T he effect of prostaglandin E1 (PGE1) on osmotic water flow across toad bladder and cyclic AMP content of the mucosal epithelial cells has been determined under basal conditions and in the presence of either theophylline or antidiuretic hormone (ADH); Under basal conditions and with PGE1 concentrations from 10(-8) to 10(-5) M no evidence of stimulation of water flow was observed, and with 10(-7) M PGE1 a significant inhibition was foundmcyclic AMP content under control conditions was 8 pmol/mg protein. It was 9 at 10(-8) M PGE1, 13 at 10(-7) M, 16 at 10(-6) M, and 23 at 10(-5) M. In the presence of theophylline, 10(-8) and 10(-7) M PGE1 inhibited the theophylline-induced water flow as expected. In contrast, 10(-6) and 10(-5) M PGE1 enhanced the rate of water flow. Theophylline increased cyclic AMP content from 8 to 18 pmol/mg protein. PGE1 in the presence of theophylline caused marked increases in cyclic AMP content; The content was 23 at 10(-7) M, 41 at 10(-6) M, and 130 at 10(-5) M; Thus PGE1 stimulates theophylline-induced water flow at cyclic AMP concentrations somewhere between 23 and 41 pmol/mg. Further evidence along these lines was obtained from experiments in which the effects of PGE1 on ADH-induced water flow were studied. Inhibitory effects of PGE1 were not observed at concentrations of PGE1 which raised the level of intracellular cyclic AMP to 30 pmol/mg protein or higher. These results were obtained despite the fact that all four concentrations of PGE1 tested were found capable of inhibiting ADH-induced water flow under appropriate conditions or, in other words, were inhibiting the adenylate cyclase controlling water flow, Thus the increase in cyclic AMP content in response to PGE1 is not derived from this enzyme. Thus the stimulation of water flow by PGE1 in the presence of theophylline is thought to be caused by cyclic AMP spilling over from one compartment to the water flow compartment. No evidence was obtained to directly suggest spillover into the sodium transport compartment. Furthermore evidence is discussed to suggest that most of the cyclic AMP generated in the tissue does not originate from the enzyme controlling sodium transport. As cyclic AMP-stimulated water flow and sodium transport are thought to occur in one cell type, the granular cells, distinct pools of cyclic AMP are thought to be present in one and the same cell type. Thus one pool controls water flow and one controls sodium transport. With high concentrations of PGE1 in the presence of theophylline or high concentrations of ADH, the adenylate cyclase responsible for water flow is inhibited; However, PGE1 can stimulate a tissue adenylate cyclase to sufficiently high levels that cyclic AMP spills over into the "water flow compartment" and thus stimulates water flow.