For half a century dinosaur lovers have been fed a lie, or at least an unsupported claim. The early carnivore Coelophysis bauri was not necessarily the cannibal that museums and books have portrayed it to be, according to the first rigorous evaluation of its stomach contents. In fact only one of three dinosaur species thought to have yielded evidence of cannibalism truly deserves the designation, researchers say.

The Coelophysis fossils in question were unearthed in 1947 from a bed of hundreds of the skeletons, which date back 210 million years to the upper Triassic period. Based on the presence of small reptile bones lodged in what appeared to be the dinos' body cavities, a curator at the American Museum of Natural History in New York City declared at the time that the beasts had fed on their young counterparts. Nobody had ever conducted a detailed study of the remains, however, and in recent years some researchers had begun to express skepticism at the claim.

Earlier this year, studying a cast of the alleged cannibal in the subway station beneath the museum, paleontology graduate student Sterling Nesbitt started to doubt the claim as well, recalls fellow Columbia University student Alan Turner, who works at the museum. "You can see the gut contents" in the cast, Turner says. "Right from the beginning we realized this didn't look like a dinosaur" in its belly. When Nesbitt, Turner and their co-workers examined the actual fossils, they noted that one of the supposed cannibals had missing and distorted rib bones, suggesting its stomach had ruptured before burial, leaving no reason to assume the stomach contents had been preserved. The bones of a juvenile fossil could have simply mingled with its ribs, creating the impression of a last meal, the group points out in the current issue of Biology Letters. The other skeleton contains a cluster of bones in the right spot to represent a meal, but the femur heads of these remains and their tissue patterns are characteristic of crocodile relatives, not dinosaurs, they report.

"This is a beautiful study, and it needed to be done. These specimens have been touted for decades as evidence of dinosaur cannibalism," says paleontologist Raymond Rogers of Macalester College. Turner says the reappraisal leaves only one credible example of dinosaur cannibalism. Three years ago Rogers and his colleagues identified two specimens of the carnivore Majungatholus atopus that show bite marks matching the species' own teeth. Claims of cannibalistic bite wounds in Tyrannosaurs are doubtful because the telltale fossil was buried among many other similar species, Turner explains.