Pre- and early puberty seem to be the most opportune times for exercise to improve bone strength in girls, but few studies have addressed this issue in boys. This study investigated the site-, surface-, and maturity-specific exercise-induced changes in bone mass and geometry in young boys. The osteogenic effects of loading were analyzed by comparing the playing and nonplaying humeri of 43 male pre-, peri-, and postpubertal competitive tennis players 10-19 yr of age. Total bone area, medullary area, and cortical area were determined at the mid (40-50%) and distal humerus (60-70%) of both arms using MRI. Humeral bone mass (BMC) was derived from a whole body DXA scan. In prepubertal boys, BMC was 17% greater in the playing compared with nonplaying arm (p < 0.001), which was accompanied by a 12-21% greater cortical area, because of greater periosteal expansion than medullary expansion at the midhumerus and periosteal expansion associated with medullary contraction at the distal humerus. Compared with prepuberty, the side-to-side differences in BMC (27%) and cortical area (20-33%) were greater in peripuberty (p < 0.01). No differences were found between peri- and postpuberty despite longer playing history in the postpubertal players. The osteogenic response to loading was greater in peri- compared with prepubertal boys, which is in contrast with our previous findings in girls and may be caused by differences in training history. This suggests that the window of opportunity to improve bone mass and size through exercise may be longer in boys than in girls.