CHEMISTRY OF NITROGLYCERINE. 73 



Nitrate of potassium is, as I have mentioned, present in considerable 

 quantity, and with the nitrite constitutes the larger portion of the reddish-brown, 

 partly viscid, partly crystalline precipitate, which is formed in the decomposing 

 fluid. I have not estimated the nitrate quantitatively ; but there is every 

 reason to believe, from the estimation of the total nitrogen by Schloesing's 

 method and from other circumstances, that the amount of the nitric acid cor- 

 responds very closely to one-third of the nitrogen present in nitroglycerine. 



But nitrite and nitrate of potassium are not the only substances formed. I 

 have also proved the presence of acetate of potassium, oxalate of potassium, 

 and, doubtfully, of formate of potassium, neither of the two latter appearing to 

 be present in large quantity. There is also a small amount of ammonia, and of a 

 reddish-brown, resinous body, which imparts the dark colour to the decomposing 

 fluid. There is likewise present a very curious and very interesting substance, 

 which possesses the unusual power of forming a firm jelly with a very large 

 proportion of absolute alcohol. In contradiction to Railton and Williamson, 

 and previous investigators, I have found no glycerine, or only the merest trace 

 of it. This is a new and most important fact. 



In order to ascertain the presence and nature of these various decomposition- 

 products, 6 67 grammes of pure nitroglycerine were decomposed with excess of 

 potash in the usual manner, the mixture being allowed to boil for five minutes, 

 and afterwards set aside for one day, to permit of the deposition of certain of 

 the substances dissolved in the hot absolute alcohol. The supernatant fluid, 

 which was transparent and of an orange colour, was then decanted, and the 

 residue was again boiled with a fresh quantity of absolute alcohol, and again set 

 aside for a day, when the alcohol was decanted and added to the previously 

 decanted alcohol. This process was repeated a third time. The mixed alcoholic 

 fluids ought to have contained the excess of caustic potash and the whole of 

 the glycerine, were any present ; and in the deep reddish-brown residue I 

 expected to find nearly all the colouring matter, and. all the salts insoluble in 

 absolute alcohol, as the nitrite and nitrate of potassium. 



The mixed alcoholic fluids were neutralised with. alcoholic sulphuric acid in 

 order to precipitate the potash as sulphate of potash. A voluminous white 

 precipitate formed, which, after standing for some hours, was separated by 

 filtration. The filtrate, although faintly acid, yielded another tolerably copious 

 precipitate of sulphate of potash on the further addition of sulphuric acid, which 

 was now added in distinct excess. Salts of potash were evidently present, dis- 

 solved in the alcohol; certainly, amongst others, the nitrite and acetate, as proved 

 by testing. The precipitate was again removed by filtration, and the alco- 

 holic fluid was now distilled fractionally in order to remove the more volatile 

 substances, as the alcohol and acetic, formic and nitrous acids, the less volatile 

 glycerine, if present, remaining in the retort. Distillation was continued until 



