Recent investigators have indicated that bypass breathing achieved by chronic implantation of an upper respiratory bypass cannula in the sheep resulted in an increase in the temperature of the venous blood draining from the nasal venous plexuses to the cavernous sinus. This increase in the venous blood temperature has also been demonstrated to affect the efficiency of the countercurrent heat exchange between the arterial blood to the brain and the venous blood in the cavernous sinus with resultant increases in brain temperature and cerebrospinal fluid pressure. However, no reports exist so far on the influence of bypass breathing on the morphology and function of the eye of the sheep;The purpose of this study was to test the hypothesis that interruption of normal nasal breathing would affect the morphology and function of the eye;Chronic upper respiratory bypass cannula, with reversible inserts to allow the sheep to breathe normally through the nostrils and the tracheal opening, was surgically implanted in 38 sheep of varying ages and body weights. Changes in retinal vasculature were studied in 14 animals using an ophthalmoscope and fundic camera normal and bypass breathing. The effects of bypass breathing on the components of the electroretinogram were studied over two trials in five animals. The arterial Pco(,2), Po(,2) and pH, as well as the intraocular pressure, were measured during normal and bypass breathing. The histology of the optic nerve head and retinal layers in 22 bypassed and 10 nonbypassed (control) animals were investigated. Vascular changes observed during bypass breathing were compared to vascular changes commonly observed in mongoloids (Down's syndrome in man);Results indicated moderated congestion and engorgement of retinal veins during bypass breathing. Electroretinogram recordings showed an increase in the amplitude (P > 0.005) and a decrease in the duration of the b-wave during bypass breathing. There was no statistical difference (P > 0.51) in the intraocular pressure during normal and bypass breathing. No changes occurred in the arterial Pco(,2) (P > 0.985), Po(,2) (P > 0.98) and pH (P > 0.298) during both normal and bypass breathing. Histologic studies showed mild to marked elevation of the optic disc as well as lateral displacement of the retina during bypass breathing. Engorged blood vessels penetrated the entire layers of the retina establishing contact with the pigment epithelium in bypassed animals. The vascular patterns observed during bypass breathing were not comparable to those reported for mongoloids.