoid relatives were doing during the same time.

Rather than being a runner, Boyd says, Eoarctos had the proportions of a climbing animal. The carnivoran likely foraged on the ground but could readily run up a tree if a larger predator came too close. In fact, Boyd points out, the mammal’s proportions most closely resemble those of a living arctoid, the raccoon, so Eoarctos likely behaved in a similar way. But not everything about Eoarctos is familiar. The animal’s jaws left the paleontologists with a puzzle.

The lower jaw of the best Eoarctos skeleton was damaged. Something had happened during the animal’s life that caused it to lose several of its lower cheek teeth. The injury wasn’t an anomaly. Other Eoarctos jaws show the same damage pattern, an indication that the animals were doing something that broke their teeth and ground down their molars.

“The missing teeth and bone infections were very surprising,” Boyd says. Paleontologists find broken teeth or cavities now and then, but this degree of dental damage is unusual. “If it was one specimen, I’d shrug and say we found a unique individual,” Boyd says. “But with so many specimens and such extensive infection with evidence of healing, it’s clear this was routinely happening to these animals.”

It had to be something the mammals were eating. So far, paleontologists haven’t found gut contents or fossilized feces that can offer direct evidence of their diet. But other fossils found at the site offer a clue.

“At the Fitterer Ranch, and elsewhere in the same formation, decent-sized snails are commonly found,” Boyd says. A future find might change the story, but the anatomy of Eoarctos resembles that of other shell-eating mammals, and the snails were both abundant and hard enough to cause the damage seen in the fossils.

The authors make a good case that Eoarctos was chomping on snails, Werdelin says, and notes that such extensive damage likely had deeper consequences for these mammals. “We may in some sense be watching a snapshot of evolution in action,” he notes, as Eoarctos individuals with more resistant teeth would have likely lived longer and suffer fewer infections. Whether Eoarctos left a descendant species with stronger jaws and teeth is as yet unknown, but it’s possible such a fossil is out there.

Go Further