The presynaptic terminals at crayfish (Procambarus spp.) opener neuromuscular junctions were voltage clamped. Calcium currents were measured during (ICa) and following (tail ICa) presynaptic depolarizations; EPSPs or IPSPs were simultaneously recorded from the (postsynaptic) muscle fibre directly beneath the presynaptic impalement. 2. For short (< or = 6 ms) presynaptic depolarizations, most of the transmitter release occurred during the tail ICa. EPSP or IPSP amplitudes at the end of the 6 ms pulse (end EPSP or end IPSP) increased monotonically with the integral of the ICa ([symbol: see text]ICa). The suppression potential for transmitter release was near the apparent reversal potential for ICa. 3. When the end EPSP or end IPSP amplitude was plotted against the peak ICa elicited during a presynaptic pulse (peak ICa), large and small depolarizations which evoked the same peak ICa evoked different amounts of transmitter release. The differences in transmitter release were eliminated when end EPSP amplitude was plotted against [symbol: see text] ICa, suggesting that transmitter release during a depolarization depends only upon calcium current and not upon a subsequent voltage-dependent step. 4. The synaptic transfer function of various measurements of EPSP or IPSP amplitude vs. [symbol: see text]ICa evoked during a presynaptic depolarization was a power function having an exponent of about 3. Similar measurements of EPSP amplitude vs. [symbol: see text]tail ICa evoked following a presynaptic depolarization had an exponent of about 2. 5. Facilitation of an EPSP or IPSP was not due to increases in calcium current at the test depolarization. 6. When the conditioning depolarization was increased and the test depolarization remained constant, EPSP amplitude at the test depolarization and facilitation increased . When the conditioning depolarization remained constant and the test depolarization was increased, EPSP amplitude at the test depolarization increased, while facilitation decreased. 7. Our data suggested that transmitter release at crayfish neuromuscular junctions is a non-linear function of calcium influx, and that facilitated release utilizes intracellular calcium differently from non-facilitated release. These data contradict simple models of facilitation which combine the residual calcium hypothesis with the calcium co-operativity hypothesis of non-facilitated release.
