Randall Munroe, the creator of the webcomic XKCD, isn't going to like this one bit. Fear of attack by Velociraptor is a running theme in the science-themed series—lazy computer programmers should be
We don't know for sure what dinosaur eyes looked like. The soft-tissue structures rotted away between the time of death and preservation. But there was one feature of the skull that allowed paleontologists Ryosuke Motani and Lars Schmitz to approach the question of whether some dinosaurs were active in the dark—a circle of bones called the sclerotic ring.
Though relatively rare in the dinosaur fossil record, sclerotic rings can give paleontologists a general picture of eye size and shape. This is because the bone surrounded the pupil and the iris of the eye. Birds, lizards, and other vertebrates have this feature, too, and the details of the sclerotic ring are closely associated with the light conditions when an animal is active.
Modern-day nocturnal animals tend to have wide sclerotic rings with a very large aperture in the middle relative to eye size. Animals that are more active during the day (diurnal), on the other hand, have smaller apertures relative to their eye size. By tracking this association, Motani and Schmitz were able to detect that dinosaurs were active during all times of the day.
(The study also included analysis of pterosaurs and other archosaurs, but I am going to restrict my comments to the findings about dinosaurs here.)
As a group, the dinosaurs did not all neatly fall into nocturnal and diurnal groups. Herbivorous dinosaurs, in particular, appear to have been cathemeral—they would have been active over short periods of time during the day and night. Rather than foraging continuously from dawn until dusk, herbivorous dinosaurs such as the hadrosaurs Corythosaurus and Saurolophus, the small ceratpsian Protoceratops, the sauropodomorph Plateosaurus and the sauropod Diplodocus were probably most active during the early, cool parts of the day and then again around twilight.
Small, predatory dinosaurs were different. Almost all the carnivorous dinosaurs that were examined had sclerotic rings consistent with a nocturnal lifestyle, including Juravenator, Microraptor and—you guessed it— Velociraptor. Based upon the inferred night-hunting habits of Velociraptor and the cathemeral pattern of Protoceratops, Motani and Scmitz suggest that the deadly encounter between the two species immortalized in the " fighting dinosaurs" specimen probably happened at twilight or in low-light conditions.
Not all theropod dinosaurs stalked prey by night, though. The small predator Sinornithosaurus appears to have had the more varied schedule seen among the herbivores, and this was also found for the omnivorous "ostrich mimic" dinosaurs Garudimimus and Ornithomimus. Early birds—the descendants of small, feathered theropods—were different. Every species in the study— Archaeopteryx, Confuciusornis, Sapeornis and Yixianornis—had eyes specialized for daytime activity. Perhaps, during early bird evolution, there was a transition from nocturnal ancestors to flying descendants active during the day.
These findings change our perspective of what Mesozoic life was like. Dinosaurs were thought to be mostly active during the day, with small mammals—including our ancestors and cousins—coming out at night. Now it seems that the Cretaceous nights were not as safe as had been presumed. With so many agile predatory dinosaurs around, mammals would have much to fear during the nighttime hours.
Then again, the idea that Mesozoic mammals scurried through the night is an assumption based upon the idea that dinosaurs were stomping around during the day. Studies of the mammals themselves will be needed to see how their activity overlapped with that of the dinosaurs. Since mammals lack sclerotic rings, though, some other technique will have to be used. Further studies of dinosaurs will be required, too. Conspicuously missing from the study were large-bodied predators akin to Allosaurus and Albertosaurus. When these giants hunted, and when the mammals under their feet were active, awaits future study.
For more, see Schmitz's own post on the research at his blog and Ed Yong's report at Not Exactly Rocket Science.
Motani, R., & Schmitz, L. (2011). PHYLOGENETIC VERSUS FUNCTIONAL SIGNALS IN THE EVOLUTION OF FORM-FUNCTION RELATIONSHIPS IN TERRESTRIAL VISION Evolution DOI: 10.1111/j.1558-5646.2011.01271.x
Schmitz, L., & Motani, R. (2011). Nocturnality in Dinosaurs Inferred from Scleral Ring and Orbit Morphology Science DOI: 10.1126/science.1200043