1. Whole-cell Ca2+ currents in cultured bovine chromaffin cells were studied using patch-clamp electrophysiology. With Ba2+ or Ca2+ as the current carriers, two separate components of whole-cell current could be distinguished by biophysical and pharmacological criteria. These components of Ca2+ current were different from T- or N-type Ca2+ channels previously described, as they were not inactivated at a holding potential of -60 mV. 2. Depolarization of the cells pats -20 mV in 10 mM-Ba2+ activated a single component of Ca2+ current, called the 'standard' current. This current showed no detectable voltage-dependent inactivation, but did show marked current-dependent inactivation as steady-state inactivation (H-infinity) plots obtained in the presence of Ba2+ were quite different from those obtained from Ca2+. 3. In most chromaffin cells large pre-depolarizations or repetitive deplolarizations in the physiological range activated a second component of Ca2+ current called 'facilitation'. Facilitation was observed with either Ca2+ or Ba2+ as the charge carrier. Recruiting facilitation increased whole-cell currents by an average of 60%. 4. Pre-pulses to +120 mV lasting 200 ms completely activated facilitation. Pre-pulses longer than 800 ms started to inactivate facilitation, while pre-pulses longer than 2500 ms completely inactivated this component of Ca2+ current. Because only outward currents were recorded at +120 mV, it is likely that facilitation inactivated in a voltage-dependent manner. 5. When the extracellular Ba2+ concentration was increased in the range from 2 to 90 mM activation of both facilitation and standard Ca2+ currents shifted in the depolarizing direction. In 2 mM-Ba2+ facilitation activated at potentials 10 mV more negative than the standard component, while in 90 mM-Ba2+, facilitation activated at a potential about 10 mV more depolarized than the standard component. Thus, the voltage sensor for the facilitation Ca2+ current appeared to sense more surface charge than did the standard Ca2+ current. 6. Tail currents measured at -20 and -30 mV in the absence of facilitation (without pre-pulses) showed one time constant for current deactivation. Tail currents measured with both facilitation and standard currents activated showed a significantly slower deactivation rate than that seen with the standard current alone. 7. The dihydropyridine antagonist nisoldipine (1-mu-M) completely suppressed the facilitation Ca2+ current even when cells were held at negative holding potentials (-80 mV). In contrast, the standard current was unaffected by 1-mu-M-nisoldipine, even at depolarized holding potentials (-20 mV). The dihydropyridine Ca2+ channel agonist Bay K 8644 did not affect the standard Ca2+ current but activated the facilitation Ca2+ current in the absence of pre-depolarizations. The addition of 1-mu-M-Bay K 8644 greatly increased whole-cell Ca2+ currents and shifted the voltage dependence in the hyperpolarized direction. 8. [Ca2+](i) transients recorded with the Ca2+ indicator dye Fura-2 demonstrated that nisoldipine (1-mu-M) suppressed only part of the [Ca2+](i) signal. Long pre-depolarizations in Ca2+-free media also produced a clear inhibition of part of the [Ca2+](i) transient, most probably via voltage-dependent inactivation of the facilitation Ca2+ current. The effects of pre-depolarizations and nisoldipine were not additive suggesting that both treatments suppressed the same facilitation Ca2+ current. 9. These results suggest that Ca2+ currents in bovine chromaffin cells result from the activity of two separate and distinct types of Ca2+ channels. The facilitation Ca2+ current appears to reflect the activity of a channel similar to L-type, while the standard Ca2+ current apparently arises from a novel type of Ca2+ channel. The role of these two different Ca2+ channels in the regulation of [Ca2+](i)-sensitive processes, such as catecholamine secretion, remains to be determined.