Little is known about the relationship among training, energy expenditure, muscle volume, and fitness in prepubertal girls. Because physical activity is high in prepubertal children, we hypothesized that there would be no effect of training. Forty pre- and early pubertal (mean age 9.1 +/- 0.1 yr) nonobese girls enrolled in a 5 day/wk summer school program for 5 wk and were randomized to control (n = 20) or training groups (n = 20; 1.5 h/day, endurance-type exercise). Total energy expenditure (TEE) was measured using doubly labeled water, thigh muscle volume using magnetic resonance imaging, and peak O(2) uptake (VO(2 peak)) using cycle ergometry. TEE was significantly greater (17%, P < 0.02) in the training girls. Training increased thigh muscle volume (+4.3 +/- 0.9%, P < 0.005) and VO(2 peak) (+9.5 +/- 6%, P < 0.05), effects surprisingly similar to those observed in adolescent girls using the same protocol (Eliakim A, Barstow TJ, Brasel JA, Ajie H, Lee W-NP, Renslo R, Berman N, and Cooper DM, J Pediatr 129: 537-543, 1996). We further compared these two sample populations: thigh muscle volume per weight was much lower in adolescent compared with prepubertal girls (17.0 +/- 0.3 vs. 27.8 +/- 0.6 ml/kg body mass; P < 0.001), and allometric analysis revealed remarkably low scaling factors relating muscle volume (0.34 +/- 0.05, P < 0.0001), TEE (0.24 +/- 0. 06, P < 0.0004), and VO(2 peak) (0.28 +/- 0.07, P < 0.0001) to body mass in all subjects. Muscle and cardiorespiratory functions were quite responsive to brief training in prepubertal girls. Moreover, a retrospective, cross-sectional analysis suggests that increases in muscle mass and VO(2 peak) may be depressed in nonobese American girls as they mature.