1. Whole-cell-voltage-clamp techniques were used to study voltage-activated transient potassium currents in a large sample (n = 143) of granule cells (GrC) from rat cerebellar slices. Tetrodo-toxin (TTX; 0.1 muM) was used to block sodium currents, while calcium current was too small to be seen under ordinary conditions. 2. Depolarizing pulses from -50 mV evoked a slow, sustained outward current, developing with a time constant of 10 ms, inactivating over a time scale of seconds and which could be suppressed by 20 mM tetraethylammonium (TEA). By preventing the Ca2+ inflow, this slow outward current could be further separated into a Ca2+ -dependent and a Ca2+ -independent component. 3. After conditioning hyperpolarizations to potentials negative to -60 mV, depolarizations elicited transient outward current, peaking in 1-2 ms and inactivating rapidly (approximately 10 ms at 20-degrees-C), showing the overall kinetic characteristics of the A-current (I(A)) The current activated following third-order kinetics and showed a maximal conductance of 12 nS per cell, corresponding to a normalized conductance of 3.8 nS/pF. 4. I(A) was insensitive to TEA and to the Ca2+-channel blockers. 4-Aminopyridine (4-AP) reduced the A-current amplitude by approximately 20%, and the delayed outward currents by >80%. 5. Voltage-dependent steady-state inactivation of peak I(A) was described by a Boltzmann function with a slope factor of 8.4 mV and half-inactivation occurring at -78.8 mV. Activation of I(A) was characterized by a Boltzmann curve with the midpoint at -46.7 mV and with a slope factor of 19.8 mV. 6. I(A) activation and inactivation was best fitted by the Hodgkin-Huxley m3h formalism. The rate of activation, tau(a), was voltage-dependent and had values ranging from 0.55 ms at -40 mV to 0.2 ms at +50 mV. Double-pulse experiment showed that development and removal of inactivation followed a single-exponential time course; the inactivation time constant, tau(ha), was markedly voltage-dependent and ranged from approximately 10 ms at -40 and -100 mV and 70 ms at -70 mV. 7. A set of continuous equations has been developed describing the voltage-dependence of both the steady-state and time constant of activation and inactivation processes, allowing a satisfactory numerical reconstruction of the A-current over the physiologically significant membrane voltage range. 8. In terms of neuronal functioning, I(A) dominates the other potassium currents on account of its faster activation kinetics and for maximal conductance, which is more than twice the total conductance of the remaining potassium currents together. Because a significant fraction of I(A), is deinactivated at the physiological resting membrane potentials, we conclude that I(A) is likely to play a preeminent role in regulating excitability and in vivo repolarization.