The effect of height on nerve conduction velocity (NCV) was studied in 41 normal subjects. Peroneal and sural NCV correlated inversely with height and with estimated axonal length, whereas median motor and sensory NCV failed to show any significant relationship to height. These findings are consistent with the hypothesis of abrupt distal axonal tapering in the lower extremities. Recognition of height effect in the legs is clinically significant, since diagnostic sensitivity of NCV determinations is improved when corrected for height differences. Multiple regression equations derived from the sample data allow satisfactory estimation of peroneal NCV, sural NCV, H-reflex latency, and F-wave latency from height (or axonal length), age, and other important predictors.