The effects of amoxapine (10(-7)-10(-4) M) have been studied in rat atrial fibres obtained from untreated animals and animals pretreated for 28 days with amoxapine (10 mg kg-1, i.p.). In untreated atria amoxapine reduced atrial rate, contractile force and df/dtmax, prolonged the sinus node recovery time and decreased atrial excitability. Amoxapine also decreased amplitude and Vmax of the upstroke, prolonged the duration of the action potential (APD) and effective refractory period (ERP) and reduced the resting membrane potential. During the treatment with amoxapine behavioural and cardiovascular adverse effects, including hypotension, tachycardia and prolongation of the Q-Tc, were observed. However, with the exception of the ERP which was significantly prolonged in pretreated atria, pretreatment with amoxapine did not modify the control values of the measured parameters compared to those obtained in untreated atria. Further addition of amoxapine produced similar changes in both pretreated and untreated atria. However, in contrast to untreated atria, in pretreated atria the prolongation of the ERP produced by amoxapine exceeded the prolongation of the APD and thus, the ERP/APD ratio increased. The decrease in atrial excitability was also more marked in pretreated than in untreated atria. Amoxapine inhibited the slow action potentials and contractions induced by isoprenaline in K-depolarized atria. It is concluded that the electrophysiological effects of amoxapine on rat atrial fibres are similar to those described for other tricyclic antidepressants. Possible explanations for the lower cardiodepressant activity of amoxapine are discussed.