Binocular disparity, resulting from the projection of a three-dimensional object on the two spatially separated retinae, constitutes one of the fundamental cues for stereoscopic perception. The binocularity of cells in one hemisphere stems from two sources: i) from the ipsilateral ganglion cells in the temporal retina which converge with inputs coming from the contralateral nasal retina; the latter axons cross at the chiasma; ii) from inputs originating in the opposite hemisphere which cross in the corpus callosum. The objective of this study was to demonstrate that interactions from both types of inputs can result in the formation of disparity sensitive neurons and presumably that either type could mediate stereoperception based on disparity cues. Two types of disparity sensitive neurons were found in the normal cat: one type, showing maximal interactive effects around zero disparity responded with strong excitation or inhibition when the stimuli were in register. These neurons are presumed to signal stimuli situated about the fixation plane. The other type, also made up of two subtypes of opposed valencies, gave maximum responses at one set of disparities and inhibitory responses to the other set. These were presumed to signal stimuli situated in front of or behind the fixation plane. In the split-chiasm cat, whose cortical binocularity is presumably assured by converging ipsilateral and callosal inputs, three of the four subtypes of disparity sensitive neurons were found, the uncrossed disparity cells being absent in these animals. Moreover, stimulating each eye individually indicated that nearly 80% of the cells in normal and about 40% in split-chiasm cats were binocularly driven. However, both these figures underestimated the amount of binocular interaction in the callosal recipient zone, since stimulating both eyes simultaneously showed that a proportionately larger number of cells were binocularly driven. Disparity sensitive cells also varied as a function of ocular dominance, i.e., cells signaling the fixation plane tended to have balanced dominance whereas units preferring stimuli situated in front of or behind the fixation plane were dominated by the ipsilateral and contralateral eyes, respectively.