Jurassic Lark

Scientists in China recently published a report in Nature that details their study of a newly discovered fossilized dinosaur species. What makes this new species (which they plan to name Tianyulong confuciusi) interesting is that it appears to have had structures on its skin that resemble feathers. From The Economist:

That birds are the descendants of dinosaurs is now accepted by almost all evolutionary biologists. The clinching discovery was of animals that were clearly dinosaurs, and clearly could not fly, but which had feathers. That did raise the question, though, of why one twig of the great dinosaur tree had developed such strange outer vestments, even before it developed wings.

This isn't the first time that a feathered dinosaur fossil was discovered. What's noteworthy about this particular fossil, though, is that it's from a group of dinosaurs that aren't "supposed" to have feathers based on our current understanding of dinosaur evolution.

The new species belongs to a group of dinosaurs called the Ornithischia. This group, which includes such famous names as Stegosaurus and Triceratops, is one of the two clades into which the dinosaurs are divided. The other is the Saurischia, whose famous members include Diplodocus and Tyrannosaurus. The Saurischia also include all the immediate ancestors of birds.
[...]
Such “protofeathers” have been found on other dinosaurs, but until now those species have, like those that sport true feathers, all belonged to a part of the Saurischia that includes the birds. Yet the split between Ornithischia and Saurischia goes back to the very beginning of the dinosaurs, 80m years before the first birds and about 100m years before T. confuciusi. So if the feather-like structures of T. confuciusi really have the same origin as feathers, then such structures must have been there from the outset, and be characteristic of dinosaurs as a whole.

To illustrate this, draw a capital "Y". The stem of the "Y" represents an ancestor common to all dinosaurs. The arms of the "Y" represent the two groups of dinosaurs that evolved from that common ancestor. Until now, all known feathered dinosaurs have belonged to the group on one arm of the Y. This has led scientists to the conclusion that birds evolved from this group of dinosaurs (the Saurichia). This new fossil (T. confuciusi), however, belongs to the group of dinosaurs on the OTHER arm (the Ornithischia). This leaves us with a few possibilities:
1. Feathers evolved independently in both groups of dinosaurs, or
2. Feathers were a characteristic of the ancestral species, or
3. The current model for dinosaur evolution and classification is completely wrong.

The simplest of the explanations is that early feathers were present in the ancestral group, and these early feathers were passed on, through evolution, to both the Saurichia and the Ornithischia. Most dinosaur species in each group evolved to be featherless, but one or a few species in each group also apparently retained the early feathers. The trait of feathers eventually died out on one arm of the "Y" (the Ornithischia). On the other arm (the Saurischia), the feathered species of dinosaurs were the earliest ancestors of modern birds.

This conclusion does lead to a problem with classification: instead of thinking of birds as dinosaurs, it may become necessary to start thinking of dinosaurs as specialized birds:

Taxonomically, the very definition of a bird was until recently an animal that has feathers. Now, taxonomists argue that since birds are descended from dinosaurs they should be classified merely as a subgroup of the Dinosauria. But if feathers truly are the diagnostic criterion, then perhaps things should be the other way round, and Stegosaurus, Triceratops, Diplodocus, Tyrannosaurus and their kin should no longer be thought of as terrible lizards, but as overweight, flightless birds.