Re: Giganto characters
HARRY R. ERWIN (herwin@osf1.gmu.edu)
3 Aug 1995 18:48:33 GMT
alex duncan (aduncan@mail.utexas.edu) wrote:
: Harry,
:
: I just had another look into your database (actually, I downloaded it and
: played w/ it some). I'm disturbed by the number of character states that
: you have entered for Gigantopithecus. For example:
: incisive canal: overlap (pongo)
--fixed
: maxillary sinus: absent
--found in a source
: brow ridge: none
--fixed
: glabella: minor
--fixed
: facial shape: muzzle
--fixed
: cranial capacity: various character states
--fixed
: brain case wall: thin
--fixed
: carotid artery entry: ventral
--fixed
: vault profile: primitive (small)
--fixed
: basicranial flexion: absent
--fixed
: separate os centrale: present
--fixed
: curved metacarpals: present
--fixed
: FM [foramen magnum?] placement: quadrupedal
--fixed
: prehallux: absent
--fixed
: etc. etc.
--mostly fixed
I've been doing a clean-up of almost all of that over the last few days,
but haven't uploaded the modified database (and won't until I've validated
_everything_). A good many of those values were plugged in as hominoid
defaults early on and were never checked for Gig. Simons claims that the
Indian Gig. looks like a big Siva, but provides no data I can use other
than a profile of the jaw. After cleaning the DB up, Gig. didn't move, but
I assume that's because size increase in Pongo and Gig. resulted in a good
deal of homoplasy, since the selective gradients imposed on other features
by the size increase would have been the same.
The thing that really drove me out of my skull was Simons indicating Gig.
was 800 pounds and 8 feet tall. Those figures are basically off-scale for
apes. The canine tooth dimorphism was from Oxnard. Again off-scale.
: I wonder how much of an impact these have on your placement of
: Gigantopithecus?
nil.
: I realize that we really do "know" what a lot of these
: character states are (although we shouldn't make these assumptions about
: the ones i've listed above), but when you make these assumptions in your
: data base you jump ahead in the process one step. For example, you have
: "maxillary sinus: absent" as a character/state. My impression is that
: there are no Giganto fossils that preserve this anatomy. You may have
: good reasons for assuming the maxillary sinus is absent, but that should
: come as a result of your cladistic exercise, and not as an assumption you
: begin with.
I've got a source for that one, since it wasn't defaulted. I'll look at
home tonight. I suspect it was Oxnard, since I don't remember seeing it in
Simons or Groves last night. In any case, it has no effect, since the rest
of the apes in that little clade have the opposite polarity.
: It would be interesting to take your database and remove all data that
: are not concretely supported by the fossil record (even stuff like "tail:
: absent" for Giganto). And then see what comes out of it. We all know
: damn good and well Giganto didn't have a tail, but it biases your
: results. I suspect the result will be a lot less confidence in the
: placement of some taxa, but that is a price that has to be paid to do
: good science.
I agree, and that's what the current exercise is about. Unfortunately,
some (most?) of my sources make the same error from time to time. I believe
it's called 'begging the question.'
: I've agonized over this myself a great deal. We teach an intro course in
: which we have students use MacClade to do an exercise w/ representative
: groups of modern primates. We have to keep characters simple, so the
: students will have some understanding of them. We throw in a few
: interesting fossil taxa (Siva, Australopithecus afarensis, Proconsul) and
: ask the students to predict character states for them after completing
: the exercise.
: We can be pretty confident that Siva didn't have a tail, because its a
: 'noid, and 'noids don't have tails. But we don't have any fossils that
: bear on the issue. So we leave a "?" in our file for the character
: state, and ask students to "predict" for us whether or not Siva had a
: tail, based on its position in the cladogram.
: To get back to the case of Giganto, you're assuming that it has a lot of
: character states that are similar to those seen in Pongo and Siva, even
: though these parts aren't represented in the fossil record. I feel
: confident that those assumptions are effecting the position of Giganto in
: your cladogram.
It didn't. Sorry.
--
Harry Erwin
Internet: herwin@gmu.edu
Home Page: http://osf1.gmu.edu/~herwin (try a couple of times)
PhD student in comp neurosci: "Glitches happen" & "Meaning is emotional"
|