926) that certain endothermic substances which, if heated gradually, would either volatilise or decompose, can be made to explode by throwing them into a vessel previously raised to a suitably high temperature. This result was observed with trinitrophenol, mane, di-, and tri-nitronaphthalene, and potassium chlorate. The authors, following up these observations of Berthelot, find that for each endothermic substance there is a temperature below which explosive decomposition does not oocur, and

more »

... s definite point is termed the temperature of explosion. The procedure adopted was to drop a small quantity of the substance t o the bottom of a clean test-tube immersed i n a bath of potassium hydrogen sulphate just when the temperatmure of the bath has attained the temperature particular to each substance, as otherwise a simple decomposition would take place. The temperature of this bath, up to 500°, was recorded on a mercury thermometer ; at higher temperatures, a thermo-couple was employed. The experiment was carried out in an atmosphere of carbon dioxide, but any other indifferent gas, such as nitrogen, may be used, and explosive decomposition occurs with equal readiness in a vacuum. From the results recorded below for the temperatures of explosion of various substanme, one or two conclusions of a general kind may be drawn. So far as nitro-derivatives are concerned, it appears that, of the disubstituted derivatives of benzene, the orthocompound, as a rule, has the lowest, and the para-compound the highest, temperature of explosion. Thus the temperatures of explosion for 0-, m-, and pchloronitrobenzene are 614O, 638O, and