Bevantolol was more potent in blocking the chronotropic than the hypotensive effects of isoprenaline in pithed rats. Bevantolol itself induced bradycardia, so that it was not possible to estimate the pA2 from nonparallel dose-response curves relating isoprenaline concentration to tachycardia. Bevantolol caused hypertension in pithed rats, an effect attenuated by phentolamine, implying that bevantolol may be an alpha-adrenoceptor agonist. Bevantolol potentiated the pressor effects of noradrenaline, the maximum potentiation equalling that produced by prior chemical sympathectomy with guanethidine, implying that bevantolol may block noradrenaline uptake. In isolated atria bevantolol-induced bradycardia was associated with a positive shift in take-off potential, a reduction in the maximum rate of depolarization (Vmax), and a lengthening of action potential duration (APD). No change in the slope of the slow diastolic depolarization occurred except at the highest concentration (18 mumol l(-1). In atrial and ventricular muscle bevantolol reduced Vmax and overshoot potential, implying reduction of fast inward sodium current (Class I antiarrhythmic action). In pithed rats bevantolol lengthened the P-R interval in the ECG, and produced atrioventricular (A-V) block, and bundle-branch block. In isolated A-V nodal preparations, intranodal conduction time was greatly increased, implying restriction of inward current through calcium channels responsible for nodal depolarization. Bevantolol had no negative inotropic effect in pithed rats, or in isolated atria, and did not alter the positive inotropic effect of raised extracellular calcium concentration, implying absence of restriction of current through calcium channels controlling contraction of the myocardium.