The whole-cell patch-clamp technique was used to study the effects of holding potential and frequency on the Ca2+ current in frog ventricular myocytes. INa was blocked by TTX, and ica was activated with depolarizing clamps from different holding potentials. Variation of the holding potential revealed three new effects on ica: (1) At -40 mV iCa declined with a time constant of 15 min, while at -90 mV, this irreversible decline (run down) in iCa did not occur. (2) The decline of iCa at -40 mV was biphasic: run down was preceeded by a slow inactivation with a time constant of 40 s, which was reversible upon returning the holding potential to -90 mV. (3) Increasing the frequency of the clamp pulses from 0.1 to 1 Hz led to a rapid decline of iCa when the holding potential was positive to -60 mV, but at -90 mV had either no effect or increased iCa by 35%, if c-AMP was included in the dialyzing solution. On the other hand, c-AMP did not alter the time course of the run down and the slow inactivation. Replacement of extracellular Ca2+ by Ba2+ markedly slowed iCa kinetics, but did not change the very slow inactivation or the frequency-induced enhancement of iCa. Injection of c-AMP led to a transient increase of iCa. The phosphodiesterase inhibitor theophylline enhanced the amplitude of the transient and slowed its decay. This effect was mimicked by increased frequency. It is concluded that frequency-induced enhancement of iCa is highly dependent on the holding potential, independent of Ca2+, and may involve elevation of the intracellular level of c-AMP via inhibition of phosphodiesterase activity. The new type of very slow inactivation is probably under direct voltage control and independent of Ca2+ and c-AMP.