1. Muscarinic cholinergic actions were investigated in a population of large multipolar spinal cord neurons in primary dissociated cell culture using conventional intracellular recording and single-microelectrode voltage-clamp techniques. 2. Cholinergic agonists were applied to the surface of neuronal somata by pressure ejecting drug-containing bathing medium from small blunt (2-10 microns) glass micropipettes. Atropine was applied by diffusion from large (20-30 microns) blunt micropipettes positioned near the soma. 3. Muscarine increased action-potential firing and evoked slow sustained membrane depolarization. Action potentials but not slow membrane depolarizations were eliminated by the sodium channel blocker, tetrodotoxin. Muscarine-induced depolarizing responses were unaffected by the calcium channel blocker, cadmium. 4. Depolarizing responses evoked by selective and nonselective muscarinic cholinergic agonists were dose dependent, reversibly antagonized by atropine, and did not desensitize. 5. Muscarine depolarized neurons and decreased membrane conductance during recording with both 3 M KCl- and 4 M KAc-filled intracellular recording micropipettes. When membrane potential was held constant using the single-electrode voltage-clamp technique (KCl-filled micropipettes), muscarine and gamma-aminobutyric acid (GABA) evoked inward currents at resting membrane potential. GABA-induced inward current responses were decreased by depolarization and had reversal potentials near -30 mV, consistent with GABA increasing chloride conductance. Muscarine-induced inward current responses were increased by depolarization and had extrapolated reversal potentials near -80 mV, consistent with muscarine decreasing a potassium conductance. 6. Unlike GABA-induced currents, muscarine-induced currents evoked in normal Tris-buffered saline (5 mM potassium) did not vary as a linear function of membrane potential and did not reverse polarity in six of seven neurons near potassium equilibrium potential. However, in high-potassium medium (15 mM) muscarinic responses did reverse polarity and current was linearly related to membrane potential. Thus, the apparent voltage dependence of muscarine responses was probably due to voltage dependency of the potassium conductance and not due to potassium channel rectification. 7. Preliminary evidence (37) indicates that muscarine decreases a time- and voltage-dependent potassium current in some cultured spinal cord neurons. Whether reduction of m current can completely account for muscarine postsynaptic actions in these cells remains unclear. Muscarine may also block a small population of non-voltage-dependent potassium channels in addition to reducing m current.