tooth counts in systematics

[smithjb@sas.upenn.edu (Joshua Smith)]

chris brochu wrote:
> 
> Dinogeorge@aol.com wrote:
> > 
> > This is indeed strange: You have only one specimen, and it sometimes has 10
> > teeth and at other times 13 teeth? Yes, I would agree that >this< kind of
> > variability would make tooth count a bad character...
> 
> 
> Not having seen the specimen, my guess is that this is a case of
> asymmetry - ten on one side, thirteen on the other.  It happens in
> crocs, though usually the difference is a single alveolus.

        BINGO.  That is the situation.  It is the only example that I 
cited.  However, it is not the only example that I have seen.  So, I 
guess we could safely say that it happens in crocs and it happens in 
theropods.

        Yeah, but I think even the systematists would probably agree that 
maxillary and mandibular counts are bad characters given intraSPECIMEN 
variation in the state...

> Part of my interest in "Nanotyrannus" (which I seriously think is an
> immature T. rex) came from statements that Alligator does not lose
> premaxillary teeth during ontogeny.  The person who said that is right -
> A. mississippiensis does not lose premaxillary teeth.  But it's just
> about the only living croc that does not lose premaxillary teeth during
> ontogeny.
> 
> Which raises the issue - just how reliable are tooth counts in archosaur
> phylogenetics?  I've been pretty careful to use characters in my own
> work that I'm pretty sure are ontogenetically invariant - e.g.,
> Paleosuchus starts out with four premaxillaries and stays that way
> throughout life - but however variable they might be, tooth counts could
> still preserve a phylogenetic signal.
> 
> This is a point the herpetologists are already way ahead of us on. 
> Weins, for example, has explored methods for using scale counts in
> lizards phylogenetically - scale counts vary within populations even
> more than tooth counts, and yet he didn't want to just throw them out
> when they could help out at some level.  The challenge is to include the
> amount of variation and its nature (ontogenetic, sexually dimorphic,
> interpopulational, whatever) in the characters - whether this can be
> done with theropods is another matter.
> 
> 
        Given the above, I will echo something I just sent to Chris 
concerning this:

        On the one hand I hope that premax count at least works for 
theropods, because I thought we might finally have a character there that 
didn't suck.

        and on the other hand I hope that it doesn't work, because then I 
can be vindicated in my visceral feeling that cladistics sucks and needs 
improvement at the very basic level of character selection.  

        Hmmm...maybe back to the drawing board.




-- 
__________________________
Josh Smith
University of Pennsylvania
Department of Earth and Environmental Science
471 Hayden Hall
240 South 33rd Street
Philadelphia, PA  19104-6316
(215) 898-5630 (Office)
(215) 898-0964 (FAX)
smithjb@sas.upenn.edu