bb. A permanent precipitate is formed. Add to the fluid a little yellow sulphide of sodium, drop by drop, until the metals of the fourth and fifth groups present in the alkaline solution are just thrown down, heat moderately, filter, and treat the filtrate as directed 219. Dissolve the washed precipitate in nitric acid, which may leave sulphide of mercury behind, and examine the solution for copper and lead, as well as for zinc and other metals of the fourth group, which may, in the same way as copper, have passed into the alkaline solution, by the agency of organic

matters.

B. To test the alkaline fluid, which now also contains some 219 sulphide of an alkali metal, for mercury (which may be present, as its sulphide is soluble in sulphide of potassium), and for metals of the sixth group, mix with a sufficient quantity of water, then with dilute sulphuric acid to acid reaction, and if the fluid does not smell strongly of hydrosulphuric acid, add some more of the latter reagent.

and

Absence of mercury
Pass on to 220.
Filter, wash the precipitate,

aa. No precipitate is formed.
the oxides of the sixth group.
bb. A precipitate is formed.
then examine it for mercury and the metals of the sixth
group according to § 191.

7. The fluid, acidified with sulphuric acid, may still con- 220
tain those metals which in combination with cyanogen form
compound radicals (iron, cobalt, manganese, chromium),
and, besides these, also alumina. You have to test it also for
cyanogen, ferrocyanogen, cobalticyanogen, &c., and for
other acids. Divide it therefore into two parts, aa and bb.
Treat it according to § 199 or, as the case may be,
§200 to detect the acids.* (Cobalticyanogen may be re-
cognised by giving a greenish precipitate with nickel salts
and white precipitates with zinc and manganese salts,
which may be proved to contain cobalt by means of the
borax bead.)

aa.

bb. Evaporate it nearly to dryness, add some pure concentrated sulphuric acid and heat till the free acid is for the most part expelled. Dissolve the residue in water and test the solution for iron, manganese, cobalt, alumina, and sesquioxide of chromium according to § 194. 2. Decompose another portion by continued heating with pure concentrated sulphuric acid, remove all other bases and then test for alkalies.

II. ANALYSIS OF SILICATES.
$205.

Whether the substance is a silicate or contains one, is ascer- 221 tained by the preliminary examination with phosphate of soda and ammonia before the blowpipe; since in the process of fusion the

* It must be remembered that ferricyanogen may have been converted into ferrocyanogen by the oxidation of certain bases, &c., thus:-Cy, Fe, 3 K+K O, HO +2 Fe 0=2 (Cy, Fe, 2 K) + Fe, 03 + HO.

metallic oxides dissolve, whilst the separated silicic acid floats about in the liquid bead as a translucent swollen mass (§ 150, 8).

The analysis of silicates differs, strictly speaking, from the common course only in so far as the preparatory treatment is concerned, which is required to separate the silicic acid from the bases, and to obtain the latter in solution.

The silicates and double silicates are divided into two classes, which require different methods of analysis; viz., (1) silicates readily decomposable by acids (hydrochloric acid, nitric acid, sulphuric acid), and (2) silicates which are not, or only with difficulty, decomposed by acids. Many rocks consist of mixtures of the two

classes of silicates.

To ascertain to which class a given silicate belongs, reduce it to a very fine powder, and digest a portion with hydrochloric acid at a temperature near the boiling point. If this fails to decompose it, try another portion by long-continued heating with a mixture of three parts of concentrated sulphuric acid and 1 part of water. If this also fails, the silicate belongs to the second class. Whether decomposition has been effected by the acid or not, may generally be learned from external indications, as a colored solution forms almost invariably, and the separated gelatinous, flocculent, or finely pulverulent hydrate of silicic acid takes the place of the original heavy powder which grated under the glass rod with which it was stirred. But whether the decomposition is complete, or extends only to one of the components of the rock, may be ascertained by boiling the separated hydrate of silicic acid, after washing, in a solution of carbonate of soda. If perfect solution ensues, complete decomposition has been effected; if not, the decomposition is only partial. The results of these preliminary tests will show whether the silicates should be examined according to § 206, or § 207, or § 208.

Before proceeding further, examine a portion of the substance also for water, by heating it in a glass tube. If the substance contains hygroscopic moisture, it must first be dried at 100° for a long time. Apply a gentle heat at first, but ultimately an intense heat; you may also conveniently combine with this a preliminary examination for fluorine (§ 146, 8).

