63

riatic acid gas, gave a distinct flash of lightunder the same circumstances, and theycombined with visible inflammation whenthe spark was passed through them, the ex-haustion being to one-twenty-fourth.

The experiment on the flame of hydro-gene with the wire of platinum, and whichholds good with the flames of the othergases, shows, that by preserving heat in ra-refied air, or giving heat to a mixture, in-flammation may be continued when, undercommon circumstances, it would be extin-guished. This I found was the case inother instances, when the heat w r as diffe-rently communicated: thus, when camphorw 7 as burned in a glass tube, so as to makethe upper part of the tube red hot, the in-flammation continued when the rarefactionwas 9 times, whereas it would only con-tinue in air rarefied 6 times, when it wasburned in a thick metallic tube which couldnot be considerably heated by it.

By bringing a little naphtha in contactwith red hot iron, it produced a faint lam-bent flame, when there remained in the re-ceiver only one-thirtieth of the originalquantity of air, though without foreign