Right atria from excised beating dog hearts were immersed in Tyrode solution with the endocardial surface of the anterior wall exposed. Glass microelectrodes were used to impale fibers in different anatomical areas of this preparation. Fibers with action potentials similar in contour to those of ventricular Purkinje fibers were found along the caval border of the crista terminalis in the area of the posterior internodal tract. The action potential of these fibers (plateau fibers) had a resting potential of -85 to -95 mv, a sharp spiked overshoot, a long plateau phase, and inherent diastolic depolarization. The maximum velocity of the action potential upstroke of plateau fibers was consistently greater than that of simultaneously recorded fibers which lacked a plateau phase (regular fibers). Epinephrine or isoproterenol produced an increase in both rate and magnitude of diastolic depolarization of plateau fibers and on occasion converted them to true pacemakers. Acetylcholine accelerated repolarization of plateau fibers with disappearance of the plateau. Increases in extracellular concentration of potassium ions from 2.7 to 10.8 mM rendered regular atrial fibers inexcitable, but plateau fibers continued to show action potentials. Plateau atrial fibers possess several characteristics exhibited by specialized conducting and impulse generating fibers. The possibility that these fibers constitute the posterior internodal tract and function in preferential conduction of excitation to the A-V node was discussed. ADDITIONAL KEY WORDShigh extracellular potassium concentration posterior internodal tract epinephrine diastolic depolarization acetylcholine latent pacemaker activity isoproterenol• While it is generally accepted that fibers of the sJnoatrial (S-A) node serve as the pacemakers in the normal mammalian heart, the manner in which excitation spreads from this node through the atria and to the atrioventricular (A-V) node is not completely settled. Lewis et al. (1) concluded from studies using surface electrodes that excitation spreads radially from the S-A node through a syncytium composed of homogeneous tissue. The concept of radial spread of the impulse was challenged by Eyster and Meek (2-5) who presented evidence for spread of excitation from the S-A node to the From the Department of Physiology, University of Wisconsin Medical School, Madison, Wisconsin 53706.This investigation was supported in part by grants from the Wisconsin Heart Association and U. S. Public Health Service Grants 5-T1-HE53705-08 and 2-T1-HE5540-06.Accepted for publication June 27, 1968.CiratlM 'on esearch, Vol. XXIII, September 1968 A-V node along a preferential pathway. Also evidence for rapid spread of the impulse from the right to the left atrium was given by Bachmann (6) who described a special bundle of tissue linking the atria ventrally. Both anatomical and electrophysiological evidence has accumulated to support the contentions of Eyster and Meek for a preferential internodal pathway. In a recent review, Robb and Petri (7...

Spontaneously beating Purkinje-papillary preparations stimulated transiently at 60-180 beats/ min exhibited typical overdrive suppression of spontaneous rate in normal Tyrode's solution. When superfused with ouabain (2 X KH g/ml), the same preparations showed a progressive diminution of overdrive suppression and, ultimately, an acceleration of spontaneous rate in the immediate postoverdrive period. Usual effects of stimulation frequency on electrogenic activity associated with automaticity in the Purkinje fiber were altered significantly during the administration of ouabain. With ouabain, rapid stimulation reduced the maximum diastolic potential and the threshold potential by approximately the same amount and increased the rate of diastolic depolarization. During overdrive, the minimum diastolic potential decreased as a function of maximum diastolic potential, rate of diastolic depolarization, and cycle length. The magnitude of all effects increased directly with the stimulation frequency and the exposure time. At any given time, frequency-induced changes were quickly reversible on return to a lower frequency. However, a single early depolarization during recovery transiently reevoked changes in diastolic depolarization similar to those seen during rapid drive. The increase in the rate of diastolic depolarization appears to be the major change responsible for frequency-dependent enhanced automaticity. The effects observed during transient rapid stimulation resemble the electrophysiological alterations responsible for the final toxic effects of ouabain but occur well before any glycoside-induced change in the spontaneous rate. These effects appear to explain the frequency-dependent acceleration of ventricular pacemakers and the phenomenon of repetitive ventricular response previously described in intact animals during digitalization. KEY WORDSoverdrive suppression postoverdrive acceleration reentry ventricular automaticity digitalis intoxication repetitive ventricular response From the

Spontaneous changes in ventricular automaticity were determined from changes in idioventricular rate in eight dogs with heart block. The mediating effects of imposed rate were assessed in terms of (1) changes in post-pacing depression after ventricular pacing in 14 dogs and (2) changes in vagally induced ventricular escape time after atrial pacing in five dogs with intact AV conduction. Spontaneous ventricular automaticity decreased throughout the major portion of digitalization. The transition to a ventricular tachycardia was late and abrupt. Transient increases in rate transiently enhanced ventricular automaticity throughout digitalization. In animals with heart block, this was recognized at 38 ± 3% (SE) of the toxic dose of ouabain as blunting of postpacing depression, i.e., a significant decrease in the depression of the first postpacing beat. Blunting was progressive and was replaced at 72 ± 3% of the toxic dose by post-pacing acceleration, i.e., the cycle length of the first post-pacing beat became shorter than the average pre-pacing cycle length. In vagal stimulation experiments ventricular escape time decreased earlier during atrial pacing at 180-190 beats/min (74 ± 3% of the toxic dose) than during atrial pacing at 140-150 beats/min (92 ±3%). Early ventricular escape was not seen after pacing at 100 beats/min. These findings indicate that heart rate is a critical determinant of early increases in ventricular automaticity during digitalization.