Abstract
Morphological data are often used to investigate differences among groups of individuals which are genetic or taxonomic in nature. A statistical analysis is presented for longitudinal data which consists of q different size measurements recorded at p successive ages. In such cases differences may be examined not just in terms of size and shape but also in terms of the ontogenesis of size and shape. A growth curve model is first used to express the differences among the groups as a polynomial in time. The estimated parameters of the growth curve model are then employed to investigate the group differences. A canonical analysis of the parameters is suggested as a means of identifying the static and ontogenetic nature of the morphological differences. To illustrate the growth curve model and the canonical analysis of the parameters, data from 25 strains of chickens are analyzed.

This publication has 1 reference indexed in Scilit: