The main purpose of this study was to correlate the tangential distributions of visual callosal and striate-extrastriate connections in the rat. Cells of origin and terminations of the visual callosal pathway of one hemisphere were labeled by the anterograde and retrograde transport of horseradish peroxidase (HRP) after multiple injections of this enzyme in the contralateral hemisphere, while ipsilateral striate-extrastriate projection fields were revealed by using the autoradiographic method following single injections of 3H-proline in striate cortex. A remarkable complementarity in the distribution of both cortico-cortical pathways was revealed by superimposing in a camera lucida the patterns of callosal and striate-extrastriate projections from consecutive tangential sections processed for HRP and autoradiography, respectively. Projections from striate cortex are distributed into multiple extrastriate fields which are partially or totally surrounded by cortical strips containing dense and overlapping accumulations of labeled callosal cells and terminations. In addition to projections to the following striate recipient areas described in previous reports: posterior (P), posterolateral (PL), lateromedial (LM), laterolateral (LL), anterolateral (AL) and anteromedial (AM); projections to laterointermediate (LI), laterolateral anterior (LLa), anterior (A), medial (M) and pararhinal (PR) areas were defined in the present study. Striate-extrastriate projection fields occupy only a portion of the acallosal islands that contain them, and the location of the fields within these islands correlates with the retinotopic location of the isotope injection in striate cortex. When compared to previous physiological and anatomical maps of extrastriate visual areas in the rat, the present results indicate that the distribution of callosal connections correlates with the borders of extrastriate visual areas, and that the projection from striate cortex into these areas is retinotopically organized. Surprisingly, a direct projection from striate cortex to the head representation region in somatosensory cortex was labeled, a finding that challenges the view that primary sensory areas do not connect directly.