The purpose of this study was to analyse the effect of a prior quadriceps contraction on knee flexor torque in normal subjects and multiple sclerosis patients. Twelve subjects (10 females, two males, mean age 34.8 years) with spastic paraparesis secondary to multiple sclerosis and 12 healthy adults, matched by age and sex, with no history of knee pathology participated in this study. A Cybex II isokinetic dynamometer and chart recorder were used to measure the torque and angle of peak torque produced during knee flexion at 30, 90 and 180°/s. The subjects performed two tests of knee flexion. In test one, the subjects maximally flexed their knee three times from a passively extended position. During the second test, the subjects maximally extended and then flexed their knee three times. The tests were randomly ordered and both lower extremities were tested. The subject's limbs were also passively dropped from an extended position at three preselected velocities (30, 90, 180°/s). The Cybex goniograms which were produced during passive and active movement were compared. The results of this study show that in subjects with spastic paraparesis secondary to multiple sclerosis, contraction of the quadriceps femoris muscle is a primary impediment to the development of knee flexor torque. In addition, the patients produce low torque, and restraint to active movement is greater than restraint to passive movement.