1. Acetylcholine (ACh), 7.5 x 10(-5) M, and carbachol, 5 x 10(-6) M (CCh) depressed the frequency of miniature endplate potentials (m.e.p.ps) in the frog (Rana temporaria) sartorius neuromuscular junction with active acetylcholinesterase to about 50-55% of the controls. 2. A similar depression was produced by the nicotinic agonists, nicotine, suberyldicholine and tetramethylammonium. 3. The muscarinic agonists, oxotremorine, methylfurmethide and methacholine were without effect on m.e.p.p. frequency. The muscarinic antagonist, atropine and the nicotinic antagonist, (+)-tubocurarine, had no effect on the depression of m.e.p.p. frequency evoked by CCh. 4. The ganglionic blockers, benzhexonium and IEM-1119, were also without effect on the CCh-evoked depression of m.e.p.p. frequency. 5. Pretreatment of muscles with anticholinesterases did not prevent the CCh-induced drop in m.e.p.p. frequency. 6. The effect of CCh was proportionally the same as in the controls in preparations where the m.e.p.p. frequency was changed by elevation of K+ and in the presence of theophylline, noradrenaline, dibutyryl adenosine 3':5'-cyclic monophosphate (db cyclic AMP) and db cyclic GMP. 7. An inhibitor of Na+,K(+)-ATPase, ouabain, 5 x 10(-5) mol l-1, prevented or reversed the depression of m.e.p.p. frequency by CCh. However, the depression was present in a nominally K(+)-free medium. Insulin and adrenaline, which are considered to be Na+,K(+)-ATPase activators, were without effect on depression of m.e.p.p. frequency. 8. The depression of m.e.p.p. frequency by 5 x 10(-6) M CCh was the same at temperatures between 5 and 30 degrees C with a Q10 near to 1.0. When threshold amounts of CCh were used (6 x 10-7 and 3 x 10-7 M), the depression was less at higher temperatures.9. The receptive structures responsible for the CCh (or ACh)-evoked depression of m.e.p.p. frequency differ pharmacologically from muscarinic, nicotinic ganglionic and neuromuscular junction ACh-receptors as well as from the synaptic cholinesterase, in contrast to previous reports (Duncan & Publicover, 1979).The low temperature-dependence points to the possibility that physical rather than biochemical processes are limiting in this presynaptic effect of cholinomimetics.