1. Whole-cell membrane currents in voltage-clamped single isolated cells of longitudinal smooth muscle of guinea-pig ileum were studied at room temperature using patch pipettes filled with either high-K+ solution or high-Cs+ solution, to suppress K+ outward current, and containing 0.3 mM-EGTA. 2. In the presence of high-K+ solution in the pipette, membrane depolarization from the holding potential of -50 mV evoked an initial inward calcium current (I(Ca)) followed by a large initial transient outward current and a sustained outward current with spontaneous oscillations superimposed. Prolonged depolarization above -20 mV produced a late transient outward current which reached a maximum (up to several nanoamps at + 10 mV) within approximately 1 s and lasted several seconds. 3. The late outward current (I(LTO)) was voltage dependent and reversed at the E(K) (potassium equilibrium potential) in cells exposed to high-K+ external solution. It was blocked by TEA+ (tetraethylammonium) or Ba2+ applied externally (calculated K(d) (dissociation constant) values were 0.67 and 4.43 mM, respectively) or by high-Cs+ solution perfusing the cell. The removal of extracellular Ca2+, application of Ca2+ channel blockers (3 mM-Co2+, 0.2 mM-Cd2+ or 1-mu-M-nifedipine) or perfusion of 5 mM-EGTA inside the cell also abolished the current. Thus, the current seems to be a Ca2+-activated K+ current. 4. There is a great discrepancy between the time course of the I(Ca) and that of the late I(LTO), which suggests that Ca2+ release from intracellular storage sites may contribute to the generation of the I(LTO). 5. Bath application of caffeine (10 mM) during the development of I(LTO) enhanced the current. However, in the presence of caffeine I(LTO) was inhibited. Moderate inhibition of I(Ca) by caffeine was also observed. 6. Ryanodine (5-mu-M) applied to the bathing solution completely inhibited I(LTO) within 3.5 min; however, it had no or little effect on the I(Ca). 7. Ruthenium Red (10-mu-M) completely blocked the I(LTO) and slightly and more slowly inhibited the I(Ca). 8. Increasing Mg2+ concentration in the pipette solution from 1 to 6 mM abolished the I(LTO). 9. It was concluded that the I(LTO) was activated mainly by Ca2+ released from the intracellular storage sites following Ca2+ entry, presumably by a Ca2+-induced Ca2+ release mechanism.