Re: pelicanomimus

[Stan Friesen <swf@ElSegundoCA.NCR.COM>]

 From: tholtz@geochange.er.usgs.gov
 > >
 > >Ah, this is important.  Assuming for now that this is a valid
 > >synapomorphy, not a convergence, it forbids placing Pelecanimimus
 > >is a stem group ("Coelurosauria") AND Harpymimus in Ornithomimosauria.
 > >[That would make either the "stem" group or the ornithomimosaurs
 > >polyphyletic, which no biologist will accept].
 > 
 > What the hell are you talking about?  It does nothing of the sort.  A whole
 > range of synapomorphies place ornithomimosaurs in Coelurosauria (see
 > Gauthier) and additional characters show that Ornithomimosauria is a member
 > of the coelurosaurian subgroup Maniraptora (the aforementioned ref. 23). 

The difference here is the difference between the stict cladistic
approach to taxonomy and the non-cladistic approach.  I treat
Maniraptorea as a taxon that may, or may not, include all of the
descendents of the latest common ancestor of its members.

What I was trying to get at was that if Pelecanimimus is closer
to Ornithomimidae than Harpymimus is, then there are three valid
alternatives, unless one is a cladist:
        1) place both Harpymimus and Pelecanimimus in
           Ornithomimosauria
        2) place both in the paraphyletic stem group, whatever
           one wishes to call it.
        3) place Pelecanimimus all by itself in a more derived
           group than Ornithomimosauria; which seems like excessive
           splitting to me.

At the time I wrote the above, I was using "Coelurosauria"
as the name of the stem group for the entire nexus under
consideration.  However, as it really only includes Compsognathidae,
that is probably a poor choice of name.

After evaluating the new evidence, I am considering a classification
something like the following:

        Taxon A (?Coelurosauria)
            Stem taxon (?what to name it?)
                Compsognathidae
            Maniraptorea
                Dromaeosauidae
                Oviraptoridae
                Avimimidae
                Caenagnathidae (= Elmisauridae)
                Troodontidae
            Ornithomimosauria
                Harpymimidae
                Pelecanimimidae
                Ornithomimidae
            Segnosauria
                Alxasauridae
                Therizinosauridae (= Segnosauridae)

But, I am having a hard time coming up with good names for
some of these groups.  Given the way the names are used now,
I suppose Coelurosauria is a usabel name for the largest group,
but I do not really like using it for a group that includes the
dromaeosaurs as the most "typical" members.  And the name
Dromaeosauria doesn't seem right for a group containing the
Oviraptorids and Avimimids.

[I am also considering placing Compsognathidae in the subgroup
Maniraptorea, making that the "stem" group, as well as the
core group, of "Coelurosauria"].

> Thus, Pel. & Harpy. are ornithomimosaurs AND maniraptorans AND
 > coelurosaurs. 

Only if you use a strict cladistic classification, *or*
you decide the groups should be defined that inclusively.

 >  The hand of ornithomimosaurs are thus transformed from the
 > basal maniraptoran condition.  Given the structure of the hand of
 > Harpymimus (which retains a primitively short mc I), it would appear that
 > the hand of Pelecanimimus is either convergent with Ornithomimidae, or that
 > Harpymimus represents an outgroup to the Ornithomimidae-Pelecanimimus
 > clade.

Quite.  If it is convergent, then inclusion of Pelecanimimus
in Ornithomimosauria is not necessary, as it is an "outgroup"
under that situation.


In fact, of the four major subgroups of Theropoda, I am
satisfied with the names of only two (Herrerasauria and
Ceratosauria).  I think both of the names "Carnosauria"
and "Coelurosauria" should be retired as formal names.

swf@elsegundoca.ncr.com         sarima@netcom.com

The peace of God be with you.