morphmet escreveu: > differ too? I am aware about Rohlf's warnings (Syst, Biol.) on the > use of single partial warps as homologous characters, but I am not > yet convinced this apply to my case.
They should apply, I regret to say. This is a consequence of the very mathematical definition of the partial warps and not of any particular dataset. Analyses that may seem interpretable in this context (PWs as putative homologous characters) may notwithstanding produce misleading results. Jim Rohlf made this point quite clear in his 1998 SB paper. As of the orthogonality of PWs, they are onthogonal in the geometrical sense - not in the statistical one, hence the need to use dimension-reducing methods as relative warpes analysis (a PCA of the PW scores), that produce statistically orthogonal axes that can be interpreted as such. Hope this helps... :-( With warmest regards, -- Replies will be sent to the list. For more information visit http://www.morphometrics.org
