Most paroxysmal forms of clinical supraventricular tachycardia (SVT) are likely due to re-entrant excitation. Electrophysiologically demonstrated mechanisms for re-entrant SVT include, in descending order of importance, atrioventricular (AV) node, AV node and accessory (AV bypass) pathway, sinus node, or atrial re-entry. Except for sinus node re-entry, none of these mechanisms for re-entrant SVT can be reliably reproduced in animal models. The authors suspected, however, that anesthetic effects on atrial and AV nodal electrophysiologic properties might be used to predict their actions against suspected re-entrant SVT. Awake-to-anesthetized (1.2 and 1.6 MAC) comparisons for the effects of enflurane (ENF), halothane (HAL), and isoflurane (ISO) on atrial and AV nodal electrophysiologic properties were made in ten chronically instrumented dogs. Studies were carried out with and without pharmacologic autonomic blockade with atropine, propranolol, and hexamethonium. By ANOVA, significant (P less than 0.05) effects of the anesthetics included: prolongation of AV nodal conduction time and the Wenckebach point in dogs with autonomic blockade (ENF, HAL, ISO); increased atrial effective and functional refractory periods in dogs without autonomic blockade (ENF, ISO); increased atrial functional refractory period in dogs without autonomic blockade (HAL); increased AV nodal functional refractory period in dogs with and without autonomic blockade (ENF, ISO), or with autonomic blockade (HAL). Sinus node re-entry, manifest by atrial echo beats during high right atrial stimulation, could be demonstrated in several dogs of each anesthetic test group during awake electrophysiologic testing. All anesthetics, with or without autonomic blockade and autonomic blockade in awake dogs, invariably abolished such re-entry. It is concluded that any anesthetic that increases atrial and AV nodal refractoriness should not be conducive to SVT caused by AV node or atrial re-entry. All of the anesthetics tested also appear effective against sinus node re-entry in dogs in which this mechanism can be demonstrated. Finally, no conclusions can be reached concerning anesthetic effects on re-entry requiring participation of both AV node and AV bypass pathways, since anesthetic effects on AV bypass pathways were not tested.