[en] Mouse islets were used to test the hypothesis that the B cell membrane must be depolarized for acetylcholine to increase insulin release. The resting membrane potential of B cells (at 3 mM glucose) was slightly decreased (5 mV) by acetylcholine, but no electrical activity appeared. This depolarization was accompanied by a Ca-independent acceleration of ⁸⁶Rb and ⁴⁵Ca efflux but no insulin release. When the B cell membrane was depolarized by a stimulatory concentration of glucose (10 mM), acetylcholine potentiated electrical activity, accelerated ⁸⁶Rb and ⁴⁵Ca efflux, and increased insulin release. This latter effect, but not the acceleration of ⁴⁵Ca efflux, was totally dependent on extracellular Ca. If glucose-induced depolarization of the B cell membrane was prevented by diazoxide, acetylcholine lost all effects but those produced at low glucose. In contrast, when the B cell membrane was depolarized by leucine or tolbutamide (at 3 mM glucose), acetylcholine triggered a further depolarization with appearance of electrical activity, accelerated ⁸⁶Rb and ⁴⁵Ca efflux, and stimulated insulin release. Acetylcholine produced similar effects (except for electrical activity) in the presence of high K or arginine which, unlike the above test agents, depolarize the B cell membrane by a mechanism other than a decrease in K+ permeability. Omission of extracellular Ca abolished the releasing effect of acetylcholine under all conditions but only partially decreased the stimulation of ⁴⁵Ca efflux. The results show thus that acetylcholine stimulation of insulin release does not result from mobilization of cellular Ca but requires that the B cell membrane be sufficiently depolarized to reach the threshold potential where Ca channels are activated. This may explain why acetylcholine alone does not initiate release but becomes active in the presence of a variety of agents