This cross-sectional study examined the relationship between arm and leg muscle mass and isometric muscle strength in 465 well-functioning women and 439 well-functioning men from the NuAge cohort, age 67-84 years. Leg and arm muscle mass and body fat were measured by dual-X-ray absorptiometry. Maximum voluntary isometric strength of knee extensors and elbow flexors was measured using the belt-resisted method and a handheld dynamometer, respectively. The regression model including leg muscle mass, physical activity level, age, height, and body fat explained 14% of the variance in quadriceps strength in men and 11% in women (p < .001), whereas the model including arm muscle mass and the same covariates elucidated 40% and 28%, respectively, of the variance in biceps strength (p < .001). These results suggest that muscle mass does not play a crucial role in the variations of isometric muscle strength in well-functioning elderly.