Untested assumptions (was Re: segnosaurs)

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

At 06:24 PM 7/31/97 -0400, Dinogeorge wrote:
><< There are some gross excesses of other kinds, of course. Does the most
>parsimonious explanation have to be the correct one? No. Do those employing a
>cladistic methodology say that it is? Of course not. >>
>
>Of course they do, character versus character! Otherwise, why are they using
>parsimony in their algorithms? PAUP: Phylogenetic Analysis >Using Parsimony<.
>Unless they're engaged in the pursuit of meaningless cladogram generation.

I have gone over this enough on the net, and the systematic literature is
full of information clarifying this, that we need not elaborate the point
here, beyond saying: as in all Science, the most parsimonious explanation is
prefered as it requires the fewest assumptions.  (Example from
morphometrics: don't assume a twelfth-order differential equation is the
best explanation for a plot of hindlimb length vs. mass when a simple
allometric equation will suffice).

><<     As for the bit about postcranial characters being the most important-
>that's out and out ludicrous.  [snip]>>
>
>Let me go over this one more time. >Cranial< characters are most useful in
>sorting genera within more inclusive clades; >postcranial< characters, being
>more conservative and less variable, are most useful in defining the larger
>clades.

Perhaps the readership here might be interested in some actual experimental
runs which test this supposition?  (Maybe not, but here's hoping...).

Duke University zoologists M.R. Sanchez-Villagra and B.A. Williams have
actually bothered to TEST whether cranial, dental, or postcranial characters
are more homoplastic.  They examined 41 published data sets of various
mammalian taxa, from within-genus studies to studies including most
mammalian orders.  They found that there was no (count them, no)
statistically significant differences between the amount of homoplasy found
in skull characters, in teeth, or in the postcranium.  Their main
conclusion: "Our analysis of existing data suggest that characters from the
dentition, the cranium, or the postcranium are in general equally
homoplastic and none should be dismissed or disregarded in phylogenetic
analyses without a close analysis of the data."

So, could we PLEASE leave our untested assumptions aside?

Source: Sanchez-Villagra, M.R. and Williams, B.A.  1997.  Levels of
homoplasy in the evolution of the mammalian skeleton.  Journal of Morphology
232: 319.

(Unfortunately, just an abstract: I grealy look forward to the final paper!).

Thomas R. Holtz, Jr.
Vertebrate Paleontologist     Webpage: http://www.geol.umd.edu
Dept. of Geology              Email:th81@umail.umd.edu
University of Maryland        Phone:301-405-4084
College Park, MD  20742       Fax:  301-314-9661