129 mothers referred to specialist psychiatric services in Birmingham and Christchurch were interviewed with the Birmingham Interview. Anxiety disorders were more frequent than depression during pregnancy, and equally frequent after delivery. The focus of pre- and post-partum anxiety may be important for psychological treatment. At a severe level, the most common prepartum theme was fear of foetal death; this was associated with a history of reproductive losses or infertility. After delivery the commonest themes were the pathological fear of cot death and fear of the criticism of mothering skills (which was a clue to a disordered mother-infant relationship). Clinicians should be vigilant for obsessional disorders, querulant (complaining) disorders, post-traumatic stress disorder, conjugal jealousy and dysmorphophobic states, which are all quite common. Patients with "postpartum depression" usually had at least one other (co-morbid) disorder, and 27% had two or more. These findings emphasize the diversity of postpartum psychiatric illness.

SummaryThe reproductive histories of 777 women suffering from migraine were compared with 182 non-migrainous women. The incidence of miscarriage, stillbirth and toxaemia of pregnancy was very similar in both groups and there was no increase in the number of congenital malformations in the children born to women who suffered from migraine compared with the control group or with the national average. It was concluded that women suffering from migraine did not have an increased risk of giving birth to children with deformity and it was unlikely that drugs most commonly used in the treatment of migraine were teratogenic.

Identifying risk factors for postpartum relapse enables us to individualise the estimation of a woman's risk and modify care plans accordingly. Duration of wellness prior to pregnancy is not associated with a lower risk of postpartum illness and so it is imperative that all women with BPAD receive referral in pregnancy.

revealed antibody and therefore a state of inm unity. The seronegative cases have remain.d seronegaFive after Ig prophylaxis and have delivered normal children with no increased levels of IgM in the cord sample. On the other hand if seroconversion had occurred following the administration of Ig the patient and her medical practitioner could be advised accordingly, but this has not occurred in our experience. It would probably be preferable, as suggested by Drs. Forrest and Menser, to have an Ig preparation with a high antibody titre, but as this is not generally available and becaus,e there is some variation in the antibody titre of the normal pooled Ig product, we have used a large dose of 3 g administered in two separate doses. Though this is considerably greater than what is advocated we think it is justifiable in the circumtances.-I am, etc.,