1. The preceding paper (Van der Kloot, 1988) described a method for estimating the timing of quantal releases during an end-plate current. This period of elevated quantal release is called the early release period or ERP (Barrett & Stevens, 1972b). In the present paper, this deconvolution method is used to study the effects of varying quantal output by extracellular ions, stimulus patterns and drugs. 2. The data were obtained by voltage clamping end-plates in low-Ca2+ high-Mg2+ solutions, or in solutions containing tubocurarine (measuring the decay of the miniature end-plate currents (MEPCs) before curarization and assuming a value for MEPC amplitude after curarization). Data were also obtained by extracellular recording in Ca2+-free solution, using a recording pipette filled with CaCl2 and regulating Ca2+ release with a bias current. The three approaches led to similar conclusions. 3. Quantal release rose during the ERP along a sigmoid curve and reached a maximum after about 1.4 ms at 10 degrees C. This is called the time to peak. Quantal release then fell, following an exponential time course with a time constant of about 1.2 ms (10 degrees C). This is called the time constant for decline. 4. The ERP was followed by further, elevated quantal release, at a much lower rate, which declined over a longer time course. This is called late release. The magnitude of late release appears to be almost independent of the magnitude of release during the ERP, although the deconvolution method is a poor one for determining late release. The remainder of the results therefore focus on the ERP. 5. Increasing [Ca2+]o increased quantal output, and the rate of quantal output. It did not change the time to peak or the time constant of decline. Similarly, replacing Ca2+ with Sr2+ did not alter the time course of the ERP. 6. Two-pulse facilitation increased quantal output without changing the time to peak or the time constant of decline. 7. Quantal output was enhanced still more following a brief series of repetitive nerve stimulations. There was a lengthening of the time to peak; there was no change in the decline. The depression produced by longer series of repetitive stimulations did not change the time course of the ERP. 8. 4-Aminopyridine (4-AP) and dimethylsulphoxide (DMSO) increased quantal output and lengthened the time to peak, without altering the time constant for decline. 9. Adenosine decreased quantal output without altering the time course of the ERP.(ABSTRACT TRUNCATED AT 400 WORDS)