and similar oxidizing bodies, moreover the salts of mercury and, probably, some other metallic salts entirely prevent its generation. It is, for this reason, advisable to prepare the liquid to be introduced into the apparatus, exactly as above described; we have then only to guard against the presence of nitric acid. Hydrochloric acid does not influence the formation of arseneted hydrogen, but, since this acid is very volatile, it is better to exclude it, especially because, according to Wackenroder, it sometimes gives rise to the production of spots in absence of arsenic (zinc spots).

§ 23. The liquid which Marsh originally recommended as fit to be introduced into the apparatus, was prepared by simply extracting the organic matters with dilute hydrochloric or sulphuric acid, and, if necessary, concentrating by evaporation. But against the applicability of this liquid serious objections may be raised; the extract may contain substances which may prevent the evolution of the arseneted hydrogen; during the evaporation, chloride of arsenic may be volatilized; the organic matters contained in the liquid usually produce a considerable intumescence, whereby the contents of the bottle are caused to enter the tubes, and thus thwart the experiment. The method of allaying the intumescence by pouring oil, or alcohol, on the liquid, as has been proposed, cannot be relied on; and since, moreover, experience has shown that hydrogen gas, evolved from liquids which contain organic matters, is capable of producing brown spots on porcelain, the direct application of the extract under consideration is usually inadmissible. Nor can we, for the reasons

above stated, recommend the use of the liquid which is prepared by treating the substances with hydrochloric acid and chlorate of potassa.

§ 24. If the precipitate which is thrown down by sulphureted hydrogen is treated with a dilute solution of caustic potassa, the sulphide of arsenic, if present, enters into solution, and, on boiling the liquid with protoxide of copper, sulphide of copper and arsenate of potassa are formed. The liquid, after having been acidulated with sulphuric acid, is well adapted for Marsh's apparatus, but offers no advantages over the solution prepared after our method.

If, in the substances under examination, solid arsenious acid has been detected, and it is thought desirable to subject it to Marsh's test, it is only necessary to dissolve it in pure water, or water acidulated with sulphuric acid, and to pour this solution into the gasbottle.

§ 25. A short time after the publication of the process of Marsh, Pfaff and Thompson made, independently of each other, the interesting discovery that there exists an antimoneted hydrogen, formed under the same circumstances as arseneted hydrogen, and exhibiting the same phenomena on being heated, burnt, etc. Thus, for example, if we add to the zinc and dilute sulphuric acid of our apparatus some oxide of antimony, or chloride of antimony, or tartar emetic, the escaping hydrogen becomes mixed with antimoneted hydrogen; the flame of the burning gas appears white, spots of metallic antimony are deposited on porcelain which is held into the flame, a mirror of metallic antimony is formed within the ignited tube, and, if the gas is allowed to pass

through a solution of nitrate of silver, a black preIcipitate is thrown down. Quite the behavior of arsenic !

This discovery of Pfaff and Thompson caused the test of Marsh to be considered very fallacious; the more so, as in a case of suspected poisoning tartar emetic is frequently administered as an emetic. The whole process, therefore, ought to be condemned, unless means are found either wholly to exclude the antimony, or to show, with certainty, that the reactions are produced by arsenic.

If the precipitate thrown down by sulphureted hydrogen (§ 10), is treated directly with nitric acid, and the residue with sulphuric acid, the resulting liquid will, of course, contain all the antimony which might have occurred in the substances. But if the treatment with nitric acid is preceded by a treatment with dilute ammonia, or with a solution of carbonate of ammonia, the antimony is almost wholly, even to a trace, excluded; and if, moreover, the residue resulting from the evaporation of the ammoniacal solution is not treated with nitric and sulphuric acid, but oxidized with nitric acid, and afterwards fused with nitrate and carbonate of soda, the filtered watery solution of the fused mass contains no antimony, this metal remaining behind after the treatment with water as insoluble antimonate of soda (§ 13).

But, even if all these precautions, which effectually exclude the presence of antimony, have been observed, it is necessary to prove the reactions obtained by the process of Marsh to be owing to arsenic, and

*it will be the duty of the judge to ask for these proofs. A vast number of characteristic tests for the discrim- . ination of arsenic from antimony have been advanced, in the course of time, so that it is almost impossible to confound the two metals. We shall now proceed to discuss them.

§ 26. The arsenic-mirror, as deposited in the reduction-tube, is highly lustrous, of blackish-brown or brownish-black color, and, when seen against a sheet of white paper, perfectly transparent brown in places where the deposit is not too thick. The antimony-mirror is, next to the heated spot, partly fused to small globules, and there it appears of almost silver-white color; more distant from the flame it is nearly black, and forms, when in very thin layers, not a continuous lustrous brown and transparent coating, but the tube appears, rather, as if dulled by a brownish-black dust.

Antimoneted hydrogen is decomposed at a lower temperature than arseneted hydrogen. Hence it is, that when hydrogen containing some of the former gas passes through the ignited tube, a metallic mirror is deposited before and beyond the heated spot, provided the quantity of the antimoneted hydrogen be not too small. Arseneted hydrogen never produces a mirror before the heated spot.

The arsenic-mirror is easily driven from one place to another, when gently heated in the current of the hydrogen gas. Hereby a considerable quantity of arsenic is carried off with the gas; and, if the latter is allowed to escape unburnt, the characteristic arsenic-odor very distinctly perceived. But if the

gas is kindled, the flame instantly assumes a bluish-white color, and spots are formed on porcelain which is held into the flame.

The antimony-mirror requires for its vaporization a much higher temperature, and, during vaporization, spots on porcelain can only be obtained when the evolution of the gas is very brisk; for in this case alone, the antimony is not completely deposited in the tube. The escaping gas, if not kindled, is perfectly inodorous. Previous to vaporization, the antimony-mirror changes its appearances at the heated spot; it fuses into small lustrous globules, which may, in all cases, be seen with the aid of a lens (Wöhler); arsenic exhibits no sign of fusion.

If the glass-tube containing the arsenic-mirror is cut off and heated in the flame of a spirit-lamp, the characteristic garlic-odor is evolved. The antimonymirror volatilizes without any odor.

§ 27. The spots which, on incomplete combustion, arseneted hydrogen deposits on porcelain, are blackish-brown and lustrous, and, when in thin layers, translucent, with a brown, or light-brown color. The antimony-spots, if not too thin, are not lustrous, of velvet-black color, and have, usually, a white deposit in the centre; when in very thin layers, they exhibit lustre; but, even then, their color is not brown, but iron-black, or dark graphitecolored, and only at the extreme edges possessed of a brownish-grey tinge.

The arsenic-spots, when moistened with a solution of hypochlorite of soda, are dissolved; the antimonyspots are not affected by the reagent. This is a very excellent distinguishing test, for which we are in