Introduction: Several studies have analyzed the relationship between physical activity and bone density. However, the prescription of exercise is not entirely clear as to the type, quantity and intensity. Objective: The objective of this study was to determine if there is a relationship between the amount of exercise and changes in bone mineral density. Methods: Fifty-two women, members of the Municipal Program of Physical Activity for Seniors, voluntarily underwent two ultrasonographies of the calcaneus within a 6-month interval. During this period, all physical activity was recorded. Afterwards, a lineal correlation study was carried out between the amount of exercise and bone changes, expressed as T-Score variation, first in total number of participants and then in groups. Considering the average body weight obtained for all women, two groups were created ("light" < 69 kg and "heavy" > 69 kg). Later, women who had participated in less than 72% of the targeted program were excluded from both groups, and the differences between the groups "light and trained" and "heavy and trained" were analyzed. To do so, the nonparametric Mann-Whitney U test was used. Results: A significant relationship of r= -0.59 was found between the total amount of exercise and the T-Score variation in the group of women above 69 kg. Significant differences were found between the "light and trained" group and the "heavy and trained" group with respect to the variation of T-Score. Conclusion: The effect of exercise on bone mineral density is determined, somehow, by body weight. This interaction is due, possibly, to mechanical demands difference.