The relation between diet and bone mineral density in premenopausal women was evaluated in a cross-sectional study of 139 women aged 30–39 years. The population consisted of volunteers recruited in Rockland County, New York, between September 1988 and August 1992. A food frequency questionnaire was used to determine nutrient intake for both the year prior to bone density measurement and for ages 13–17 years. Physical measurements included height, weight, grip strength, and percent body fat. Bone mineral density was measured in the lumbar spine, hip, and forearm. Multiple regression equations were used to relate nutrient intake to bone density while controlling for age, height, weight, and grip strength. There were no relations between lumbar spine or distal forearm bone density and any nutrient studied from either the current or teenage diet. Current dietary calcium intake was modestly related to hip bone density (β = 0.077; p = 0.074). When fiber intake was added to the multiple regressäon model, the association between calcium and hip bone density was strengthened (β = 0101; p = 0.037); this would be expected, because fiber interferes with calcium absorption. In the teenage diet, phosphorus and calcium intake were related to hip bone density. A higher lifetime calcium intake was associated with a higher hip bone density compared with low lifetime calcium intake. An increase in teenage calcium intake from 800 to 1,200 mg per day is estimated to increase hip bone density by 6 percent.