It was the object of this study to measure the time course of the action potential in individual human sensory nerve fibres in relation to conduction properties of the axons. For this purpose, the technique of percutaneous microneurography was combined with intradermal electrical stimulation of distal portions of the axons. Recordings were made at the wrist level from 57 type-identified mechanoreceptive median nerve afferents [mainly rapidly adapting (RA) and slowly adapting type I (SAI)] innervating the glabrous skin of the hand. Measurements were made of the duration and time-to-peak of the positive peak of the diphasic (large positive phase followed by smaller, slower negative phase) action potential typically recorded using microneurography. Durations ranged from 0.31 to 0.75 ms (mean 0.50 ms) and times-to-peak from 0.12 to 0.45 ms (mean 0.21 ms), with no difference between afferent categories (RA, SAI). Time-to-peak was strongly positively correlated with duration (linear r = 0. 81). Conduction velocity was measured over the distance extending from the point of intradermal stimulation (typically in the fingertips) to the point of recording at the wrist (distal conduction velocity). Absolute refractory period was measured using paired stimuli applied at the point of intradermal stimulation, within the receptive field of the afferent (distal absolute refractory period). Distal conduction velocities ranged from 15 to 60 m/s (mean 33 m/s), and distal refractory periods from 0.7 to 4.5 ms (mean 2.1 ms), with no difference between afferent types (RA, SAI). Distal absolute refractory period was inversely correlated with distal conduction velocity. The data were slightly better described assuming a non-linear (exponential) relationship; the non-linear correlation coefficient was −0.77. The time course of the action potential varied inversely with distal conduction velocity and directly with distal absolute refractory period. The time-to-peak versus conduction velocity data were slightly better described by a power than a linear relationship. Coefficients of correlation were: duration versus conduction velocity, linear r = −0.76; time-to-peak versus conduction velocity, non-linear r = −0.64; duration versus absolute refractory period, r = 0.70; time-to-peak versus absolute refractory period, r = 0.76. Extensive intercorrelation between the variables duration, distal conduction velocity and absolute refractory period was revealed by multiple correlation techniques. Inter- and intra-subject skin temperature variation was within 5°C. Correcting the time course, conduction velocity and absolute refractory period values for temperature variation within this limited range did not affect the results. The results demonstrate that the time course of the action potential varies systematically with distal axon conduction properties, in normal human subjects. The present approach for study of action potential time course and conduction properties of single axons in man promises to provide a sensitive means for assessing the impact of nerve disease on neural impulse transmission.