A method of determining the lung volume has been described in the preceding paper (Herrald and McMichael 1938). During the course of an investigation on chronic pulmonary disease in the South Wales coalfield, sixty-six normal adult male subjects were studied as controls, and the results of this part of the investigation are here presented. The only previous work on comparable numbers was carried out in Rochester, New York, by Hurtado and his co-workers, who examined a hundred young adult normals (fifty males and fifty females) and, more recently, fifty middle-aged normal males, in the recumbent position (Hurtado and Boller 1933; Hurtado, Fray, kaltreider and Brooks 1934; Kaltreider, Fray and Hyde 1938). They also contrasted the results on males with ten subsidiary observations, made in the sitting posture, in each of the two series (Hurtado and Fray 1933 b; Kaltreider, Fray and Hyde 1938). These workers point out that no satisfactory standards of normality for total lung volume together with its subdivisions had previously been available. Their own standards would be generally acceptable were it nor for the fact that doubt has been thrown on the validity of the method used (Lassen, Cournand and Richards 1937). The source of error in the method is briefly discussed in the preceding paper, and it is shown to be most pronounced at functional residual air values of over 3.5 l. Since most normal values of this measurement are below this figure it is not surprising that our present results show remarkable general agreement with the figures of the Rochester workers. 1.Terminology Throughout this paper we use the following terminology: 2. Methods (a)Lung volume determinations The technique adopted was that described in the preceding paper. By this method the possible error of the original Christie technique is avoided.