The TECHNIQUE of determining motor conduction velocities in human nerves is now extensively used in the investigation of neuromuscular disorders. The standard method requires a supramaximal stimulus for motor fibers in order to obtain a reproducible conduction time which is as short as possible. This method, therefore, measures the conduction velocity in only the fastest motor fibers and gives no information concerning slower conducting fibers. Although this technique is helpful is diagnosing peripheral nerve dysfunction even in patients without signs and symptoms, there are some patients who have peripheral nerve symptoms without a reduction in the speed of motor nerve conduction.1-5 It is known from studies in experimental animals that the speed of conduction determined in whole nerve trunks need not represent the rate in the fastest single fibers due to shunting and dispersion of the activity in these few large fibers.6 Therefore, additional techniques have been