Evidence that birth weight is related to bone mass in later life suggests that the intrauterine environment programs the trajectory of subsequent bone development. To explore this hypothesis, we examined whether maternal diet in pregnancy, as assessed by the maternal food frequency questionnaire (FFQ) completed at 32 weeks gestation, is related to bone mass of the child, as measured by total body DXA carried out at age 9 years in the Avon Longitudinal Study of Parents and Children (ALSPAC). Diet records were linked to DXA scan results for the total body and spine sub-region and pooled between pre- and early pubertal boys and girls (n=4,451). Regression analysis was carried out between DXA values and dietary factors following adjustment for social and other confounding factors. Maternal magnesium intake was related to total body BMC (beta=4.9, 7.4-23.1; g) and BMD (beta=4.9, 2.5-7.3; g/cm2 x10(3)) (standardized regression coefficient with 95% confidence limits; P<0.001). An equivalent relationship was no longer observed after adjusting for the height of the child, to which magnesium intake was also related (beta=0.48, 0.20-0.77; cm; P=0.001). Maternal intake of potassium was related to spinal BMC (beta=1.8, 0.8-2.9; g) and BMD (beta=10.5, 4.9-16.0; g/cm2 x10(3)) (P=0.001), which was no longer observed after adjusting for the weight of the child, to which potassium intake was also related (beta=0.52, 0.16-0.88, P=0.005; kg). A significant association was also observed between maternal dietary folate intake and spinal BMC adjusted for bone area using a linear regression model (beta=0.55, 0.16-0.94; g; P=0.006), which persisted after adjusting for height and weight. Our observation that constituents of maternal diet are related to DXA measures at age 9 is consistent with the hypothesis that the trajectory of bone development in childhood is programmed by early life factors.