Study of the microcirculation by iontophoresis is potentially confounded by any non-specific effects of current application. Laser Doppler flow (LDF, mean ± SD; arbitrary units; AU) was recorded on the forearms of healthy volunteers during and 20 min following application of 0.10-mA current for 1, 3 and 5 min, using deionised water as a vehicle. Local heating to 44°C was then applied for 24 min to assess maximal vasodilation. Cathodal current applications resulted in delayed and prolonged vasodilation (peak values: 78 ± 29, 75 ± 19, 80 ± 37 AU) whereas anodal peak LDF was 13 ± 6, 27 ± 34 and 72 ± 40 AU for 1-, 3- and 5-min periods of current applications, respectively. From current onset, inflexion points in the responses to 3- and 5-min anodal current applications occurred at 4.5 and 6.5 min, respectively, and at ∼1.5 min for all cathodal current applications. For 5-min current applications: a preliminary tourniquet ischaemia neither changed the time course nor the amplitude of the response to current application. In this situation, local anaesthesia abolished the current-induced vasodilation. Chronic capsaicin pretreatment decreased the amplitude of the vasodilation. Pretreatment with 500 mg oral aspirin decreased the cathodal vasodilation and abolished the anodal vasodilation, even in the absence of preliminary ischaemia. We conclude that vasodilation to prolonged application of 0.10-mA continuous monopolar current after transient tourniquet ischaemia cannot be exclusively the result of an axon reflex initiated by current onset. This current-induced vasodilation is at least partly dependent on capsaicin-sensitive afferent fibres and relies on aspirin-sensitive mechanisms at both polarities.