1. Whole-cell patch clamp techniques were used to record transient outward potassium currents in embryonic rat neocortical neurons maintained in culture. The effect of lidocaine and its quaternary derivative QX222 on this transient outward current (TOC) was examined. 2. Extracellular application of lidocaine produced a progressive decrease in peak TOC amplitude with no change in the overall current waveform. Peak amplitude was reduced to 74 ± 6% (mean ± SE) of control by 0.5 mM and to 48 ± 2% by 2 mM lidocaine. The effect of lidocaine was fully reversible. Intracellular application of QX222 also resulted in a concentration-dependent reduction of the TOC with 0.5 mM reducing the peak amplitude to 73 ± 8% of control and 1 mM reducing it to 41 ± 4%. 3. Lidocaine reduced TOCs when applied intracellularly but the rate of block was considerably slower than with extracellular application. QX222 had no effect on the TOC when applied extracellularly. 4. Lidocaine (1 mM) induced a 4-5 mV hyperpolarizing shift in the voltage-dependence of activation and steady-state inactivation with no change in the slope factor. This small hyperpolarizing shift could not account for the >30% reduction in peak amplitude produced by 1 mM lidocaine. Hyperpolarizing shifts were not seen with 0.5 mM intracellular QX222. The lack of effect on the slope of the activation and steady-state inactivation curves indicates that local anesthetic (LA) actions were not voltage-dependent. 5. Lidocaine or QX222 did not produce a significant change in the TOC decay time-constant. The lack of any significant change in the voltage-dependence of steady-state inactivation or time course of recovery from inactivation indicates that LAs were not altering inactivation mechanisms. 6. Application of lidocaine or QX222 during a 60-s period, when the TOC channels were kept in the resting state, reduced the peak amplitude of the first evoked TOC. The reduction was not as large as when TOCs were evoked at regular 0.1-Hz intervals during drug application. This indicates that the amount of tonic block induced by the LA is dependent on channel activation. Increasing the activation frequency to 1 Hz after a steady-state had been obtained with lidocaine or QX222 did not result in additional reductions in the TOC. 7. It is proposed that lidocaine and QX222 tonically block TOCs by binding to a site within the channel protein complex, rendering the channel non-conductive. The apparent activation-dependence of the tonic block could be due to an increase in accessibility of the binding site as a result of channel activation. The binding site is accessible to both charged and uncharged LAs via different pathways.