A. SILICATES DECOMPOSABLE BY ACIDS.

$206.

a. Silicates decomposable by hydrochloric or nitric acid.*

1. Digest the finely pulverized silicate with hydrochloric acid at 222 a temperature near the boiling-point, until complete decomposition is effected, filter off a small portion of the fluid, evaporate the remainder, together with the silicic acid suspended therein, to dryness, heat the residue at 100° (scarcely above) with constant stirring, until hardly any more hydrochloric acid fumes escape, allow to cool, moisten with hydrochloric acid or, as the case may be, with nitric acid, afterwards add a little water, and heat gently for some

time.

* Nitric acid is preferable to hydrochloric acid where compounds of silver or lead are present.

This operation effects the separation of the SILICIC ACID, and the solution of the bases in the form of chlorides or nitrates. Filter, wash the residue thoroughly, and examine the solution by the common method, beginning at § 189, II. or III. The residual silicic acid must always be tested, as it cannot under any circumstances be considered pure. It frequently contains titanic acid, occasionally sulphate of baryta, possibly also sulphate of strontia, and often a little alumina.. It is best tested by repeated heating in a platinum dish with hydrofluoric and sulphuric acids, until all the silicic acid is removed. in the form of fluoride of silicon. The residue is ignited, fused with bisulphate of potassa and then treated with cold water. If anything insoluble now remains, it is filtered off and tested according to § 99 for SULPHATE OF BARYTA (and strontia). The dilute aqueous solution is tested by long boiling for TITANIC ACID* (104, 9), and the filtrate therefrom is tested by ammonia for ALUMINA. (Should there be any chance of the presence of chloride of silver in the silicic acid, digest a portion with ammonia, filter and examine the filtrate by supersaturation with nitric acid.)

2. As in silicates, and more particularly in those decomposed 223 by hydrocholoric acid, there are often found other acids, as well as metalloids, the following observations and instructions must be attended to, that none of these substances may be overlooked :

a. CARBONATES are detected in the process of treating with hydrochloric acid. SULPHIDES are often detected in the same operation, otherwise they are tested for according to § 156, 8.

B. If the separated silicic acid is black, and turns white upon ignition in the air, this indicates the presence of CARBON or of ORGANIC SUBSTANCES. In presence of the latter, the silicates emit an empyreumatic odor upon being heated in the glass tube.

7. Test the portion of the hydrochloric acid solution filtered. off before evaporating for SULPHURIC ACID, PHOSPHORIC ACID, and ARSENIC ACID for sulphuric acid by diluting and adding chloride of barium; for arsenic acid by heating the solution to 70° and conducting sulphuretted hydrogen into it; for phosphoric acid by adding nitric acid, evaporating to dryness on the water-bath, warming the residue with nitric acid, filtering and adding molybdate of ammonia. Where arsenic is found, phosphoric acid is tested for in the fluid filtered from the sulphide of arsenic.

8. BORACIC ACID is best detected by fusing a portion of 224 the substance in a platinum spoon with carbonate of soda and

*If the silicic acid has been separated by evaporation on the water-bath, only a small part of the titanic acid will be found remaining with it, the rest will pass into the hydrochloric acid solution and will be precipitated by ammonia in conjunction with the sesquioxide of iron and alumina. To find this, fuse the dried precipitate with bisulphate of potassa, dissolve the fusion in cold water, filter if necessary, dilute considerably, pass sulphuretted hydrogen until all the sesquioxide of iron is reduced, and (without filtering off the sulphur) keep the fluid boiling for half an hour with a constant current of carbonic acid passing through it. Filter, wash and ignite; the sulphur will burn off, the titantic acid will remain. Should it still contain iron, redissolve it by fusion with bisulphate of potassa and treatment with cold water, and precipitate by boiling with hyposulphite of soda.

་

potassa, boiling the fused mass with water, and testing the
solution by§ 144, 6.

e. With many silicates, boiling with water is sufficient to
dissolve the CHLORIDES present, which may then be readily
detected in the filtrate by nitrate of silver; the safest way,
however, is to dissolve the mineral in dilute nitric acid, and
test the solution with nitrate of silver.

. FLUORIDES, which often occur in silicates in greater or smaller proportion, are detected by § 146, 6.

b. Silicates may resist the action of hydrochloric acid, but are decomposed by concentrated sulphuric acid.

Heat the finely pulverized mineral with a mixture of 3 parts 225 of concentrated pure sulphuric acid and 1 part of water (best in a platinum dish), finally drive off the greater portion of the sulphuric acid, boil the residue with hydrochloric acid, dilute, filter, and treat the filtrate as directed § 190; and the residue, which, besides the separated silicic acid, may contain also sulphates of the alkaline earths, &c., as directed § 206, 1. If you wish to examine silicates of this class for acids and salt radicals, treat a separate portion of the substance according to § 207.

