PLATE XXXIII.

The principal part of the apparatus consists of a tube made of difficultly fusible glass, and closed at one end; this tube should be 1 foot in length, and from 3 to 4 lines in diameter. Dry hydrate of lime is introduced into this tube, and pushed down to the end, so as to occupy about two inches in length (from a to b). The mercurial compound to be analyzed is intimately mixed in a mortar with an excess of soda-lime, (vide § 40, 4,) and then introduced into the tube, when it is made to occupy the place from 6 to c; those portions of the mixture which adhere to the mortar are removed from the latter with the aid of soda-lime, and the soda-lime which has been used for this purpose is likewise introduced into the tube, so as to occupy the place from c to d; a layer of pure soda-lime is then placed from d to e, and a loose stopper of pure asbestos is pushed down the tube, so as to occupy the place between e and f. The anterior end of the tube is finally drawn out, and bent at a somewhat obtuse angle. The manipulations requisite to perform these various processes, will be found described in detail, § 149, (elementary organic analysis). A few gentle taps upon the table suffice to shake together the contents of the tube in such a manner as to leave a free passage above them in the whole length of the tube.

The tube thus prepared and arranged is now introduced into a combustion furnace, and its point thrust into a receiving flask, filled to about one-half with water; the point is made to rest upon the surface of the water in the flask, in such a manner as partially to close the aperture of the tube. The tube is then surrounded with live coals, proceeding slowly from e towards a, as in organic analysis; the last traces of mercurial vapor are finally expelled the tube by heating the hydrated lime in the end of the

tube. Whilst the tube still remains red-hot, the neck is cut off at f, and carefully rinsed with a syringe bottle, transferring the rinsing water into the receiving flask; the small globules of mercury which have distilled over into the latter, are united into a large one, by agitating the flask, and, after the lapse of some time, the perfectly clear water is decanted or removed by means of a syphon, and the mercury projected into a weighed porcelain crucible, when the water still adhering to it is removed with blotting-paper. The mercury is finally dried under a bell-jar, by the side of a vessel containing concentrated sulphuric acid, and this drying process continued-(but entirely without the application of heat)—until the weight of the mercury ceases to vary. For the properties of this metal, vide § 58. This method, if very properly and carefully executed, yields accurate results. This is more particularly the case with the somewhat more complicated modification adopted by Erdmann and Marchand for the determination of the atomic weight of mercury and of sulphur. (Journal für praktische Chemie XXXI. page 385.-Pharmaceutisches Centralblatt, 1784, page 354.)

b. In the humid way.

The solution of the compound under examination is, in the first place, treated with hydrochloric acid, and evaporated, and this process repeated several times, to ensure the expulsion of the nitric acid which may be present; the solution is then introduced into a flask, free hydrochloric acid added, and subsequently a clear solution of protochloride of tin in excess, which must likewise contain free hydrochloric acid; the mixture is then boiled for some time, and subsequently allowed to cool. After having been left standing at rest for some time, the perfectly clear supernatant fluid is decanted from the metallic mercury, which, if the process has fully succeeded, will be found united into one globule; should this be the case, the globule of mercury may be washed at once by decantation, first with water acidulated with hydrochloric acid, and finally with pure water. The washed mercury is then further treated and determined as

sub. a. If, on the other hand, the particles of mercury have not united into one globule, they are (after the decantation of the clear supernatant fluid) to be kept boiling for a few minutes, with a small amount of moderately diluted hydrochloric acid, which will generally suffice to attain the desired end. For the properties of metallic mercury, vide § 58.

Phosphorous acid, sulphurous acid, and other reducing agents, may, in this process, be substituted for the protochloride of tin.

This method yields accurate results, but it requires the very greatest care in the performance of the various operations. The resulting figures fall generally somewhat short of what they ought to be, according to theoretical calculation. (Compare Experiment No. 61, made by one of my pupils.) This, however, is entirely owing to defective execution of the necessary operations, and not by any means to defects inherent in the method, and the generally received notion, viz. that a small portion of mercury escapes during the operations of boiling and drying, is quite erroneous. (Experiment No. 42.)

2. Determination as protochloride of mercury.

The solution of the compound under examination is mixed with hydrochloric acid, should it not already contain this acid; solution of potass is then added, until the excess of the acid is nearly neutralized; the solution is now mixed with formiate of soda in excess, and kept standing at rest for four days, at a temperature of from 140° to 176°. After the lapse of this time, the fluid is filtered off from the precipitated protochloride of mercury, and the latter collected upon a weighed filter, which has previously been dried at 212°. The filtrate is again kept standing for twenty-four hours, at a temperature of from 140° to 176°; and should a new precipitate form, this is added to the first, and the same process repeated, until the filtrate remains perfectly clear. The whole of the precipitate is then washed, dried at 212°, and weighed. This method is extremely tedious, and requires moreover the strictest care and attention in the performance of the various operations; its application is therefore almost exclusively confined to the sepa

ration of mercury from certain metals. The operator must take particular care to confine the temperature to 170°, since otherwise metallic mercury might separate; should this be the case, the precipitate will exhibit a greyish appearance, and the experiment must, under such circumstances, be considered a failure. 3. Determination as bisulphuret of mercury.

a. The solution of the persalt of mercury under examination is free from nitric acid.

The solution is slightly acidified with hydrochloric acid, should it not already be so, introduced into a flask with ground stopper, and mixed with a recently prepared clear saturated solution of sulphuretted hydrogen in slight excess, so that the odor of sulphuretted hydrogen is clearly perceptible, after agitating the flask; the latter is then stoppered, and the precipitated bisulphuret allowed to subside.

b. The amount of mercury present is so considerable, that its precipitation would require a very large portion of sulphuretted hydrogen water. Washed sulphuretted hydrogen gas (§ 36, 2) is substituted for the water; the gas is transmitted through the moderately dilute solution of the compound under examination. c. The solution contains nitric acid.

Potass is added to the solution until the acid is nearly neutralized; the fluid is then mixed with a clear solution of cyanide of potassium in excess, and the mercury finally precipitated either with sulphuretted hydrogen water, or colorless hydrosulphuret of ammonia, or by transmitting sulphuretted hydrogen gas through it.

The precipitate formed in either case is allowed to subside, collected upon a weighed filter, quickly washed with cold water, dried at 212°, and weighed,

Should the precipitate happen to contain free sulphur, (owing to the presence of peroxide of iron, chromic acid, &c., &c., or to whatever other cause,) it is to be introduced, whilst still moist upon the filter, into a small flask, hydrochloric acid added, heat applied, and nitric acid dropped into the mixture until the

sulphur which remains undissolved appears of a pure yellow color. The mixture is then diluted with water, the fluid filtered off from the sulphur, and the filtrate treated as directed sub. c.

For the properties of the precipitate, vide § 58.

This method, if properly and carefully conducted, yields exceedingly accurate results, and is in my opinion preferable to the other methods.

Most of the compounds of oxide of copper dissolve in water. Metallic copper, oxide of copper, and those of its salts which are insoluble in water, are to be dissolved in dilute nitric acid. Sulphuret of copper is heated with moderately dilute nitric acid until the sulphur which separates appears of a pure yellow color; the addition of hydrochloric acid promotes this decomposition greatly.

b. Determination.

Copper is usually weighed in the form of oxide, (§ 59.) Into this form it is either converted by direct precipitation or ignition; or it is, in the first place, precipitated as sulphuret of copper. The amount of copper present in a solution may also be determined by means of metallic copper; it is in this case, inferred from the amount of metal required to convert a solution of oxide into a solution of sub-oxide of copper.