it is necessary, before performing the experiment, to convert all such substances into oxides. This is effected by calcination, on Ch or in an open glass tube. The finely powdered assay is placed on Ch and alternately treated with the O F1 and R Fl, and this process is repeated until the substance no longer emits, while in the incandescent state, the odor of sulphur or arsenic. The heat must never be raised so high as to cause fusion, and between every two succeeding calcinations the assay should be taken from the Ch and freshly powdered.

The experiment is generally made on platinum wire, where the color of the bead is more readily observed; Ch is used only in such cases where the substance under examination contains metallic oxides which are easily reduced. It is not sufficient to observe the color of the bead after cooling; but all changes of color which take place during the action of the flame, and through all the various stages of cooling, should be carefully noticed.

Some substances possess the property of forming a limpid glass with borax, which preserves its transparency on cooling, but which, if slightly heated in the O Fl, becomes opaque, when the flame strikes it in an unequal or intermittent manner. This operation has received the name of "flaming," and any substance thus acted upon is said to become "opaque by flaming."

The third and fourth columns of Tables I and II exhibit the behavior of the most important oxides to borax and salt of phosphorus.

In Table III the oxides are arranged with reference to the color which they impart to the beads in O Fl and R Fl.

Examination with Carbonate of Soda.

§ 38. In subjecting a body to the treatment of Sd we have to direct our attention to two points.

Some substances unite with Sd to fusible compounds, others form infusible compounds, and others again are not acted upon at all; in the last case the Sd is absorbed by the Ch and the assay is left unchanged. With Sd unite to fusible compounds with effervescence: § 39. Silicic acid; it fuses to a transparent glassy bead which,

after cooling, remains transparent if the Sd has not been added

in excess [No. 54].

Titanic acid; it fuses to a transparent glassy bead which, when cold, is opaque and of crystalline structure [No. 65].

Tungstic and molybdic acids; the mass, after the union has been effected, is absorbed by the Ch [No. 31 and No. 34].

The salts of baryta and strontia form with Sd fusible compounds which are absorbed by the Ch [No. 57 and No. 58].

§ 40. The second point to be observed is the elimination of metallic matter. Of the metallic oxides, when treated with Sd on Ch in R Fl, are reduced: the oxides of the noble metals and the oxides of arsenic, antimony, bismuth, cadmium, copper, cobalt, iron, lead, mercury, nickel, tin, zinc, molybdenum, tungsten, and tellurium. Of these, arsenic and mercury vaporize so rapidly that frequently not even a coating is left on the Ch. Antimony, bismuth, cadmium, lead, zinc, and tellurium are partly volatilized and form distinct coatings on the Ch. The non-volatile reduced metals are found mixed up with the Sd. To separate them from the adhering S and Ch powder, we may proceed in the following manner:

The fused mass of Sd and metal, and the portion of the Ch im. mediately below and around the assay, is placed in the little agate mortar, rubbed to powder, the powder mixed with a little water, and stirred up. The heavy metallic particles settle to the bottom, part of the Sd dissolves, and the Ch powder remains suspended in the water. The liquid is carefully poured off, and the residue treated repeatedly in the same manner until all foreign matter is removed. The metal remains behind as a dark heavy powder or, when the metal is ductile and easily fusible, in the shape of small flattened scales of metallic lustre. If the substance under examination contains several metallic oxides, the metallic mass obtained is usually an alloy, in which the several metals may be recognized by processes to be described hereafter. It is only in some exceptional cases that separate metallic globules are obtained, ex. gr. in substances containing iron and copper.

For a more detailed account of the behavior of the various metallic oxides under this treatment, see the second column of Tables I and II.

§ 41. The examination with Sd is usually performed on Ch in the R Fl, and, as a general rule, the flux is added successively in small portions. This is particularly necessary when the assay is to be tested for its fusibility with Sd, since a great many minerals, &c., behave very differently with different quantities of the flux.

