Exercise is thought to have considerable potential as a preventive for osteoporosis. We critically examined 27 studies that address the prophylactic role of exercise in osteoporosis. The results from both cross-sectional and longitudinal studies showed that differences in bone mass were more pronounced in the axial skeleton as opposed to the peripheral compact skeleton. The 17 cross-sectional studies demonstrated greater bone mass among highly trained athletes compared with sedentary subjects, while results among recreational athletes were inconsistent. The 10 prospective investigations examining the effect of exercise on bone mass yielded conflicting results; only one study of six found an overall positive response in compact bone mass at the radial site, and only one study examining the spine showed a significant gain among the exercisers. Additionally, all the prospective investigations included serious methodologic flaws; most failed to employ a randomized design, appropriate estimates of sample size were lacking, none provided information on blind outcome assessment, and most studies were of short duration. Current evidence suggests that exercise may have only limited value in affecting bone mass in the short term and widespread recommendations for the prophylactic use of exercise should await further validation using better methodological rigor.