1. Using the single-electrode voltage clamp technique, three calcium current components were recorded at 35 degrees C from mouse dorsal root ganglion (DRG) neurones in culture. A transient low-threshold calcium current (T current) was recorded at clamp potentials (Vc) positive to -60 mV. Holding potentials (Vh) at or negative to -90 mV were required to fully remove inactivation. A large transient high-threshold calcium current component (N current) was recorded at Vc positive to -40 mV. Vh at or negative to -80 mV removed all steady-state inactivation. A slowly inactivating high-threshold calcium current component (L current) was recorded at Vc positive to -30 mV. Inactivation was removed by Vh at or negative to -60 mV. When currents were evoked at Vc positive to -20 mV from Vh negative to -60 mV, all three calcium current components were present. 2. Pentobarbitone (500 microM) had no effect on the isolated T current, but reduced the isolated L current 50-100% when evoked at Vc of -20 to 0 mV from Vh of -50 mV. Pentobarbitone had voltage-dependent effects on calcium currents containing all three calcium current components. Pentobarbitone produced small and equal reductions of the peak and late (greater than or equal to 300 ms) calcium currents evoked at -20 to 0 mV from Vh at or negative to -80 mV, but at more positive Vh there was a greater reduction in the peak current. The rate of current inactivation was increased in the presence of pentobarbitone. 3. Current-voltage plots were constructed from currents recorded in the absence and presence of 500 microM-pentobarbitone. Pentobarbitone reduced the magnitude of the calcium current without affecting the voltage dependence of the current-voltage relation. 4. Calcium current traces were fitted with a multiexponential function to determine the amplitudes and inactivation time constants (tau i) of the three calcium current components. Inactivation time constants decreased with more positive Vc for all three calcium current components. Pentobarbitone reduced only those tau i corresponding to the N current. 5. Recovery from inactivation of the N current was determined using a two-pulse protocol. In control neurones, recovery from inactivation occurring at 0 mV was slower at Vh = -65 mV than at Vh = -80 mV. In the presence of pentobarbitone, recovery from inactivation was faster, and occurred at a similar rate at both potentials. 6. Steady-state inactivation curves for the N current were derived from neurones in the absence and presence of pentobarbitone.(ABSTRACT TRUNCATED AT 400 WORDS)