This investigation was undertaken to determine whether acute psychiatric admissions to hospital were in any way influenced by the premenstruum or by menstruation. For many years it has been believed that psychiatric illness was apt to exacerbate at the time of the menstrual period, and in this paper an attempt is made to determine if such a relationship exists. It has been prompted by the recognition of the premenstrual syndrome and its successful treatment (Greene and Dalton, 1953).As there are many ways in which the acutely ill can be admitted to hospital it was decided to examine admission records of two large mental hospitals (2,470 and 2,290 beds) and the mental observation ward of a London teaching hospital, and to confine the investigation to those patients whose illness was so acute that immediate admission was sought. This meant that patients who were admitted when a bed vacancy occurred were excluded, as were those admitted through an out-patient clinic. By this means the investigation was restricted to those whose admission was obtained because of the acuteness of their illness at the time of admission.The sister in charge of admissions recorded the date of the patient's last menstrual period at the time of her admission. The data were obtained from statements of patients suffering from a variety of psychiatric illnesses and excluding those too ill to give reliable information. All patients with amenorrhoea exceeding 28 days were excluded.Results For the interpretation of results the 28 days of the menstrual cycle were divided into seven four-day periods, in which days 1-4 corresponded to menstruation, days 13-16 covered ovulation, and days 25-28 the premenstruum. It was recognized that menstruation may exceed four days, but the full menstrual flow with relief of water retention usually occurs during the first four days. The relative incidence rate represents the number of actual admissions during a four-day period compared with the expected number of admissions during a four-day period (or one-seventh of total admissions). Fig. 1 gives the time in the menstrual cycle of psychiatric admissions for 276 patients and shows that the expected rate of admissions is doubled during menstruation. Fig. 2 gives the time of admissions of 36 attempted suicides and shows a relative incidence rate during menstruation of 2.75 or 39%. This figure of 39% of attempted suicides during menstruation compares with the figure of 36% reported by Heller in 1900, 22%by Ollerdorf in 1905, and 33% by Slavik in 1910 (quoted by Gregory, 1957, and supports the findings of MacKinnon and MacKinnon (1956) that successful suicides seem commoner during the luteal phase. Figs. 3 and 4 show that admissions during menstruation are greatly increased both for 185 patients with depression and for 114 patients with schizophrenia.The influence of menstruation on psychiatric admissions was evenly distributed among single, married, and widowed; among nulliparous and multiparous; and among those admitted for first and for recurrent attacks. The...