The effects of autoimmune (or idiopathic) thrombocytopenic purpura (ATP) on 51 pregnancies in 29 women is presented. There is no convincing evidence that the clinical course of ATP is influenced by pregnancy. There was no increased incidence of obstetric complications. No problems were encountered following spontaneous vaginal delivery or low forceps delivery. There were no perinatal deaths. Twenty newborn infants were studied in detail and significant thrombocytopenia was present in half of them. The severity of thrombocytopenia in the newborn correlated closely with the severity of maternal disease; those women in clinical remission following splenectomy but with presumed high levels of antiplatelet antibodies were those most likely to be delivered of affected newborn infants. The cord blood platelet count was the most reliable guide to the potential severity of the neonatal thrombocytopenia. Platelet counts were lowest between the second and fourth days of neonatal life. The management of the severely affected infant is discussed.