Diet and exercise are two manageable factors that affect bone remodeling. Experiments were designed to test the hypothesis that bone density and calcium status would be affected by conditioning and deconditioning, and by dietary Ca and cationanion difference (DCAD). In experiment one, 12 Arabian horses were conditioned for 12 wk on a high speed equine treadmill. Diets were designated LH, which contained low Ca (.35%)-high Cl (.6%), LL, low Ca-low Cl (no added Cl), HH, high Ca (.7%)- high Cl, and HL, high Ca-low Cl. Data were collected while horses were at rest every 21 d. Serum and plasma were analyzed for total and ionized Ca, P, Mg, Cl, total protein, albumin, parathyroid hormone (PTH), osteocalcin, and hydroxyproline. Radiographs of the left third metacarpal bone were taken. An aluminum step wedge exposed simultaneously was used as a reference standard for estimating bone mineral content (BMC) with an image analysis program. Bone measurements, including BMC, bone and medullary width, and cortical area, were taken in the proximal diaphyseal and proximal metaphyseal area. Bone variables increased with training and with high Ca diets as compared to horses fed low Ca diets. Serum Ca decreased and serum PTH increased with training. Hydroxyproline was unchanged in horses fed the low Ca diets and decreased in horses fed the high Ca diets. Serum osteocalcin increased then decreased with training. The high Ca diet appeared to facilitate bone remodeling in response to training, but dietary Cl, hence DCAD, had no effect. In experiment two, 11 conditioned Arabian horses were taken out of training and placed in stalls for 12 wk. They were walked on a mechanical walker in two 30 min sessions 7 d/wk. Diets were designated LC (.35% Ca) and HC (.7% Ca). Data collection and analysis were identical to those in experiment one. Bone variables decreased with deconditioning but were unaffected by dietary Ca. Serum Ca increased with deconditioning and was greater in horses fed the HC diet, but PTH did not change. Horses fed the HC diet had greater serum osteocalcin, which decreased with deconditioning. Dietary Ca influenced bone remodeling in response to training, but did not have an effect on bone response to deconditioning. Loss of BMC during 12 wk of stall confinement may predispose horses to skeletal injuries when training is resumed.