The location, topographic organization, and function of the middle temporal visual area (MT) in the macaque monkey was studied using anatomical and physiological techniques. MT is a small, elliptically shaped area on the posterior bank of the superior temporal sulcus which can be identified by its direct inputs from striate cortex and by its distinctive pattern of heavy myelination. Its average surface area is 33 mm2, which is less than 3% of the size of striate cortex. It contains a complete, topographically organized representation of the contralateral visual hemifield. There are substantial irregularities in the detailed pattern of topographic organization, however, and the representation is significantly more complex than that found for MT in other primates. Much of MT is devoted to the representation of central visual fields, with the emphasis on central vision being similar to that found in striate cortex. Electrophysiological recordings have confirmed previous reports of a high incidence of direction selective cells in MT. The transition in functional properties, from cells lacking direction selectivity outside MT to direction selective cells within, occurs over a distance of 0.1-0.2 mm or less along the lateral border of MT. Such a transition does not occur along the medial border, however, as the cortex medial to MT contains many cells with strong direction selectivity. Nevertheless, this region differs from MT in its myeloarchitecture, its lack of inputs from striate cortex, and the large size of its receptive fields. These results demonstrate the existence of three distinct visual areas on the posterior bank of the superior temporal sulcus which can be distinguished on the basis of both physiological and anatomical criteria.

The distribution of interhemispheric connections was studied in extrastriate visual cortex of the macaque monkey. Callosal fiber terminations were identified by staining for anterograde degeneration following transection of the splenium of the corpus callosum. Retrogradely labeled cell bodies of callosal projection neurons were identified histochemically following application of horseradish peroxidase to the cut surface of the callosum. Results were displayed on unfolded, two-dimensional representations of the cortex, which permitted spatial and topological relationships between callosal recipient and callosal free cortical regions to be discerned readily.The overall pattern of callosal inputs to visual cortex can be subdivided into nine callosal recipient zones which surround seven callosal free regions in the occipital, temporal, and parietal lobes. This pattern provides reliable and useful landmarks for identifying the borders of at least five topographically organized extrastriate visual areas. There is a pronounced dorsoventral asymmetry in callosal projections, not recognized in previous studies, which reflects an asymmetry in the organization of visual areas in dorsal versus ventral halves of the occipital lobe.The pattern of callosal fiber terminations is mirrored by a very similar distribution of callosal projection neurons. There are significant regional differences in the laminar distribution of callosal projection cells, and these differences may reflect functionally distinct cortical subdivisions.A considerable degree of individual variability was found in the relationship of callosal connections to gyral and sulcal landmarks as well as in the fine structure of individual callosal recipient strips, suggesting that each animal has a unique "callosal fingerprint." These findings emphasize the usefulness of the callosal pathway in elucidating the functional organization of extrastriate visual cortex.

No abstract