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

Re: tooth counts in systematics

You've got some great points here, but they actually focus on two
independent issues. 

 One of them is intraspecific variation.  Norell's talk at the Ostrom
Symposium gently chastized all other previous theropod systematic
studies for ignoring (or miscoding) interspecific variation, e.g. by
coding a large multispecific group with a single character state, and I
think we'll see more rigorous approaches to this in the future.
Intraspecific variation is a different matter, and as we've discussed,
there are ways of handling it - provided it's made explicit in the first

Your other concern involves explicit, unambiguous character state
definitions, which can be viewed as an independent problem.  Part of the
problem, I think, is that most theropod phylogenies after Gauthier 1986
focused on discussions of the tree, with minimal character discussion. 
(And you theropod people are not alone - I can think of a couple of croc
papers by a certain Field Museum employee that are just as guilty.) 
It's possible, or even probable, that the author of such characters has
clear states in mind, but a two-sentence descriptor makes it difficult
for others to use the matrix.  Adding lengthy character discussions
makes papers VERY long and hence difficult to publish.

I learned my lesson last year, after the Gavialis and Leidyosuchus
papers came out.  A couple of people tried to code new taxa on my
matrix, and in some cases they were clearly misunderstanding the
characters I used.  And this was my fault, not theirs - I should have
published figures along with each character, specifying what I meant.   
(This situation is partially rectified in a forthcoming alligatoroid
phylogeny, which should be out this summer.  The other portions of the
tree will have to await completion of Sue - and any characters I use in
the Sue analysis will be thoroughly figured.)  

A possible third issue, which is also independent, is hinted at in what
you wrote - the problem of multiple primitive states.  ALL character
states in a character should be explicitly defined.  "Orbit keyhole
shaped versus not keyhole shaped" leaves open the possibility that one
of your states could actually be many states - "not keyhole shaped"
could mean oval, circular, or horizontally slitlike.  (For what it's
worth, it would be better to define this character with respect to the
presence or absence of an anterior process on the postorbital - but
that's another discussion.)  This is variation, but of a very different

The linkage between your points comes with the explicit discussion of
the characters.  Any character discussion should include intraspecific
variation, if any; its limits, nature, and any reasons for not coding
it; and so on.


Joshua Smith wrote:

>         I was advocating a rigorous examination of the inter and
> intrataxonomic variation of the selected characters.  It has
> been my experience that most analyses (we are talking theropods only
> here--) don't really examine the amount of variation that exists within
> the characters they describe and use, but rather describe the variation such 
> as
> "premaxilla very deep subnarially" or "key-hole" or "light bulb" shaped
> orbits or lower temporal fenestrae "very large".
>         So, the other condition is lower temporal fenestrae very small?
>         What does that mean?  How is large defined vs. small?  Is it
> repeatable?  Is my definition of "large" the same as yours?  How can I go
> into my newly excavated skull and determine if the lower temporal
> fenestra is large or small when it looks to me to in the middle of the two?
> Seems really ambiguous to me.

Christopher A. Brochu
Department of Geology
Field Museum of Natural History
Roosevelt Road at Lake Shore Drive
Chicago, IL 60605

voice:  312-922-9410 x469
fax:  312-922-9566
electronic:  cbrochu@fmppr.fmnh.org