Forty-eight sedentary and 39 quite active or well-trained men participated in this study. Muscle biopsy samples were taken from the vastus lateralis for the determination of fiber type composition (I, IIa, IIb), fiber type area, and assay of the following enzymes: malate dehydrogenase (MDH), 3-hydroxyacyl CoA dehydrogenase (HADH) and oxoglutarate dehydrogenase (OGDH). Maximal oxygen uptake (VO2max) was determined with a progressive cycle ergometer test, while endurance performance or maximal aerobic capacity (MAC) was defined as the total work output during a 90-min cycle ergometer test. Correlation analysis revealed no evidence of association between fiber type composition and VO2max kg-1 or MAC kg-1 in sedentary subjects, while active men exhibited significant correlation between % type I (r = 0.52), % type IIb (r = 0.31) and VO2max kg-1. Enzyme activities were not significantly correlated with MAC kg-1 and VO2max kg-1 in sedentary men while active men exhibited significant correlation for the three enzymes (0.37 less than or equal to r less than or equal to 0.51) with VO2max kg-1. These results show that the contribution of muscle fiber type and enzyme activities to aerobic performance may be inflated from a statistical point of view by the training status heterogeneity of subjects. They also suggest that variation in these muscle characteristics does not account for the individual differences in aerobic performance of subjects who have never trained before. Therefore, the assessment of muscle characteristics is not as useful as originally thought for the detection of individuals with a high potential for endurance performance among untrained subjects.