No subject

[Don.Colless at CSIRO.AU]

John Grehan wrote:

"It seems to me to be a corollary to the current problem of genetic
similarity analysis not being able to designate only shared derived
states prior to analysis."

I'm a bit surprised that John sees this as a problem for only the
moleculars. And it doesn't really matter: an UNROOTED minimum length
tree can be computed (and in principle often is, by morphologists, too)
without any polarisation of characters, and the root then estimated by
attaching a plausible outgroup. Shared derived states can then be read
off the cladogram - and preferably evaluated for THEIR plausibility. A
lot more analysis could (and IMHO should) follow (but that's another
argument!).

Don C.


Don Colless,
Div of Entomology, CSIRO,
GPO Box 1700,
Canberra. 2601.
Email: don.colless at csiro.au
Tuz li munz est miens envirun