[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

RE: comments in this week's Science about the Thermopolis Archaeopteryx

Thomas R. Holtz, Jr wrote-

Two brief articles in the latest Science (actually, a Technical Comment and Response) about the Mayr et al. article on the
Thermopolis Archaeopteryx:

Corfe, I.J. & R.J. Butler. 2006. Technical Comments: Comment on "A Well-Preserved Archaeopteryx Specimen with Theropod Features".
Science 313:1238.
and the response:
Mayr, G., and D.S. Peters. 2006. Response to Comment on "A Well-Preserved Archaeopteryx Specimen with Theropod Features". Science

Corfe and Butler (2006) find that Confuciusornis is in the Archaeopteryx-Rahonavis clade in trees one step longer than the tree published by Mayr et al. (2005), where it was sister to Microraptor. They altered codings in the Buitreraptor paper after Mayr et al. and found that it did not change Confuciusornis clading with Archaeopteryx and Rahonavis clading with dromaeosaurids. The recoded two characters Mayr et al. changed in Archaeopteryx, coding the scapulocoracoid and metatarsus as fused. I'm not sure which coding is correct, as I haven't looked into either character yet. I think Mayr and Peters (2006) are right to defend against the assertion they suggested flight originated twice (as opposed to being lost). However, they then state-

"We are not aware of any derived character shared by Archaeopteryx and Confuciusornis that is not also present in the deinoychosaur Microraptor. However, and as noted above, Microraptor shares derived characters with Confuciusornis that are absent in Archaeopteryx, and we thus do not agree with Corfe and Butler's comment that "the hypothesis of a polyphyletic Aves is no better supported by available data than that of a monophyletic Aves." One of the essentials of phylogenetic systematics, which makes it superior to other methods of phylogenetic reconstruction, is the naming of apomorphies of the clades in question. Purely statistical
comparisons reduce it to just another numerical approach. Unless Corfe and Butler can present unambiguous apomorphies of a clade including Archaeopteryx and Confuciusornis, to the exclusion of deinonychosaurs, we do not believe that their argument poses a robust challenge to our hypothesis of bird ancestry."

Yet the entirity of the evidence is what's important in cladistic analyses, not just comparisons of two nodes in competing trees. While the Archaeopterx-Confuciusornis node may only have been supported by two characters, and the Microraptor-Confuciusornis node supported by five, the former tree was only one step longer than the latter tree, so additional support for the whole topology must be present at other nodes. For instance, placing Confuciusornis in Dromaeosauridae necessitates that it reverse dromaeosaurid synapomorphies or that Microraptor evolve them in parallel to Sinornithosaurus and other dromaeosaurids. This isn't needed in the Archaeopteryx-Confuciusornis tree, so probably helps it be shorter than it would be otherwise.

Also, there are derived characters shared by Archaeopteryx and Confuciusornis to the exclusion of Microraptor. The potential list is shorter than if other dromaeosaurids were used, in part because Microraptor is more aerial/birdy, but importantly also because it lacks well preserved/described cranial remains. So the palatal and braincase characters shared by Archaeopteryx and birds to the exclusion of dromaeosaurids aren't known in Microraptor, though they would still count against Mayr et al.'s topology (since Microraptor is still a dromaeosaur). They aren't known in Confuciusornis either, as all described remains are extremely crushed. The potential list of derived characters is also shorter because confuciusornithids have heavily modified skulls and hands compared to other basal birds and paravians. So characters like Archaeopteryx's serrationless teeth can't be included to indicate it is more birdlike than Microraptor, as confuciusornithids lack teeth. But despite these handicaps, there are some obvious derived characters shared by Archaeopteryx and Confuciusornis that aren't found in Microraptor. For one thing, the external naris is larger, the dorsal premaxillary process longer and the anterior margin of the external naris is placed further posteriorly. Also, the tail is shorter (less than four times femoral length) and made of less vertebrae (<24). Posterior dorsals have lateral central fossae in the two birds, but not Microraptor. Finally, when viewed edge-on, the coracoids are bent (at the level of the coracoid tubercle) more in Archaeopteryx and Confuciusornis than in Microraptor.

Of course the TWG matrix isn't the best to solve this issue, as it was designed with deeper relationships in mind. One very important point is that this isn't a three taxon problem. The character distributions of taxa like omnivoropterygids, Shenzhouraptor, Jixiangornis, Dalianraptor, Jinfengopteryx, scansoriopterygids, Buitreraptor and undescribed Ukhaa basal troodontids(?) are important to resolving the phylogeny in this area of the tree. I wouldn't be surprised if none of these taxa (or confuciusornithids and Archaeopteryx) were closer to birds than Dromaeosaurus, Troodon or Oviraptor are.

Mickey Mortimer