1. Seven to ten days after sectioning their axons, rat sympathetic neurons were studied using intracellular recording techniques in an in vitro preparation of the superior cervical ganglion. 2. In 75% of axotomized cells, an after-depolarization (ADP) was observed following spike firing or depolarization with intracellular current pulses. Discontinuous single-electrode voltage-clamp techniques were employed to study the ADP. When the membrane potential was clamped at the resting level just after an action potential, a slow inward current was recorded in cells that showed an ADP. 3. In the presence of TTX and TEA, inward peaks and outward currents were recorded during depolarizing voltage jumps, followed by slowly decaying inward tail currents accompanied by large increases in membrane conductance. The inward peak and tail currents activated between -10 and -20 mV and reached maximum amplitudes around 0 mV. With depolarizing jumps to between +40 and +50 mV, net outward currents were recorded during the depolarizing jumps but inward tail currents were still activated. 4. In the presence of the Ca2+ channel blocker cadmium, or when Ca2+ was substituted by Mg2+, the ADP disappeared. In voltage-clamped cells, cadmium blocked the inward tail currents. The reversal potential for the inward tail current was approximately -15 mV. Substitution of the extracellular NaCl by sucrose or sodium isethionate increased the amplitude of the inward tail current, and displaced its equilibrium potential to more positive values. Changes in extracellular [K+] did not appreciably affect the inward tail current amplitude or equilibrium potential. Niflumic acid, a blocker of chloride channels activated by Ca2+, almost completely blocked the tail current. 5. No ADPs were observed in non-axotomized neurons, and when depolarizing pulses were applied while in voltage clamp no inward tail currents were evoked in these normal cells. 6. It is concluded that axotomy of sympathetic ganglion cells produces the appearance of a Ca(2+)-dependent chloride current responsible for the ADP observed following spike firing.