Isokinetic muscle testing has become an increasingly popular method for measuring muscle performance. However, the relationship of isokinetic strength measurements to static strength measurements remains unclear. The relationship between maximal isokinetic and maximal isometric torque was investigated at angular velocities of 24, 48, 96, and 192 degrees/s for both flexion and extension of the elbow joint. Twenty-four subjects were tested bilaterally, and discrete torque values at the positions of 60 and 90 degrees of flexion were obtained from the continuous isokinetic records for comparison with isometric torques recorded at the same two positions. Data were analyzed using an analysis of covariance with isokinetic torque, the dependent variable, expressed as a function of isometric torque and as a function of the design variables side and position. The results demonstrated significant effects of side and position and also demonstrated the dependence of isokinetic strength on velocity. Confidence intervals were determined for the prediction of isokinetic torque from observed isometric torque for normal individuals, providing a potentially useful tool to aid in the diagnosis of muscular disorders.