Maximal oxygen uptake (VO(2max)) is commonly divided by body mass or fat-free mass (body mass minus fat mass) in order to make it size independent so that comparisons among persons of different size can be made. However, numerous studies have shown that the ratio created is not size-independent. Analysis of covariance (ANCOVA) allows a dependent variable to be compared between groups at a common value of a covariate. The purpose of this study was to compare VO(2max) at the same fat-free mass (FFM) in 230 sedentary subjects (half men) who ranged in age from 20 to 70 years. The subjects underwent maximal cardiopulmonary exercise testing on a cycle ergometer as ventilation and the expired gas fractions were being measured. Two ANCOVA models were evaluated. The dependent variable, fixed factor and covariate(s) in the linear model were VO(2max), sex and FFM, respectively. The corresponding terms in the log-linear model were ln VO(2max), sex, and ln FFM and age. Sex made a significant contribution to both models. In the linear model, the mean VO(2max) at the same FFM was 27% higher in men (2,444 versus 1,929 ml min(-1); P<0.001). In the log-linear model, the corresponding value at the same FFM and age was 32% higher in men (2,368 versus 1,794 ml min(-1); P<0.001). The goodness of fit indices of squared multiple correlation coefficient and standard error of estimate were significantly better for the log-linear model. We conclude that VO(2max) at the same FFM is considerably higher in men than in women who have a sedentary lifestyle.