00 MATTHEW HAY AND ORME MASSON ON THE 



of traces of impurities (oxidised derivatives of glycerine) irremovable by the 

 water with which the nitroglycerine was washed. 



We have also estimated the nitrogen in other samples of nitroglycerine, 

 prepared with different proportions of acid, in order to ascertain whether a 

 difference in the method of preparation causes any corresponding difference in 

 the composition of the liquid. The results show that it is not so. In these 

 cases Dumas's method was not employed, but a modification of Schloesing's 

 process, which we have found to give equally good results, in spite of the 

 contrary experience of Hess and Schwab and of Sauer and Ador. 



The weighed quantity of nitroglycerine was dissolved in absolute alcohol 

 and decomposed by boiling for ten minutes with excess of an alcoholic solution 

 of caustic potash. Water was then added, and the whole of the alcohol was 

 driven off by evaporation ; after which the fluid was made up to a given 

 volume, of which a measured portion was taken for experiment. The volume 

 of nitric oxide evolved by the reducing action of the ferrous chloride and 

 hydrochloric acid (the reagents employed in Schloesing's method) was, in each 

 case, compared with the volume of gas obtained under precisely the same 

 conditions of temperature and pressure from a standard solution of pure 

 potassium nitrate, aud also, in certain instances, of pure sodium nitrite ; and 

 from these data the percentage of nitrogen was calculated. A correction is, 

 however, necessary, inasmuch as a small portion of the nitrogen is always 

 evolved as ammonia on boiling the nitroglycerine with potash. The amount 

 of this was determined in a preliminary experiment, as follows : — 



1*1533 grm. of nitroglycerine was dissolved in 5 c.c. absolute alcohol and 

 a solution of 1*5 grm. of potash in absolute alcohol added, the flask being 

 immediately connected with a modified Boussingault's apparatus and then 

 boiled. After half an hour an equal volume of water was added, and boiling 

 then continued three-quarters of an hour more. The distillate was received 

 into standard acid ; and, at the end of the operation, this was titrated with 

 standard soda. By this means it was found that *0053 grm. of ammonia had 

 been evolved ; so that nitroglycerine loses nitrogen during decomposition with 

 potash to the extent of 0"38 per cent, of its weight. This amount was, 

 therefore in each case, added to the percentage of nitrogen found by 

 Schloesing's method. 



In the following table the letters in the second column refer to the various 

 samples of nitroglycerine employed. A detailed description of their manu- 

 facture will be found in the preceding paper,* where they are lettered in the 

 same order. Experiment IX. was made with nitroglycerine obtained from 

 a Nobel's dynamite cartridge by displacement with water and desiccation over 

 the water-bath. 



* "A Contribution to the Chemistry of Nitroglycerine," by Matthew Hay. 



