The Hoffmann (H) reflex was measured from a grid of 31 locations on the calves of 20 healthy volunteers. In 17 cases the maximum amplitude of response was found at a site more distal than at the conventional midcalf recording location. Three quarters of the distance between the popliteal crease and the flare of the medial malleolus, the waveform was found most consistently triphasic. At this site the latency was, on the average, 1.4msec longer than at the midcalf location. These results may be related to the more distal origin of the H-reflex dipole when compared to the M (motor) wave or T (tendon) reflex response, and indicate that a distal recording site is preferred if the H-reflex per se is to be assessed.