_Achillobator giganticus_ and homoplasy

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

        No, you probably haven't heard of it. It's the recently named giant
dromaeosaur from Mongolia (Perle, Norell, and Clark 1999), published in an
obscure place in Mongolia.
        Femur is 500 mm in length: a pretty big sucker. The sonovagun has a
slightly *propubic* pelvis, and really narrow pubes- reminds me of
tyrannosaur or ornithomimid or Coelurus pubes in that respect. As for
anterior expansion of the pubic boot: It would put any self-respecting
oviraptorid to shame. Positively to shame. And don't even get me started on
the *proximally* situated obturator process and extremely long pubic apron
(these characters are probably correlated).
        The paper suggest that it's related to Dromaeosaurus although I
don't think the paper goes into much detail about why. It's got a pretty
darn deep maxilla, too; this would have given the animal a vaguely
Allosaurus-like profile I imagine. The ilium is weird; it is very deep and
the posterior is enormously expanded.
        My first thought was, "this can't be a dromaeosaur", but it is.
Posterior serrations on the teeth bigger than anterior ones, anteriorly
forked chevrons, hyperelongate prezygs, killer-claw, a lot of medial
closure to the acetabulum, big ol' posterolaterally located antitrochanter,
sharp tip to the end of the the ilium, short anterior iliac blade with a
big anterior notch, V-shaped ventral margin to the pubic peduncle. Most of
the weird features are presumably derived relative to the other members of
dromaeosauridae, one feature which is probably primitive is the small but
undeniably present Proximodorsal Process of the ischium. It's supposed to
be in Sinornithosaurus too if I recall, but it's easier to see here.

        I guess that having been given yet another swift kick in the head
by evolution ( "wake up, stupid! This is evolution, and I do things
COMPLICATED! You think you understand me? Think I fit in your neat little
naive schemes? Think again!")  I am beginning to feel even more hesitancy
towards assigning much confidence to our lovely little cladograms. Sure,
some of the general picture stuff- troodonts up somewhere near
dromaeosaurs, tyrannosaurs as coelurosaurs, etc.- we are beginning to get,
but I feel uneasy about some of the other stuff- e.g. how the various
sickle-claw groups relate to each other and birds, if oviraptorids are
really more primitive than dromaeosaurs or just more "degenerate", where on
earth alvarezsaurs go- let alone figuring out the higher-level
relationships of any highly fragmentary specimen (e.g. Unenlagia or
Rahonavis) that doesn't immediately fit glove-tight into one of our
existing monophyletic clades, or of figuring out the relationships of an
family that isn't represented by a fairly good sampling of genera. I really
doubt anyone could have figured out that Achillobator was a dromaeosaur
based on the pubis alone, for example, or even maniraptoran.
        I mean, there was a thread a while back about assigning
Sinosauropteryx to this or that higher-level taxon based on various little
characters, and there is some discussion here about what Itemirus and
Stokesosaurus were. But honestly, what's the point? Achillobator takes what
seemed to be a bunch of really pretty good typical dromaeosaurid
characters, to me anyways- like an ischium only 2/3 the length of the
pubis, lack of substantial anterior expansion of the pubic boot, a distally
placed obturator process- and just tears through them like so much tissue
paper. It's homoplasy up the wazoo. But it's still a dang dromaeosaur.
        Dunno. I'm not saying it isn't useful to go out and produce
cladograms (or I wouldn't be spending so much time at it), it just seems
that sometimes there are arguments out there based too heavily on the
unquestioned assumption that, well, because we got an answer back from
PAUP, that answer is right or pretty close to it.  Because it happens to be
a step or two shorter than (literally) dozens or perhaps hundreds of other
answers. Or that any cladogram no matter how lousy is better than an
explicit, well-thought-out and well-argued hypothesis that hasn't been run
through PAUP.  Two or three- maybe even half a dozen- characters do not a
robust clade make. I guess what I would like to see in the dino field is
more worrying about how robust our results really are, and how much stuff
like intraspecific and interfamilial variation, long-branch attraction,
taxon and character sampling, accidental miscoding, workers' coding biases,
and missing evidence screw with them.
        But I guess that's what they make journals like "cladistics" for. I
should really get over to the library more often.
        My 2¢ for what it's worth.  Maybe Chris Brochu or Tom Holtz would
have some more intelligent stuff to say about this. I wouldn't mind hearing
about what people think about the patterns of homoplasy seen in other taxa
besides theropods , which is all I ever think about. Are theropods really
more homoplastic, or have we just not looked at the others in enough depth
to really uncover the homoplasy?

        -N