Weight-bearing endurance activity and resistance exercise are recommended to help preserve bone health during adulthood. However, the effects of resistance training relative to those of weight-bearing endurance activity often are confounded by body weight and composition. The purpose of this study was to determine the effects of long-term running, cycling, and resistance training on whole-body and regional bone mineral density (BMD), adjusting for body weight and composition. Cyclists (CYCLE; n = 19), runners (RUN; n = 10), and resistance trained men (RT; n = 13) ages 19-45 years participated in this cross-sectional study. Current and lifetime bone loading was calculated using ground-reaction force values of the reported physical activities. Whole-body and regional BMD and body composition were assessed using dual X-ray absorptiometry. Bone turnover markers and hormones were measured in fasting serum samples. The RT athletes had significantly greater body weight, lean body mass (LBM), and fat mass than CYCLE and RUN athletes; percent body fat did not differ among groups. Unadjusted BMD at all sites was significantly greater in the RT compared with CYCLE and RUN. After adjusting for LBM, RUN had significantly greater spine BMD than CYCLE. Subjects' LBM was a significant predictor of BMD in RT and CYCLE but not in RUN, suggesting that high-impact activity may override the benefits of LBM on BMD. Current bone loading was positively associated with serum osteocalcin concentrations (r = 0.480, p = 0.002). In conclusion, the results of the present study demonstrate that long-term running and resistance training increase BMD compared with cycling. However, it seems that high-impact activities, such as running, have a greater positive effect on BMD than resistance training.