Smithsonian National Museum of Natural History


Dinosaur Diversity

and the

Fossil Record


More than 1200 species of non-avian dinosaurs have been named since 1824, when William Buckland described Megalosaurus, the first dinosaur known to science. They are now known from every continent and every terrestrial environment that is preserved, from the beginning of the Late Triassic Period (235 mya) to the close of the Late Cretaceous (65 mya).

This impressive record tells the story of a dominant terrestrial vertebrate group from its origins as small, predatory bipeds through a rapid radiation into a wide range of sizes, feeding habits, and body forms. My research addresses the reliability of the fossil record, and how its particular imperfections, as well as the biases imposed by how fossils are collected, may impact how well this story is understood. Fundamentally, I am interested in how much of the true picture of dinosaur diversity can be extracted from the fossil record.

A straightforward view of this story seems to show that dinosaurs diversified gradually through the Jurassic Period, reaching a peak before declining somewhat in the Early Cretaceous, and then rebounding to their highest diversities at the end of the Cretaceous.

Since 2000 I have been assembling a database of all known dinosaur occurrences on Earth, down to the level of individual collections. It forms part of the Paleobiology Database [link to], and therefore can be studied in light of corresponding data on taphonomy, sedimentology, stratigraphy, taxonomy, and other fossil organisms. Currently there are more than 5000 collections included, representing about 10,000 occurrences worldwide. Every red dot on this map shows a location where dinosaur fossils have been found.

These data show an interesting correlation between the diversity of dinosaurs and the number of collections made. Specifically, the more collections that have been made in a particular time interval, the better the chance that more dinosaur taxa are also known.

This suggests that we cannot ignore the effects of sampling when assessing the dinosaur fossil record, and further that the actual pattern of dinosaur diversity is likely to be quite different from that inferred previously. I am currently collaborating with colleagues on a project to compare dinosaur diversity with patterns in lithology and sea level change, two other factors that may have important effects on diversity.

This has important further implications for studies of dinosaur distribution and biogeography, which can be seriously hampered by incompletely understanding the true geographic ranges of taxa. For many time intervals, there is simply no fossil record for dinosaurs in particular regions. Even in the best-sampled time period, the Late Cretaceous, many areas have produced no dinosaur fossils at all.

Finally, by documenting the record as completely as possible, we can better target sampling-poor regions and time intervals for future collecting efforts.

[ TOP ]