The effects of glucagon on tissue and plasma cyclic AMP levels have been investigated in rabbits anesthetized with urethane. Glucagon (2 nmole/kg.) caused at least a twofold increase in hepatic cyclic AMP, which reached a peak within two minutes and declined to basal values after 40 minutes. Plasma cyclic AMP also increased at least twofold, reaching a peak at 10 minutes and declining to basal values after 60 minutes. Glucagon (20 nmole/kg.) stimulated hepatic and plasma cyclic AMP in a manner indistinguishable from that observed at the lower dose. Hepatectomy abolished the plasma cyclic AMP responses to glucagon, and no significant stimulation of cyclic AMP concentration was noted in the heart, adipose tissue, small bowel, or kidney. Cyclic AMP hydrolysis was estimated in blood taken before and after administration of glucagon. Glucagon (2 nmole/kg.) increased cyclic AMP hydrolysis slightly, but this was explained by the raised cyclic AMP levels. By contrast, cyclic AMP hydrolysis increased two-to-threefold in blood taken 20 and 40 minutes after glucagon (20 nmole/kg.). The higher dose of glucagon also stimulated cyclic AMP hydrolysis in crude liver homogenate, which could not be explained by increases in cyclic AMP concentration. The increase in cyclic AMP hydrolysis observed in blood and liver may partly explain the failure to show additional stimulation of hepatic and plasma cyclic AMP levels with the higher dose of glucagon. Despite the changes in cyclic AMP hydrolysis, a highly significant correlation was observed in individual rabbits between the hepatic and plasma cyclic AMP responses to glucagon (2 and 20 nmole/kg.), when these were calculated as incremental areas above mean basal levels. It is suggested that measurement of plasma cyclic AMP levels after stimulation by glucagon may be an accurate index of the hepatic cyclic AMP response to glucagon in vivo.