This study quantified and compared the cross-sectional and longitudinal influence of age, self-report physical activity (SR-PA), and body composition (%fat) on the decline of maximal aerobic power (&OV0312;O2peak). The cross-sectional sample consisted of 1,499 healthy men ages 25–70 yr. The 156 men of the longitudinal-sample were from the same population and examined twice, the mean time between tests was 4.1 (±1.2) yr. Peak oxygen uptake was determined by indirect calorimetry during a maximal treadmill exercise test. The zero-order correlations between &OV0312;O2peak and %fat (r = −0.62) and SR-PA (r = 0.58) were significantly (P< 0.05) higher that the age correlation (r = −0.45). Linear regression defined the cross-sectional age-related decline in &OV0312;O2peak at 0.46 ml-kg−1·min−1· yr−1· Multiple regression analysis (R = 0.79) showed that nearly 50% of this cross-sectional decline was due to %fat and SR-PA, adding these lifestyle variables to the multiple regression model reduced the age regression weight to −0.26 ml-kg−1·min−1· yr−1· Statistically controlling for time differences between tests, general linear models analysis showed that longitudinal changes in aerobic power were due to independent changes in %fat and SR-PA, confirming the cross-sectional results.