SummaryElectrical stimulation through an acupuncture needle in a Chinese point of the human hand decreased the subjectively estimated sensation elicited by noxious and by tactile stimulation of the ipsilateral finger. A hyperbolic relationship was found between the stimulus duration (T, ses) required for the initiation of hypesthesia and the intensity of current (I, mA) passed through the acupuncture needle: I=0.11+18/T Electroacupuncture, a pain relieving technique, utilizes electrical pulses for stimulating various loci through needle electrodes. It produces hypalgesia or a rise in pain threshold within a circumscribed area, this effect outlasting for some minutes the stimulation Cl, 9, 12, 14) .To provide a modern basis for acupuncture analgesia, evidences based mainly on neurophysiological (2, 5, 10, 15) , humoral (11,14,16) and psychophysical (3,4,6,8) investigations have been presented. However, little information is available concerning the stimulus factors, such as strength and duration, to effect the sensory depression.We have investigated the relationship in man between these parameters, in regard to depression of the sensation of touch-hypesthesia and experimental painhypalgesia.The eyperiments were performed on 13 male subjects 18-49 years old. A sterilized acupuncture needle was inserted into 'Ho-Ku' point located between metacarpal bones I and II of the right hand. The stimulator employed for electroacupuncture had constant voltage output with the current variable from 0 to 2 mA by means of a variable series resistor between output and needle. It could generate, at a rate of 2/3 sec, a negative going square wave of 1 sec in duration on which brief positive going pulses of 667 Hz were superimposed, the polarity referring to the needle electrode. The indifferent electrode was a metal plate placed on the ipsilateral hand plam. To elicit tactile sensation the right index finger was tapped with a small hair bundle (1. 5 mm2, in tip dimension) driven by an electromagnet. For experimental pain, the nail bed of the same finger was stimulated by a steel needle which was driven by anther electromagnet. These mechanical stimulations were applied every 2-3 sec before, during and after acupunctural sitmulation. All the subjects were instructed to compare the subjectively estimated magnitude of each experimental sensation with that of the immediately preceding one. They had to indicate their estimates by a set of 3 buttons connected to different potential outputs; the output level 3 (vertical scale in Fig. 1) was used for the