Neck bones on the menu: fossil vertebrae show species interaction.

Three fossil neck bones from an ancient flying reptile--one of them with the broken tip of a tooth embedded in it--indicate that the winged creatures occasionally fell victim to meat eaters.

The telltale vertebrae were found within a limestone nodule unearthed from 100-million-year-old sedimentary rocks in northeastern Brazil. Each about 4 centimeters long, the neck bones represent the fourth, fifth, and sixth cervical vertebrae of a pterosaur that would have had nine such bones. The size of the vertebrae suggests that the creature had a wingspan of about 3.3 meters, says Eric Buffetaut, a paleontologist at France's National Center for Scientific Research in Paris.

As paleontologists were removing rock from around the pterosaur bones, they found a tooth tip embedded in the foremost vertebra. The 1-cm-long fragment is cone shaped and has a thin, smooth enamel with no serrations along its cutting edges. That combination of features suggests that the tooth belonged to the spinosaur species known as Irritator challengeri, a 10-m-long, bipedal carnivore whose bones appear in local rocks of the same age.

Preserved in a lifelike arrangement, the pterosaur neck bones show no signs of having been digested. So, the spinosaur probably didn't consume the neck segment. The animal that lost this tooth didn't suffer any permanent damage because spinosaurs constantly replaced their teeth.

Spinosaurs had long, crocodile-like snouts, says Buffetaut. The body cavities of some spinosaur fossils have contained fish scales apparently etched by stomach acid, so paleontologists had suggested that the animals primarily consumed piscine prey. However, British paleontologists had previously unearthed a spinosaur that appeared to have preyed upon or scavenged the remains of a juvenile Iguanodon, a herbivorous dinosaur.

The new finding, described by Buffetaut and his colleagues in the July 1 Nature, bolsters the notion that spinosaurs had a more catholic diet. It's impossible to determine whether the Brazilian spinosaur preyed on alive pterosaur or merely scavenged a carcass, Buffetaut notes.

It's "surprising, but obviously not impossible" that a spinosaur could break its tooth on such a light, thin-walled bone, says Angela C. Milner, a vertebrate paleontologist at the Natural History Museum in London.

However, that anatomical feature probably explains why this particular tooth tip ended up stuck in the unlucky pterosaur's neck bone, says Alexander W.A. Kellner, a paleontologist at the National Museum in Rio de Janeiro. If the neck bone's wall had been thick, the spinosaur's tooth would have broken off before it could have pierced the surface.