We compared three different methods of measuring bone mass for their diagnostic value and their usefulness in follow-up measurements. The three methods were: measurement of (1) bone mineral content in the distal forearm by single photon (125I) absorptiometry, (2) bone mineral content and bone mineral density of the lumbar spine measured by dual photon (153Gd) absorptiometry, and (3) total body bone mineral and total body bone density also measured by dual photon (153Gd) absorptiometry. The diagnostic validity was evaluated from measurements on healthy premenopausal women, and three groups of postmenopausal osteoporotic women (prior forearm fracture (N = 45), prior spine fracture (N = 46), or prior hip fracture (N = 27]. The forearm measurement separated all three osteoporotic groups from the premenopausal women at least as well as the spine measurement. The value of follow-up procedures was estimated using data from a clinical trial on prevention of postmenopausal bone loss by sex hormones. Fewer participants are needed in clinical trials when a forearm scanner is used instead of a spine scanner, because of the better precision of the former. The forearm scanner seems to be the best tool of the three for bone mass measurements, in both clinical practice and for research purposes.