the quantity of the product, and produces a chemically pure article, as is shown by the fact that the Nitro-Glycerin thus produced is perfectly colorless, congeals uniformly at the same degree of temperature and produces, when exploded, no offensive vapors deleterious to the health of the miners using it. Moreover, as, in my opinion, these nitrous fumes tend to induce decomposition in the Nitro-Glycerin and thus to render it unstable, dangerous, and liable to spontaneous explosion, as is demonstrated to be the case in the analogous substance gun cotton, the introduction, in the method of Mowbray, of cold, dry, compressed air into the mixture, in order to get rid of these nitrous fumes, must be regarded as a substantially new invention. In my opinion, the character of the Nitro-Glycerin is determined by the strength of the acids used in its preparation; the stronger the acids, the purer the product and the more efficient. I verily believe this: first, because it is true of the precisely analogous compound gun cotton, which is prepared in the same way; Hadow having proved, as stated in his paper published in the Quarterly Journal of the Chemical Society of London in 1854, Volume VII., page 201, that at least three products are obtained by acting upon cotton by a mixture of sulphuric and nitric acids, the most explosive being always produced by the strongest acids; and 2nd, because of similar differences observed in Nitro-Glycerin made by different experimenters, and believed by them to be due to like differences in composition; Railton obtained by analysis, as stated in his paper in the Quarterly Journal of the Chemical Society of London for 1854, Volume VII., page 222, the composition now universally adopted as that of Tri-Nitro-Glycerin. De Vrij believes the product he obtained, Journal de Pharmacie, series III., Volume XXVIII., page 38, 1855, to be Tri-Nitro-Glycerin, and Liecke in Dingler’s Polytechnisches Journal, Volume CLXXIX., page 157, 1866, gives methods by which Mono-Nitro-Glycerin, Di-Nitro-Glycerin and Tri-Nitro-Glycerin may be produced, the essential difference in these methods being only the strength of the acids employed. Gladstone’s Report of the British Association for 1856, page 52 (Notices and Abstracts), has shown that different samples of Nitro-Glycerin differed in properties according to the amount of water contained in the Glycerin; this water, by diluting the acids, making them weaker. Moreover the physiological properties of Nitro-Glycerin have been found by different experiments to differ widely. Sobrero, its discoverer, says a very small quantity taken upon the tongue produces a severe headache for several hours, whence he concludes that it is poisonous. De Vrij in 1851, says that it is not poisonous, and in 1855 that it produces headache, though ten drops caused no symptoms of poisoning in a rabbit. Dr. Herring, in 1849, reported in the American Journal of Science and Arts, series II., Volume VIII., page 257, observed