OBJECTIVE To compare the maximum oxygen uptake V0(2max) of elite endurance athletes and to explain why the body mass exponent, necessary to render V0(2max) independent of body mass, appears to be greater than 0.67. METHODS Study 1: V0(2max) of 174 international sportsmen and women was assessed. Athletes were recruited from seven sports (middle- and long-distance runners, heavyweight and lightweight rowers, triathletes, and squash and badminton players). Study 2: calf and thigh leg muscle masses were estimated in 106 male and 30 female athletes from 11 sports. Differences in V0(2max) and leg muscle masses between "sports" and "sex" were analyzed independent of body mass by using allometric log-linear ANCOVA. RESULTS Heavyweight rowers had the greatest V0(2max) when expressed in L.min but long-distance runners had the highest V0(2max)in mL.kg.min. However, the ANCOVA identified no difference in "mass independent" V0(2max) between the five "pure" endurance sports (runners, rowers, and triathletes) (P > 0.05) with the two racket sports being significantly lower. The body mass covariate exponent was inflated, estimated as 0.94. The results from study 2 estimated calf and thigh leg muscle masses to increase in proportion to body mass, and, respectively. CONCLUSIONS After having controlled for differences in body mass, V0(2max) did not differ between pure endurance sports (P > 0.05). Assuming that athletes' thigh muscle mass increases in proportion to body mass as observed in study 2, a similar disproportional increase in V0(2max) would be anticipated, providing a plausible explanation for the inflated mass exponent associated with V0(2max) identified in this and other studies.