§ 42. Instead of carbonate of soda, the neutral oxalate of potassa or cyanide of potassium may be advantageously used for all experiments of reduction, since these reagents exercise a more powerful reducing action than common Sd. They are, for this reason, frequently employed when the presence of such metallic oxides is suspected, whose conversion into metals require high temperatures and the aid of a very efficient deoxidizing agent.

Examination with Solution of Cobalt.

§ 43. A few substances, when moistened with a solution of nitrate of cobalt and exposed to the action of the O Fl, assume a peculiar color. The use of this test is, however, very limited, since the reaction can only clearly be seen in such bodies which, after having been acted upon by the O Fl, present a white appearance, or nearly so.

§ 44. Substances which are sufficiently porous to imbue a liquid, are merely moistened with a drop of S Co, placed into the platinumpointed pincers, and treated with the O Fl. Other substances must be powdered, the powder placed on Ch, wetted with a drop of S Co, and treated as above. The color can only be distinguished after cooling. A bluish color, of more or less purity, indicates the presence of alumina [No. 21]; and a pale-reddish color [flesh-color] that of magnesia [No. 59]. It must, however, be borne in mind, that the alkaline and some other silicates, when heated with S Co to a temperature above their fusing point, also assume a blue color, owing to the formation of silicate of cobalt. In testing for alumina, therefore, the heat must not be raised so high as to cause fusion of the assay. In testing for magnesia this precaution is not necessary; on the contrary, the color will appear the brighter and the more distinct, the higher the temperature to which the assay was exposed.

§ 45. Among the oxides of the heavy metals, those of zinc and tin assume characteristic colors with S Co. The reaction is best seen when the assay, alone or mixed with Sd, is exposed to the R Fl on Ch. The ring of oxide which is deposited around the assay is then moistened with S Co and treated with the O Fl. Oxide of zinc takes a fine yellowish-green, and oxide of tin a bluish-green color [No. 36 & No. 33].

§ 46. Besides the compounds above mentioned there are some others which, when exposed to the action of S Co and O Fl, experience a change of color. These bodies are either of very rare occurrence, or the change produced in them is not sufficiently distinct. It will, therefore, be sufficient merely to mention the names of the compounds and the color which S Co imparts to them:

Baryta [brownish-red], tantalic acid [flesh-color], zirconia and phosphate of magnesia [violet], titanic acid, niobic acid, and antimoric acid [green], strontia, lime, glucina, and pelopic acid [gray].

THIRD CHAPTER.

SPECIAL REACTIONS FOR THE DETECTION OF CERTAIN SUBSTANCES WHEN IN COMBINATION WITH OTHERS.

§ 47. THE preceding chapter and accompanying table show the changes which many of the simple chemical compounds undergo when heated, or when treated with the usual blow-pipe reagents. The reactions are sufficiently characteristic to distinguish the various compounds from each other, so that, when any of the above named substances in a pure state is under examination, there is no difficulty to determine its nature. This, however, is not of frequent occurrence, and in the majority of cases the body to be tested will be of a more complex nature. The results of the experiments will vary accordingly. For instance, an ore of cobalt, containing iron, will not impart to the bead of Bx or S Ph in the O Fl a blue color, but a green one, resulting from the mixture of the blue of cobalt and the yellow of iron; lead, when accompanied by antimony, deposits a dark-yellow coating on Ch resembling that of bismuth, &c. In such cases we may often, by attentively observing all the phenomena which present themselves, and by carefully comparing the results obtained by the various experiments, detect many, if not all, of the components of the substance under examination. Sometimes we attain this end quicker by varying the order, or by introducing auxiliary agents into the series of experiments; and in other cases, again, it is only to be arrived at by subjecting the assay to treatments different from those mentioned in the preceding pages.

This chapter contains the principal reactions for the detection of substances which require the application of peculiar agents, and the methods for ascertaining the presence of certain bodies when in combination with others. The alphabetical arrangement will be found of practical use.

(39)