Elderly women are at increased risk for bone loss and fractures. In previous cross-sectional and longitudinal studies of women residing in northern latitudes, bone loss was most pronounced during winter months and in those consuming less than 1 g calcium per day. In this study we sought to test the hypothesis that calcium supplementation by either calcium carbonate or dietary means would prevent seasonal bone loss and preserve bone mass. Sixty older postmenopausal women without osteoporosis were randomized to one of three treatment arms: Dietary milk supplementation (D-4 glasses of milk/day), Calcium carbonate (CaCO3-1000 mg/day in two divided doses), or placebo (P). After 2 yr, placebo-treated women consumed a mean of 683 mg/day of calcium and lost 3.0% of their greater trochanteric (GT) bone mineral density (BMD) (P < 0.03 vs. baseline); Dietary supplemented women averaged a calcium intake of 1028 mg/day and sustained minimal loss from the GT (-1.5%; P = 0.30), whereas CaCO3-treated women (total Ca intake, 1633 mg/day) suffered no bone loss from the GT and showed a significant increase in spinal and femoral neck BMD (P < 0.05). Femoral bone loss occurred exclusively during the two winters of the study (i.e. total loss, -3.2%; P < 0.02 in placebo-treated women) with virtually no change in GT BMD during summer. Serum 25-OH vitamin D declined by more than 20% (P < 0.001) in all groups during the winter months but returned to baseline in summer; PTH levels rose approximately 20% (P < 0.001) during winter but did not return to baseline during the summers. Urine N-telopeptide and osteocalcin levels increased significantly but only in the P-treated women and only during winter. Serum insulin growth factor binding protein 4, an inhibitory insulin growth factor binding protein, rose 15% (P < 0.03) from summer to winter, but this increase was significant only in those women consuming <1000 mg/day of calcium. By multivariate analysis, total calcium intake was the strongest predictor of bone loss from the hip. Urinary N-telopeptide also closely correlated with GT BMD but only during winter (P = 0.003). We conclude that calcium supplementation prevents bone loss in elderly women by suppressing bone turnover during the winter when serum 25-OH vitamin D declines and serum PTH increases. The precise amount of calcium necessary to preserve BMD in elderly women requires further studies, although in this study, at least 1000 mg/day of supplemental calcium was adequate prophylaxis against femoral bone loss.