Adrenergic transmission was investigated in the dog mesenteric vein by recording electrical responses of single smooth muscle cells to perivascular nerve stimulation. Perivascular nerve stimulation generated an excitatory junction potential (e.j.p.) and a slow depolarization of the membrane. The amplitude of the e.j.p. was increased by increasing the stimulus intensity, and at high intensity, a spike potential was generated. Repetitive stimulation of the nerves showed facilitation of e.j.ps and enhanced the amplitude of slow depolarization. A linear relationship was observed between the amplitude of the e.j.p. and of slow depolarization. The slow depolarization was inhibited by application of yohimbine or phentolamine, but not by prazosin. The amplitude of e.j.p. was increased by prazosin and was decreased by yohimbine. Both e.j.p. and slow depolarization were inhibited by guanethidine or tetrodotoxin. Exogenously applied noradrenaline depolarized the muscle membrane and, in high concentrations (greater than 10(-7)M), generated slow waves. These effects of noradrenaline were blocked by yohimbine. High concentrations of prazosin (greater than 10(-6)M) showed weak inhibitory effects on the noradrenaline actions. The amplitude of e.j.p. was decreased by exogenously applied noradrenaline in a dose-dependent manner. The inhibitory effect of noradrenaline on the e.j.p. was suppressed by yohimbine, but not by prazosin or phentolamine. Phentolamine, but not prazosin, enhanced the facilitation process of e.j.ps. This effect was not suppressed by exogenously applied noradrenaline. Application of neostigmine but not of atropine, reduced the e.j.p. amplitude without affecting the slow depolarization. It was concluded that, in the dog mesenteric vein, perivascular nerve stimulation produced three types of electrical responses of the smooth muscle membrane, i.e., e.j.p., slow depolarization and spike potential. The slow depolarization was generated by activation of alpha 2-adrenoceptors. Exogenously applied noradrenaline reduced the e.j.p. amplitude through activation of prejunctional alpha 2-adrenoceptors, but the reduction may not involve alpha-autoinhibitory mechanisms.