A 32-year-old chronically relapsing depressed male patient with a mild mental handicap had tried different forms of pharmacotherapy which were either not tolerated or failed to prevent recurrences of episodes of psychotic depression. The use of maintenance ECT as the mainstay of the therapeutic regime led to a marked consistent clinical improvement.

Units to admit psychiatrically ill mothers with their babies to psychiatric hospitals were pioneered follow ing Main's description (1958) of the treatment of mothers with their children at the Cassel Hospital. Particular interest in puerperal psychosis and dis orders of bonding and attachment led to units usually providing for babies below six or 12months. Suggesting older infants should also be admitted, Main referred to the benefits in admitting parents to paediatric wards with ill children, and described how toddlers had been admitted with their mothers at the Cassel. Another early unit at Banstead Hospital cared for schizophrenic mothers with their babies and found that joint admissions lead to faster recovery, lower relapse rates and more babies continuing to live with their mothers (Baker et al, 1961).

We introduce the idea of a professionally isolated and apparently disconnected agency. We describe how health professionals working in settings with good multi-agency links may not recognize the effect of the isolated agency system on the presentation of problems. In particular, we describe the setting of the Sexually Transmitted Diseases clinic (STD) in which confidentiality can be turned into secrecy by the patients at the service of their denial of the seriousness of their problems. The isolation of the STD clinic as an agency resonates with the patients' isolation from their own feelings and emotional contact with people. It also isolates the professional and the therapeutic network. By way of clinical examples, we describe how this may deskill the staff and lead to dangerous situations.

No abstract

istic vaginal bleeding associated with cervical cancer.' In our experience substantial antepartum haemorrhage rather than a bloody cervical smear is the usual presenting symptom when cervical cancer complicates pregnancy.In the National Maternity Hospital in Dublin between 1984 and 1990, 1479 There were no significant differences between the groups in any of the factors examined. In particular, the age at the menopause and the range of ages at the menopause were not significantly different between the left handed and right handed groups.We conclude that there is no evidence of a relation between handedness and age at the menopause in a white population.