The frequency and seventy of hypoxaemia during induction of anaesthesia in neonates and small infants at the Norfolk and Norwich Hospital, a district general hospital, was compared, using pulse oximetry, with that of the nearest specialist hospital, the Queen Elizabeth Hospital for Sick Children in London. There were differences in staffing and the choice of anaesthetic techniques between the hospitals. One third of the patients in both hospitals experienced desaturation of more than 5% (moderate or severe hypoxaemia) at one or more recordings during induction. The highest incidence of hypoxaemia was associated with awake intubation. There was no statistical difference in the incidence or severity of hypoxaemia between the hospitals. In the district general hospital, moderate or severe hypoxaemia of greater than 30 s duration was more likely if an anaesthetist with a regular paediatric operating list was not present at induction (p < 0.01). Copyright © 1992, Wiley Blackwell. All rights reserved