Re: Dinosaurs with Pygostyles

[Nick Longrich <nrlongri@midway.uchicago.edu>]

>At 05:33 AM 16/01/2000 -0800, Jaime A. Headden wrote:
>>Barsbold R.; Currie, P.J.; Myhrvold, N.P.; Osmólska,
>>H.; Tsogtbaatar K.; and Watabe M. 2000. A pygostyle
>>from a non-avian theropod. Nature 403: ???-???.
>
>Of course, given the date of the conference, these papers are reflective of
>the pre-Sinosauropteryx etc. state of affairs.  Nonetheless dinopersons
>will probably want to read at least the concluding section on Mesozoic
>Birds and Early Avian Evolution, consisting of seven papers and reports on
>four roundtable discussions.
>
>Anyway, one of these is a paper by Andrzej Elzanowski, "A comparison of the
>jaw skeleton of theropods and birds, with a description of the palate in
>the Oviraptoridae".  The abstract reads:
>
>"Similarities to birds in the structure of the jaws and palate suggest that
>oviraptorosaurs (oviraptorids and caenagnathids), therizinosauroids, and
>ornithomimosaurs are the closest theropodan  relatives of birds, which is
>in conflict with recent phylogenetic reconstructions based on postcranial
>evidence. No specific avian similarities could be found in the jaws and
>palate of dromaeosaurids. The ectopterygoid of the oviraptorids connects
>the lacrimal to the palatine, as does the avian uncinate (lacrimopalatine).
>This and other cranial similarities between the oviraptorosaurs and
>omithurine birds raise the possibility that oviraptorosaurs are the
>earliest known flightless birds. With Archaeopteryx and the theropods
>providing evidence of plesiomorphic conditions, similarities in the
>mandibles, teeth, and tooth implantation in the Ichthyomithidae and
>Hesperornithidae may be interpreted as synapomorphies supporting monophyly
>of the Odontognathae."
>
>Specifically, the author notes four characters shared by oviraptors and
>ornithurines not found in Archaeopteryx: Articular with Iateral process;
>Articular with medial process; Mandibular symphysis fused; Jugal bar
>rod-shaped.  He concludes that: "These suggest that the oviraptorosaurs
>branched off after Archaeopteryx and thus represent the earliest known
>flightless birds. Except for the elongate forelimbs (which become shortened
>in all flightless forms), the postcranial skeleton of Archaeopteryx does
>not have any avian traits that would be absent in the oviraptorids .
>Therefore, if flightlessness had evolved at a stage of avian evolution
>close to Archaeopteryx this would be extremely diff

        Some complications:
        The jugal bars in both Caudipteryx and the therizinosaurs are the
typical, thick structures seen in most theropods, not like the thin ones in
Oviraptoridae (and the avimimus skull material, make your own decision
about what that means...). So I think that the rodlike jugal bar is best
interpreted as convergent. Nor do either Caudipteryx or therizinosaurs
possess dental symphyses, so again I'd doubt it's a synapomorphy.
        So I disagree with Elzanowski's arguments; I'm not so sure that
some of his conclusions couldn't be correct. In a number of ways
oviraptorosaurs are less birdlike than dromaeosaurs- the sternum is
shorter, they haven't completely lost the anterior portion of the pubic
boot, the transverse processes in the anterior tail are placed fairly
dorsal, the posterior tip of the ilium lacks the typical pointiness seen
especially in archaeopteryx, rahonavis, modern birds, and somewhat in
dromaeosaurs.
        But in a number of other ways they are disturbingly advanced. The
separated acromion, the shortening of the distal tail, the fan instead of
frond in the tailfeathers. And of course that pygostyle. There appears to
be a lot of convergence and reversal and I'm not sure where they all happen.

        The problem of secondary flightlessness is very complicated and, in
my opinion, much abused. It's hard to even get the hypothesis considered as
a rational possibility. Usually some appeal to phylogenetics is brought up,
but one of the biggest problems is that we don't yet understand the
phylogenetics! How many studies actually *test* whether dromaeosaurs and
oviraptorosaurs are inside an Archaeopteryx+modern birds clade?  Gauthier
1986? Nope. Holtz 1994? Nope. Makovicky and Sues 1998? Nope. Sereno 1999?
Forster et al (1998) and Novas' works are the only ones that actually allow
for the possibility. What is generally done is to either code only
Archaeopteryx, or to assume that Archaeopteryx and modern birds form a
clade exclusive of everybody else. This is not testing hypotheses of avian
evolution, it is assuming them. I don't mean to knock people's hard work on
phylogeny. But I can't agree with the way Greg Paul's hypothesis has been
treated. It may be right, it may be wrong. It may be right for reasons we
haven't thought of, or wrong for different reasons than we thought, but
we've got to figure out a way to test it rationally. Until that's done the
science is not going to advance; personally I think that maybe
overemphasizing characters and underemphasizing what these characters
changed for exactly is aat the heart of it. But then I tend to see
everything through a functionality lens.

>end rant<

        -N