THE CHEMISTRY OF Till-: ANIMAL BODY. 581 



All proteids yield histidin, arginin, and lvsin on decomposition. As 

 regards the composition of the proteid molecule, Kossel pictures a protamin 

 nucleus like sturin, to which may be attached leucin, tyrosin, glucosamin, or 

 glycocoll, and to these again sulphur, iodine, or iron. Treatment of proteid 

 with 20 per cent, hydrochloric acid or tryptic digestion may break it up into 

 leucin, tyrosin, histidin, argenin, lvsin, etc. Kossel speaks of histidin, arginin, 

 and lysin as hexon-bases, since they (and leucin also) contain six atoms of 

 carbon, and he calls attention to the fact that in this respect they are similar 

 to the carboydrates. Just as carbohydrates exist as poly-hexoscs, so prota- 

 mins and proteids may be built up as poly-hcxon-bases. Cohn has found 

 that proteid may yield as much as 50 per cent, of leucin. 



The products derived from proteid in metabolism are different from the 

 above. Thus it has been found that the body's proteid, the proteid from 

 meat, and gelatin, may all yield about 60 per cent, of dextrose in diabetes. 1 

 It has been further shown 2 that the metabolism of the body's proteid, of 

 casein, and of gelatin yields between 3 and 4 per cent, of glycocoll, which 

 may be eliminated as hippuric acid. It is possible to conceive of a carbo- 

 hydrate portion united to a protamin nucleus and to amido bodies such as 

 glycocoll 3 (see p. 558). 



Miiller and Socman 4 have declared that the source of the sugar in 

 diabetes must be the hexon-bases and leucin, but Halsey 5 has shown that 

 feeding leucin will not increase the sugar in diabetes. Halsey suggests a 

 synthetic formation of sugar from lower proteid decomposition-products, but 

 a synthetic formation of sugar in the animal has never been shown. It must 

 be admitted that we are still in the dark regarding even the simplest expres- 

 sion of the constitution of proteid. 



It has been impossible within the limits set to do more than to glance at 

 the proteid bodies. Many facts concerning the behavior of proteids have 

 been mentioned throughout the text, and cannot be classified here. 



The size of the proteid molecule must be very great, and one computation 

 shows the following figures 6 (see also p. 577) : 



C204H322N.r )2 O 66 S 2 . CV 26 H 1I71 Ni 94 214 S 3 . 



Egg-albumin. Proteid from haemoglobin (dog). 



It is well, perhaps, finally, to speak of experiments which, however incom- 

 plete, at least throw some light on the possibilities of the problem of the syn- 

 thesis of proteid. Lilienfeld 7 through the condensation of the ethyl-ester of 



1 Keilly, Nolan, and Lusk : American Journal of Physiology, 1898, vol. i. p. 395. 



2 Parker and Lusk : I hid., L900, vol. iii. J>. -J7-J. 



3 Ray, McDermott, and Lusk : Ibid,, 1S99, vol. iii. p, 153. 

 * Deutsche medicinisehe Wochenschrift, 1899, S. 209. 



5 Sitzungsberichte der Gesellschaft zur Befordeiimg der gesammten NaturvnssenwhafienjMU Marburg, 

 1899, S. L02. 



6 Bnnge : Physiologische Chemie, 3d ed., 1893, S. 56. 



7 Verhandlungen der Berliner physiologischen Gesellseliaft, Archiv fur Physiologic, 1894, 

 S. 383. 



