In voltage clamp studies of cardiac Purkinje fibers, a large early outward current is consistently observed during depolarizations to voltages more positive than -20 mV. After the outward peak of the current, the total membrane current declines slowly. Dudel et al. (1967. Pfluegers Arch. Eur. J. Physiol. 294:197--212) reduced the extracellular chloride concentration and found that the outward peak and the decline of the current were abolished. They concluded that the total membrane current at these voltages was largely determined by a time- and voltage-dependent change in the membrane chloride conductance. We reinvestigated the chloride sensitivity of this current, taking care to minimize possible sources of error. When the extracellular chloride concentration was reduced to 8.6% of control, the principal effect was a 20% decrease in the peak amplitude of the outward current. This implies that the membrane chloride conductance is not the major determinant of the total current at these voltages. The reversal potential of current tails obtained after a short conditioning depolarization was not changed by alterations in the extracellular chloride or potassium concentrations. We suspect that the tail currents contain both inward and outward components, and that the apparent reversal potential of the net tail current largely reflects the kinetics of the outward component, so that this experiment does not rule out potassium as a possible charge carrier. The possibility that potassium carries much of the early outward current was further investigated using tetraethylammonium, which blocks potassium currents in nerve and skeletal muscle. This drug substantially reduced the early outward current, which suggests that much of the early outward current is carried by potassium ions.