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

RE: Livezey and Zusi's big bird morph analysis [...]

John Scanlon (riversleigh@outbackatisa.com.au) wrote:

<With regard to the putative morphological convergence of foot-propelled divers
(and other dinosaurian examples mentioned), I'm reminded of a far more
pervasive convergence within snakes, for burrowing. You can remove any
convergence between two groups (and hence error it may cause) in an analysis by
deleting one of them; i.e. if there are n putative convergers, run the analysis
n times including one at a time along with 'generalised' taxa. I did this with
snake morphological data in an unpublished bit of my PhD, and essentially the
same method was later proposed by Siddall and Whiting (1999) to beat LBA in
sequence data.>

  I think while this idea has merit, to test the quality of the characters on
the matrix and the positions of the taxa, both taxa should be in the matrix at
the time of any test. To accomplish this, assume you know what characters are
acting on the matrix to pull two taxa closer than one suspects or as a test of
a character's efficacy, and delete it. Run the matrix without any given number
or set of numbers of characters, and analysis the effect of the
"foot-propelled" features or "burrowing" features have on the taxa themselves.
Since these are the operating convergent qualities, those features still
present not affecting those habitae should still provide an effect on the
matrix. One can also artificially alter a known outlying taxon and code its
features for the shared grebe-loon states, and see if that effects the tree
negatively or positively, as well as coding, say, the grebes for
non-foot-propelled features, say as a diminished/basal phoenicopteriform-ally
as has recently been proposed. This might produce further interesting results.


Jaime A. Headden

"Innocent, unbiased observation is a myth." --- P.B. Medawar (1969)

We won't tell. Get more on shows you hate to love 
(and love to hate): Yahoo! TV's Guilty Pleasures list.