Retrospective studies have suggested that the prepubertal years may be an important window of opportunity to increase bone mass, but there have been few direct studies and little exploration of gender differences in this age group. In this study, we report the associations among physical activity measures, sunlight exposure, body composition, and bone mass in 8-yr-old children. We studied 330 children in 1996 (115 girls and 215 boys; response rate, 60%) who had previously taken part in a cohort study of cot death in 1988. They had measurement of anthropometrics (height, weight, and body composition), sunlight exposure (by questionnaire), and physical activity [questionnaire, muscle strength by dynamometry, and bicycle ergometric physical work capacity at a pulse of 170 beats/min (PWC170)]. Bone mineral density (BMD) was assessed at the spine, femoral neck, and total body by a Hologic QDR2000 densitometer. In females only, PWC170 [hip, 2.4%/quartile (95% confidence interval (CI), 0.3-4.5); spine, 1.7%/quartile (95% CI, 0.0-3.4); size adjusted] was associated with BMD, whereas in males only, BMD was associated with both sports participation (hip, 4.2% (95% CI, 1.1-7.3); spine, 4.3% (95% CI, 0.9-7.7)] and muscle strength [hip, 1.7%/quartile (95% CI, 0.0-3.4); but not spine; size adjusted]. Winter sunlight exposure was associated with BMD in girls [hip, 2.9%/category (95% CI, 0.7-5.0); spine, 3.6%/category (95% CI, 1.4-5.8)], but not in boys [hip, 0.3%/category (95% CI, -1.4 to +2.0); spine, 1.4%/category (95% CI, -0.7 to +3.5)]. Males and females were very similar in body size. However, males had higher size-adjusted BMD at the hip (9.6%; 95% CI, 6.9-14), whereas females had higher size-adjusted BMD at the spine (3.2%; 95% CI, 0.8-5.6%). In conclusion, this study has suggested that physical activity and exposure to sunlight are important in the bone mineralization of prepubertal male and female children. The magnitude of both gender and environmental differences in bone mass in this age group is substantial, suggesting that modification at this stage of life may influence peak bone mass and possibly fracture risk in later life.