B. SILICATES WHICH ARE NOT DECOMPOSED BY ACIDS.*

§ 207.

As the silicates of this class are most conveniently decomposed 226 by fusion with carbonate of soda and potassa, the portion so treated cannot, of course, be examined for alkalies. The analytical process is therefore divided into two principal parts, a portion of the mineral being examined for the silicic acid and the bases, with the exception of the alkalies, whilst another portion is specially examined for the latter. The mineral must also be examined for other acids.

1. Detection of the silicic acid and the bases, with the exception of the alkalies.

Reduce the mineral to a very fine powder, mix this with 4 227 parts of carbonate of soda and potassa, and heat the mixture in a platinum crucible over a gas or Berzelius spirit-lamp until the mass is in a state of calm fusion. Place the red-hot crucible on a thick cold iron plate, and let it cool there: this will generally enable you to remove the fused cake from the crucible, in which case break the mass to pieces, and keep a portion for the examination for acids. Put the remainder, or, if the mass still adheres to the crucible, the latter with its contents into a porcelain dish, pour water over it, add hydrochloric acid, and heat gently until the mass is dissolved, with the exception of the silicic acid, which

It will be understood, from what has been stated § 205, that these are not decomposed by heating with hydrochloric acid and sulphuric acid in open vessels; but by heating them, reduced to a fine powder, in a sealed glass tube, with a mixture of 3 parts of concentrated sulphuric acid and 1 part of water, or with hydrochlorie acid, to 200°-210°, most of them are decomposed, and may accordingly be analysed also in this manner (AL. MITSCHERLICH).

separates in flakes. Remove the crucible from the dish if necessary, evaporate to dryness, and treat the residue as directed 222.

2. Detection of the alkalies.

To effect this the silicate must be decomposed by means of 228 a substance free from alkalies. Hydrofluoric acid or a metallic fluoride answers this purpose best; but fusion with hydrate of baryta will also answer the purpose.

a. DECOMPOSITION BY MEANS OF A METALLIC FLUORIDE.-Mix 1 part of the very finely pulverized mineral with 5 parts of fluoride of barium, or pure, finely pulverized fluoride of calcium, or with 3 parts of fluoride of ammonium, stir the mixture in a platinum crucible with concentrated sulphuric acid to a thickish paste, and heat gently for some time in a place where the fumes will be well carried away; finally heat a little more strongly, until the excess of sulphuric acid is completely expelled. Boil the residue with water, add chloride of barium cautiously as long as a precipitate continues to form, then baryta-water to alkaline reaction, boil, filter, mix with carbonate of ammonia and some ammonia as long as a precipitate forms, and proceed as directed 168.

b. DECOMPOSITION BY MEANS OF HYDRATE OF BARYTA.— 229 Mix 1 part of the very finely pulverized substance with 4 parts of hydrate of baryta, expose the mixture for half an hour in a platinum crucible to the strongest possible heat of a good Berzelius or gas-lamp, and treat the fused or agglutinated mass with hydrochloric acid and water until all the bases are dissolved; precipitate with ammonia and carbonate of ammonia, filter, evaporate to dryness, ignite, dissolve the residue in water, add a little pure lime, boil, filter, precipitate again with ammonia and carbonate of ammonia, filter, evaporate, ignite, and test the residue for potassa and soda as directed §197.

3. Examination for fluorine, chlorine, boracic acid, phosphoric acid, arsenic acid, and sulphuric acid.

Use for this purpose the portion of the fused mass reserved in 230 227, or, if necessary, fuse a separate portion of the finely pulverized substance with 4 parts of pure carbonate of soda and potassa until the mass flows calmly; boil the fused mass with water, filter the solution, which contains all the fluorine as fluoride of sodium, all the chlorine as chloride of sodium, all the boracic acid as borate, all the sulphuric acid as sulphate, all the arsenic acid as arsenate, and at least part of the phosphoric acid as phosphate of soda, and treat as follows:

a. Acidify a small portion with nitric acid, and test for CHLORINE with nitrate of silver.

b. Test another portion for BORACIC ACID as directed § 144, 6.

c. To detect FLUORINE, treat a third portion as directed § 146, 7.

d. Acidify the remainder with hydrochloric acid and test a small portion with chloride of barium for SULPHURIC ACID; heat the remainder to 70°, and test with hydrosulphuric acid