Dynamics of the Golgi method: a time-lapse study of the early stages of impregnation in single sections

This paper describes the early stages of impregnation by the Golgi method. Sections of aldehyde-fixed and potassium dichromate-treated cerebral cortex were mounted on glass slides and cover slipped. The dichromate solution was replaced by silver nitrate solution and events in the section were followed and recorded by time lapse microphotography and video recording until stopped by replacement of silver nitrate solution by glycerol. The sections were subsequently prepared for electron microscopy (EM) study. In sections about 2 x 2 mm and 100 microns thick a fine, dark granular precipitate formed at the edges within the first minutes of exposure to silver nitrate and a wave of brownish colouration spread to a depth of about 0.3 mm. After approximately 7 min, shrub-like focal precipitates (nucleation centres) appeared in the sections. From these nucleation centres (but also from the section edges) thread-like 'outgrowths', usually identified as dendrites, spread into somata. Sometimes impregnation began within the soma and spread into dendrites. The rate of impregnation (i.e., of silver chromate deposition within dendrites) was typically 1-7 microns min-1, faster in the earlier stages (up to 3 microns s-1) and very slow after 30 min, by which time many neurons were more or less fully impregnated. The dimensions of the section, the width of an agar frame around the sections, and the frequency with which the silver nitrate in the sections was replenished all affected the extent and time course of the impregnation. By EM the earliest intracellular deposits consisted of tubulolamellar formations which did not cross plasma or endocellular membrane boundaries and which contained irregularly shaped and scattered granules, initially about 10 nm in diameter. The latter progressively enlarged and coalesced as the tubulolamellar formations extended, eventually to fill the cross-sectional area of neuronal processes and cell bodies. These observations shed light on why so few neurons become impregnated with the Golgi method. Impregnation occurred only in those cells a part of which was within a nucleation centre.