noasaurids

[Nick Longrich <longrich@alumni.princeton.edu>]

Regarding Noasauridae:

Is it certain that there actually is a "Noasauridae" 
(Noasaurus+Masiakasaurus or anything else)? The problem I see is that 
the systematics of the "Ceratosauria" are very poorly understood, 
partly because many of the animals, like Noasaurus, are very 
fragmentary, partly because so many of them have yet to be adequately 
described, and partly because even today there is a tendency to throw 
Ceratosaurus in with Coelophysis without actually testing whether 
they're a clade or not (although a lot of people are starting to 
reassess this).
	At least one character contradicts "Noasauridae", and that's 
the form of the epipophyses. The epipophyses of Noasaurus' cervicals 
are anteroposteriorly elongated and form a distinct anteriorly 
directed prong- so that other than being longer and lower, the 
vertebra is remarkably similar to Carnotaurus. Masiakasaurus does not 
have this feature as far as I know. Given this, referring 
Masiakasaurus to a "Noasauridae" is premature, in my mind. We aren't 
even sure what that digit is- is it really a second pedal digit, or 
could it be some other pedal digit, or could it even be a manual? Are 
we even certain all the material is one animal?
	This brings up another question- should we use theropods in 
studies of biogeography? The problem is that theropods are currently 
very unstable in their phylogenetic relationships. So one day our 
Antarctic theropod is falling out with a European theropod, arguing 
for an Antarctica-Europe connection, and the next day it's falling 
out with a Mongolian theropod, and there's a volcanic island arc 
linking Outer Mongolia to Antarctica? I'm not saying this is totally 
useless- for example, we've clearly got Alvarezsauridae in North 
America, Asia, and South America, and this has interesting 
implications for the group- just I think we need more skepticism, 
particularly in cases involving really fragmentary material.

nick longrich