838 THE HOLE OF NITROGEN, ETC., II., 



excess of the reagent. It was first used by Schjerning(8), and is 

 the chief method in use for measuring the progress of digestion- 

 experiments. Hedin(9) uses for this purpose a solution containing 

 tannin 7, salt 10, acetic acid 5, per cent. Bigelow and Cook(10> 

 have shown that none of the amino-compounds are precipitated 

 except creatin and cliphenylamine. Alkaloids, and possibly other 

 plant-bases as well, are precipitated, and these are present in the 

 majority of seeds. 



(3) Alcohol precipitates proteins, proteoses and peptones com- 

 pletely^), and does not precipitate any other nitrogenous 

 compounds. Mellanby's experiments(7) show that for each 

 different protein there is a critical temperature below which the 

 precipitate formed remains quite soluble, and above it the proteins 

 are chemically changed and rendered insoluble; only the proteoses 

 and peptones are precipitated in the soluble form. The precipi- 

 tation with hot alcohol certainly appears the most convenient for 

 use with plant-extracts; the tannin-salt solutions are very difficult 

 to decompose in the Ivjeldahl process, as they require careful 

 watching on account of the constant frothing. The precipitation 

 by copper hydroxide, though it is the method which has been 

 almost exclusively used by Schulze and all the early workers, is- 

 evidently too indefinite except for comparative purposes; we 

 hardly know how it acts, especially towards those obscure plant- 

 bases which it is necessary to keep in solution. 



The following preliminary experiments were carried out in. 

 order to test the difference in the above methods and for purposes 

 of comparison. 



Stutz -r's reagent was prepared by precipitating a solution of 

 copper sulphate and glycerol with potassium hydroxide, and the 

 green ppt. washed free from alkali. The ppt. was then suspended 

 in water with 10 % of glycerol added. Each 1 cc. contained 

 01 gm. of copper hydroxide. The tannin solution contained 

 10 % of tannic acid and 10 % of sodium chloride. 



An aqueous extract of the seeds of Acacia pycnantha was made, 

 and equal volumes taken for comparison of the above three 

 reagents. The proteins were precipitated, and the nitrogen in 



