This study examined the validity of an anthropometric estimate of thigh muscle cross-sectional area using magnetic resonance imaging (MRI). The anthropometric model assumed that a cross section of the thigh could be represented as a circle with concentric circular layers of fat-plus-skin, muscle, and bone tissue. On 18 healthy, active men and women (mean +/- SD age = 23 +/- 5 yr), total thigh circumference (CT) was measured with a fiberglass tape, fat-plus-skin thickness was measured over the quadriceps (SQ) using calipers, and the distance across the medial and lateral femoral epicondyle (dE) was measured with calipers. Direct measurements of each tissue were obtained by planimetry of an MRI image taken at the same site as the circumference and skinfolds. Thigh muscle cross-sectional area (AM) was estimated as follows: [equation: see text] Mean +/- SD AM from MRI and anthropometry were 121.9 +/- 35.1 cm2 and 149.1 +/- 34.1 cm2 (r = 0.96, SEE = 10.1 cm2), respectively. Errors in the anthropometric approximations of AM were due to an overestimate of the total thigh cross-sectional area and an underestimate of fat-plus-skin compartment. Because of the close relationship between MRI and anthropometric estimates of AM, zero-intercept regression was used to produce the following final equation, applicable for use in populations studies of young, healthy, active men and women: [equation: see text]