The purpose of this study is to examine the stability of somatotypes of boys with different activity levels followed longitudinally for eight years. Thirty-nine boys ated 11 were somatotyped annually (until 18 years) using the Heath-Carter anthropometric rating method. Three sub-groups were formed according to their levels of systematic physical activity designated High, Moderate, and Low Activity. There were no differences in somatochart distributions of the Total Group, but there were considerable individual changes which cancelled each other in group comparisons. All boys changed their somatotype ratings at least once and 67% changed in component dominance. The inter-age correlations for the components were generally of poor prediction value, and the somatotype migratory distances showed considerable variation. The Total Group means differed on three components and this was attributed to differences within the Moderate Activity Group, the only sub-groups to differ. It was concluded that changes occur in somatotypes of adolescent boys, but the influence of physical activity on the changes is inconclusive.