Phylogeny talks at Ostrom Symp.

["Thomas R. Holtz, Jr." <th81@umail.umd.edu>]

Okay, a overview of the most important part... of Day One (the phylogeny
part) of the Ostrom Symposium:

Sereno:  Update of the phylogeny presented in his recent papers (e.g., the
1997 Annual Review of Earth & Planetary Sci. paper and the one in N. Jb.
Geol. Paleont. Abh. 210: 41-83).  For those who don't recall from memory
(i.e., basically anyone on this list who ISN'T as obsessed with theropod
phylogeny as me), the structure of maniraptoriform relationships here is:
(ornithomimids + therizinosauroids) + (tyrannosaurids + (oviraptorosaurs +
((dromaeosaurids + troodontids) + birds))).  Added to this are
alvarezsaurids, which fall out in his cladogram as the sister taxon to
ornithomimids, and _Caudipteryx_, an oviraptorosaur.  I agree with many of
Sereno's observations of manual characters potentially uniting
alvarezsaurids and ornithomimosaurs (I coded many of them the same way), but
other features happened to unite them with birds (as per Norell and Chiappe)
in my own study.  Sereno also pointed out that support for the position of
therizinosauroids and (to a lesser extent) troodontids as he presented
wasn't as strong as some of the other clades: I agree with him on this whole
heartedly.

Me: The NEW version of the Gaia cladogram, plus a reduced analysis (which
added the feathered forms).  In the maniraptoriform part of the cladogram, I
found the basic structure: (ornithomimosaurs + tyrannosaurids) +
((therizinosauroids + oviraptorosaurs) + (dromaeosaurids + birds)).
_Microvenator_ bounced around within the oviraptorosaurs, and Troodontidae
(bless its homoplastic little heart...) could equally parsimoniously occupy
one of two positions: sister taxon to Ornithomimosauria, or sister taxon to
Dromaeosauridae + Avialae.  Ugh.  Homplasy, thy name is "theropod"...  :-)
Okay, and one step futher out, and you get Troodontidae as the sister group
to Therizinosauroidea + Oviraptorosauria...  All based on real, observed,
independantly verified data.  Also, I pointed out that using _Allosaurus_ as
the immediate outgroup to Coelurosauria in studies (an almost universal
practice), although nice from the standpoint that we really know its
anatomy, has the drawback that it is a derived carnosaur with many features
which seem to be convergent with advanced coelurosaurs in more global analyses.

In fact, in the culled analysis (the one whose data matrix will be provided
in the Ostrom Symposium volume), if you use _Allosaurus_ as the closest
sister taxon to coelurosaurs,  it does nasty things (like fail to recapture
the two basic positions for Troodontidae, and pop inside the ingroup, and so
forth...).  Using the less complete but more primitive _Monolophosaurus_
helps out here.  Addition of the feathered guys from Liaoning finds
_Confuciusornis_ closer to ornithothoracine birds (enantiornithines +
ornithurines) than my other OTUs, and _Protarchie_ and _Caudi._ in several
different positions immediately outside the ((therizinosauroid +
oviraptorosaur) + (dromaeosaurid + bird)) clade *OR* with _Caudipteryx_ as
the sister group to (therizinosauroids + oviraptorosaurs) and _Protarchie_
as the sister group to (dromaeosaurids + birds).

I used a lot more taxa in there, but these are probably the main ones of
interest to people.

Also, incidentally, if the Yunnan therizinosauroid IS a therizinosauroid,
and our phylogenies are close to the actual tree, then the basal divergence
within Maniraptoriformes occurred by the Early Jurassic, and possibly
earlier.  Time to set Jim Kirkland's fossil-finding sense to "Early or
Middle Jurassic" and let him loose!!

Norell, presenting work by Norell, Makovicky & Clark:  They've done what a
lot of us would have liked to have done, but were afraid to due because of
operational speeds: code individual specimens, not taxa.  They stayed within
Coelurosauria for this study, with _Allosaurus_ for an outgroup.  Their
results, if memory serves: Tyrannosauridae as the sister taxon to other
coelurosaurs, then _Scipionyx_, then (I think) _Sinosauropteryx_ and
_Compsognathus_, then ornithomimosaurs, then a paraphyletic Dromaeosauridae
plus monophyletic Avialae (including Alvarezsauridae), then a troodontid +
caenagnathid + therizinosauroid + _Caudipteryx_ + oviraptorid clade.
Culling the matrix for only specimens with greater than 66% of the
characters scorable cleared  up some mess, but left the basic structure
intact.  A new serious contender for relationships in maniraptoriforms.  As
I upgrade my work comptuer, I'll be getting a copy of NONA (the
phylogenetics software they are using): it can do some real nice things with
matrix searching that I want to try.  They also presented some other
important new ways of dealing with missing data which should prove fruitful
in the future.  As I also pointed out, there is a lot of variation WITHIN
the classic operational taxonomic units of our studies, which must be
accounted for in analyses.

Martin:  _Caudipteryx_ and _Protoarchaeopteryx_ are flightless saururine
birds, not non-avian dinosaurs.  There is no way a "teeter-totter" dinosaur
could become a vertically oriented bird.  Some of his main functional
arguments were based on illustrations from Greg Paul's PDW which Greg no
longer endorses (i.e., the position of the hands in avetheropod dinosaurs,
which based on newer specimens he now illustrates in bird folded fashion
rather than bunny folded fashion).

Currie:  No explicit cladogram (bother...), but _Sinosauropteryx_ is a
compsognathid, _Protarchaeopteryx_ is a maniraptoriform of some kind, and
_Caudipteryx_ is an oviraptorosaur.  Some nice anatomical details.

Chiappe: Presented some basal bird work (surprise).  In his study,
alvarezsaurids dropped two nodes to fall immediately outside _Archaeopteryx_
+ later birds, but still closer to Aves than to other "typical" theropod
taxa.  Chiappe agreed that _Iberomesornis_ is an enantiornithine, and
mentioned that _Sinornis_ and _Cathayornis_ may be the same taxon.  Once
again showed that Martin et al.'s formulation of "Saururae"  (a clade
including Archie and enantiornithines, but not modern birds) is
paraphyletic, and is based on primitive characters.

Cracraft: Spoke about early neornithine evolution and relationships among
major living bird groups.  Lots of cladograms, developed under different
methods.  Hard to summarize overall: maybe Ron Orenstein or Chris Brochu
could give a good overview.  Some good morpho and molecular evidence for a
major radiation of seabirds among neognaths, not at the base of the tree
(the Feduccia model), but among the more derived clades.

I (unfortunately) missed Groth & Livezey's talks (for reasons which might
become apparent on the Discovery Channel on Friday, although maybe not...):
again, hopefully Ron or Chris can summarize.

I did catch the latter half of Per Ericson's talk on relationships among
birds: if I heard him correctly, he was placing _Ichthyornis_ AMONG neognath
birds (teeth notwithstanding).

The Discussion Section:  A more formalized discussion section than (say)
SVP.  People were brought up in groups: Sues, Sereno, me and Norell as one
group; Martin, Currie & Chiappe as a second; and the modern bird guys as a
third.  Because of the fairly congruent results of our talks, we didn't have
much of the way of detailed questions during our section.  We got some
functional questions on flight and on dino tails (which we deferred to the
next day, which is when those papers were presented): afterwards I
predicted, correctly, that there would be taxonomic or systematic questions
asked of the functional folks...  The Martin, Currie, Chiappe trio got the
most questions, most of which were asked of Martin.

I'll get to the second day of talks later: hopefully Ornestein, Brochu,
Kirkaldy, Chapman, or any of the other list members who were there will fill
in important gaps I left out from day one.

Oh, and John Ostrom got a very well deserved standing ovation.

Thomas R. Holtz, Jr.
Vertebrate Paleontologist     Webpage: http://www.geol.umd.edu
Dept. of Geology              Email:tholtz@geol.umd.edu
University of Maryland        Phone:301-405-4084
College Park, MD  20742       Fax:  301-314-9661