_Process by Liquid Reagents._—The first method is by the employment of several liquid tests, which cause in the solution peculiar precipitates. Many such tests have been proposed; but the most characteristic and precise are _hydrosulphuric acid_, _ammoniacal nitrate of silver_, and _ammoniacal sulphate of copper_. The indications of each of the three tests must concur, otherwise, in a medico-legal case, no one can be entitled to speak with certainty to the existence of arsenic. But when they do concur, the evidence is unimpeachable. When this method of analysis is followed, corresponding experiments ought always to be made with the water that is used for diluting or otherwise preparing the subject of examination, or with distilled water, if the article be already sufficiently aqueous. This precaution is necessary on account of the risk of accidental impregnation of the water or other reagents with arsenic.[515] _Hydrosulphuric acid_ [sulphuretted-hydrogen] is obtained by decomposing proto-sulphuret of iron with diluted sulphuric acid in such an apparatus as is represented at Fig. 5. And the gas may be either applied directly to the suspected fluid, or condensed in distilled water, and thus kept in store for occasional use in the liquid shape. Before applying this test, the suspected fluid must be acidulated with acetic or hydrochloric acid; because an excess of alkali prevents the action. And if an acid be indicated by litmus in the fluid, neutralization, or slight supersaturation, with potash must be effected, before adding acetic or hydrochloric acid; for if the acidity should happen to be owing to an excess of sulphuric or nitric acid, the test is decomposed, and yellowish-white sulphur deposited.—These precautions being taken, hydrosulphuric acid occasions a sulphur-yellow or lemon-yellow precipitate. If the arsenical solution, however, be very weak, a yellow colour merely is struck, because the precipitate, which is sesqui-sulphuret of arsenic, is dissolved by the excess of the test; but it separates after ebullition, or a few hours’ exposure to the air. Co-existing animal and vegetable principles sometimes enable the fluid to retain a minute portion even after ebullition, so as to acquire a yellow milkiness; but they do not in any case prevent the test from producing the yellow colour. Acidulation with acetic or hydrochloric acid favours its subsidence in all cases; and according to Mr. Boutigny, alkaline sulphates, muriates and nitrates have the same effect.[516] Hydrosulphuric acid is so delicate as to act on the oxide in a hundred thousand parts of water. The proper colour of the precipitate is lemon or sulphur-yellow; which, when vegetable or animal matter is present, acquires a shade of white or brown.