To determine the effects of a tower climbing exercise on mass, strength, and local turnover of bone, 70 9-wk-old Sprague-Dawley rats were assigned to seven groups: a baseline control and three groups of sham-operated sedentary, orchidectomized (ORX)-sedentary and ORX-exercise rats. Rats voluntarily climbed a 200-cm tower to drink water from a bottle set at the top. At 4 wk, the periosteal bone formation rate (BFR), moment of inertia, bone mineral content, bone mineral density, and bending load at the midfemur were maintained in ORX-exercise rats, whereas these parameters were reduced in ORX-sedentary rats. At 8 wk, the periosteal mineral apposition rate and BFR in ORX-exercise rats were significantly higher, whereas the parameters in ORX-sedentary rats did not differ compared with sham-sedentary rats. In ORX-exercise rats, the trabecular mineralizing surface, BFR, and bone volume of the lumbar vertebrae were maintained at the same levels as those in the sham-sedentary group, whereas the osteoclast surface decreased compared with the ORX-sedentary group. However, the climbing exercise did not affect bone mineral content, bone mineral density, or the compression load of the lumbar vertebrae. These results show that, in the midfemur, the voluntary climbing exercise maintained cortical bone mass and strength by stimulating periosteal bone formation and partially prevented ORX-induced trabecular bone loss, depressing the elevation of turnover. Interestingly, in ORX rats, the climbing exercise had the opposite effect on bone formation at the periosteal femoral cortical bone, where the exercise increased the bone formation compared with vertebral trabecular bone, where the exercise decreased it.