Background and aims: In experiments designed to dissect the signalling pathways in beta cells, the depolarizing effect of glucose metabolism is often replaced by a strong K + depolarization. Recent observations raised doubt that high K + is an appropriate experimental substitute for the physiologically induced depolarization. Methods: Using mouse islets insulin secretion was measured by batch perifusion and ELISA and the free cytosolic Ca 2+ concentration ([Ca 2+ ] i ) by the Fura technique. Primary mouse beta cells were used for electrophysiological experiments. Results: At basal glucose 40 mM K + induced a massive monophasic response, whereas 15 mM K + had only a minimal insulinotropic effect, although the increase in [Ca 2+ ] i was not inferior to that by 20 mM glucose. In voltage-clamp experiments Ca 2+ influx appeared as nifedipine-inhibitable inward action currents in the presence of sulphonylurea plus TEA to block compensatory outward K + currents. Under such conditions 15 mM K + prolonged the action currents and 40 mM K + transformed them into a continuous inward current. Correspondingly, 15 mM K + led to an oscillatory increase, 40 mM K + to a plateau of [Ca 2+ ] i superimposed on the [Ca 2+ ] i elevated by sulphonylurea plus TEA. Raising K + to 15 or 40 mM in the presence of sulphonylurea (+/- TEA) led to a fast further increase of insulin secretion, which could be antagonised by nifedipine or CoCl 2 . Conclusion: Raising extracellular K + does not simply activate voltage-dependent Ca 2+ channels, but concentration-dependently transforms the pattern of Ca 2+ influx into the beta cell and may thus generate stimuli of supraphysiological strength for insulin secretion.