Picric acid was not considered to be an explosive, properly so called, for a long time after its discovery, but the disastrous accident which occurred at Manchester (_vide_ Gov. Rep. No. LXXXI., by Colonel (now Sir V.D.) Majendie, C.B.), and some experiments made by Dr Dupre and Colonel Majendie to ascertain the cause of the accident, conclusively proved that this view was wrong. The experiments of Berthelot (_Bull. de la Soc. Chim. de Paris_, xlix., p. 456) on the explosive decomposition of picric acid are also deserving of attention in this connection. If a small quantity of picric acid be heated in a moderate fire, in a crucible, or even in an open test tube, it will melt (at 120 deg. C. commercial acid), then give off vapours which catch fire upon contact with air, and burn with a sooty flame, without exploding. If the burning liquid be poured out upon a cold slab, it will soon go out. A small quantity carefully heated in a tube, closed at one end, can even be completely volatilised without apparent decomposition. It is thus obvious that picric acid is much less explosive than the nitric ethers, such as nitro-glycerol and nitro-cellulose, and very considerably less explosive than the nitrogen compounds and fulminates. It would, however, be quite erroneous to assume that picric acid cannot explode when simply heated. On the contrary, Berthelot has proved that this is not the case. If a glass tube be heated to redness, and a minute quantity of picric acid crystals be then thrown in, it will explode with a curious characteristic noise. If the quantity be increased so that the temperature of the tube is materially reduced, no explosion will take place at once, but the substance will volatilise and then explode, though with much less violence than before, in the upper part of the tube. Finally, if the amount of picric acid be still further increased under these conditions, it will undergo partial decomposition and volatilise, but will not even deflagrate. Nitro-benzene, di-nitrobenzene, and mono-, di-, and tri-nitro-naphthalenes behave similarly. The manner in which picric acid will decompose is thus dependent upon the initial temperature of the decomposition, and if the surrounding material absorb heat as fast as it is produced by the decomposition, there will be no explosion and no deflagration. If, however, the absorption is not sufficient to prevent deflagration, this may so increase the temperature of the surrounding materials that the deflagration will then end in explosion. Thus, if an explosion were started in an isolated spot, it would extend throughout the mass, and give rise to a general explosion.