VI. and VII.), the student has been somewhat prepared for this system, which, although at first sight it may appear to be a waste, is in reality an economy of time. The only true source of certainty in analysis consists, in fact, in this application of tests in a certain fixed order. The first step must always be to separate a group of substances (such as a subdivision of basic radicals) the members of which may always be removed together as one precipitate by the employment of a certain reagent. The labour is thus simplified: for if such a reagent fail to produce a precipitate, we know that an entire group of bodies is absent; if, on the contrary, a precipitate is produced, we have an assurance that it can only contain certain members, and know therefore the limit of our search in that direction. This accomplished, a second group-test is applied, and the same course followed, until, having exhausted group-tests, we have ascertained in what subdivision or subdivisions the substance or substances sought for exist. Then, by the application of less general tests, we continually subdivide the groups until the isolation of the individual members is accomplished. This principle applies to the analysis of simple salts or mixtures-to the detection of basic or acidradicals. And the student cannot be warned too early of the extreme folly of what may be termed analytical "angling "—of the promiscuous employment of tests which, when properly applied, are extremely effective in detecting individual substances. Complicated results may thus ensue, to unravel which may baffle all the ingenuity of the student; many substances, too, may thus be entirely overlooked, in consequence of the special test employed acting upon bodies other than the one sought for, and in a way not remembered.

To take an instance from a frequent occurrence in the laboratory. The student has a solution to analyse, the colour of which is green. He instantly concludes that the base is copper; and instead of employing the ordinary sequence of group-tests, he devises a short and easy method. Knowing that hydrate of ammonium gives a characteristic reaction with copper salts, he adds it a green precipitate is formed, and then redissolved; but the solution does not present the deep blue colour of cuprammonium salts. Thus he is disappointed; but, still under the impression that he is dealing with a

copper salt, he tries the action of hydrate of potassium: a green precipitate occurs, somewhat pale, it is true, but the student nevertheless regards it as confirmatory of his original supposition. Ferrocyanide of potassium is next added; and the green precipitate which follows is a new perplexity. In despair at these results so conflicting and so contradictory of his original idea, he adds sulphide of ammonium: the black sulphide formed confirms his first supposition; the doubts consequent upon the previous reaction clear away; and he definitely pronounces the solution to contain copper. Why has nickel been thus obviously mistaken for copper? Because the experimenter, in defiance of the conflicting evidence which the special tests have afforded, has entirely neglected all proper sequence of experiment, and omitted to apply the test which claims precedence of all-the group reagent, hydrosulphuric acid in an acid solution, by the employment of which he would at once have been able to decide the question about which so much time has been wasted.

The student, however, should be equally careful not to trust. too implicitly to a mechanical familiarity with the routine of analysis, and should learn by repeated trials to assign to each indication no more than its due value. The accumulation of evidence is also of the highest importance.

Next to the accurate recollection of reactions characteristic of each basic and acid-radical, a scrupulous attention to the orderly application of tests is, as we have seen, most necessary to success in analysis; and so, in order that the student may be well practised in this method, it is advisable that he should at first test substances for basic radicals only, reserving for a second stage the analysis of salts with a view to the detection of their acid-radicals: for these latter are, for the most part, identifiable only by a chain of circuitous evidence. And it must be remembered that, upon the knowledge of the basic radical present in a salt, the mode of testing for its acid constituent is frequently founded; indeed, the presence of the latter is often ascertained in the examination for the former constituent. We now proceed, therefore, to details.

CHAPTER II.

QUALITATIVE ANALYSIS OF A SINGLE SALT.

SECTION I.-Examination for the basic constituent.

THE student will have learned, from the analytical schemes already given (Part I. Chapter VI., pp. 81, 97, 150, 190, 245), what basic radicals he should be prepared to find in any salt. In order to facilitate analysis, it has been found advisable to perform in the first place a few experiments, chiefly with the blowpipe, upon the dry salt. These experiments constitute what has been termed "the preliminary examination ;" and the results which it yields with a simple salt are often so decisive as at once to determine the nature of the basic radical present. Skill in this blowpipe analysis is of great service to the travelling mineralogist, since the apparatus necessary for conducting it may be packed in a small compass, and the results obtained afford a tolerably certain and valuable guide in examining the minerals of the country through which he passes. The student, however, who has the means of controlling these results by further analysis of a solution of the substance, or "analysis by the wet method," should always consider the latter, as in truth it is, the mode of procedure most to be relied on, and should then employ the blowpipe examination only as the preliminary step to a more accurate analysis. A convenient form of this examination is given in the following Table, in which, and in the other Tables which succeed, symbols are often employed in the place of words.

Simple soluble salts for analytical examination may be selected from those mentioned under each basic radical in Chapter VI. For acid-radicals, their combinations with sodium, potassium, or ammonium may in general be used. Examples of insoluble salts may be obtained from those printed in antique type.

Preliminary examination for the detection of the basic radical.

* We cannot be sure that the substance belongs to B, and not to C, until we have also heated it on charcoal with Na2CO3 or KCy.

Most salts containing water of crystallization fuse as soon as heated, from the solvent action of the water; but the substance under examination can only be pronounced fusible when it continues so under the prolonged action of heat.

The student should make careful and ample notes of all observations made, drawing up his results in the tabular form given above; and having done this, he should proceed to make a solution of the substance under examination, preparatory to the actual analysis.

ACTUAL ANALYSIS FOR THE DETECTION OF ONE BASIC RADICAL.

If the substance be soluble in water, that liquid is the best solvent that can be employed, since all others entail more or less trouble in the subsequent treatment; and so desirable is it that water should if possible be used, that in analysing a substance known to be a single salt only, if it is but slightly soluble in water, that aqueous solution may be taken in preference to an acid one. The solubility or insolubility of any substance in a given liquid may be ascertained very readily by boiling the solid and liquid together, allowing the mixture to cool, and then filtering it. A drop of the filtrate is now evaporated on a bright piece of platinum foil or in a watch glass, when, if a tolerable residue be visible, enough substance is contained in the solution for the purposes of the analyst. But if the residue be scarcely appreciable, or if none be present, the substance must be considered insoluble, and another course adopted.

If water fails to dissolve the substance, acids must be employed. The incautious use of acids, however, leads not unfrequently to failure in analysis; great judgment is therefore necessary in dealing with these solvents. The acids most commonly employed are hydrochloric and nitric, or a mixture of these; but before proceeding to the selection of one of these solvents, the student must recall the results of his preliminary examination, if indeed the substance has been found insoluble in water. If, in the blowpipe experiments, a malleable globule of the lustre and colour of silver or lead, or a sublimate indicative of mercury has been obtained, nitric, not hydrochloric acid, must be employed. The reason of this is apparent. Hydrochloric acid, in dissolving silver, mercurous, or lead salts, produces an insoluble chloride of those metallic radicals, which will baffle the student's