1 Experimental and clinical studies suggest that class I and class III antiarrhythmic drugs may be subject to pharmacological tolerance during long term treatment, leading to loss of therapeutic e ectiveness. 2 The aim of this study was to ascertain whether prolonged in vivo treatment with the Class Ia agent quinidine can modify cardiac (electrical and mechanical) responses to the drug. 3 A group of guinea-pigs (n=7) were treated intraperitoneally (q.d.) for 6 days with 75 mg kg 71 quinidine sulphate. Preliminary pharmacokinetic experiments indicated that this dose could attain plasma concentrations similar to those that are therapeutic in man (2 ± 5 mg l 71 ). A control group (n=7) received a saline solution for the same period. 4 Twenty-four hours after the last administration hearts were removed and retrogradely perfused at constant¯ow (stimulation frequency: 2.5 Hz). The following parameters were measured: maximal derivative of intraventricular pressure (dP/dt max ); coronary perfusion pressure (Cp); PR, QRS and JT intervals, on surface ECG. The e ects of quinidine on these parameters were measured at di erent concentrations (2, 4, 8, 12, 16, 20 mM) and compared in the two experimental groups. 5 In the control group quinidine decreased in a dose-dependent manner dP/dt and increased PR and QRS intervals. JT interval was increased at the lowest concentrations and decreased at the highest (biphasic e ect). Cp did not change signi®cantly. 6 In the pretreated group quinidine qualitatively produced the same e ects on dP/dt and ECG intervals as in control group. Also the magnitude of these e ects was not signi®cantly di erent between the two groups. In contrast with ®ndings in control experiments, Cp was signi®cantly decreased by increasing quinidine concentration. Mean baseline Cp was higher in pretreated than in the control group (though not signi®cantly, P=0.072) and quinidine addition abolished this di erence. Thus, it is suggested that quinidine withdrawal induced a rebound increase in coronary tone, due to the unmasking of vasoconstrictor homeostatic mechanisms elicited by the in vivo vasodilating e ect of the drug. 7 In conclusion, our data do not support the possibility that tolerance ensues during long term quinidine treatment, at least as far as electrophysiological and contractility e ects are concerned. Further experimental work is needed to explain the appearance of a coronary vasodilating e ect in pretreated hearts.