& W. Johnson — Estimation of Nitrogen in Nitrates. 261 



tillation to remove, although Thorpe makes no such suggestion, 

 I added a quantity of newly fused, pure caustic potash, and 

 more water and distilled again. 



This second distillation gave additional ammonia equal to 

 0-0227 of nitrogen. 



The total nitrogen thus obtained was : 



2d " 0-02275 



0-05222 instead of 0-09408 

 Or a deficiency of 45 per cent. 



In a second trial, 0-6689 grams of sodium nitrate were taken. 

 The conditions of reduction were as near as practicable the 

 same as before, but on completing the first distillation the odor 

 of nitric oxide was plainly evident in the receiving vessel, and 

 the amount of free acid in the latter was greater than at the 

 outset, the standard acid, not only being not neutralized by 

 ammonia coming from reduction, but made more acid by the 

 reaction of nitric oxide upon the oxygen and water of the con- 

 densing vessels. 



In a third experiment with 04193 grams of pure potassium 

 nitrate, 40 grams of cleansed zinc were covered with concentrated 

 copper sulphate solution for fifteen minutes, the precipitated 

 copper was washed with great care to avoid as much as possible 

 detaching it from the zinc. 40 c. c. of water and the nitrate 

 were added, the mixture heated very gradually and distilled 

 finally with addition of more water, for one hour. The result was 

 like that of experiment 2. . 



A fourth trial resulted in the same increase of acid in the 

 receiver. In all these cases the copper was very loosely at- 

 tached to the zinc, so that by mere washing much of it, and on 

 boiling most of it separated. „ . . . , , 



I now referred to Gladstone's description of his method of 

 preparing the zinc-copper couple, and made a fifth experiment, 

 following his different directions as follows : 



A weak solution of copper sulphate (l'o per cent UifeUJ 



was employed, and the zinc was let remain in it for some time 



until the color of the solution was very nearly discharged. 



The distillation was then conducted as Thorpe directs. 



0-5338 grru. KN0 3 containing 13-84 per cent N. 



yielded J^^_ JN ' 



deficiency= 8-64 " N. 



Thorpe mentions no difficulty in the application of his pro- 

 cess except in case of ammonium nitrate. With reference to 

 its analysis he says, "the discordances appear to .depend ^ upon 

 differences in the strength of the solution and the amounts of 



