A pattern of alternating dark and pale bands was observed in the striate cortex of the macaque monkey. The bands, which ran parallel to the surface, were seen in tangential sections stained with a reduced silver method for normal fibers and were most clear in layer 4C alpha, immediately deep to the line of Gennari. The dark bands were about 300 mu wide and showed blind endings and bifurcations. The light bands were about 50 mu wide and did not branch or terminate within area 17. Because the dark bands were similar in width to the bands of terminal degeneration which have been shown to result from single-layer lesions of the lateral geniculate body, it seemed possible that they corresponded to ocular dominance columns. To test this idea, the boundaries of ocular dominance columns were marked in a physiological experiment: tangential electrode penetrations were made in an anesthetized monkey and, as the electrode was advanced horizontally in the fourth layer, the eye preference of single units and of the background activity was monitored. Small electrolytic lesions were placed at the points where a change in eye preference occurred. The brain was subsequently fixed, sectioned tangentially and stained with the silver method. All the lesions--total of 12 --fell directly on the pale bands. Moreover, the electrode had not passed over any pale bands without a lesion being placed. It was concluded that the dark bands do correspond to single ocular dominance columns and the pale bands to the boundaries between columns. The banding appearance is due to a greater density of tangential fibers within columns than at the borders of columns. These tangential fibers are in part the preterminal arborizations of geniculocortical axons, since some of them have been shown to degenerate after geniculate lesions. The ocular dominance columns were mapped for most of the striate cortex, using serial tangential sections stained with the silver method. The overall pattern was ,imilar in several monkeys, though the details of the branching arrangements varied from animal to animal. The columns met the 17-18 border at right angles. On the outer surface of the hemisphere the columns converged from the 17-18 border, turned medially with repeated fusions of columns, and streamed over the lip of the calcarine fissure. In the roof of the fissure they met a second system of columns oriented parasagittally. In terms of the visual field, the columns ran roughly horizontally for the central 10 degrees of the field, and circumferentially beyond that. The columns were not mapped in the stem of the fissure, the area corresponding to the far periphery of the field. The constancy of column width across the cortex probably allows a functional matching between ocular-dominance and orientation columns.