The mechanism of transmitter release by intracellular Ca has been explored by recording presynaptic Ca concentration ([Ca2+]i) with Ca-sensitive fluorescent dyes and by controlling [Ca2+]i with photosensitive Ca chelators. [Ca2+]i decays slowly (in seconds) after presynaptic action potentials, while transmitter release lasts only a few ms after each spike at fast synapses. Simulations of Ca diffusing from Ca channels opened during action potentials suggest that transmitter is released by brief, localized [Ca2+]i reaching about 100 microM ('Ca domains'). Several indirect measures of [Ca2+]i levels achieved at release sites are in agreement with this estimate. Synaptic facilitation is a short-term synaptic plasticity in which transmitter release is enhanced for up to 1 s following prior activity. This seems to be due to the residual effect of Ca bound to a different site from the multiple fast, low-affinity binding sites that Ca must occupy to trigger secretion. The release of transmitter by localized Ca domains explains the variable degree of apparent cooperatively of Ca action obtained when relating transmitter release to Ca influx. Increasing Ca influx by elevating extracellular [Ca2+] increases the [Ca2+]i in each Ca domain, and release increases with a high-power dependence on Ca influx because of a high degree of Ca cooperativity. However, prolonging presynaptic spikes or using depolarizing pulses of increasing amplitude increases Ca influx by opening more Ca channels and increasing the number of Ca domains locally triggering release. Partial overlap of these domains results in a slightly greater than linear dependence of release on total Ca influx. Post-tetanic potentiation (PTP) is a minute-long form of synaptic plasticity that correlates with measures of residual presynaptic [Ca2+]i. The linear relationship between PTP and residual [Ca2+]i suggests that, as in synaptic facilitation, Ca seems to act at a different site from those that directly trigger release. Presynaptic sodium accumulation also contributes to PTP, apparently by reducing the Na gradient across the presynaptic membrane and impeding the removal of presynaptic Ca accumulated in the tetanus by Na/Ca exchange. Transmitter release at crayfish motor nerve terminals can be reduced by presynaptic inhibition, which reduces the Ca influx into terminals. Serotonin enhances transmitter release without increasing either resting [Ca2+]i or Ca influx during spikes, apparently operating at a site 'downstream' of Ca to modulate release. Spikes transiently accelerate transmitter release triggered by elevation of [Ca2+]i using photosensitive chelators, even in low-[Ca2+] media that blocked detectable transmitter release.(ABSTRACT TRUNCATED AT 400 WORDS)