Cortical subdivisions based on cytoarchitectural and myeloarchitectural observations of normal tissue were correlated with the topography of corticocortical connections in the visual system of the Pakistani hedgehog. Large subpial aspiration lesions were made in both visual and non-visual cortical regions to determine the areal limits of the corticocortical connections of the visual cortex. Subsequently, discrete electrolytic lesions were placed within the visual cortex. After appropriate survival periods, the brains were processed and stained with the Fink-Heimer technique. The results of these studies show that the visual cortex may be subdivided into four distinct regions from lateral to medial; the lateral parastriate cortex, the lateral and medial part of striate cortex, and the medial parastriate cortex. Within these regions, interhemispheric connections between visual cortices arise mainly in the lateral striate and lateral parastriate regions and terminate in a single band within the lateral portion of the cytoarchitecturally defined striate cortex. These corticocortical projections, therefore, substantially overlap with the geniculostriate projections. Lateral striate cortex and lateral parastriate cortex project in a reciprocal fashion that correlates well with the physiologically defined mirror image representation of two retinotopic maps of the binocular visual field on cortex. These connections are reflected about a line that is closely correlated with the medial edge of the band of commissural axon terminals that is located within the lateral striate cortex, instead of corresponding exactly with the striate-parastriate border as they do in other mammals. Medial striate cortex projects to medial parastriate cortex, indicating that the monocular portion of V I is related to a separate secondary area of cortex on the medial wall of the hemisphere.