Several studies have reported time of birth is associated with differences in obstetric practice. We investigated the relationship between timing of birth and obstetric and neonatal outcomes, to help plan working patterns under European Working Time Directive (EWTD) legislation. This was a retrospective observational study undertaken in a tertiary-level university teaching hospital. Data were derived from the labour ward register of births for all women who delivered after 24weeks gestation in 2004. Births during on-call hours refer to those that occurred at weekends and after 1630 and before 0830 on weekdays. The majority of infants, 67.3%, were born in on-call hours. Infants were more likely to be delivered by ventouse(p<0.0001), but there was no difference in caesarean section(CS) rates. 83.0% of operative deliveries performed for failure to advance in the second stage of labour took place in on-call hours, as did 77.5% of emergency CS for fetal distress. 38.9% of infants born during on-call hours on weekdays followed induced labours, compared to 24.7% of births at weekends and 17.7% of births in non on-call hours(p<0.001), while 80.0% of deliveries by emergency CS after induction occurred during on-call hours. The majority of perinatal deaths occurred among infants born during on-call hours, even when excluding congenital malformations, and most infants with low Apgar scores were born during on-call hours. Complicated deliveries were more likely to occur in on-call hours. This study confirms previous reports that time of birth impacts on neonatal outcome. Increased demands on staff working out-of-hours have implications for healthcare, staffing and implementation of new working hours under EWTD legislation.