Little data exist for upper and lower body mechanical power capability of adolescent athletes. This study compared arm (A) and leg (L) anaerobic peak and mean power (PP and MP) of 20 male and 20 female adolescent athletes after normalization for body mass (BM), fat-free mass (FFM), and lean A and L cross-sectional area (CSA). Power outputs were assessed by the Wingate anaerobic test. FFM and CSA were estimated via anthropometry. No significant (P > 0.05) differences existed between the sexes in Tanner sexual maturity, chronological age, or overall training activity. Males had higher (P < 0.001) absolute PP (W) (L 694 vs 442; A 494 vs 309) and MP (L 548 vs 307; A 337 vs 214). Ratio normalization and ANCOVA were used to remove the influence of body size differences. Ratio normalization showed that males had greater leg PP/BM, MP/BM, MP/FFM, MP/CSA, as well as arm PP/BM and MP/BM, whereas all leg and arm PP and MP ANCOVA adjusted means for BM, FFM, and CSA, except arm MP adjusted for FFM, were significantly (P < 0.01) higher for males than females. We conclude that factors other than muscle mass, possibly qualitative in nature, are responsible for the sex difference in anaerobic performance of adolescent athletes.