In order to examine the significance of blood pressure elevation during electroconvulsive therapy (ECT) for posttreatment cognitive impairment, routine treatments were compared with treatments where the blood pressure elevation was attenuated by administration of trimethaphan (Arfonad). As a reliable, valid, and sensitive measure of cognitive impairment, the forgetting score in four memory tests was used. Both retrograde and anterograde amnesic effects were examined. In spite of a substantial attenuation of the blood pressure response, there was no decrease of the amnesic effects. It is concluded that low-pressure narcosis is not successful in alleviating memory disturbance after ECT. There is no support to the concept that the blood pressure elevation is relevant to the memory disturbance. Since there was a tendency toward decreased seizure duration, low pressure narcosis may be inappropriate because it counteracts a physiological compensatory mechanism to meet the increased metabolic demands during the seizure.