Objective To assess the maternal and neonatal effects of upright compared with recumbent positions during delivery, in terms of defined outcome variables. Design A randomised controlled trial. Setting St Monica's Nursing Home, a midwife based maternity unit in Cape Town, South Africa. Participants Five hundred and seventeen women of low obstetrical risk assigned to deliver at the nursing home. Results The trial showed that women who adopted the upright posture for delivery experienced less pain, perineal trauma and fewer episiotomies than those who delivered in the supine position. Conclusion The data suggest that in women of low obstetrical risk, choice of posture during delivery may be encouraged.