1. Presynaptic or simultaneous pre- and postsynaptic voltage-clamp protocols were implemented in the squid giant synapse in order to determine the magnitude and time course of the presynaptic calcium current (ICa) and its relation to transmitter release before and after presynaptic injection of proteins. These included several forms of synapsin I, calcium-calmodulin-dependent protein kinase II (CaM kinase II) and avidin. 2. The quantities and location of these proteins were monitored by fluorescence video-enhanced microscopy during the electrophysiological measurements. 3. Presynaptic injection of dephosphorylated synapsin I inhibited synaptic transmission with a time course consistent with diffusion of the protein through the terminal and action at the active release zone. A mathematical model relating the diffusion of synapsin I into the terminal with transmitter release was developed to aid in the interpretation of these results. 4. Synapsin I inhibition of transmitter release was reversible. 5. The action of synapsin I was highly specific, as phosphorylation of the tail region only or head and tail regions prevented synapsin I from inhibiting release. 6. Injections of heat-treated synapsin I or of avidin, a protein with a size and isoelectric point similar to those of synapsin I, had no effect on transmitter release. 7. CaM kinase II injected presynaptically was found to facilitate transmitter release. This facilitation, which could be as large as 700% of the control response, was related to the level of penetration of the enzyme along the length of the preterminal A mathematical model of this facilitation indicates a reasonable fit between the distribution of CaM kinase II within the terminal and the degree of facilitation. 8. The overall shape of the postsynaptic response was not modified by either synapsin I or CaM kinase II injection. 9. The data suggest that, in addition to releasing transmitter, calcium also penetrates the presynaptic cytosol and activates CaM kinase II. When activated, CaM kinase II phosphorylates synapsin I, which reduces its binding to vesicles and/or cytoskeletal structures, enabling more vesicles to be released during a presynaptic depolarization. The amplitude of the postsynaptic response will then be both directly and indirectly regulated by depolarization induced Ca2+ influx. This model provides a molecular mechanism for synaptic potentiation.