Maturity-associated variation in peak O2 uptake was considered in a longitudinal sample of 47 boys and 40 girls who were enrolled in sports schools. The children were followed annually from 11 to 14 years of age. O2 uptake and heart rate were measured during a maximal exercise test on a cycle ergometer. For boys, individual velocity curves were used to operationally define maturity groups: early-decreasing velocities from 11 to 14 years, n = 9; average-velocities reaching a peak and then decreasing, n = 7; and late-increasing velocities from 11 to 14 years, n = 31. The distributions of stages of genital and public hair development were consistent with early, average and late maturity status designation based on velocities of stature growth. Prospectively collected ages at menarche were used to define maturity groups in girls: early--< 12.0 years, n = 7, 10.8 +/- 0.6 years; average--12.0.12.9 years, n = 20, 12.4 +/- 0.3 years; and late-- > or = 13.0 years, n = 13, 13.5 +/- 0.4 years. Early maturing boys had a greater O2 uptake at each observation period. Early and average maturing girls did not differ in maximal O2 uptake, but both had greater O2 uptake than late maturers. When expressed per unit body mass, differences among the three maturity groups of boys were reduced and not significant. Late maturing girls tended to have greater maximal O2 uptake per unit body mass than early and average maturing girls, but the differences were not significant at all ages. However, with body mass at the first observation as the covariate in analyses of covariance, the three maturity groups of boys differed significantly in peak VO2 at each observation, while the three maturity groups of girls did not. Thus, removing the confounding effect of body mass on O2 uptake by simply dividing the measured values by mass is of limited utility.