1. Calcium and sodium currents and non-linear capacitive currents were recorded from isolated ventricular cells from neonatal rats, using the whole-cell patch-clamp technique, usually with a holding potential of -100 mV. 2. When recording with internal and external solutions designed to suppress virtually all ionic currents except the calcium current, careful subtraction of all linear capacitive and ionic currents revealed that depolarizations elicited a small transient outward current which preceded the inward calcium current. This outward current was discernible just below the threshold potential for the calcium current and increased with larger depolarizations to a maximum for potentials of about +30 mV and above. 3. Elimination of the calcium current revealed that at each potential the transient outward current was accompanied by a roughly equal transient inward current upon repolarization. The properties of these currents indicate that they are non-linear capacitive currents. Best-fit Boltzmann curves of the 'on' charge (integral of the transient outward current) gave values for qmax, V and k of 3.9 nC/microF, -29.3 mV and 15.5 mV with internal Cs+. The maximum 'on' charge is similar to that found with calcium currents (4.3 nC/microF). Similar values were obtained with internal TEA+. 4. Boltzmann fits of conductance vs. voltage for the calcium channel gave mean values of -15.5 and 13.3 mV for V and k (with internal Cs+); the corresponding values for the sodium channel were -49.9 and 5.4 mV. 5. Pre-pulses (20 ms) to -60 mV inactivated 77% of the peak sodium current, but only inactivated about 10% of the peak calcium current and reduced the maximum 'on' charge (moved at potentials positive to -60 mV) by 19%. 6. With a holding potential of -100 mV, 10 microM-nifedipine blocked 89% of the calcium current, but had little effect on the amount of 'on' charge. The 'off' charge appeared to be slower in the presence of nifedipine. 7. These results and consideration of the number of calcium channels and high-affinity binding sites for dihydropyridines (DHP), suggest that a large part of the charge movement may be related to DHP binding sites and involved with gating calcium channels. Comparison with skeletal muscle suggests similarities in the mechanisms involved in excitation-contraction coupling.