DISTRIBUTION OF NITROGEN 87 



A. DISTRIBUTION OF THE NITROGEN IN THREE 



GROUPS. 



The differentiation of proteins by the estimation of the various 

 groups of the units was first attempted in Hofmeister's laboratory by 

 Hausmann [1899, 1900], who estimated amide nitrogen, di-amino 

 nitrogen and mono-amino nitrogen. The protein (i gram) was 

 hydrolysed with 20 c.c. of concentrated hydrochloric acid by boiling 

 for five hours under a reflux condenser ; the solution was diluted and 

 distilled with excess of magnesia, and the ammonia, which was 

 liberated, was collected in excess of standard acid ; the solution was 

 then acidified with acid and precipitated with phosphotungstic acid ; 

 after twenty-four hours the precipitate was filtered off, washed with 

 phosphotungstic acid, dissolved in alkali and the nitrogen estimated 

 in an aliquot portion by Kjeldahl's method ; the filtrate was made 

 up to a definite volume and nitrogen estimated in an aliquot portion. 



Numerous objections to the accuracy of the data were raised. 

 Henderson [1899] maintained that the amount of amide nitrogen varied 

 according to the strength of the acid employed in the hydrolysis and 

 the time of hydrolysis; Kutscher [1900], and also Chittenden and 

 Eustis [1900], showed that the precipitation of the di-amino acids was 

 not complete, and Schulze and Winterstein [1901, 2] found that certain 

 mono-amino acids, e.g., phenylalanine, were precipitated by phospho- 

 tungstic acid under certain conditions. Hart [1901] preferred barium 

 carbonate to magnesia for distilling off the ammonia. 



Osborne and Harris [1903], Giimbel [1904], and also Rothera 

 [1904], critically examined the various objections. The amount of 

 amide nitrogen was not found to vary as Henderson stated ; if similar 

 conditions are always maintained in the precipitation with phospho- 

 tungstic acid most valuable comparative results can be obtained ; the 

 errors of incomplete precipitation of the di-amino acids and precipita- 

 tion of mono-amino acids almost compensate each other. 



Adopting Gumbel's suggestion of distilling off the ammonia in 

 vacuo at 40 and Osborne and Harris' procedure, the process may be 

 carried out as follows : 



I. About I gram of protein is boiled with about 100 c.c. of 20 

 per cent, hydrochloric acid in a 500 c.c. round bottom flask under a 

 reflux condenser until the solution no longer gives the biuret reaction, 

 usually from seven to ten hours. It is then evaporated in vacuo at 



