Thirty-two females, ASA class I, scheduled for termination of pregnancy before the 13th week participated in the study. After induction of anaesthesia, a soft plastic catheter was inserted in the oropharynx and the patients were postoperatively monitored with a capnograph, which sampled in- and expired air from the catheter. In a randomized cross-over design the patients received oxygen therapy by the Hudson face-mask and a nasal cannula. Humidified and heated oxygen was given in both cases, using a flow of 15 litres per minute. The nasal cannula resulted in a significantly higher expired end-tidal O2-fraction (FETO2) compared to the Hudson mask. During the nasal cannula period the respiratory rate was significantly lower than during the face mask period. The end-tidal CO2 pressure (PETCO2) remained constant.