The etiology of age-related bone loss is unclear but both lack of exercise and dietary calcium deficiency have been implicated in its causation. This 2-year randomized placebo-controlled study was designed to examine the effects of increased dietary calcium and exercise in 168 women who were more than 10 years postmenopausal. The subjects were randomized into one of 4 groups: placebo, milk powder containing 1 g of calcium, calcium tablets 1 g/night, and calcium tablets 1 g/night and an exercise regimen. The exercise group aimed to undertake 4 h of extra weight-bearing exercise per week and were undertaking 10% more activity than other groups at 2 years. Bone mineral density at the lumbar spine, three hip sites, and two sites of the tibia close to the ankle joint were measured at 6 month intervals. Dietary intake was evaluated by a weighed food record, exercise was evaluated by an exercise diary, and blood and urine samples were obtained to examine effects on calcium homeostasis. Individual data points were compared using repeated measures ANOVA and least squares regression. Calcium supplementation by either the calcium tablets or the milk powder resulted in cessation of bone loss at the intertrochanteric hip site (placebo, calcium tablets, calcium and exercise, milk powder -0.81, +0.17, +0.23, and +0.07% per year, respectively; p < 0.05 for all supplementation groups compared with placebo) with similar results at the trochanteric hip site. The calcium and exercise group had less bone loss at the femoral neck site when compared with calcium supplementation alone (placebo, calcium tablets, calcium and exercise, milk powder -0.67, -0.18, +0.28, and -0.18% per year, respectively; p < 0.05 for calcium and exercise compared with calcium alone). There was a significant reduction in the rate of bone loss at the ultradistal site of the tibia (placebo, calcium tablets, calcium and exercise, milk powder -2.5, -1.6, -1.0, and -1.5% per year, respectively; p < 0.05 for all supplementation groups compared with placebo). There was no significant bone loss at the spine site in any group.(ABSTRACT TRUNCATED AT 250 WORDS)