By use of highly sensitive radioimmunoassays, 3':5'-cyclic AMP (cAMP) and 3':5'-cyclic GMP (cGMP) were measured in individual layers of light-and dark-adapted rabbit retinas, and the effects of ischemia were determined. In lightadapted retinas, cGMP levels ranged 50-fold, with over 90% of the total concentrated in the photoreceptor cells. The layer of outer segments contained 95 gmol/kg of dry weight, or three times the concentration present in the remainder of the photoreceptor cell layers. By contrast, levels of cAMP varied only 4-fold; the lowest level (6 gmol/kg of dry weight) was found in the outer segment layer and the highest level (22 ;mol/kg of dry weight) in the inner segment layer of the photoreceptor cells. Dark adaptation elevated cGMP levels only in retinal layers containing photoreceptor cells, and the greatest proportional increase was observed in the synaptic layer of photoreceptor cells. Dark adaptation also caused increases of cAMP that were restricted to the outer plexiform and outer nuclear layers. Ischemia lowered cGMP levels, but only in retinal layers containing photoreceptor cells, and elevated cAMP levels, primarily in the inner layers of the retina. emia was studied. The findings demonstrate that the cyclic nucleotides are discretely localized in retinas and that both light exposure and ischemia produce regionally selective changes in their levels. MATERIALS AND METHODS Tissue Preparation. Randomly bred, 6-month-old pigmented rabbits were obtained locally. Light adaptation consisted of 1 hr of exposure to two 100-watt tungsten bulbs about 25 cm above the rabbits. After the photopic stimulation, the rabbits were killed by decapitation. The eyes were removed, and after various ischemic periods in room light, were quickly frozen in liquid N2 cooled to its freezing point by partial evaporation under reduced pressure. Consequently, the liquid N2 does not boil when the eyes are immersed, and freezing is quicker. Dark adaptation consisted of an overnight period in cages wrapped with black cloth and placed in a completely dark room. After decapitation in this room, the dark-adapted eyes were dissected free with the aid of an infrared light source and an infrared image converter, and after various periods, were frozen in the dark as above. Under the most ideal conditions it required 1 min (46-69 sec) to remove and freeze either the light-or dark-adapted eyes. All eyes were stored at -70°until sectioned. Tangential sections (6 ,m) of the retinas were cut at -20°and dried at -400 under reduced pressure. Sections were taken from the paracentral regions of the retina, and care was taken to avoid that portion underlying the myelinated fiber bundle. Paracentral areas of rabbit retina are poorly vacularized (9). Samples (0.03-0.9 ,g dry weight) were dissected from the individual retinal layers and weighed on a quartz-fiber balance (10). Stained, undissected retinal sections were used to help identify individual layers.Assay of cGMP and cAMP. The procedures used for extraction and radioimmunoas...