50 THE CHEMICAL CONSTITUTION OF THE PROTEINS 



and alanine can be obtained from protein cystine, which decomposition 

 may take place according to Gabriel in the following way : 



CH 3 



H 2 | 



> CH . NH 2 



COOH 



^~^^' 

 CH 2 SH 



CH.NH 2 

 COOH ^ 



-Ar/T^ CH OX CH 3 



I > H 2 | 



CH' > CH.SH 



COOH COOH 



Thus, the work of Friedmann on the constitution of cystine, its synthesis 

 by Erlenmeyer jun. and by Fischer, definitely show that it has the 

 composition 



HOOC . CH(NH 2 ) . CH 2 S-S CH 2 . CHfNH^'. COOH. 



The proof that bromophenylmercapturic acid is derived from cysteine, 

 the formation of a-thiolactic acid from cystine derived either from 

 stones or proteins, and the identity of protein cystine with stone cystine 

 show that cystine is the only sulphur-containing compound in the pro- 

 tein molecule : and also that the number of sulphur atoms in the protein 

 molecule is two or a multiple of two, instead of the variable number 

 which had been determined by the earlier workers upon the sulphur in 

 the protein molecule. This work was commenced by Mulder, who was the 

 first to observe that albumin, caseinogen, etc., when heated with alkali 

 gave off hydrogen sulphide ; in consequence of this he regarded these 

 compounds as composed of sulphur and protein in various proportions. 

 Fleitmann, a pupil of Liebig's, in 1847 then showed that this view of 

 the constitution of albumin, etc., was erroneous, for he found that only a 

 portion of the sulphur was spilt off by alkali, and that a portion still 

 remained combined with the protein. The later investigators upon this 

 question Nasse, Danilewsky, Kruger, Suter, Malerba, Schulz con- 

 firmed Fleitmann 's results, and in addition they determined the ratio of 

 total sulphur to loosely bound sulphur, as this sulphur easily split ofT by 

 alkali was called. Their results varied considerably, and this was due 

 to the different methods which they employed. In some proteins, e.g n 

 serumalbumin, the ratio of loosely bound sulphur to total sulphur was 

 as 2 : 3, in others I : 2 or 5 : 3. From these values determinations were 

 made of the molecular weight : thus serumalbumin was given a mole- 



