Pulmonary gas exchange is the primary function of the lung, and during my lifetime, its measurement has passed through many stages. When I was born, many physiologists still believed that the lung secreted oxygen. When I was a medical student, the only way we had to recognize defective gas exchange was whether the patient was cyanosed. The advent of the oximeter soon showed that this sign could be very misleading. A breakthrough was the introduction of blood gas electrodes that could measure the PO2 , PCO2 , and pH of a small sample of arterial blood. It was soon recognized that the commonest cause of hypoxemia was ventilation-perfusion inequality, and that this could also be responsible for CO2 retention. In the early days, the understanding of the mechanisms of pulmonary gas exchange relied on graphical analysis because the oxygen and carbon dioxide dissociation curves are nonlinear and interdependent which precluded algebraic methods. However, with the introduction of digital computing, problems that had hitherto been impossible to tackle became amenable to study. A key advance was the development of the Multiple Inert Gas Elimination Technique. Now, noninvasive methods for measuring gas exchange show promise, and the whole subject continues to develop.