To study the effect of calcium supplementation on perimenopausal bone loss, 295 women were randomized into a control group and 2 supplementation groups receiving, respectively, 1000 and 2000 mg elemental calcium/day for a period of 2 yr. We observed a significant decrease in lumbar bone loss in relation to the calcium supplementation (mean loss after 2 yr of 3.5% in the control group vs. 1.3% and 0.7% in the 1000 and 2000 mg groups, respectively), a significant increase in urinary calcium excretion, and a significant decrease in the urinary hydroxyproline/creatinine ratio, serum alkaline phosphatase, osteocalcin, and 1, 25-dihydroxyvitamin D. The effect of calcium supplementation on lumbar bone loss was significant in the first year of supplementation, but not in the second. However, the urinary hydroxyproline/creatinine ratio and the serum alkaline phosphatase level remained significantly decreased in the treatment groups at the end of the study; this was not the case for serum osteocalcin. Calcium supplementation did not have a significant effect on metacarpal cortical bone loss. The difference in biochemical parameters between the 2 supplementation groups was small. No significant interaction was observed between the menopausal status of the subjects and the effect of calcium supplementation. We conclude that calcium supplementation retards lumbar bone loss in the first year of calcium supplementation by reducing bone turnover. However, the effect on lumbar bone loss over a longer time span is still uncertain. © 1991 by The Endocrine Society.