1. Spontaneous inhibitory postsynaptic currents (IPSCs) and evoked IPSCs were recorded by a whole-cell patch-recording technique from cultured Purkinje cells of the rat. The size of spontaneous IPSCs, after a train of depolarizing pulses was applied to the Purkinje cells, increased to 163 +/- 6% (mean +/- S.E.M., n = 7 cells) of the control levels measured before the stimulus train. 2. The GABAergic postsynaptic currents were recorded under voltage clamp from the synapse formed between two Purkinje cells. These IPSCs increased to 218 +/- 31% (n = 4) of control levels after depolarizing stimulation was applied to the postsynaptic Purkinje cells. Size-increased IPSCs were observed as long as recording continued and the phenomena will be called potentiation in this paper. 3. Intracellular application of Ruthenium Red (20 microM) did not block the potentiation of spontaneous IPSCs induced by the depolarizing stimulus (165 +/- 9%, n =6), but heparin (2 mg ml-1) partially blocked the potentiation (123 +/- 10%, n = 6). Heparin applied together with Ruthenium Red (20 microM) blocked potentiation completely (96 +/- 5%, n = 8) at concentrations higher than 1 mg ml-1. 4. Intracellular free calcium concentrations ([Ca2+]i) was monitored as the ratio of fura-2 fluorescences excited at 340 and 380 nm. In control cells, [Ca2+]i was increased by each depolarizing pulse. When Purkinje cells were dialysed with heparin or heparin with Ruthenium Red, the rise in [Ca2+]i was suppressed. 5. Bath application of thapsigargin (1 microM) blocked the potentiation (99 +/- 2%, n = 4) and suppressed the rise in [Ca2+]i. 6. When 30 mM BAPTA was applied intracellularly, a train of depolarizing pulses failed to induce potentiation of IPSCs and failed to raise [Ca2+]i. The results from points 3-6 suggest that the increase in [Ca2+]i, most probably coupled with the release from intracellular stores especially from the inositol trisphosphate (IP3)-sensitive stores, is crucial for the potentiation of IPSCs. 7. Bath application of a metabotropic glutamate receptor activator (t-ACPD, 200 microM) increased both the amplitude and frequency of spontaneous IPSCs and increased the [Ca2+]i slightly in dendrites. The inward current induced by the puff-applied GABA (2 microM) was increased, after t-ACPD application, to 186 +/- 36% of the control level (n = 3). Bath application of quisqualate (2 microM) caused a rapid increase in [Ca2+]i in dendrites and in the cell body and increased both the amplitude and frequency of spontaneous IPSCs. 8. The bath application of an inhibitor of phospholipase C (PLC), U73122 (1 microM), suppressed a rise in [Ca2+]i and blocked the potentiation (106 +/- 3%, n = 5). The inactive form, U73343 (1 microM), did not affect the potentiation (151 +/- 11%, n = 7) or the rise in [Ca2+]i. These observations suggest a possible involvement of the mechanism of Ca2+ activation of PLC and the IP3-induced Ca2+ release in the induction of IPSC potentiation in Purkinje cells.