With the aim of studying peripheral physiological mechanisms involved in transcutaneous electrical nerve stimulation (TENS) for the suppression of pain, the authors have examined 14 healthy volunteers in the perceptual part and 8 in the microneurographic part of this study. For pain suppressing stimulation they have used ring electrodes and stimulators capable of producing rectangular, sine wave, triangular and hybrid wave forms. Pain was induced with electrical stimuli on the distal phalanx of the middle finger. TENS with rectangular, sine wave and triangular pulses activates mainly A beta, but also some A delta fibres. Pain suppressing capacity of different wave forms used was the same and was achieved with stimuli close to the threshold for pain. Pain was most effectively diminished when noxious and pain suppressing stimuli were applied to the same finger; this effect was smaller with the two stimuli on neighbouring fingers and very weak when pain suppressing stimulus was applied to the contralateral hand. Increases in stimulus frequency resulted in an increased and variable latency as well as intermittent blocking of those spike components which had been near threshold at low frequency stimulation. This interpreted as being due to local changes in excitation threshold. In surface-recorded averaged electroneurograms this effect was seen as amplitude decrement and increased response duration, but was not due to impulse transmission failure in A delta fibres as seen microneurographically. On continuous high frequency stimulation, the intensity of sensation elicited diminished, although there was no change in the microneurogram. The authors conclude that the peripheral contribution to the analgesic effects of TENS seems to be unimportant.