In 1971, a Polish-Mongolian joint expedition made a spectacular discovery: the exquisitely preserved skeletons of a Velociraptor and Protoceratops locked together. These animals—popularly known as the "fighting dinosaurs"—had died in the midst of combat and have often been taken as an indication that Protoceratops was a regular food source for Velociraptor. But while certainly the most fantastic, this is not the only evidence of a predator-prey relationship between these dinosaurs.
During the field seasons of 2008 and 2009, paleontologists collected numerous dinosaur bone fragments from the Cretaceous rock of Bayan Mandahu, Inner Mongolia. Among the lot were the remains of a horned dinosaur and two teeth of a dromaeosaurid dinosaur. Given the scrappy nature of these remains it was impossible to be absolutely certain of their identity, but given their age, anatomy and the place in which they were found, it is likely that the fossils represent Protoceratops and Velociraptor.
Toothmarks on the Protoceratops bones may explain why the skeleton was not found in better condition. At least eight fragments of bone showed clear signs that they had been bitten, and three different toothmark patterns were visible. There were shallow grooves made in the surface of the bone, two deeper punctures, and one piece of bone had toothmarks on both sides. Regardless of whether the specific identification of the dinosaurs turns out to be correct, the bones show that a Velociraptor-type dinosaur fed upon Protoceratops or a very closely related horned dinosaur.
When the Velociraptor fed upon the Protoceratops is another matter. Given the state of the material, it is impossible to tell whether the horned dinosaur was killed by the predator or whether the meat-eating dinosaur was scavenging. In either case, however, the toothmarks left on the bone were made long after the Protoceratops was killed. The teeth and jaws of Velociraptor were not suited to crushing bone, and so it is reasonable to hypothesize that it would have fed on all the available soft tissues first. The toothmarks on the bone mean that there was relatively little flesh left and the feeding Velociraptor was scraping whatever it could off the tattered carcass. From a paleontologist's perspective, this also accounts for why the Protoceratops skeleton was so scrappy—by the time it was buried, it had already been torn apart.
For more on this research, see this post on Archosaur Musings by one of the study's authors, Dave Hone.
Hone, D., Choiniere, J., Sullivan, C., Xu, X., Pittman, M., & Tan, Q. (2010). New evidence for a trophic relationship between the dinosaurs Velociraptor and Protoceratops Palaeogeography, Palaeoclimatology, Palaeoecology, 291 (3-4), 488-492 DOI: 10.1016/j.palaeo.2010.03.028