1. Whole-cell voltage clamp techniques were used to examine the properties of voltage-dependent Ca2+ channel currents in single smooth muscle cells enzymatically dissociated from guinea-pig urinary bladder. Potassium currents were blocked with intracellular Cs+. A holding potential of -60 mV was normally applied. 2. When the membrane potential was returned to the holding potential after a depolarizing step, tail currents were seen after depolarizations to positive potentials, and the size of the tail current increased with increasing positivity of the preceding depolarization. 3. After depolarization to +80 mV (a potential at which little inward current flowed through the Ca2+ channels) tail currents on returning to the holding potential increased in size as the duration of the depolarization was increased. 4. Investigation of the mechanism mediating the tail currents showed that they were not flowing through non-selective cation channels, and had no contribution from Ca(2+)-activated Cl- channels or Na(+)-Ca2+ exchange. 5. The tail currents and the inward currents evoked by a simple depolarizing test potential were equally decreased by nifedipine in a dose-dependent manner. This suggests that L-type Ca2+ channels are responsible for both of the two types of inward currents. The inward currents were also inhibited in a similar manner when caffeine was applied. 6. Although the tail currents evoked on stepping from +80 mV to a holding potential of -60 mV increased in size with the duration of the conditioning potential, the total membrane Ca2+ conductance did not increase, since the inward currents evoked on stepping to +20 mV (a potential at which the Ca2+ channels are still fully activated) did not change with time. 7. The amplitude of the inward current evoked by a simple depolarizing test potential was similar to that evoked on stepping to the same test potential after preconditioning at +80 mV, if the test potential was higher than +20 mV. However, following repolarization to the holding potential, the amplitude of the subsequent tail current was larger and the deactivation time constant longer, after the conditioning depolarization. These results suggest that the voltage-dependent Ca2+ channels have at least two open states with different time constants, the tail current being the result of a long open channel state induced by large depolarizations. 8. When variable repolarizing potentials were applied after n +80 mV depolarization (5 s), the current-voltage relationship of the tail current was nearly linear between -60 and +30 mV. The deactivation was faster when a larger repolarization step was applied.(ABSTRACT TRUNCATED AT 400 WORDS)