As I recall, assuming you ran your PCA on the covariance matrix (not the correlation matrix, and hopefully using log-transformed data), check the stats on PC1. If all of the correlations with characters have the same sign (+/-), then PC1 is a general component. Assuming ontogeny is the primary driver of variation in the data, then this component may be interpreted as representing general size, effectively the ontogenetic trajectory of the data. Scores along this component may be considered a proxy for relative progression along the trajectory (relative size/ age). The formula for this line, derived from the "a values) can then be used to describe the trajectory and predict further scores; although the "a-s" are not always part of the output of a PCA program.

Please take this with a grain of salt, it has been two decades since I first learned morphometrics. I use it often, but my explanation may be lacking... @richard_strauss at Texas Tech is one of the top experts in this type of analysis, and he may be able to recommend a recent summary paper to help guide you.

Jonathan Wagner thank you very much. You helped me a lot ! I think i got this. =)