[91] The most vivid effects of combustion known are those produced by the _condensation_ of oxygen, or chlorine: but in this instance, a violent explosion with heat and light is produced by their separation, and _expansion_; a perfectly novel circumstance in chemical philosophy. The results of its explosion indicate its composition to be one atom of chlorine, and one of oxygen. None of the metals that burn in chlorine act upon this gas at common temperatures; but when the oxygen is separated, they then inflame in the residual chlorine. This fact Davy illustrated by a series of experiments, one of which, from its extreme beauty, I shall here relate. If a glass vessel, containing copper-leaf, be exhausted, and the euchlorine afterwards admitted, no action will take place; but throw in a little nitrous gas, and a rapid decomposition will ensue, and the metal will burn with its accustomed brilliancy. The discovery of this interesting gas, and that of the facts connected with it, not only confirmed the novel views with regard to the elementary nature of chlorine, but they reconciled the contradictory accounts of different authors respecting the properties of that body. The weak attraction subsisting between the elements of this compound gas, which by a comparatively low temperature are made repulsive of each other, confirms also the supposition of Davy, that oxygen and chlorine belong to the same class of bodies. The discovery of the _peroxide of chlorine_ was made during an examination of the action of acids on the _hyper-oxymuriates_ of Chenevix, undertaken by Davy in consequence of a statement of M. Gay Lussac, that a peculiar acid, which he called _chloric acid_, might be procured from the _hyper-oxymuriate of baryta_ by sulphuric acid. With regard to this acid, which its discoverer considered as composed of one atom of chlorine and five atoms of oxygen, Davy entered into a warm controversy, affirming that the fluid in question owed its acid powers to combined hydrogen; and that it was analogous to the other hyper-oxymuriates, as being triple compounds of inflammable bases with chlorine and oxygen, in which the two former determine the character of the compound: this opinion, however, he afterwards abandoned, and I have reason to believe that he regretted ever having advanced it. Amidst these new views, it became necessary to alter our opinions with regard to many of those compounds which have been termed _muriates_, but which, it would appear, contain neither muriatic acid nor oxygen, but are, strictly speaking, combinations of metals with chlorine, held in union by a very powerful affinity, since chlorine is capable of expelling the whole of the oxygen from any metallic oxide, and of taking its place; even those metals that are most distinguished by their affinity for oxygen, abandon it whenever their oxides are heated in chlorine, in which case oxygen gas is disengaged.