A group of 68 premenopausal women participated in a controlled 12 month exercise program. Two groups were matched according to age, body size (body mass index), and typical activity level. Data collection included bone mineral density (BMD) of the lumbar spine with dual-photon absorptiometry and of the os calcis with single-photon absorptiometry, lean body mass, urinary calcium/creatinine, and urinary gamma-carboxyglutamic acid (Gla). Subjects were given a daily 500 mg supplement of elemental calcium. There was no significant difference between groups in terms of diet, in urinary calcium/creatinine or Gla, or in lean body mass. The weight lifting group had a nonsignificant increase in mean lumbar BMD of 0.81% and the control group exhibited a nonsignificant decrease of 0.5%. However, a paired t-test revealed a significant change in the means in either group or as matched pairs. The relatively small change seen as a result of this modified Nautilus exercise program may prevent moderate weight lifting from being a practical answer for osteoporosis, even in a highly motivated population.