A voltage-clamp study was made of a slow excitatory post-synaptic potential (slow e.p.s.p.) that can be elicited in the medial cells of the left pleural ganglion of Aplysia californica by the firing of at least three different presynaptic neurones (labelled I, II and III). Each of these three neurones elicits other permeability changes in addition to the slow e.p.s.p., and all elements of these synaptic responses were shown to be mediated monosynaptically. The slow e.p.s.p., associated with an increase in membrane resistance, was shown to be due to a decrease in K permeability. When the slow e.p.s.p. was present spontaneously, it could be blocked by three compounds (tetraethylammonium (TEA), phenyltrimethylammonium (PTMA), or methylxylocholine (beta-TM 10], all previously shown to block the cholinergic receptor that mediates an increase in K conductance in the medial cells (see Kehoe, 1972b). Furthermore, in ganglia in which no slow e.p.s.p. was seen in response to firing of the neurones I, II, and III, such a response became manifest when agonists capable of activating the cholinergic receptor were applied (e.g. acetylcholine (ACh), carbachol, arecoline, or F2268). The slow e.p.s.p. thus appears to result from the reduction, induced by any one of three 'blocking neurones', of a cholinergically controlled K conductance. Finally, when presynaptic neurone I (the only neurone tested) was fired shortly before or during the activation of presynaptic neurone IV, previously shown to be cholinergic (Kehoe, 1972b), the K component of the cholinergic post-synaptic inhibitory potential was markedly reduced. The concentration at which a given agonist caused the manifestation of the synaptic diminution in K conductance (i.e. the slow e.p.s.p.) was found to be the same as that at which it caused a reduction in the synaptically activated, cholinergic, K-dependent conductance elicited by presynaptic neurone IV. Intracellularly injected adenosine 3',5'-cyclic monophosphate (cyclic AMP) imitated the effect of the 'blocking neurones' on the K conductance activated by bath-applied cholinomimetics. This effect was superimposed on a cyclic-AMP-induced, voltage-dependent inward current that disappeared when the cell was bathed in Na-free sea water, or when the extracellular Ca concentration was increased to 60 mM. The effect of cyclic AMP on the cholinergic K conductance remained even after this cyclic-AMP-activated inward current was eliminated.(ABSTRACT TRUNCATED AT 400 WORDS)