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

RE: Who says dromaeosaurs can't fly?

> -----Original Message-----
> From: owner-dinosaur@usc.edu [mailto:owner-dinosaur@usc.edu]On Behalf Of
> Jeff Hecht
> Sent: Friday, September 13, 2002 9:14 AM
> To: Ronald Orenstein; GSP1954@aol.com; dinosaur@usc.edu;
> P2Christiansen@zmuc.ku.dk; CMARTIN@sil.si.edu; gleahy@harper.cc.il.us
> Subject: Re: Who says dromaeosaurs can't fly?
> At 11:31 PM -0400 9/12/02, Ronald Orenstein wrote:
> >
> >Umm - I still don't see this.  All cladistics does is produce a
> >phylogenetic tree based on coded characters.  It says absolutely
> >nothing about when, or how many times, flight evolved.  I don't see
> >why you need to change the cladistically-derived tree at all based
> >on flight or non-flight, as long as you are willing to assume that
> >flight evolved within the maniraptorian complex more than once, or
> >that it evolved more basally on the tree and was lost several times
> >in the various cladistic branches.  Where is the contradiction?
> >
> I think the problem may be that cladistics implicitly assumes that a
> character should have evolved only once -- not many times as you
> suggest. As I understand it, this is the base of creating a "most
> parsimonious" tree that is considered the most plausible one.

A point of clarification: cladistic analyses (or at least maximum parsimony
analyses) assume that *all other things being equal* the same morphology or
behavior or gene sequence or whatever are the results of the same
evolutionary event.  However, all other things are not equal, and thus any
realistic analysis results in copious amounts of homoplasy.

However, to assume otherwise (that is, to assume a priori that a particular
set of identical morphologies, behaviors, gene sequences, etc. are
convergences or reversals) is putting the conclusion before the analysis; it
is assuming the answer!  If we actually knew the phylogenetic relationships
of a set of taxa, we wouldn't be doing the analysis! (We would also have
god-like powers, as the rest of the mere mortals are stuck having to infer
these historical relationships by the bits and pieces that remain and have
been recovered).

> It's a problem because developmental genetics is showing that a lot
> of major traits are controlled by two genetic switches. One gene
> turns on a second gene -- that is, one gene controls whether or not a
> second gene will be expressed, and that second gene controls (for
> example) limb development. We have only a very primitive
> understanding of that process at present, but it's central the
> emerging discipline of Evo-Devo -- evolutionary development.
Indeed, an extremely interesting field to study.  It is from these works
that morphological phylogenetics may actually, finally begin to be able to
do what the molecular guys do and have some empirical basis for weighting
one set of changes as more likely than another set.  Gould's latest tome has
an excellent section on the implications of evo-devo for evolutionary

                Thomas R. Holtz, Jr.
                Vertebrate Paleontologist
Department of Geology           Director, Earth, Life & Time Program
University of Maryland          College Park Scholars
                College Park, MD  20742
Phone:  301-405-4084    Email:  tholtz@geol.umd.edu
Fax (Geol):  301-314-9661       Fax (CPS-ELT): 301-405-0796