[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: *Gansus*



----- Original Message ----- From: "Harris, Jerald" <jharris@dixie.edu>
Sent: Monday, March 12, 2007 6:03 PM


We simply used Clarke & Norell's matrix, with a few changes and additions -- we didn't try to change or add codings from anything we hadn't personally seen (well, Luis & Jingmai probably saw, and they did contribute to our changes for a few things), including _Enaliornis_. This wasn't because we were/are lazy -- we are trying hard to get the funding we'll need to go see as many of these things in person as humanly possible -- but because our analysis was preliminary.

That's clear. That's why I asked -- instead of assuming an answer. :^)

Rest assured also that a brand-new data matrix is being compiled, based in part on Clarke & Norell and others but with lots of new/changed characters and more accurate codings.

Looking forward to it!

Still, it does beggar the question of exactly how much faith one can put in a computer-generated analysis when the human mind says something else...are we Borg or are we capable of abstracting beyond a bunch of statistics?

Well... subjective assessments of which characters are "important" have so often been shown to be wrong by now that I don't think we can abstract beyond the statistics, despite all the well-meaning talk that a cladistic analysis should be the starting point rather than the end of a phylogenetic study. On the other hand, of course, subjective judgments are inevitable at the other end of the analysis -- how to code a character.


And such judgments are again dependent on the data. Think of the sternal characters in Mickey's analysis that I bemoaned yesterday (and every few years again). I think that what is actually phylogenetically informative about them is the point in the ontogeny at which the bone or process in question ossifies or fuses. To figure that point out we'd need growth series that we simply don't have -- apparently not even for *Confuciusornis* (though someone should have a look at a couple hundred specimens and...).