In 1785 he demonstrated the nature and proportion of the constituents of ammonia, or volatile alkali. This substance had been collected in the gaseous form by the indefatigable Priestley, who had shown also that when electric sparks are made to pass for some time through a given volume of this gas, its bulk is nearly doubled. Berthollet merely repeated this experiment of Priestley, and analyzed the new gases evolved by the action of electricity. This gas he found a mixture of three volumes hydrogen and one volume azotic gas: hence it was evident that ammoniacal gas is a compound of three volumes of hydrogen and one volume of azotic gas united together, and condensed into two volumes. The same discovery was made about the same time by Dr. Austin, and published in the Philosophical Transactions. Both sets of experiments were made without any knowledge of what was done by the other: but it is admitted, on all hands, that Berthollet had the priority in point of time. It was about this time, likewise, that he published his first paper on chlorine. He observed, that when water, impregnated with chlorine, is exposed to the light of the sun, the water loses its colour, while, at the same time, a quantity of oxygen gas is given out. If we now examine the water, we find that it contains no chlorine, but merely a little muriatic acid. This fact, which is undoubted, led him to conclude that chlorine is decomposed by the action of solar light, and that its two elements are muriatic acid and oxygen. This led to the notion that the basis of muriatic acid is capable of combining with various doses of oxygen, and of forming various acids, one of which is chlorine: on that account it was called _oxygenized muriatic acid_ by the French chemists, which unwieldy appellation was afterwards shortened by Kirwan into _oxymuriatic acid_. Berthollet observed that when a current of chlorine gas is passed through a solution of carbonate of potash an effervescence takes place owing to the disengagement of carbonic acid gas. By-and-by crystals are deposited in fine silky scales, which possess the property of detonating with combustible bodies still more violently than saltpetre. Berthollet examined these crystals and showed that they were compounds of potash with an acid containing much more oxygen than oxymuriatic acid. He considered its basis as muriatic acid, and distinguished it by the name of hyper-oxymuriatic acid.