Exercise interventions are known to stimulate bone mass acquisition in childhood. However, it is unclear whether low levels of physical activity to which children in the UK are generally exposed also affect bone development. Whether physical activity affects biomechanical strength of the skeleton, thereby reducing the risk of fractures in later life, is also uncertain. We recently found a positive relationship between habitual levels of physical activity as assessed by accelerometer recording, and total body bone mineral content as measured by DXA, in 4457 11-year-old children from the unique Avon Longitudinal Study of Parents and Children birth cohort. Equivalent associations were observed for height-adjusted bone area, suggesting that physical activity may affect cortical geometry and hence biomechanical strength of the skeleton. To test this hypothesis, we will examine prospectively the relationship between physical activity at age 11, 13 and 15, and cortical geometry parameters derived from pQCT measurements of the tibia at age 15. We will also examine interactions between physical activity, cortical geometry, fat mass and lean mass, based on our recent observation that the relationship between physical activity and total body DXA parameters is modified by body composition.