Seizure durations in electroconvulsive therapy (ECT) were determined by both EEG and BP cuff methods in 20 patients receiving 225 seizures. The estimates were highly correlated, with cuff estimates 10% shorter than those by EEG. Routine monitoring of seizure duration by the BP cuff method is recommended, particularly in unilateral ECT. Monitoring seizure time should resolve controversies over the efficacy of different forms of ECT, as well as improve clinical practice where disturbing reports of missed seizures are now documented.