In the six years that have elapsed since the tranquillizing drugs have been introduced for the treatment of mental illness, psychiatrists have been overwhelmed by the multiplicity of drugs, each claimed by its manufacturer to be more effective and less toxic than all the others. Contemporaneously, there has been a demand for the better assessment of the value of therapy in psychiatry with less reliance on clinical impression and more on the controlled trial (Eysenck, 1957; Foulds, 1958). It is quite impossible for each clinician to familiarize himself with each new drug, and it is difficult to carry out a fully-controlled double-blind trial on more than a small proportion of them. Moore (1959) has suggested a centrally-controlled trial of the tranquillizing drugs in schizophrenia by a body such as the Medical Research Council (1954) similar to the study on aspirin and cortisone in rheumatoid arthritis. Difficulties concerning diagnosis, and, probably more important, variations in the environment of the treatment situation which are of so much importance in determining the outcome of a mental illness, militate against the successful pursuance of such a scheme, and do not appear to have been overcome so far.