On a practical note, it's not always useful to eliminate allometry (e.g., by regressing shape variables on centroid size and analyzing the residuals). Sender: [EMAIL PROTECTED] Precedence: bulk
A few years ago I did a study of the shape differences between two species of clams, Mercenaria mercenaria and Mercenaria campechiensis, and their hybrids. If size and allometry were removed, the clams were unable to be differentiated based on shape. However, they had different "allometric trajectories"; as clams of each species grow in size they become more and more different in shape. Clams of the same size but of different species were able to be differentiated based on shape when allometry was left in. In my experience I have found that many studies automatically eliminate allometry as a matter of course. In my case, the differences between the species were entirely due to allometry, but the differences in shape were not artifactual. I think this is an obvious point for morphometricians, but I wanted to add my two cents. Matt [EMAIL PROTECTED] == Replies will be sent to list. For more information see http://life.bio.sunysb.edu/morph/morphmet.html.
