Maybe I'm missing something here, or perhaps I'm showing a bit of ignorance, but if you are concerned about allometry why not regress measures of "shape" (relative warp scores, for example) against centroid size first to see if there is a relationship. If a significant relationship exists couldn't you then conduct subsequent analyses on the residuals?
kw --------------------------------------------------------- Kenneth I. Warheit, Ph.D. Senior Research Scientist Washington Dept. Fish and Wildlife 360-902-2595 (voice) 360-902-2774 (lab) 360-902-2162 (fax) [EMAIL PROTECTED] --------------------------------------------------------- == Replies will be sent to list. For more information see http://life.bio.sunysb.edu/morph/morphmet.html.
