Quantal release of neurotransmitter: an iterative method for the automatic computation of binomial distribution parameters.

A computational method is presented by which, when the amplitude-frequency histogram of the excitatory post-synaptic potentials shows a binomial distribution, an accurate evaluation of the statistical parameters p and n may be obtained. The entire procedure is a combination of 3 basic methods: steepest descent, parabolic interpolation and Montecarlo technique. The two statistical parameters are evaluated independently with respect to each other, which makes an effective control of the accuracy of the calculation possible by comparing the p by n product with the average number of quanta released in response to each nerve impulse, conventionally computed. Applications to quantal release studies in both rat and guinea-pig superior cervical ganglia are also presented.