This essay summarizes 16 reports, published since 1956, that describe the effects of hyperoxia on exercise endurance in persons with COPD who have severe airflow obstruction (ie, FEV(1) < 1.0 L or < 39% of predicted) and mild hypoxemia at rest (ie, PaO(2) > 62 mm Hg or arterial oxygen saturation [SaO(2)] measured by pulse oximetry of > 91%). The term hyperoxia is used because, in a proportion of study participants, oxygen administration increased exercise endurance in a dose-dependent fashion, up to a fraction of inspired oxygen of 0.5 or a flow of 100% O(2) of 6 L/min. The process appears to be dependent on an increase in PaO(2) rather than on the restoration of SaO(2) to normal levels. The results of pulmonary function tests were not predictive of response. Increased exercise performance was associated with a decrease in dyspnea, respiratory frequency, and minute ventilation. The slowing of respiratory frequency and the decrease in pulmonary air trapping likely accounted for the decrease in dyspnea. Slowing of the respiratory rate, which occurred at the expense of the retention of CO(2), is most likely due to a hyperoxia-induced decrease in chemoreceptor ventilatory drive from the aortic and carotid bodies. Research is called for to determine the following: (1) the prevalence of COPD patients who have severe airflow limitation accompanied by mild hypoxemia; (2) the proportion of these patients who show improvements in exercise performance during a test of hyperoxic exercise; and (3) whether enhanced exercise performance during a brief test translates into a meaningful increase in the ability to perform the activities of daily living.