Fifty ASA grade 1 children, who presented for dental outpatient extraction were studied. They were randomly allocated to two groups after induction: group 1 had conventional nasal mask anaesthesia and group 2 anaesthesia with a laryngeal mask. Group 2 had fewer hypoxic episodes and significantly better arterial oxygen saturations (p less than 0.01). There was no difference between the groups as regards surgical access, difficulty of extraction or bleeding. The laryngeal mask appears to provide an alternative to conventional nasal mask anaesthesia, with better overall oxygenation and would seem particularly suitable for prolonged or difficult extractions.