1. Two glass micro-electrodes were inserted into neighbouring cells from rat or mouse pancreatic segments, superfused in vitro. The tip of a third glass micro-electrode, filled with 2 M-AChCl, was placed just outside the acinus under investigation. Membrane potential and resistance, and changes in these parameters in response to short pulses of ACh stimulation, were recorded.2. The resting current-voltage relationship, obtained by injecting 100 msec depolarizing or hyperpolarizing current pulses through one of the intracellular micro-electrodes and recording the membrane potential with the other intracellular electrode, was linear within the range -5 to -60 mV.3. Injecting depolarizing or hyperpolarizing current (d.c.) through one of the intracellular micro-electrodes, the membrane potential (as measured with the other intracellular micro-electrode) could be set at various levels. The effect of ACh at different membrane potentials was investigated. When the acinar cell membrane was hyperpolarized, the amplitude of ACh-evoked depolarization was increased, while ACh-evoked depolarization was reduced when the membrane potential was reduced by depolarizing current, and finally changed into a hyperpolarization at very low membrane potentials. In each acinus investigated (rat and mouse), there was a linear relationship between amplitude of ACh-evoked potential change (DeltaV) (+ value or - value according to polarity) and resting membrane potential. During superfusion with control solution, the value of the membrane potential at which ACh did not evoke a potential change (E(ACh)) was about -15 mV in the mouse and about -20 mV in the rat. During superfusion with a chloride-free sulphate-containing solution (steady state), a linear relationship between DeltaV and resting membrane potential was again found but E(ACh) (mouse) was about +10 mV.4. A continuous rough estimate of E(ACh) was obtained by injecting repetitively depolarizing current pulses (100 msec) through one intracellular micro-electrode; in this way, the effect of ACh measured by the other intracellular electrode could be assessed simultaneously at the spontaneous resting level, and at a depolarized level. The direction of change in E(ACh) following acute changes in the superfusion fluid ion composition was assessed. Replacing extracellular chloride by sulphate caused an immediate change in E(ACh) in the positive direction. Re-admission of chloride, after a long period of chloride ion deprivation, caused an immediate sharp change in E(ACh) in the negative direction. Replacing extracellular sodium by Tris caused an immediate transient negative change in E(ACh). In contrast, taking away extracellular calcium changed E(ACh) in a positive direction. Augmenting extracellular potassium concentration to 40 mM caused a change in E(ACh) in the positive direction.5. At a membrane potential (V) equal to E(ACh) the sum of ionic currents evoked by the action of ACh is zero. Using the Goldman treatment, it appears that ACh increases membrane Na, K and Cl permeability. The approximate relative ion permeabilities of the pathways opened up by ACh are: P(Na)/P(K) = 2.5 and P(Cl)/P(K) = 5. At V = E(ACh), the approximate relative sizes of the ACh-evoked currents are: I(Na)/I(K) = 2.6 and I(Cl)/I(K) = 1.6 ACh, therefore, causes influx of Na and Cl and a small efflux of K.