Neurotransmitter and neuromodulator release takes place by exocytosis; influx of calcium into the cell or nerve terminal triggers the fusion of the storage granule with the cell plasma membrane. These events can be modelled by studying the fusion of artificial or biological membranes with each other. Using a specially constructed stopped-flow mixing apparatus, our previous studies have shown that the kinetics of both aggregation and fusion of small vesicular structures (artificial lipid vesicles, neurotransmitter storage granules, etc) follow second order kinetics with fusion being aggregation rate limited. Our original stopped-flow assay for membrane fusion, based on resonance energy transfer between fluorescent phospholipids was subject to artifacts arising from the interactions of the probes with calcium. We have developed a new assay, based on changes in pyrene-labelled phosphatidyl choline fluorescence which not only solves these problems but allows for rapid, easy calculation of expected results. This assay has been applied to the magnesium-promoted fusion of artificial vesicles as well as protein-catalysed fusion. A stopped-flow study of cobalt ion transport across the membranes of small unilamellar vesicles shows that they leak profusely during fusion while larger vesicles with less radical changes in surface curvature do not. We ascribe this to defects in the packing structure of the membrane phospholipids. A method for measuring the membrane potential of vesicular structures and changes in potential by various treatments has been derived from this experimental setup. Various proteins and polypeptides can catalyse fusion of artificial and biological membranes. Some have known functions in biological systems, e.g., the spike proteins from rhabdoviruses; therefore in vitro studies of these fusion mechanisms may have clinical relevance. Polylysine will fuse small unilamellar vesicles under conditions similar to spike protein-mediated virus/cell membrane fusion. Studies indicate that polylysine-mediated fusion is not aggregation rate limited and resembles that seen for in vitro fusion of chromaffin granules.