[Taxacom] Underlying synapomorphy

[Richard Zander]

I think the problem is that one should not accept one explanation when there is an alternative. If there are two alternatives, scientists act on explanations when the alternatives are improbable or are at a defined level of improbability, or entertain as good working hypotheses those that are far more probable than the others. Notions of atavism and orthogenesis have been dusted off recently because actual demonstrated evo-devo mechanisms throw doubt on some aspects of using parsimony with morphological traits. Not only can traits be generated by reactivated suppressed genes, but linkage during selection makes two or more traits no less probable than one as a synapomorphy.
 
Given this, it remains true that cladograms with morphological traits are commonly similar to those of molecular traits. This is amazing, and it means that parsimony is pretty good but may occasionally fail (for perfectly good theoretical reasons) at least in details of separating similar species, and whenever it is most important to you to get a good answer.
 
One should not accept good resolution by ignoring problems but perhaps accept a given degree of bad resolution given the data and present methods of analysis.
 
_______________________
Richard H. Zander
Missouri Botanical Garden
PO Box 299
St. Louis, MO 63166 U.S.A.
richard.zander at mobot.org
 

________________________________

From: taxacom-bounces at mailman.nhm.ku.edu on behalf of Kipling (Kip) Will
Sent: Sat 3/7/2009 8:37 PM
To: taxacom at mailman.nhm.ku.edu
Subject: [Taxacom] Underlying synapomorphy



"Underlying synapomorphy" is more or less equivalent to the notion of
atavism and largely overlapping with orthogenesis. Taken as hypothesized
explanations of observed patterns these are relatively harmless as long
as one recognizes what is actual observation and what is a good story.
But confusing pattern and process is a venerable part of our field I fear.

If taken as a priori truth, a model of how character state distributions
came to be as they are, or how they "should" be distributed, these can
lead to a dangerous ability to explain away any pattern one does not
like. Or perhaps more accurately, a preferred grouping is determined and
any conflict (patterns of homoplasy)can be dismissed ad hoc. "These
belong in a group because they share an underlying synapomorphy" could
be used to justify a preferred grouping sans any observable evidence
supporting such a grouping. Studies of development or genetics could
provide actual synapomorphies, shared and observable. No one would need
to call them "underlying synapomorphies."