io8 THE CHEMICAL CONSTITUTION OF THE PROTEINS 



histidine and lysine figures are higher. These data are found by dif- 

 ference and depend upon the accuracy of the determination of arginine, 

 of amino N and of total nitrogen. The figures for the di-amino acids 

 are in general higher than the figures given by the Kossel and Patten 

 method (p. 55). Thus in gram per 100 grams of protein : 



Both methods give the same value for arginine, the value for lysine is 

 higher and more accurate, while the value for histidine is more consistent 

 by the Kossel-Patten method. These and other data are given on 

 page 132. 



The method has been of service to Hartley [1914] in his study 

 of the proteins of ox and horse serum. His results confirm the obser- 

 vations of previous writers that the albumin and globulin of serum ex- 

 hibit well-marked differences in composition. Albumin contains more 

 di-amino nitrogen than globulin, which is in agreement with the data by 

 Hausmann[i899], Gumbel [1904], and Gibson [1912]. Albumin also 

 contains more nitrogen as cystine than globulin, a fact which agrees 

 with the data of By waters and Tasker [1913-14] who found that albumin 

 contained about 2 per cent, of sulphur and 0-25 per cent, of carbohy- 

 drate, while globulin contained 1-2 per cent, of/sulphur and 3-2 per 

 cent, of carbohydrate. Globulin was also found to contain phosphorus 

 while this element is absent from albumin. The chief difference found 

 by Hartley was the presence of more lysine in albumin than in globu- 

 lin. The chemical differences between these two proteins are against 

 the supposition of Moll [1904, 1906] that albumin can be converted 

 into globulin. There is so little difference between the globulins that 

 the chemical composition supports Chick's suggestion [1914] that 

 under certain conditions pseudo-globulin may undergo a process of 

 denaturation and that a substance is formed which is very similar to 

 euglobulin. This substance may be a mechanical complex formed by 

 the interaction and mutual precipitation of the two colloids, pseudo- 

 globulin and lipoid. The euglobin of serum is a protein-lipoid com- 

 plex. 



Crowther and Raistrick [1916] used the method for the analysis 

 of the proteins of colostrum and milk and showed that the casein- 



