DINO SORE: The Dawn Bird

There's a bit of news going around the interwebz today regarding the phylogenetic positioning of Archaeopteryx and its recently discovered relative, Aurornis, the 'dawn bird.' The basic problem boils down to this: is Archaeopteryx a bird or not?

I'm still a bit on the fence here, but I'm mainly with the 'not quite a bird' camp. When Archie was first discovered in Germany in the nineteenth century, it was clearly the most birdlike fossil ever seen (disregarding those that were straightforward, clear cut modern birds). It had beautifully preserved feathers, including what were clearly flight feathers forming broad wings. Yet it was, skeletally, very much like a small theropod dinosaur. It was clear cut: this was the first bird, and its closest relatives were small, predatory dinosaurs like Compsognathus.

Archaeopteryx lithographica

Later discoveries complicated matters. Indisputable bird fossils kept being found that pushed their lineage further back, encroaching on the 'first bird's' territory. The dinosaur renaissance helped bolster the idea that birds are specialised dinosaurs, particularly Bakker's studies on the Deinonychosauria. Sickle-clawed predators like Deinonychus, Velociraptor and Troodon became clear as the most birdlike of all non-avian dinosaurs. That birds are dinosaurs is no longer really in question, except for some iconoclastic paleos with funny ideas. It's the exact structure of the family tree that's the problem. Archaeopteryx is in danger of losing its crown as first bird; indeed, it may not be a bird at all.

I'm pretty much in agreement with Xu Xing, whose 2011 analysis removed Archie from the direct line to birds and placed it elsewhere in the paravian lineage. To me, just looking at the fossil with an amateur's eye, Archaeopteryx looks for all the world like a troodontid: small, slender, feathered and with a sickle-claw on each foot. While the exact positioning is questionable, it looks like it belongs somewhere in the Deinonychosauria, outside of the Avialae, which is to say, birds (the exact definition of bird is always a bugger anyway).

It's possible to retain this positioning and keep Archie as a bird, but that would require expanding the Avialae  to include the Deinonychosaurs. It's easy to keep pushing back like this, but with more and more birdlike features becoming apparent in theropods, it risks making theropod and bird mere synonyms, and therefore pretty useless as terms in the argument. There are some issues with this, including that it implies that powered flight evolved twice; alternatively, it was secondarily lost in some Deinonychosaurs. There are shades of the anti-dinobird argument here, in that Deinonychosaurs are essentially being referred to as flightless birds; however, there is no claim that they don't belong in the Dinosauria.

Aurornis xui

This year's analysis of the recently found Aurornis fossil muddies the waters again. Pascale Godefroit and his group find Aurornis to be earlier in the bird lineage than Archie, in a phylogenetic analysis that restores it to true birdhood.  It removes Archaeopteryx from the Deinonychosauria, bu makes Troodontidae a sister group to the Avialae. I'm not entirely convinced, but it is compelling. A good reshuffling could allow the best of both worlds regarding Archie. In this new phylogeny, Archie is not the first bird. That honour, presuming we accept that Avialae=birds, goes to Aurornis. This system avoids the need for two origins of powered flight, and tidies up the temporal spread of the Paraves.

There are some issues. Luis Chiappe disagrees with the analysis of Aurornis, suggesting that its features are not birdlike enough to support its positioning as a true bird. It also buggers about with the epidendrosaurs (Scansoriopterygidae), but they are an unusual group whose position in the Avialae is debatable. Nonetheless, it's a compelling argument, if not wholly convincing.

Links:

BBC News article
Nature article preview
Wiki entry on Aurornis