Fig. 8.

Although the separation of arsenic from antimony is the chief object in making use of the apparatus proposed by Flandin and Danger, it is evident that this result is not fully accomplished, since a small portion of arsenious acid remains in the tube D (Fig. 7), together with the intermediate oxide of antimony. The following method secures the complete separation of these metals: An amalgam of sodium and mercury is introduced into the flask A, (Fig. 8), which is provided with two openings. The tube B, terminating in a funnel at its upper extremity, passes through one of these orifices. The other aperture contains a cork enclosing the small tube C, which is bent at a right angle and communicates, by means of a cork, with the larger tube D filled with cotton or asbestus. A set of Liebig's bulbs, E, containing a solution of nitrate of silver, is attached to the other extremity of this tube. The apparatus being mounted, the solution under examination is slightly acidulated and introduced by means of the tube B into the flask A: the disengagement of gas begins immediately. If arsenic and antimony are contained in the solution, arsenetted hydrogen and antimonetted hydrogen are evolved. Both gases are decomposed in passing through the solution of nitrate of silver contained in the Liebig bulbs: the arsenetted hydrogen causes a precipitation of metallic silver, all the arsenic remaining in solution as arsenious acid; the antimonetted hydrogen is decomposed into insoluble antimonate of silver. After the operation has continued for several hours, the apparatus is taken apart, the nitrate of silver solution thrown on a filter, and the precipitate thoroughly washed. An excess of hydrochloric acid is then added to the filtrate, and the precipitate formed separated from the solution by filtration, and well washed. The wash-water is added to the solution, and the whole then examined for arsenic by means of Marsh's test.

The precipitate formed in the nitrate of silver solution, which contains antimonate of silver, is well dried, mixed with a mixture of carbonate and nitrate of soda, and calcined in a porcelain crucible for about three-quarters of an hour. The crucible is then removed from the fire, and the cooled mass treated with hydrochloric acid until a drop of the filtered fluid ceases to give a residue when evaporated upon a watch-glass to dryness. A current of sulphurous acid is now conducted through the filtered solution until the odor of this gas remains persistent. The excess of acid is then removed by boiling, and the solution placed in Marsh's apparatus and tested for antimony.

DETECTION OF MERCURY.

If a mercurial salt exists in a considerable quantity in the substances extracted from the alimentary canal, or ejected either by stools or vomiting, it can be isolated by treating these materials with water, filtering the liquid, and evaporating the filtrate to dryness. The residual mass is taken up with alcohol, and the solution again filtered and evaporated. Upon dissolving the residue obtained by this operation in ether and filtering and evaporating the solution, a residue is obtained which when dissolved in water forms a fluid wherein the presence of mercury can be detected by means of the ordinary tests.

When, however, only a minute quantity of mercury is present, and this has been absorbed, its detection is more difficult. It will be necessary under these circumstances to make use of either Smithson's pile or Flandin and Danger's apparatus.

SMITHSON'S PILE.

Smithson's pile consists of a small plate of copper around which a piece of thin gold foil is wrapped. This is immersed in the solution to be tested for mercury, which has previously been slightly acidulated: if mercury be present, the plate acquires a white color which disappears upon exposure to the flame of a spirit-lamp. A similar reaction occurs in presence of tin, as this metal would likewise be deposited upon the plate, and, upon heating, would penetrate the metal and restore to it its natural color. The danger of mistake arising from this fact is obviated by introducing the copper plate into a tube closed at one end and bent at a right angle. The open extremity of the tube is drawn out to a fine point and immersed in water contained in a second tube also closed at one end. Upon heating the plate in the flame of an alcohol lamp, the white color disappears if produced by mercury, and at the same time this metal condenses in the narrow extremity of the tube. The metallic globules formed can be recognized either by the naked eye or with the aid of a lens, or by rubbing them with a piece of gold foil when the latter will acquire a white coating.

When Smithson's pile is employed, the organic substances are most advantageously decomposed by means of chlorine. It is advisable to operate with as small a quantity of fluid as possible, for, owing to the volatility of bichloride of mercury, a portion of this salt may be lost by the evaporation of aqueous, alcoholic, and even etherial solutions, and the detection of minute quantities rendered impossible.

APPARATUS PROPOSED BY FLANDIN AND DANGER.