A series of 102 patients with a multiple pregnancy (including three sets of triplets) delivered vaginally is presented. An epidural block was provided for 55 of these patients, and comparisons were made between these patients and those who did not have an epidural block. There was a higher incidence of low Apgar-minus-colour (A-C) scores at one minute among the second twins of at least 36 weeks gestation in the epidural series, but no other remarkable contrasts in the (A-C) scores of the respective group of infants. In the non-epidural series two second twins of at least 36 weeks gestational age apparently suffered severe cerebral trauma. Of the 91 liveborn infants in the non-epidural series, five died (two with intracranial haemorrhage, three with hyaline membrane disease); of the 112 liveborn infants in the epidural series, two died (both with hyaline membrane disease). The incidence of instrumental delivery of infants presenting by the vertex, and of breech extraction, was markedly higher in the epidural series. The mean interval between full dilatation of the cervix and delivery of the first twin was greater in the epidural series, but the mean interval between delivery of the first and second infant was smaller in the epidural series. It is concluded that the provision of an epidural block for labour and delivery to patients with a multiple pregnancy is beneficial to the infants.