Jeff's Lunchbreak

Around a year ago, I read Peter Wellnhofer's book, Archaeopteryx: The Icon of Evolution. I learned quite a bit, but a new paper on Xiaotingia zhengi throws a kink into archaeopteryx's relative importance in understanding bird evolution.

Here are some typical examples of the coverage the paper has received:

• Nature News - Archaeopteryx no longer first bird
• Internation Business Times - 'Oldest Bird' Archaeopteryx May Not be Bird After All
• Christian Science Monitor - Archaeopteryx may not have been a bird, but just a feathery dinosaur
  The tagline for that article was, "Archaeopteryx: The creature long believed to be the earliest known bird may not have been a bird at all, suggests a new Chinese fossil find."

Okay, first thing is to clear up some misconceptions. Paleontologists have not been suggesting that archaeopteryx was the first bird - only that it was the oldest known bird. In fact, if you go look at that review I wrote and look at the family tree from Wellnhofer's book, archaeopteryx is clearly shown on a side branch that went extinct, so nobody's even been arguing that archaeopteryx was a direct ancestor of modern birds. In fact, I'll just copy that family tree here:

What the new paper does call into question is how closely related archaeopteryx is to modern birds. First of all, it's clear that birds are a type of dinosaur, and archaeopteryx was obviously a fairly closely related dinosaur. But new fossils have been found of other closely related dinosaurs - some that could probably fly, and some that probably couldn't. Whereas previously it was assumed that flight probably only evolved once, and therefore archaeopteryx was a bird, all these new fossils are calling that into question.

The new paper suggests that archaeopteryx was more closely related to deinonychosaurs* than to modern birds. The problem is that deinonychosaurs can't fly. So, there are basically two possibilites:

1. Flight evolved multiple times in the maniraptorans. The common ancestor of archaeopteryx and modern birds was flightless, and each lineage evolved flight on its own. It's possible that powered or gliding flight evolved in other closely related lineages (such as Microraptor)
2. Flight evolved once in the maniraptorans. The common ancestor of archaeopteryx and modern birds could fly, but then some lineages, such as the deinonychosaurs, lost that ability (like ostriches with teeth).
3. Okay, I guess there's a third possibility. It's possible that the common ancestor of archaeopteryx and modern birds could fly, but that flight also evolved in another lineage of maniraptorans.
4. Well, I suppose there's a fourth possibility, as well - that the proposed phylogeny by Xing Xu et al is incorrect, and that archaeopteryx should still be considered a member of avialae.
5. I can't actually think of a fifth possibility, but I'm sure thre are some more possibilities I haven't thought of.

Here's the cladogram from the paper with the new proposed phylogeny:

(Click for larger image)

(Source: Pharyngula)

Paraves is the group that contains both archaeopteryx and modern birds. It also includes the deinonychosaurs, which couldn't fly. But in the avialae branch, it includes epidexipteryx, which also couldn't fly. We know from modern birds that losing the ability to fly has occured multiple times (ostriches, penguins, kiwis, dodos, etc.), so it's possible that the common ancestor for Paraves could fly and the deinonychosaurs and epidexipteryx both lost that ability. But it's also possible, like I wrote above, that flight evolved independently in the archaeopterygidae and the avialae.

So why doe this matter? Aside from trying to sort out the evolutionary relationships, I would think that this is relevant to the trees down vs. ground up debate on the origins of flight. As Wellnhofer pointed out, archaeopteryx was a terrestrial animal, not adapted for life in the trees. And it looked like its terrestrial characteristics were primitive, not secondarily derived from an arboreal ancestor. That was strong evidence in favor of the ground up hypothesis. But if flight evolved independently in archaeopteryx and modern birds, then archaeopteryx doesn't really tell us much about that evolution in birds. The trees down vs. ground up debate is back on.

So, what we have now are a bunch of fossils of very closely related dinosaurs, but it's not quite clear how they all fit together, and especially how the origin of flight fits in. I know what I would like for the truth to be, but reality doesn't care about my feelings. It's just going to take finding more fossils to fill in the family tree and make the picture more clear. Get to work on it paleontologists.**

For more information, here's a very good blog entry describing the new paper:
Pharyngula - Xiaotingia zhengi

*Okay, my nomenclature's not exactly right, but rather than re-type the whole thing, I'm just going to add this note. When I wrote 'deinonychosaurs' in this article, I was mostly referring to the dromeosaurs and troodontids, since this new analysis places archaeopteryx and related animals in with the deinonychosaurs.

**Actually, a comment from Tom Holtz in that Pharyngula thread (Holtz is a paleontologist at the University of Maryland) indicates that there are some new fossils that were recently discovered that should be coming to light fairly soon.

Updated 2011-07-31 - Added the fourth and fifth possibilities in the list, and added the note on my misuse of the nomenclature.