A short response. GPA superimposition removes the effect of variation in centroid size not any shape variation that may be correlated with size variation. If there is allometry then that is not removed because the superimposition does not remove variation in shape - the whole point of the GPA superimposition is to preserve variation in shape while eliminating variation in location, orientation, and scale.
If you wish to remove the effect of allometry from your data then you need a procedure that removes some of the shape variation. F. James Rohlf [EMAIL PROTECTED] 3/28/03 > -----Original Message----- > From: [EMAIL PROTECTED] > [mailto:[EMAIL PROTECTED] Behalf Of > [EMAIL PROTECTED] > Sent: Friday, March 28, 2003 9:03 AM > To: [EMAIL PROTECTED] > Subject: Allometry persists after procrustes? > > > Dear friends, > > I understand that one of the main consecuences of applying alignment > tranformations to landmark data, such as the GLS, is the elimination of > the effects of size. In this case, I also understand that size is > assumed to be isometric. Nevertheless, I found a paper explaining that > allometric residues can still persist in the aligned coordinates, since > centroid size is an isometric parameter. If this is true, then the > residues of the procrustes fit, which are used in posterior projections > and multivariate analyses, are still "contaminated" by allometric > variation. I have been trying to interpret correctly this statement, but > still have my doubts. > > I will be grateful with any advice with respect to this question. > > Thank you very much. > > Pablo > > Pablo Jarr�n > Center for Ecology and Conservation Biology > Biology Department > Boston University > [EMAIL PROTECTED] > > == > Replies will be sent to list. > For more information see http://life.bio.sunysb.edu/morph/morphmet.html. == Replies will be sent to list. For more information see http://life.bio.sunysb.edu/morph/morphmet.html.
