a paste, then place the mass on charcoal, and direct upon it a strong reducing flame. The flux will at first be absorbed; but more must be added, and the heat continued as long as any portion of the assay remains on the surface of the charcoal. The first portion of carbonate serves for collecting the metallic contents, while the latter is for completing the reduction of the remaining unreduced metallic oxide. A few drops of water are now added to extinguish the ignited parts, and then the whole of the charcoal upon which the assay rested, is cut out with a knife, and put into an agate mortar, care being taken to lose nothing in the transfer. The mass is mixed with water, and ground to a fine powder. If there be any metal, from its greater specific gravity, it falls immediately to the bottom. Repeated levigations and decantations are then resorted to, till the whole of the pulverized charcoal is removed. When the charcoal is entirely washed away, the metal, if any had been in the assay, will be found at the bottom of the mortar. If it is an infusible or a brittle metal, it appears in the form of a metallic powder; if a malleable one, it produces flat shining plates.

By this method, half a per cent. of Tin, and even a much smaller quantity of Copper, may be very easily detected. If, however, a number of metallic bodies are contained in the same substance, they will be reduced, together, to a metallic alloy. Some, notwithstanding, are obtained singly, as copper and iron, which give reguli distinct from other metals.

The metals which can be reduced by this means, besides the noble ones, are,-Molybdenum, Tungsten, Antimony, Tellurium, Lead, Zinc, Tin, Copper, Bismuth, Nickel, Cobalt, and Iron. The last three are magnetic, and can therefore be at once distinguished from the others. However, there are a few of these which are partly or altogether volatile, and coat the charcoal with sublimated oxides; namely, Antimony, Tellurium, Bismuth, Lead, and Zinc. Some other metals, as Arsenic, Cadmium, and Mercury, are also reducible, but, being extremely volatile, must be heated in glass tubes, in order to obtain their sublimates.

If, in these experiments, a metallic button, containing a number of metals, is obtained, the alloy must be examined on char

coal, with borax or microcosmic salt, as will be hereafter adverted to when treating of different qualitative metallic assayings.

When the substance under examination is a sulphide or arsenide, or contains either sulphur or arsenic, these bodies must in the first place be driven off, and the metal oxidized, before the reduction test can be made. The process is conducted as follows: the sample is ground to an impalpable powder, and pressed into a shallow cavity in the charcoal with the spatula, then acted upon by the flame. The oxidating flame is to play upon the assay first, by which means part of the sulphur is given off as sulphurous acid, and the metal oxidized, which combines with the sulphuric and arsenic acids, also formed in the process. When no more sulphurous acid is evolved, which is readily recognised by its characteristic odor, the reducing flame must be brought into action, whereby all sulphuric and arsenic acid is reduced, and the greater part of the metallic arsenic is expelled. When the alliaceous smell is no longer discernible, the oxidating flame is again resorted to, until decomposition is so far accomplished as to satisfy the operator. The assay is now turned over on the charcoal, and the other side treated in the same manner. The mass is then removed, again powdered, and once more treated on the charcoal with the oxidating and reducing flames. The sulphur is by this means entirely got rid of, but arsenic frequently remains as arsenic acid, in combination with such oxides as those of cobalt or nickel, and can only be removed with great difficulty. When the substance contains a large proportion of arsenic, which might be dangerous if it were allowed to escape in the room, previously to roasting, it should be treated in a tube, by which means the greater part of this metal sublimes, and condenses in the upper part.

If the substance contains neither cobalt nor nickel, but another metal with arsenic, the metal can be obtained by the reduction of the roasted sample with soda, perfectly free from the arsenic ; but if cobalt or nickel be present, the reduced metal will always be contaminated with arsenic, and must therefore be further treated with borax on charcoal. In this reduction experiment, an operator must allow an uninterrupted strong flame to play

A

upon the assay. Great care must be exercised in separating the charcoal which has become impregnated with reduced metal, and also in the levigation and decantation; and, lastly, the metal obtained in grains, plates, or fine powder, must not only be inspected with the naked eye, but also through a magnifying glass. When the substance is an earth, containing no coloring metallic oxides, it is to be treated with a solution of nitrate of cobalt. piece is selected, of sufficient porosity to allow the solution of cobalt to permeate, and a few drops added. It is then subjected, in the points of the forceps, to a moderate oxidating flame; during this process, the color will assume a characteristic aspect-if blue, pure alumina is indicated; if rose-red, magnesia. If the latter color is discernible, the assay must be fused, because, when magnesia is present, the rose-red tint is better brought out in a strong flame. Minerals containing lime, or an alkali, give a blue tinge, when strongly heated with cobalt; but alumina gives it by a moderate heat, and can, therefore, be readily distinguished.

