A 32-year-old woman with prolapsed mitral valve was treated with flecainide because of an episode of primary ventricular fibrillation. The drug was chosen because several hundred short runs of ventricular tachycardia unrelated to exercise were observed during 24-hour monitoring. Oral medication at a dose of 200 mg twice daily suppressed all repetitive ventricular ectopy. Eighteen months later, however, a further 24-hour recording showed a ventricular arrhythmia which was provoked early in the morning by exercise. Because the tablets were taken late in the evening and early in the morning it was suspected that toxic levels of flecainide may have produced the arrhythmia. Measurement confirmed this suspicion. Reducing the dosage abolished the arrhythmia.