AT THE onset of their research into double-blind studies the authors took particular note of a statement by the Committee on Research of the Group for the Advancement of Psychiatry.1This statement suggests that inert placebos, as they are frequently employed in double-blind studies, fail to reproduce the essential symptoms and side effects experienced by patients receiving an active drug; judges in turn perceive subtle differences between the "drug" and the "control" groups, hence eliminating from the design genuine "blindness." Other investigations concerning double-blind designs had suggested the use of an active control substance which would mimic the symptoms and side effects of the active drug.2-5The present investigators have reported their attempts to develop such an active control substance in two previous studies, the findings of which are summarized below.6,7 In the first of these studies, using an inactive control substance