It has long been assumed that the superior vocal ability of the trained professional singer arose from a higher than average breathing capacity and consequent above-normal ventilatory efficiency. However, until now, it has not been clear whether this presumed superior pulmonary capacity and breathing efficiency arose from training, from heredity, or from other factors. To clarify the role of training (and by inference that of other factors also) upon the capacity for singing, various indices reflecting static lung volumes, as distinguished from dynamic parameters measured during the act of singing, in trained professional singers, students of voice and subjects with no vocal training, were compared. Results indicated that contrary to reports by others, there were no significant differences in the total lung capacity (TLC) of the trained professional singer and that of the other two groups when allowances were made for age and sex; but when the ability to mobilize or utilize TLC was compared, it was found that the trained singer was much better able to do this than either of the other two groups. Specifically, it was found that the ratio of the residual lung volume (RV) (the amount of air remaining in the lungs at the end of a total voluntary expiration) to TLC was lower in the trained singer than in the students of voice, and that these students, in turn, had a lower RV/TLC ratio than the untrained subjects. These findings, therefore, suggest that the increased singing ability of the trained professional singer arises in large part from the ability to increase breathing efficiency by reducing the residual lung volume and, further, that this ability tends to improve with length of vocal training.