Summary1. Experimental cardiac arrhythmias were produced in dogs anaesthetized with pentobarbitone. Ventricular arrhythmias were induced by strophanthin-K, light petroleum plus adrenaline or coronary ligation procedures. Atrial flutter was induced by an injury-stimulation technique. The acetylcholine and glycogen concentrations of the atria and ventricles were estimated. 2. Physostigmine pretreatment (01 mg/kg) significantly reduced the incidence of ventricular arrhythmias after myocardial ischaemia but had no effect on any of the other arrhythmias. 3. Physostigmine markedly increased the acetylcholine concentrations of atria and ventricles in control dogs, to nearly the same extent. Physostigmine had no effect on ventricular acetylcholine concentrations in dogs treated with strophanthin-K and light petroleum plus adrenaline but in the coronary ligation group it caused a significant increase in the acetylcholine concentrations of both atria and ventricles, and of atrial acetylcholine only in the injury-stimulation group.4. All the arrhythmias produced marked glycogenolysis of both the atria and the ventricles, to nearly the same extent. Although physostigmine produced marked glycogenolysis in the control dogs it significantly inhibited cardiac glycogenolysis after light petroleum plus adrenaline, atrial glycogenolysis after strophanthin-K-induced arrhythmias and ventricular glycogenolysis after myocardial ischaemia.5. There appears to be a possible correlation between the increase in the acetylcholine concentration of the ventricles and anti-arrhythmic actions of physostigmine, but there is a less clear correlation between changes in the glycogen concentration of ventricles and the anti-arrhythmic action.