To investigate the effects of a2-adrenergic receptors on the secretions of pancreatic glucagon and insulin, clonidine, midaglizole and yohimbine were intravenously administered in four conscious sheep. Clonidine infusion at a dosage of 1.0 nmol/kg/min produced hyperglucagonemia, hypoinsulinemia and hyperglycemia. Midaglizole or yohimbine was infused for 30 min, at doses of 5, 10 and 50 nmol/kg/min during the clonidine infusion. The highest yohimbine infusion (50 nmol/kg/min) blocked the clonidine-induced responses of glucagon, insulin and glucose. On the other hand, the midaglizole (50 nmol/kg/min) infusion brought about no statistical effect on the clonidine-induced responses of glucagon, insulin and glucose. The a2-adrenergic antagonistic effect of midaglizole was clearly less than that of yohimbine in the present experiments. It is concluded that the glucagon secretion is enhanced and the insulin release is inhibited by a2-adrenergic stimulation in conscious sheep.clonidine ; midaglizole ; yohimbine ; glucagon ; insulin Recently, it has been demonstrated that adrenergic subtype receptor mechanism modulates glucagon and insulin secretion in several species. Xylazine, an a2-adrenergic agonist, caused hyperglucagonemia, hypoinsulinemia and hyperglycemia in sheep (Muggaberg and Brockman 1982). Both phenylephrine, an a1-adrenergic agonist, and clonidine, an a2-adrenergic agonist, increased plasma glucagon and lowered plasma insulin in mouse (Skoglund et al. 1987). It has been shown that yohimbine, an a2-adrenergic antagonist, eliminated the glucagon response to epinephrine infusion in rats (Patel 1984), in rabbits (Knudtzon 1984b) and in sheep (Oda et al. 1990). On the contrary, clonidine was ineffective on glucagon secretion in human (Ferlito et al. 1985), in rat islet cell (Schuit and Pipeleers 1986) and in rat pancreas ). Furthermore, studies using metoprolol, a selective /31-adrenergic antagonist, suggest that the adrenergic stimulation of glucagon secretion was mediated through /"1-receptors in rat pancreas ). Thus it may be conceivable that the conflicting results concerning the modulation of glucagon secretion with adrenergic subtype