At the level of prevention of bone mineral loss produced by ovariectomy, the aim of the present study was to determine the effect produced by supplementation of Ca in the diet and a moderate exercise programme (treadmill), simultaneously or separately, in ovariectomized rats, an experimental model of postmenopausal bone loss. Female Wistar rats (n 110, 15 weeks old) were divided into five groups: (1) OVX, rats ovariectomized at 15 weeks of age, fed a standard diet; (2) SHAM, rats sham operated at 15 weeks of age, fed a standard diet; (3) OVX–EX, ovariectomized rats, fed a standard diet and performing the established exercise programme; (4) OVX–Ca, ovariectomized rats fed a diet supplemented with Ca; (5) OVX–EXCa, ovariectomized rats with the exercise programme and diet supplemented with Ca. The different treatments were initiated 1 week after ovariectomy and were continued for 13 weeks for subgroup 1 and 28 weeks for subgroup 2, to look at the interaction of age and time passed from ovariectomy on the treatments. Bone mineral density (BMD) was determined, at the end of the study, in the lumbar spine (L2, L3 and L4) and in the left femur using a densitometer. Bone turnover was also estimated at the end of the study, measuring the serum formation marker total alkaline phosphatase (AP) and the resorption marker serum tartrate-resistant acid phosphatase (TRAP). As expected, OVX rats showed a significant decrease (P<0·05) in BMD, more pronounced in subgroup 2, and a significant increase in AP and TRAP with regard to their respective SHAM group. The simultaneous treatment with Ca and exercise produced the best effects on lumbar and femoral BMD of ovariectomized rats, partially avoiding bone loss produced by ovariectomy, although it was not able to fully maintain BMD levels of intact animals. This combined treatment produced a significant increase in AP, both in subgroups 1 and 2, and a decrease in TRAP in subgroup 1, with regard to OVX group. The exercise treatment alone was able to produce an increase in BMD with regard to OVX group only in subgroup 1 of rats (younger animals and less time from ovariectomy), but not in subgroup 2. In agreement with this, there was an increase of AP in both subgroups, lower than that observed in animals submitted to exercise plus Ca supplement, and a decrease of TRAP in subgroup 1, without significant changes in this marker in the older rats. Ca treatment did not produce any significant effect on BMD in OVX rats in both subgroups of animals, showing a decrease of AP and TRAP levels in the younger animals with no significant variations in markers of bone remodelling in the older female rats compared with their respective OVX group.