Bone mineral density is under strong genetic control and polymorphisms of the vitamin D receptor (VDR) have been suggested to account for some of the genetic variation in bone mass. However, the relationship between VDR polymorphisms and bone density is controversial and has not been confirmed by all workers. Since there is little information on the association between VDR genotype and bone mass in the UK, we studied VDR genotype, bone mineral density, and osteoporotic fracture in a cohort or pre- and postmenopausal women from the Northeast of Scotland. We found a highly significant "inverse" association between the VDR genotype and bone mineral density at the hip such that individuals of "bb" genotype had a femoral neck bone density of 0.79 standard deviation lower than individuals of BB genotype (p < 0.02). This contrasts with most previous studies in which the "bb" genotype has been associated with high bone density. A similar, but nonsignificant trend was seen for lumbar spine BMD. To study the clinical significance of this observation, we examined the distribution of VDR genotypes in a subgroup of patients with severe osteoporosis who had vertebral compression fractures (n = 44) as compared with with age-and gender-matched controls (n = 44). Despite the differences in BMD between genotypes, there was no significant excess of any specific VDR genotype in osteoporotic fracture patients, indicating that VDR genotyping may be of limited practical value in identifying patients at risk of vertebral fracture. This study confirms that there is a significant association between VDR genotype and bone mass in our population. The "inverse" relationship between VDR genotype noted in this, as compared with previous studies, would be consistent with a model whereby VDR polymorphisms are not the cause of reduced BMD, but rather, are in linkage disequilibrium with a disease-causing locus nearby.