Sahrmann and Norton [ 51 have questioned the prevalent belief that hyperactive stretch reflexes are directly related to impairment of movement in patients with the upper motor neuron syndrome: "though the premise is widely accepted that spasticity limits voluntary movement, there are no critical data to test its validity." We have recently examined a hemiparetic man whose clinical presentation and electromyographic findings provide some relevant data.A 46-year-old man had a four-year history of right hemiparesis due to a left hemisphere oligodendroglioma. H e complained that he could not use a hammer to drive nailsor to flatten objects such as aluminum cans. He had moderate weakness of the right upper limb, more pronounced distally. Tendon jerks were hyperactive on the right side. Slow movements of the elbow were done without difficulty, but rapid extensions were checked abruptly en route to the desired position. When the patient tried to strike with a hammer, movement was arrested about halfway to the target and then completed in a series of short, jerky movements.We surmised that his inability to extend the forearm rapidly was caused by hyperactivity of the stretch reflex in the elbow flexors. To test this possibility, EMG tracings were obtained from the right biceps and triceps during movement of the elbow at various rates. Action potentials, detected by means of Beckman skin electrodes, were fullwave rectified and processed by an averaging filter [21; the apparatus used to record position and velocity of the hand has been described elsewhere [l]. Signals representing the two EMGs, as well as the kinetic variables, were recorded with an Ampex tape recorder, then played back at reduced tape speed and displayed by means of a Grass polygraph.The patient was asked to extend his elbow repeatedly at various speeds, ranging from the slowest to the fastest possible. Slow movements were completed without difficulty, but faster ones were interrupted in midcourse, as seen in Figure 1. This sudden arrest of movement after about 0.25 second cannot be attributed to weakness of the triceps muscle, but rather to an excessive braking force exerted by the flexor group. After the initial arrest, extension was completed in a series of smaller accelerations, shown by rise and fall of the velocity curve.Since only the faster movements were arrested in this way, we suspected that the phenomenon was related to the rate sensitivity of antagonist muscle spindles [4]. Accordingly, the patient's right elbow was subjected to a series of 4 0 passive extensions at various rates. For each movement, the peak of the filtered EMG activity was measured as millimeters of deflection on the strip-chart recording and compared with the maximal hand velocity (Fig 2 ) . Statistical analysis showed that the two variables were positively related, the coefficient of correlation being 0.77 (Fig 3). This means that the faster the elbow was extended, the more powerfully the biceps contracted. At the highest speeds, flexor activity was strong enough to stop t...