The use of isokinetic measurements for research and clinical practice becomes more meaningful when it can be demonstrated that subjects being tested at multiple velocities are making similar efforts and when the relationships between various biomechanical measures are understood. The purpose of this study was to examine 1) integrated electromyographic activity per second of the quadriceps femoris muscle at four isokinetic speeds (30, 60, 90, and 120 degrees/sec) and 2) the relationship of power to peak torque at each speed. Two groups were tested: "normals" (healthy subjects) and patients undergoing long-term steroid therapy for rheumatic diseases. The integrated electromyographic activity per second did not vary significantly across speeds, indicating that subjects in each group made equivalent efforts at all four speeds. At each speed, for each group, high correlations were found between peak torque and power. Population-specific and speed-specific linear regression equations were calculated that allowed for predictions of power from the more easily obtained clinical measure of peak torque.