Vertebral and peripheral bone mass have been measured with single and dual photon absorptiometry and radiogrammetry in 146 male and 220 female volunteers ranging in age from 20 to 85 yr. One hundred four subjects with interfering diseases, treatment, or x-ray manifestations of lumbar osteoarthritis were excluded for purposes of this study. Patterns of age-related bone gain and diminution differed between sexes and measuring sites. The effect of menopause on the peripheral and vertebral skeleton also differed. Men, at all measured sites, have more bone than women. In the fifth decade, however, women's lumbar bone mineral content was almost equal to the value found in men. Bone loss associated with aging was more marked in women than in men and started, for the lumbar spine, at about the age of 25 yr in both women and men and, for the peripheral bones, at the age of 55 in women and 65 in men. Bone loss in the spine in women was not linear. Women in the fifth and sixth decade, who still had menstruation, differed significantly from those who had not menstruated for at least the last 6 mo. Bone diminution at menopause was twice as great in the lumbar spine than elsewhere in the peripheral skeleton, 15% versus 7%. Of the 25% total bone loss of the spine during adult life in women, 60% was lost within 10 yr after menopause. Estrogen deficiency, not aging, is the predominant cause of bone loss in the spine. For the peripheral skeleton, there is a two-component decrease, a rapid loss induced by the menopause superimposed on a slower age-related loss. Although there was a significant correlation between peripheral and vertebral bone mass indices, it was clear that observations made at one site will not necessarily reflect changes observed at another site.