We have compared the measurements obtained by different methods: dual energy X-ray absorptiometry (DXA), histomorphometry, ash weight, and two morphometric indices (robusticity and bone weight/bone length index) in the orchidectomized (ORX) rat model of male osteoporosis. We examined 144 male wistar rats: 48 sham-operated, 48 ORX, and 48 ORX-treated with a bisphosphonate (risedronate) 2 or 10 mg/kg/day, 5 days per week. Rats were sacrificed at 2, 4, 8, or 16 weeks after the beginning of the study. DXA was performed on a Hologic QDR 2000 on the whole body, whole tibia, and tibial metaphysis. Bone volumes (C.BV/C.TV, and BV/TV) were measured by histomorphometry on the proximal tibial. A significant correlation was obtained between weight measured by DXA and scale (r = 0.993, P <0.000001). However, DXA underestimated weight by 0.3%. This discrepancy was dependent on the rat's weight. The weight bone length (WL) index was linearly correlated with BMD (r = 0.86), BMC (r = 0.96), and ash weight (r = 0.97). Correlation with robusticity was lower than with the WL index. A significant correlation was found between BMC of the metaphyseal region and the bone volumes but this explained only 27% of the variance; correlation with BMD was poorer (r = 0.40). BMC and ash weight were highly correlated (r = 0.992, P <0.000001). However, DXA overestimated BMC by 11% and the overestimation was found to be clearly dependent on the net mineral content of the bone.