PC1 by definition corresponds to the direction of maximum variance in your data. Within a species and sex this usually corresponds to size and/or age if you include specimens that do vary in size. However, if you include different species and sexes or specimens that are about the same size then the direction of maximum variation could be anything depending on how you choose which species or morphs to include.
____________________________________________ F. James Rohlf, Distinguished Professor, Emeritus. Ecology & Evolution Research Professor, Anthropology Stony Brook University From: Tarek Ismail <tarekismail2...@gmail.com> Sent: Wednesday, December 26, 2018 5:17 AM To: MORPHMET <morphmet@morphometrics.org> Subject: [MORPHMET] PC1 correlation with CS and length Hi all I wish to discuss a part of my work result. Studying ontogeny of a land isopod, and after PCA analysis, I have no correlation between PC1 and body length as well as with centroid size of pooled data (adults and juveniles) and also for adults. Since PC1 usually interpreted as a size related axis, what can be the reason beyond this unexpected relationship. Gardner et al., (2016) study the Symmetry in the Core Goodeniaceae (not the case here) and found no correlation between length and centroid size. They attributed that to the Symmetrical variation rather than as a consequence of asymmetric or allometric factors. I decrease the number of landmarks, but have the same result. I attach the PCA Fig., it represent half number of specimens, F=females; M=males and U=juveniles. Thanks in advance Tarek G. Ismail Sohag University, Egypt t_gad_2...@sohag.edu.eg <mailto:t_gad_2...@sohag.edu.eg> -- MORPHMET may be accessed via its webpage at http://www.morphometrics.org --- You received this message because you are subscribed to the Google Groups "MORPHMET" group. To unsubscribe from this group and stop receiving emails from it, send an email to morphmet+unsubscr...@morphometrics.org <mailto:morphmet+unsubscr...@morphometrics.org> . -- MORPHMET may be accessed via its webpage at http://www.morphometrics.org --- You received this message because you are subscribed to the Google Groups "MORPHMET" group. To unsubscribe from this group and stop receiving emails from it, send an email to morphmet+unsubscr...@morphometrics.org.
