November 2012

Last month, paleontologists from Canada, the U.S., and Japan announced an exciting discovery: feathered dinosaur fossils in North America. When Ornithomimus edmontonicus was first studied in the 1930s, its ostrich-like skeleton earned it a name that translates to “bird mimic.” Now new fossils and a re-evaluation of old ones have revealed that its body covering also fits the moniker. A newly unearthed, year-old juvenile specimen is covered in downy, hair-like feathers, and re-examination of an adult specimen turned up traces of standard feathers with a central shaft. While most popular reporting has focused on the idea that these shafted feathers may have been used to attract mates, the real news in this research lies elsewhere …

Where's the evolution?

Since O. edmontonicus was flightless (adults would have weighed in at around 350 pounds) and since the ornithomimosaur clade of dinosaurs branched off during the early stages of the evolution of feathers, the scientists reasoned that natural selection for flying ability probably didn’t shape feathers as they first evolved. Furthermore, since only adults of O. edmontonicus bore standard feathers and juveniles had mere feather fuzz, the scientists hypothesized that shafted feathers served a uniquely adult function in this species — attracting a mate or brooding, for example. All of that is reasonable and represents evidence relevant to the question of the early function of feathers, but it is not particularly new. Scientists have long hypothesized that feathers first appeared well before birds could fly or glide and were co-opted and shaped for that function much later in the evolutionary history of dinosaurs.

The real story here is that complex, shafted feathers probably evolved much earlier than we had previously thought. Understanding how scientists reached that idea requires a bit of tree-thinking. The earliest fossil feathers yet discovered belong to Anchiornis huxleyi and date to between 151 and 161 million years ago. These were no simple filaments; A. huxleyi had a full set of elaborate, shafted feathers on its wing-like arms — and on its legs! O. edmontonicus also had shafted feathers as well, but its fossils date to only around 70 million years ago. O. edmontonicus and A. huxleyi have similar feathers, and O. edmontonicus is younger than A. huxleyi — so how did the discovery of O. edmontonicus feathers imply that this basic tool of flight evolved earlier than previously suspected? Through phylogenetic reasoning….

Phylogeny A represents what was known about the distribution of feathers before the recent investigations of Ornithomimus. Based on this phylogeny and the principle of parsimony, one would reason, first, that ornithomimosaurs probably had some sort of hair-like feather — and that’s exactly what was found. Second, one would reason that shafted feathers likely evolved some time shortly before therizinosaurs and oviraptors split from the rest of the dinosaurs, since these structures are only found among a restricted clade of the group. Phylogeny B adds the latest data on Ornithomimus into the mix. Notice that the new information suggests that shafted feathers first evolved earlier than previously thought — not because the Ornithomimus fossils are particularly old, but because ornithomimosaurs branched off of the dinosaur family tree quite early relative to other dinosaurs in which shafted feathers have been found.

The past decade has seen a spate of feathered dinosaur discoveries, and the Ornithomimus research gives us every reason to think this will continue. First, the research turned up the first dinosaur feather fossils ever found in America and the first ever associated with an ancient river, expanding the scope of places and sediments in which to seek fossil feathers. Previous feather finds had been restricted to particular lake deposits in Germany and China. Second, the study suggests that more careful observation and fossil preparation could yield additional evidence of feathers. One of the fossils that had markings indicative of feathers had been in a museum collection for years. Because the original collector and preparator had likely not been thinking of O. edmontonicus as a feathered animal or because the specimen came from sediments not believed to be capable of preserving feather traces, the significance of these marks was originally overlooked. Finally, the study suggests that additional dinosaur species likely had shafted feathers. If shafted feathers evolved early, perhaps other early branching clades like Alvarezsauroidea inherited the fancy plumage type, and we just haven’t yet found the fossils that preserve these structures. But we can be certain that paleontologists will now be on the lookout for them!