Summary The measurement of arterial pCO2 and expired CO2, by means of formula established by Enghoff, Rossier and co., Riley and co., allows to determine alveolar ventilation and physiological dead space, which represent very important functional data in respiratory physiology. The physiological and anatomical dead spaces in healthy subjects at rest are closely related, but they differ notably during hyperventilation, work and in pathological conditions. Variation of physiological dead space results from the fact that it is conditionned by anatomical dead space, by total ventilation, respiratory frequency, intensity of exchange of gases and alveolar ventilation. The existence of these various factors and their susceptibility to vary in opposite direction allows to explain that it is quite impossible to look upon absolute values of physiological dead space as an easy test of pulmonary function. On the other side, repeated measurements in the same subject on two occasions have shown that, despite the fact that some of the data used in the determination of the dead space remain constant, variation of respiratory frequency and of tidal air arc susceptible of provoking considerable modifications of the results, especially in pathological states. This emphasizes the importance of experimental conditions and decreases the value of this single determination On the other hand relationships between alveolar ventilation and global ventilation and especially relationships between ventilation of dead space per minute, and alveolar ventilation are more constant and their determination gives better account of the functional value of the lung, without however any sharp limits between physiological and pathological area. The relationships between these different data and residual air still need new investigation.