Much of the literature on the family of the schizophrenic patient has suggested that a number of common features are present in the personalities of the parents, and that these are significant in the aetiology of the illness. Most studies have been concerned with the parent-child relationship and only a few with the adult patient and his parents. Tietze's description of mothers as generally' over-anxious, obsessional, and domineering is typical (Tietze, 1949); and Fromm-Reichmann (1948) coined the term "schizophrenogenic" to describe such women. The more systematic work has indicated a high frequency of domineering and over-solicitous behaviour among mothers (Mark, 1953;Freeman and Grayson, 1955;Gerard and Siegel, 1950; Kohn and Clausen, 1956). Two studies, however, have produced negative evidence (Neilsen, 1954;Hotchkiss, Carmen, Ogilby, and Wiesenfeld, 1955) and the second of these was the only one in which the behaviour of the mother and patient was directly observed.Even if differences exist between the mothers of schizophrenics and other mothers, this need not be of aetiological importance. The possibility must first be excluded that behaviour such as "overprotectiveness" may be the result of the patient's unusual behaviour influencing the parents (Kasanin, Knight, and Sage, 1934). Once an illness has developed in one member of a family, a heightened level of tension is probably common. Once established, tense relationships in turn may have an important effect on the later stages of the illness. Some evidence that family relationships can influence the course of schizophrenia was provided by a previous study, which showed that re-admission of long-stay patients was related to the type of living group to which they returned (Brown, Carstairs, and Topping, 1958;Brown, 1959). Patients who lived with wives and parents showed a higher re-admission rate than those going to brothers, sisters, or more distant kin, or in lodgings. There was evidence that the risk of deterioration in clinical condition was increased when prolonged contact with close relatives in the house was unavoidable-when, for example, both patient and mother were unemployed. Results could not be explained entirely by the length or past severity of illness or by differences in clinical condition at the time of discharge; and it was concluded that it might not always be best for the schizophrenic patient to return to the close emotional ties of affection or hostility often found in parental and marital homes.These close emotional ties are not, of course, confined to households of any particular kinship category. It was therefore decided to continue the work by studying the relationships within each home to which discharged patients returned. The majority of patients in this second study were short-stay schizophrenics, and a different survey method was used. Patients and their families were interviewed at the time of discharge and during the year, if the patient was re-admitted, as well as at the end of the follow-up period. In this way the diff...