Outward currents activated by depolarization were studied in the neuropilar membrane of locust nonspiking local interneurons, using the single-electrode voltage-clamp technique in situ. Preliminary observation of these currents in 272 neurons revealed two families. The first and most commonly observed (85% of recordings) showed a large transient current followed by a slowly decaying/late current. The second (15% of recordings) showed an additional outward current with a slow rate of activation, a peak within 100-150 msec, and a slow rate of inactivation. Only neurons of the first type were studied further. The transient current was activated by depolarization around -60 mV, with a time to peak of approximately 11 msec at -50 mV and less than 3 msec at -20 mV. This current decayed exponentially, with a time constant of 8.1 +/- 1.6 msec (n = 8 interneurons) at -30 mV. This time constant of inactivation did not appear to depend strongly on membrane voltage, in the range in which it was studied. A second and longer time constant of inactivation of 50-400 msec could not be assigned to either of the transient and late components of the outward current. The ratio of transient-to-late current varied between 1.6 and 5.4, with a mean of about 2.5. The reversal potential for the transient current could, on average, be shifted by 14 mV by a threefold increase in the bath K+ concentration, indicating that K+ is a charge carrier for the current. The transient current became inactivated with maintained depolarization and appeared half-inactivated at about -60 mV (slope factor k1/2 = 8 mV). This current was thus not fully inactivated at "resting" potential (average, -58 mV). Recovery from inactivation followed a single exponential time course, with a time constant of approximately 100 msec at -80 mV. The time course of recovery from inactivation of the transient current was well correlated with that of the recovery of transient outward rectification, as measured in current-clamp recording. Tetraethylammonium, at a bath concentration of 10 mM reduced the transient current by 70% and the delayed current by 60%. 4-Aminopyridine, at a bath concentration of 5 mM, had a significant effect in only two of five interneurons, reducing the transient current by approximately 85% and the late current by approximately 15%. Quinidine at a bath concentration of 100 microM was ineffective. Although these blockers did not allow a clear pharmacological separation of the currents, they were effective in reducing the outward rectification observed in current clamp during step depolarization.(ABSTRACT TRUNCATED AT 400 WORDS)