gamma-Aminobutyric acid (GABA)- and piperazine-activated single-channel currents were recorded from the bag region of the somatic muscle of the nematode parasite Ascaris suum. Cell-attached and outside-out patch-clamp techniques were used. Clean membranes were routinely prepared using collagenase. GABA (concentrations greater than 1 microM) or piperazine (concentrations greater than 200 microM) applied to the extracellular surface of the patches brought about the opening of channels producing rectangular shaped current pulses of varying duration but essentially constant amplitude. The I/V relationships of the single-channel currents for both agonists were linear and had conductances in the region of 22 pS (in symmetrical 170 mM Cl-). The reversal potential was near 0 mV when Cl- was equally distributed on both sides of the membrane. Occasionally two subconductance states were seen. The mainstate single-channel permeability was estimated to be 4 X 10(-14) cm3 s-1. At low concentrations of GABA (3-4 microM), the effective mean channel open time was in the region of 32 ms (-75 mV, 22 degrees C, cell-attached patches). At low concentrations of piperazine (500 microM) the effective mean open channel lifetime was shorter, in the region of 14 ms (-75 mV, 22 degrees C cell-attached patches). For each agonist the channel open lifetime distributions were best described by the sum of two exponentials suggesting two open mainstates. Channel openings occurred as single events and in bursts with brief closed periods within bursts. The channel closed time histograms at these concentrations were best described by the sum of up to three exponentials, suggesting the presence of three closed states. Channel open times showed no appreciable voltage sensitivity. Before desensitization, increases in agonist concentration produced an increase in the probability of the channel being open. The increased probability was associated with an increase in the frequency of channel opening, an increase in the effective mean channel open time, an increase in burst duration, an increase in the number of openings per burst, together with a reduction in the proportion of brief openings. Desensitization was seen as a decline in the probability of the channel being opened during prolonged applications of agonist. It was associated with the appearance of very long (seconds) closed periods. The distributions of the closed channel times were then best described by up to four exponentials.