318 THE METABOLISM OF THE CARBOHYDRATES 



By various colour reactions, 1 Langstein has shown that pure 

 serum albumin also contains a carbohydrate group, and from one 

 preparation of it this worker was able to separate small quantities 

 of glucosamin. There still remains some doubt, however, as to 

 whether this may not have been derived from some adherent mucin 

 of which it is very difficult to remove all traces from the albumin. 



With pure serum globulin, on the other hand, much more satisfac- 

 tory results have been obtained. H. A. K. Morner and Langstein ( 2) 3 ) 

 have independently shown that both the eu- and pseudo-globulin 

 fractions of this can yield monosaccharides when hydrolysed with 

 acid, and by a careful study of the osazones, the products of oxidation 

 and the benzoyl compounds of these, and their power of fermenting 

 with yeast, they have identified these monosaccharides as dextrose, 

 glucosamin, and possibly Isevulose. By decomposing the globulin 

 with baryta, a polyose practically identical with albamin has been 

 obtained by Langstein. By decomposing this polyose with acids 

 glucosamin was obtained. The percentage of nitrogen in this albamin 

 was found to be somewhat variable, and, since dextrose has been 

 found along with glucosamin in the products of acid hydrolysis, 



1 There are certain colour reactions by which the presence of carbohydrate 

 groups in proteids may be detected. The more important of these are as 

 follows : 



1. The Reaction of Molisch. To one cubic centimetre of a one per cent, 

 suspension of proteid in water are added two drops of an alcoholic solution of 

 a-naphthol (15%) ; three cubic centimetres of concentrated sulphuric acid are then 

 run down the side of the test-tube so as to form a layer underneath the proteid 

 solution. If the proteid contain a carbohydrate group, a purple ring forms 

 where the two fluids are in contact. (The green diffuse colour which also 

 appears in the acid is no part of the reaction.) 



This reaction is so exceedingly delicate "0001 grin, dextrose or pentose gives 

 it that great caution must be taken that the proteid under examination is 

 entirely free of adherent carbohydrates, or of proteids such as mucin, which 

 contain carbohydrate groups. It is indeed too delicate a test on which to base 

 absolute conclusions. This reaction is usually described as due to the liberation 

 of furfurol by the action of the acid. 



2. The Orcin-ferric Chloride Reaction of Bial. ( 5 ) Into four cubic centimetres 

 of fuming hydrochloric acid in a small test-tube is dropped orcin till no more 

 will dissolve ; two cubic centimetres of the proteid solution are then added and 

 one drop of a dilute solution of ferric chloride. The mixture is boiled for about 

 two minutes. If hexoses be present, bluish-green flakes soon separate out, and if 

 these be dissolved by shaking with amyl alcohol the resulting solution will be 

 found to give, when examined in the spectroscope, an absorption band at the 

 beginning of the green absorbing the orange. Pentose and glycuronic acid give 

 similar bluish-green precipitates, but the orange of the spectrum is not absorbed. 



Glucosamin and Isevulose do not give this reaction. It is therefore an impor- 

 tant one for indicating what carbohydrate groups ought to be sought for by the 

 more elaborate methods 



