INTRODUCTIONAmong the aspects of ventilatory function of the lungs, about which exact knowledge is lacking, is that of the effectiveness of aeration of pulmonary air spaces during quiet breathing. Among the questions to be answered are:(1) Does the normal lung behave as a perfect mixing chamber, all portions sharing equally in each breath?(2) In abnormal lungs, particularly in emphysema, how large and important is the factor of inadequate ventilation of portions of such lungs?Regarding the second question, anatomical studies in pulmonary emphysema show that the bronchioles are narrow, in relation to the distended alveoli. Air aspirated from emphysematous bullae has relatively high carbon dioxide and low oxygen contents (1), indicating underventilation. Marked variations in the gas concentrations of the different parts of an expired alveolar specimen, in such subjects, have led to the same conclusion (2). The authors (3), in the course of analyzing the rebreathing method for measuring residual air, found that the usual alveolar samples represent an imperfect measure of average pulmonary gas concentration under the changing conditions of a rebreathink experiment. From these experiments, imperfect intrapulmonary gas mixture was recognized as a factor in pulmonary disability; but it could not be measured and compared with the other factors, namely, pulmonary overdistension, diminished vital capacity, and impaired gas diffusion, in these subjects.Regarding the question of intrapulmonary mixture in the normal lung, opinions are far from uniform. Haldane (4) recognized that the apparent uniformity of the alveolar air obtained by his method of sampling signified only that the technique yielded a representative sample of the lung gases. Sonne (2) found differences in the different fractions of the alveolar air sample, which, he believed, indicated that the central, less elastic portions of the lung were less well aerated than the peripheral areas.The fractionation of ordinary expired alveolar samples for this type of study has well-known limitations, namely: (1) that rarely more than one-third of the total alveolar air can be expired by any forced expiration, and (2) that the oxygen and carbon dioxide content of any fraction is related to the blood circulation of the alveoli from which it came, as well as to their aeration. For example, a high carbon dioxide content may indicate slow aeration or good perfusion, or both. To avoid these limitations, Siebeck (5, 6) and, more extensively, Roelson (7 to 9) used a foreign gas, hydrogen, in the breathing mixture. These experiments showed a greater degree of imperfect -mixture than those using air, indicating either that hydrogen diffused into stagnant air spaces, not emptied by a forced expiration, or that reduced perfusion of the poorly ventilated spaces had masked the inadequate ventilation in the airbreathing experiments.For the present study oxygen rather than hydrogen was chosen as the gas to be breathed since (1) it does not disturb normal lung function over short periods...