Factors influencing oxygenation of the arterial blood were studied, during routine anaesthesia, in thirty-six patients anaesthetized with halothane and allowed to breathe spontaneously, with a mean minute volume of 5 l./min. There was evidence of an unsteady respiratory state during the first hour of anaesthesia. Oxygen consumption was 87 per cent of basal. Physiological deadspace amounted to 33 per cent of the expired tidal volume (all patients intubated). At high levels of inspired oxygen concentration, the mean alveolar-arterial Po, gradient was 184 mm Hg, corresponding to a shunt of 14 per cent of pulmonary bloodflow. At lower levels of alveolar Po 2 , the alveolar-arterial Po, gradient diminished but was above the value which would be caused by a shunt of 14 per cent. This was probably due to uneven ventilation perfusion ratios (maldistribution) corresponding to a calculated venous admixture rising as high as 30 per cent. It is concluded that, to ensure the maintenance of a normal arterial Po a in the majority of patients, the alveolar Po, is required to be as high as 200 mm Hg and this needs an inspired oxygen concentration of 35 per cent under the conditions investigated in this study. Dr. Beddoes, quoted by Sir Humphry Davy The present study was designed to make a (1800), was probably the first to point out that simultaneous assessment of the principal factors the inhalation of nitrous oxide might result in influencing arterial oxygenation during anaesdesaturation of the blood. Since that day a great thesia with spontaneous respiration. Analysis of many publications have described desaturation arterial blood together with inspired and expired during the course of anaesthesia. Unfortunately, air (sampled simultaneously) has given informathese studies have seldom given sufficient infor-tion on the interplay of oxygen consumption, inmation to indicate the precise cause of the hy-spired oxygen tension, alveolar ventilation, shuntpoxia, since a considerable number of factors can ing and maldistribution on the arterial oxygen influence oxygenation. The position is somewhat tension. From the results it has proved possible clearer when artificial ventilation is employed, and to formulate some recommendations as to the studies by Campbell, Nunn and Peckett (1958), inspired oxygen concentration required under the Frumin et al. (1959) and by Stark and Smith conditions of the study. (1960) have denned some of the more important DEFINITION OF TERMS factors influencing arterial oxygenaaon under these circumstances. When the patient is allowed The terms "alveolar" and "deadspace" are now to breathe spontaneously, however, very little is confused by alternative definitions. Throughout known of the derangement of physiological fac-this P 3 !** ^ ** defincd M follows: tors influencing arterial oxygenation, although it Alveolar gas refers, not to end expiratory, but to is established that appreciable shunting may occur "ideal" alveolar gas as defined by Riley et al. (Stark and Smith, 1960) and that saturation may (1946). Ar...