There has been a great need for a satisfactory method for the determination of hydroxyproline to facilitate the study of the composition of proteins. Lang (1) and Waldschmidt-Leitz and Akabori (2) developed a calorimetric estimation involving oxidation to pyrrole with sodium hypochlorite and color formation with isatin or p-dimethylaminobenzaldehyde. However, oxidation was said to be incomplete and a correction factor was necessary. Dakin (3) and Bergmann (4) estimated hydroxyproline by isolation procedures. These procedures also required the use of large correction factors, as well as relatively large quantities of protein. McFarlane and Guest (5) devised a calorimetric method for hydroxyproline involving sodium peroxide oxidation and color formation with copper and isatin. This method yielded low values according to Devine (6), and in our hands similar results were obtained. The unique high hydroxyproline content of collagen suggests the desirability of an accurate method for the determination of this amino acid in small quantities as a means of estimating the amount of collagen or gelatin in a mixture of proteins. Because hydroxyproline so far has not been found to be a nutritive requirement for a microorganism, a chemical procedure is required.

A short and simple calorimetric method is here described that is applicable to the determination of hydroxyproline in hydrolysates of 40 to 100 y of collagen with a reproducibility of f2 per cent (Table II) and an accuracy of f2 per cent (Table I) as judged by recovery of hydroxyproline from elastin hydrolysates and from an amino acid mixture simulating collagen. Oxidation, in the manner of McFarlane and Guest with sodium peroxide, yields products that form an intense red color with p-dimethylaminobenzaldehyde. The intensity of color produced with 5 to 15 y of hydroxyproline is 4 to 5 times that formed in previous colorimetric procedures. Because of this greater intensity of color and because of a different preliminary treatment of the hydroxyproline solutions, only 1 to 2 per cent as much protein is required as in earlier methods. In acid hydrolysates of proteins, the only amino acid other than hy-