The purpose of this study was to assess an effect of genotype on the patterns of age-related changes in running performance and physical growth during adolescence. The total sample was 14 pairs of male and 25 pairs of female monozygotic (MZ) twins and 19 pairs of male and 15 pairs of female dizygotic (DZ) twins. Performance on the 50-m dash and endurance run (1,500 m for boys and 1,000 m for girls, respectively) and stature and body mass were measured at yearly intervals from 12 to 17 years of age. A principal component analysis was applied to the longitudinal data, and within-pair resemblance in the scores on the first, the second, and the third principal components was compared between MZ and DZ twins. The analysis was conducted with both sexes pooled because of the limited number of twin pairs. The first two principal components explained at least 87% of the total variance, while the third principal component (PC3) explained a smaller portion of the total variance (more than 5%) in the dash and the endurance run. However, the total variance was explained almost entirely (more than 96%) by the first two components in height and mass. The first principal component (PC1), which explained 67-84% of the total variance, was a good indicator of average performance/body size of individuals during the period of observation. The second (PC2) and the third components (PC3) could be considered as indicators of the "shape" of developmental/growth curves, which were not related to inter-individual differences in the average performance/body size. Intraclass correlations (r(i)) of the scores on PC1 were consistently higher for MZ than for DZ twins, and estimated genetic variance was significant in endurance run, stature and body mass. The situation was the same for the scores on PC2 and/or PC3. However, such a between-genotype difference in r(i) and a significant genetic variance were not observed in the dash. It was concluded that MZ twins are more similar in the patterns of age-related changes in the performance of endurance run and physical growth during adolescence than DZ twins, suggesting a genetic contribution, while there was no evidence of a genetic contribution in the dash. Am. J. Hum. Biol. 13:71-80, 2001. (C) 2001 Wiley-Liss, Inc.