The effect of exercise on castration-induced osteoporosis in 3-month-old male rats weighing 264 +/- 4 g at the beginning of the experiment was studied. A testosterone deficiency was induced by orchidectomy (ORC), and the exercise group ran 10 m/minute for 1 hour a day on a treadmill at 0% grade. There were seven groups of eight rats (n = 56) randomized into a control group killed at time 0, and sham, ORC and ORC and exercise groups killed at 4 and 8 weeks. ORC reduced body weight gain (with analysis of variance (ANOVA) P < 0.001), and at 4 weeks the body weight was 343 +/- 14 g in ORC group and 301 +/- 4 g in the ORC and exercise group (P < 0.01). The increase in femoral length was slower in the ORC+exercise groups. The ash weight of the tibia did not decrease significantly after ORC or ORC+exercise. ORC did not affect 45Ca incorporation, but exercise slightly increased it in the whole tibia 8 weeks after ORC (with ANOVA P = 0.057). ORC had significantly lowered the trabecular bone volume in the secondary spongiosa of the distal femur at 4 and 8 weeks, and exercise did not prevent this. This is an opposite finding to our previous study with ovariectomized female rats [12]. ORC also significantly had reduced the osteoblast-lined trabecular bone surface and the number of osteoclasts by 8 weeks after the operation. Exercise increased the osteoblast-lined surface and the number of osteoclasts.(ABSTRACT TRUNCATED AT 250 WORDS)