Single unit activity of 842 cells has been recorded in cat visual cortex and analyzed with respect to vestibular induced, and spontaneous saccadic eye movements performed in the dark. This study has been done in awake, chronically implanted cats, subsequently placed in "acute" conditions to achieve the precise retinotopic mapping of the cortical areas previously investigated. In areas 17 and 18, respectively, 27% and 24% of the cells tested were influenced by horizontal saccadic eye movements in the dark (E.M. cells). In the Clare-Bishop area, the proportion of E.M. cells was 12%, while only 2% of such cells were found in areas 19 and 21. The distribution of E.M. cells in areas 17 and 18 with respect to retinotopy showed that E.M. cells were more numerous in the cortical zones devoted to the representation of the area centralis (38% in area 17, 27% in area 18) than in the zones subserving the periphery of the visual field (17% and 12%, respectively). Two of the characteristics of E.M. cell activations appear dependent on the retinotopic organization. First, larger number of E.M. cells presenting an asymmetry in their responses to horizontal saccadic eye movements in opposite directions (directional E.M. cells) were encountered in the cortical representation of the peripheral visual field. 53% of E.M. cells recorded in area 17 and 71% in area 18 were directional in the cortex corresponding to the peripheral visual field. This percentage was of 23% and 25% respectively in the cortex devoted to area centralis. Second, E.M. cells were found to have a latency from the onset of the saccade systematically larger than 100 ms (i.e., they discharged at, or after the end of the eye movement) if they were located in the cortical representation of the area centralis, while E.M. cells related to the peripheral visual field displayed a wider range of latencies (0-240 ms). Results obtained in Clare Bishop area, although limited to the representation of the peripheral visual field, were quantitatively and qualitatively similar to those observed in the homologous retinotopic zones of areas 17 and 18. It is concluded that an extra-retinal input related to oculomotor activity is sent to the cat visual cortex and is organized, at least in areas 17 and 18, with respect to the retinotopic representation of the visual field. These data support the hypothesis of a functional duality between central and peripheral vision and are discussed in the context of visual-oculomotor integration.