CHEMISTRY OF NITROGLYCERINE. 69 



Action of Fixed Alkalies on Nitroglycerine. — The nitroglycerine used was 

 made by myself and not extracted from dynamite, as has been very frequently 

 the case with previous investigators. I shall afterwards describe the mode of 

 preparation employed. It is sufficient in the meantime to state, that whatever 

 was the variation practised in the method of the production of the nitro- 

 glycerine, the products were perfectly uniform in character. The action of the 

 alkali was examined both in aqueous and in alcoholic solutions. Nitro- 

 glycerine is so insoluble in water that it was decidedly preferable to make use 

 of an alcoholic solution of the ether and mix it with an alcoholic solution of 

 pure caustic potash (crystallised from its solution in alcohol). Absolute ethylic 

 alcohol was employed in every instance. 



When a moderately strong solution of caustic potash (1 in 10) is added to 

 a solution of nitroglycerine of similar strength, the following phenomena are 

 observed. The first few drops of the alkaline solution produce an orange- 

 coloured precipitate, which, on the addition of more of the potash, assumes 

 along with the whole fluid a deep reddish-brown colour. A large amount of 

 heat is developed during the mixture, amounting almost to ebullition of the 

 alcohol ; a strong aldehyde-like odour is evolved, without any perceptible 

 odour of ammonia or acrolein. The fluid quickly separates into two layers — 

 the lower, and much the smaller, being partly of the nature of a solid 

 precipitate, yet in great part syrupy and of a very deep reddish-brown colour, 

 and containing nearly all the colouring matter formed by the decomposition of 

 the nitroglycerine. The upper layer constitutes the bulk of the fluid, and is 

 yellowish in colour, and at first muddy, but, after a few minutes, becomes quite 

 transparent. The application of external heat is, as I have ascertained, quite 

 unnecessary to complete the decomposition, although in most of my 

 experiments I have with this object boiled the fluid over the water-bath 

 for several minutes, sometimes to the entire dissipation of the alcohol, water 

 being added as the alcohol is evaporated. When water is so added, the syrupy 

 precipitate, in proportion to the amount of alcohol still present, becomes 

 partially or completely dissolved, yielding a deep reddish-brown solution. If 

 sufficient water is added to the fluid without previous removal of the alcohol, 

 it is still possible to obtain a perfect solution of all the substances present. It 

 was in such diluted solutions, obtained either in the one way or the other, that 

 I estimated the amount of nitrite of potassium formed. This was effected by 

 means of starch, potassium iodide, and dilute sulphuric acid, a thoroughly 

 well-boiled 5 per cent, solution of starch, and containing 2 per cent, of 

 potassium iodide, being employed. The blue colour obtained on the addition 

 of these reagents was compared, as regards its intensity, with the colour 

 produced by a similar amount of the reagents added to standard solutions of 

 nitrite of sodium placed in test-tube's of the same diameter and used in the 



