Synaptic patterns in the visual cortex of the cat and monkey. Electron microscopy of Golgi Preparations

Abstract Primary visual cortex (area 17) from macaque monkeys, cats and kittens was stained by a rapid Golgi method modified to give optimal preservation of fine structure. Single impregnated neurons were identified in the light microscope, and then cut in serial thin sections for electron microscopy. As in conventional E.M. material, synapses in the Golgi preparations could be classified into types 1 and 2: type 1 synapses possessed round vesicles in the axon and a pronounced postsynaptic opacity; type 2 synapses had smaller, flattened vesicles in the axon and a slight postsynaptic opacity. Synapses onto impregnated cells were identified by the presynaptic opacities (dense projections), and classified as type 1 or 2 according to the shape and size of the vesicles. Pyramidal cells and spiny stellate cells (found in layer 4) had a similar synaptic environment: type 1 synapses occurred only on the spines, while type 2 synapses were found on dendritic shafts, cell bodies, and axon initial segments. Spinefree stellate cells possessed both type 1 and type 2 synapses on the cell body and on the dendritic shafts, the type 2 synapses being most frequent near to, the type 1 synapses far from, the cell body. Synapses formed by the axon terminals of nine cells were located, and were classified as type 1 or 2 on the basis of the thickness of the postsynaptic opacity. The axons of pyramidal and spiny stellate cells formed type 1 synapses; those of spinefree and sparsely spinous cells formed type 2 synapses. In the 12‐day old to 5‐week old kittens, cells corresponding to the spinefree cells of the adult cat did possess spines. These transient spines were commonest on and near the cell body, and were the sites of type 1 synapses. On these and other grounds, the spiny stellate cells of layer 4 were concluded to be closely related to pyramidal cells, and quite distinct from the class of spinefree and sparsely spinous cells. The interpretation of conventional E.M. preparations of cerebral cortex is discussed.