Most previous literature on muscle strength and bone density examined the association between specific muscle groups and adjacent bones. This study examined the association of grip strength with bone density at distant sites, such as the spine and hip, as well as at the wrist and radius. Subjects were 649 postmenopausal women aged 65 years and older. Bone mineral density was measured at the ultradistal wrist and midshaft radius in the nondominant arm using single-photon absorptiometry and at the lumbar spine and hip using dual-energy x-ray absorptiometry. Grip strength was measured in both the dominant and nondominant hands using a dynamometer. Overweight women had significantly stronger grip strength in both hands (p < 0.01 for dominant hand, p < 0.05 for nondominant hand grip strengths). Exercises had significantly stronger grip strength in the nondominant hand than nonexercisers (p < 0.05). Women reporting arthritis had significantly weaker grip strength in both hands than women not reporting arthritis (p < 0.001). There were no significant differences in grip strength by estrogen use, cigarette smoking, or thiazide use. Regression analyses adjusted for age, obesity, exercise, cigarette smoking, thiazide use, arthritis, years postmenopausal, and estrogen use indicated significant positive associations between grip strength and bone density at all sites. Regression analyses stratified by exercise indicated that only women who exercised had significant associations of grip strength with bone mineral density (all p < 0.05). In this cohort of older women, grip strength was an independent indicator of general bone density.(ABSTRACT TRUNCATED AT 250 WORDS)