There is general agreement that bone density falls with age and is higher in heavy people than light people. We have studied a variety of potential correlates of vertebral, ankle, and hip bone density to evaluate other potential influences on the skeleton. We recruited 196 healthy women who were more than 10 years past the menopause and collected a diet and activity record, a 24 h urine, and a fasting blood and urine specimen. These blood and urine samples were analyzed for factors related to calcium homeostasis. We then measured bone density at lumbar vertebrae 1-4 and the hip and the ankle bone density of the nondominant leg. Correlations between vertebral, hip, and ankle bone density and other measured variables were explored using the statistical package SPSS PC. At the vertebral site, in addition to correlations with age and body mass index (BMI), a negative correlation with a measure of bone resorption, the hydroxyproline creatinine ratio (OHPCR), was noted. At the ankle site, in addition to correlation with age, BMI, and OHPCR, a positive correlation with activity and a negative correlation with serum calcitriol were noted. At the hip site, as well as age, BMI, and OHPCR, significant correlations with GFR and dietary calcium intake were noted. These data suggest that even in women 10 years past the menopause bone resorption has a significant effect on bone density, that renal function may account for some of the variance in bone density at the hip, and that activity effects are more marked at sites of greater loading, namely the ankle.