There are many reports that, through pre-and post-junctional mechanisms, sympathetic and parasympathetic (vagal) nerves can interact in the control of heart rate. The predominant interaction is accentuated antagonism (AA), where the bradycardia produced by vagal stimulation (VNS) is amplified when heart rate has been increased by sympathetic stimulation (SNS) or beta-adrenergic agonists. The acetylcholine-activated potassium current (IK,Ach), is the primary driver of vagal bradycardia. To examine the participation of IK,Ach in AA, a series of experiments was performed on isolated, double innervated, guinea-pig atrial preparations. Vagal bradycardia was elicited by 10-s trains (1, 2, 5 and 7.5 Hz) or single bursts of VNS (3 stimuli at 50 Hz) before and during acceleration of HR by either SNS (1-3 Hz) or isoprenaline (ISO), in both absence and presence of tertiapin-Q (TQ-IK,Ach blocker). When expressed as an absolute change in HR (beats/min), bradycardia produced by VNS trains was amplified (AA) at all frequencies of VNS in ISO, and at 5 and 7.5 Hz during SNS. Bradycardia in response to 1 and 2 Hz VNS was reduced during SNS. In TQ, only the bradycardia produced by 5 and 7.5 Hz VNS in ISO was amplified. The bradycardia produced by a single burst of VNS was amplified in both ISO and SNS. After TQ the bradycardia in response to a VNS burst was unchanged in ISO, while it was reduced during SNS. When these data were adjusted to account for the increase in baseline HR brought about by SNS and ISO, there was no longer evidence of AA. Diminished responses to low frequencies of VNS (1 and 2 Hz) persisted, and were also seen during IK,Ach block by TQ. We applied the same adjustment to data from 20 published studies. In 8 studies all data indicated AA; 3 studies provided no evidence for AA, and in 9 studies evidence was mixed. There is no doubt that AA can occur in the control of heart rhythm during simultaneous SNS and VNS, but conditions which determine its occurrence, and the mechanisms involved in this interaction remain unclear.