More on the 'cladistics' of sequences

[Curtis Clark]

on 2004-06-06 19:12 HJJACOBSON at AOL.COM wrote:

> Therefore if changing outgroups changes the character
> hypotheses, we need an objective way of selecting the outgroup. I think this is what
> John is trying to say: What are your objective criteria for selecting outgroups?

The papers mentioned in an earlier post (especially Maddison et al.),
plus my own from that same distant era (Clark, Curtis and Daniel J.
Curran. 1986. Outgroup analysis, homoplasy, and global parsimony: A
response to Maddison, Donoghue, and Maddison. Syst. Zool. 35:150-154.),
address exactly that: pick as many outgroups as you can get data for.
Just as more data give a better-supported ingroup tree, so do they
better support outgroup relationships and rooting. From the analysis
standpoint, outgroups are not in any way special. The key is to be able
to support the conjecture that your outgroup node is not actually in the
ingroup. Once you do that, everything else falls into place.

Note that, although I and my coauthor disagreed with specific points of
the Maddison paper (especially wrt constrained and unconstrained
analysis), we all agreed on the basic principles.

--
Curtis Clark                  http://www.csupomona.edu/~jcclark/
Web Coordinator, Cal Poly Pomona                 +1 909 979 6371
Professor, Biological Sciences                   +1 909 869 4062