138

violence and a roaring noise, making an in-tense flame of the length of five feet. Theblow-pipe was exposed at right angles to astrong wind, and double gauze lamps andsingle lamps successively placed in it. Thedouble gauze lamps soon became red hot atthe point of action of the two currents ; butthe wire did not burn, nor did it communi-cate explosion. The single gauze lamp didnot communicate explosion, as long as it wasred hot and slowly moved through the cur-rents ; but when it was fixed at the point ofmost intense combustion, it reached a weldingheat, the iron wire began to burn withsparks, and the explosion then passed.

In a second and third set of experimentson this violent blow-pipe of fire-damp, singlelamps, with slips of tin-plate on the outsideor in the inside, to prevent the free passageof the current, and double lamps, were ex-posed to all the circumstances of the blast,both in the open air and in an engine-housewhere the atmosphere was explosive to agreat extent round the pipe, and throughwhich there was a strong current of atmo-spheric air; but the heat of the wire neverapproached near the point at which iron