The purpose of this study is to investigate the relationship between joint angle and EMG activity concerning the human knee joint and hamstring muscles. Ten healthy males participated in this study. They performed maximum voluntary isokinetic knee flexion in the flexion angle range of 0° to 120° in a prone position on a table. The EMG activities of the semitendinosus, semimembranosus, long and short heads of the biceps femoris muscles were detected by bipolar fine wire electrode, and were integrated at knee flexion range intervals of 15° from 0° to 120°. After three months, the same examination was performed with the same subjects to confirm the reliability of this study. Results showed the mean peak isokinetic torque was attained from a 15° to 45° knee flexion angle. The EMG activities of the hamstring muscles varied with changes in the knee flexion angle. The integrated EMG of the semitendinosus, semimembranosus and short head of the biceps femoris muscles in-creased gradually as the knee flexion angle was increased from 0° to 105°. The peak integrated EMG of these three muscles was attained between 90° and 105° knee flexion in both trials. On the other hand, the peak integrated EMG of the long head of the biceps femoris muscle was largest at a knee angle ranging from 15° to 30°. Peak integrated EMG decreased when the knee flexion angle increased in both trials. These results show that the electromyographic activity of agonist muscles during maximum voluntary movement varied with the change of muscle length or joint angle. Therefore, it was considered that the relationship between joint torque and joint angle is influenced not only by the sarcomere length and the moment arm but also muscle activities of agonist muscles, even if the movement is performed with maximum effort.