We addressed whether muscle quality (force per unit muscle mass) changes with age in cross-sectional and longitudinal analyses from three groups from the Baltimore Longitudinal Study of Aging: (1) Isometric arm strength studied cross-sectionally in 617 subjects with muscle mass estimated by cross-sectional area (CSA) from arm circumference and by 24-hour urinary creatinine excretion (CREAT); (2) longitudinal study for 10 to 25 years in 412 men using the same measures as the first group; and (3) isometric knee extensor strength studied cross-sectionally in 675 subjects; muscle mass estimated by CREAT, CSA from thigh circumference, and leg nonosseous fat free mass (FFM) from dual energy x-ray absorptiometry. Muscle quality declined in both arm and leg with age in cross-sectional analyses using CSA and FFM, but not CREAT. No age-associated arm muscle quality declines were observed longitudinally using CREAT or CSA. The relationship between muscle quality and age is dependent on how muscle mass is estimated and on whether subjects are studied cross-sectionally or longitudinally. In addition, CREAT may measure a muscle property not accounted for by CSA or FFM.