In rat liver perfused in situ stimulation of the nerve plexus around the hepatic artery and the portal vein caused an increase in glucose output and a shift from lactate uptake to output. The effects of nerve stimulation on some key enzymes, metabolites and effectors of carbohydrate metabolism were determined and compared to the actions of glucagon, which led to an increase not only of glucose output but also of lactate uptake. 1. Nerve stimulation caused an enhancement of the activity of glycogen phosphorylase a to 300% and a decrease of the activity of glycogen synthase I to 40%, while it left the activity of pyruvate kinase unaltered. Glucagon, similarly to nerve action, led to a strong increase of glycogen phosphorylase and to a decrease of glycogen synthase; yet in contrast to the nerve effect it lowered pyruvate kinase activity clearly. 2. Nerve stimulation increased the levels of glucose 6-phosphate and of fructose 6-phosphate to 200% and 170%, respectively; glucagon enhanced the levels to about 400% and 230%, respectively. The levels of ATP and ADP were not altered, those of AMP were increased slightly by nerve stimulation. 3. Nerve stimulation enhanced the levels of the effectors fructose 2,6-bisphosphate and cyclic AMP only slightly to 140% and 125%, respectively; glucagon lowered the level of fructose 2,6-bisphosphate to 15% and increased the level of cyclic AMP to 300%. 4. In calcium-free perfusions the metabolic responses to nerve stimulation showed normal kinetics, if calcium was re-added 3 min before, but delayed kinetics, if it was re-added 2 min after the onset of the stimulus. The delay may be due to the time required to refill intracellular calcium stores. The hemodynamic alterations dependent on extracellular calcium were normal in both cases. The activation of glycogen phosphorylase, the inhibition of glycogen synthase and the increase of glucose 6-phosphate can well explain the enhancement of glucose output following nerve stimulation. The unaltered activity of pyruvate kinase and the marginal increase of fructose 2,6-bisphosphate cannot be the cause of the nerve-stimulation-dependent shift from lactate uptake to output. The very slight increase of the level of cyclic AMP after nerve stimulation cannot elicit the observed activation of glycogen phosphorylase.(ABSTRACT TRUNCATED AT 400 WORDS)