Those crystalline rocks which will not admit of the solution permeating, must be reduced to a fine powder in an agate mortar ; then kneaded into a paste with water, and spread on the charcoal. The solution of nitrate of cobalt is now added, and the mass subjected to an incipient red heat. If decomposition does not take place, the color imparted,—blue, red, or black, proceeds from the cobalt solution only, and not from any of the ingredients. If the mass separates from the charcoal, after the moderate red heat, as a thin pellicle, it must be submitted to an intense oxidating flame, in the points of the forceps. The color now imparted must be observed on cooling, by daylight; because, when seen at candlelight, a blue often appears dirty violet, or even red, by transmitted light.

When a metallic oxide is present in a mineral, which may give a colorless result, it will not interfere with the alumina or magnesia reaction, unless a large quantity is present, and then the mass will generally become grey or black. In treating some minerals per se on charcoal, or in the reduction test with carbonate of soda, the metal often sublimes, becomes oxidized immediately, and is deposited upon the charcoal. Many such oxides

are easily expelled by an oxidating flame, others with greater difficulty, and some not at all; so that it is often almost impossible to separate the pellicle of oxide from the charcoal. If such a film rest upon the support after the experiment, a few drops of a solution of nitrate of cobalt is to be added to it, and the assay submitted to a pretty strong oxidating flame. If zinc be present, a distinct bright green color will be imparted to the substance, which should be viewed through the microscope.

The quantity of the cobalt solution to be employed depends greatly upon its strength: experience, however, very soon teaches the proper quantities to be taken, in order to obtain a decided reaction.

Minerals, et cetera, examined according to the foregoing rules, in flasks, open tubes, and with fluxes, in the different flames produced by the Blowpipe, give numerous phenomena from which results must be gathered; and attention must be given to the slightest occurrences that ensue, for such often lead to the discovery of substances, never supposed to exist in the assay under examination.

The results, however, obtained in the examination of an unknown body before the Blowpipe, can be fully appreciated only by those who are acquainted with chemistry, and, therefore, no rules can be laid down sufficient to detail all the minutia, as they can be arrived at by industry and practice only.

¶B General Rules for Qualitative Blowpipe Analysis, by means of which, partly with the assistance of the moist way, the single ingredients of compound bodies may be detected.

If by means of the Blowpipe, alone, or with glass fluxes, the constituents of a compound body are not satisfactorily determined, the moist way must be resorted to, in order to ascertain, as nearly as possible, the collective ingredients. By such an association an operator can, with accuracy and facility, examine the generality of bodies composed of different earths and metallic oxides, native silicates, slags, et cetera.

For the manipulation, 100 to 150 milligrammes of a substance are quite sufficient to take, in order to detect all the ingredients; and should the operator be an expert Blowpipe analyst, he will find that less will serve-only a small portion of a liquid reagent is necessary. When its pyrognostic qualities have been remarked, or the class to which it belongs has been determined, and whether it is an alkali,-earth, metallic salt, silicate, aluminate, compound metallic oxide, metallic oxide with an earth, metallic sulphide,—or a combination of different metals, including arsenic and selenium, the single ingredients may readily be recognized. The procedure will be particularly adverted to in the qualitative analysis of alkalies, earths, metallic and non-metallic bodies, and such substances as are met with in nature. To prevent a repetition of many manipulations, we shall previously describe the methods for melting and dissolving, et cetera, and afterwards give them separately.

§ 1. SMELTING OF THE SUBSTANCE WITH BORAX AND CARBONATE OF SODA.

A compound in which the earths cannot be detected by a single reaction, must be triturated in an agate mortar to an impalpable powder, and from 75 to 100 milligrammes employed for the analysis. If it contains sulphides or arsenides it must be roasted twice with charcoal, in a manner similar to the quantitative assaying of copper, to expel all or the greater portion of the sulphur or arsenic, and to oxidize the metals which are generally combined with traces of formed sulphuric or arsenic acids. The powder is now triturated in a mortar with carbonate of soda and borax, the proportions of these reagents being rectified according to the degree of solubility of the assay; the mixture is then pressed in a cornet of soda paper, similar to a quantitative copper assay, and operated upon in a cylindrical cavity made in a piece of charcoal, by an intense Blowpipe-flame. In general, the melting is more readily effected in the oxidating than in the reducing flame. If sulphuric acid or reducible metallic oxides are not present, the oxidating flame is always preferable, but if otherwise,