Resting potential (r.p.) and muscarinic response mechanisms were studied in Xenopus laevis oocytes using the voltage-clamp technique. Insertion of micro-electrodes into the oocyte produced a 'shunt' membrane conductance which partially sealed after a few minutes. The oocyte resting potential (measured with a single intracellular electrode) ranged from -40 to -60 mV. Ouabain and low K+ solution depolarized both follicles and denuded oocytes. The electrogenic Na+-K+ pump was more active in the latter. In the presence of ouabain, the r.p. agreed with the constant field theory. alpha (PNa+/PK+) was 0.12 in follicles and 0.24 in denuded oocytes. beta (PCl-/PK+) was 0.4 in both. At [Na+]o lower than 70 mM, the r.p. deviated considerably from the constant field predictions. The relatively large value of alpha indicated the major role of Na+ in oocyte r.p. determination. The oocyte muscarinic response was separated into four distinct components: the fast depolarizing Cl- current, 'D1'; the slow depolarizing Cl- current, 'D2'; the slow hyperpolarizing K+ current, 'H'; and the large membrane Cl- current fluctuation, 'F'. The H response reversal potential showed a Nernst relationship to [K+] and was selectively blocked by intracellular injection of tetraethylammonium (TEA). The D1 and D2 reversal potential showed a Nernst relationship to [Cl-]. In Ca2+-deficient, EGTA-containing medium, D2 and F were abolished and D1 and H were reduced. Verapamil inhibited all responses. Increasing [Ca2+]o caused a significant increase in D1, D2 and F response amplitudes. Intracellular injection of 0.6-10 pmol guanosine 3',5'-cyclic monophosphate, induced a large outward K+ current, similar to the muscarinic H response.