The existence of coated vesicles (CVs) in neurons has been known since 1961 when Gray reported ‘complex vesicles’ in mossy fibre endings of the rat cerebellar cortex. He described them as spheres, 60–80 nm in diameter, surrounded by shells consisting of closely packed 15–20 nm vesicular bodies. This unusual structural organisation was probably only an apparent one. Today the structure assigned to CVs in neurons and other cells, which is based upon detailed analyses of isolated CVs, consists of a spherical lipid-bilayer vesicle enclosed by a protein coat composed of pentagonal and hexagonal subunits (Kanaseki & Kadota, 1969; Kadota & Kadota, 1973a, b; Pearse, 1975; Crowther, Finch & Pearse, 1976; Woods, Woodward & Roth, 1978). The comparison of CVs isolated from different tissues (Pearse, 1975, 1976; Woods et al., 1978) and in thin sections of isolated tissues (Nickel, Vogel & Waser, 1967) revealed no significant differences between the structure and chemical properties of CVs in neurons and in other cells.

For a time there was some doubt whether neuronal CVs were discrete organelles. Difficulty in resolving the three-dimensional organisation of the coat by goniometry led to the idea that the CVs seen in thin sections were fixation artifacts arising from the denaturation of microtubules or of the neuronal cytonet (Gray, 1972, 1975; Westrum & Gray, 1977). This idea, however, was not supported by the studies of CVs isolated from unfixed brain tissue cited above.