Dear Catherine, In exploring the possible clinal relationship among your specimens you may also want to consider 2 Block Partial Least Squares, which can be accomplished with the tpsPLS software. You can use tpsPLS to accomplish this, where your shape data form one matrix and the geographic coordinates of your five sites form the other matrix. The technique is described by Rohlf and Corti (2000) and used to describe a cline in baboons in Frost et al., (2003) in the Anatomical Record.
You could perhaps also consider using multiple regression of your shape variables on the rank order, or even the distance from the first locality, or whatever, of your sampling localities. Sincerely, Steve Frost --- morphmet <[EMAIL PROTECTED]> wrote: > [Oops, found a couple of postings I missed. No > excuses. -Moderator > (a.k.a. dslice)] > > I'm a first year Ph.D student in the Ecology and > Evolution Department at > Leeds University, England. I'm studying a cline in > morphotype in the > inter tidal snail Littorina saxatilis which suggests > there may be > speciation in progress. I am very interested in > using geometric > morphometrics to quantify the variation in shell > shape of the snails, > although my knowledge of the field is limited at > this stage. > > I have sampled L.saxatilis from five locations along > a vertical transect > of the shore. I wish to analyse the cline in > morphotype by analysing > shell shape variation between the five locations > along the transect > containing hopefully three morphotypes (H, I and M). > I have also sampled > from a second transect to replicate the experiment. > > I have taken the route of landmarks and I am using > the software TPS. I > plan to use PCA/DFA in an attempt to discriminate > between the > morphotypes. However, I am unsure whether to use > the Partial warp > scores matrix, Relative warps score matrix or Weight > matrix from the > TPSRelw program in the analysis. > > I would be grateful for any advice. > > Catherine Sinfield > > [EMAIL PROTECTED] > > > > > > ________________________________________________________________________ > Yahoo! Mail Mobile > Take Yahoo! Mail with you! Check email on your > mobile phone. > -- > Replies will be sent to the list. > For more information visit > http://www.morphometrics.org > > __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com -- Replies will be sent to the list. For more information visit http://www.morphometrics.org
