A twelve-week study involving fifty-three patients is described as taking place in a practice with a higher than average geriatric population. The purpose of the study was to substitute flurazepam for habitually used barbiturates or methaqualone/diphenhydramine. Of the original fifty-three patients admitted to the study, fifty-one completed; the two drop-outs resulting from concomitant physical illness. Eighty-four per cent of patients were successfully changed to flurazepam. Of those who did not accept flurazepam, eight per cent accepted nitrazepam, while six per cent of patients were motivated to stop all hypnotics. During the three month period of the study none of the well-known disadvantages of the barbiturates and methaqualone/diphenhydramine were seen with flurazepam. The author found flurazepam to be a very efficient hypnotic of relatively low toxicity which could be easily substituted fro barbiturates and methaqualone/diphenhydramine in the treatment of long-term insomnia.