A comparison was made of the effects of selective ${alpha_1}-adrenoceptor$ agonist phenylephrine and selective ${alpha_2}-adrenoceptor$ agonist clonidine on endothelium-containing and endothelium-denuded rings of the rat aorta. In the case of phenylephrine, removal of endothelium increased sensitivity 2.5 fold at $EC_{50}$ level and maximum contractive response 1.4 fold. In the case of clonidine, which gave only 15% of maximum contractive response given to phenylephrine on endothelium-containing rings, removal of the endothelium increased sensitivity 5.6 fold at $EC_{50}$ level and maximum contractive response 5 fold, which was about 55% of that given by phenylephrine. In endothelium-denuded ring, phenylephrine-induced contraction tended to be more increased in tonic contraction than in phasic contraction as compared to that in endothelium-containing ring, while clonidine-induced contraction was monophasic and was increased only in tonic contraction. In the calcium-free solution or in the presence, of verapamil, contraction stimulated by clonidine was almost abolished while that stimulated by phenylephrine produced only phasic contraction. The depression of sensitivity to these agonists in rings with endothelium appeared to be due to the vasodepressor action of endothelium derived relaxing factor (EDRF), because hemoglobin, a specific blocking agent of EDRF, abolished this depression. It is unlikely that the endothelium-dependent relaxation was due to stimulation of release of EDRF, because clonidine did not produce endothelium-dependent relaxation in 5-hydroxytryptamine-precontracted ring even when its contractile action was blocked by the ${alpha_1}-adrenoceptor$ antagonist, prazosin. When the efficacy of phenylephrine was reduced to about the initial efficacy of clonidine by pretreatment with dibenamine, the contraction-response curves for phenylephrine became very similar to the corresponding curves obtained for clonidine before receptor inactivation. In the dibenamine-treated rings, contraction of phenylephrine was abolished in calcium-free solution or in the presence of verapamil like that obtained for clonidine before receptor inactivation. These results suggest that EDRF spontaneously released from endothelium depress contraction more profoundly in a case of an agonist with low efficacy and the phenylephrine-induced contraction was totally dependent on extracellular calcium as was that obtained for clonidine when the efficacy of phenylephrine was reduced to that of clonidine by irreversible inactivation of ${alpha_1}-adrenoceptor$ with dibenamine.