The

assayed both for regulus, or matt, and metal. assay for regulus has for its purpose the determination of the relative quantity of sulphuret and gangue in the ore; and, in consequence, the quantity of regulus a given quantity of ore would produce in the large way.

The assay for regulus is simple and easy. It consists in acting upon the ore with a substance capable of determining the fusion of the gangue, but exercising no decomposing action on the sulphurets. No flux fulfils these conditions better than borax. With equal parts of glass of borax, all ores to be assayed in this manner, fuse very readily at a full red heat. Naked pots are generally employed for this kind of assay; but then great precaution must be taken to collect the substance without loss; because, in the first place, it may adhere to the crucible, and because it is exceedingly brittle. This inconvenience may be avoided by the employment of crucibles lined with charcoal, which Berthier says are indispensable in the assay of very poor ores. In my own experience, I have never found any inconvenience from the use of naked crucibles, where the proper degree of heat has been employed.

If it be thought necessary to use as fluxes the same substances as those employed in the large way as lime, baryta or quartz, the assay must be made at the same temperature as an iron assay in a lined crucible -a plain one will answer every purpose.

It is essential to remark that every regulus produced is composed of sulphurets, containing the minimum amount of sulphur; since all the sulphurets contained in ores are found at a higher degree of sulphuration, a certain quantity of sulphur sublimes during fusion, so

that the amount of gangue cannot be deduced from the quantity of regulus produced.

By the humid method, however, the proportion of gangue can be readily determined, and in consequence that of the sulphuret; but the proportion of regulus producible by fusion with borax cannot be determined by this experiment, because the nature of the sulphurets being unknown, the amount of sulphur they would give off during fusion could not be found. By a combination of the two methods a complete knowledge of the nature of the substances assayed is obtained.

In order to make the assay by the humid method, the substance is pulverized and acted on by acetic acid which dissolves but the calcareous gangue. It is then dried and weighed. The residue is then boiled in aqua regia, and washed, and roasted, to burn off sulphur. The stony gangues left, after this process, are weighed, and the metallic part calculated by difference.

The process just described supposes the substance assayed does not contain a notable proportion of oxide of copper or iron; in case it did they would be dissolved by either acid, and the true proportion of the sulphurets could not be ascertained. This proportion also could not be determined by the dry way; but the proportion of regulus producible on the large scale, could be very accurately determined by fusing the ore with borax or any other non-desulphurating flux; and that is all which is generally required.

The assay of the ores of copper containing sulphur, requires necessarily a roasting; and a roasting made with such care that neither sulphur nor sulphuric acid shall remain in the substance; without which in the

subsequent fusion with a reducing flux the slag would retain traces of copper.

Berthier says, all other processes which have been attempted to be employed have been without success, and present more difficulties than the usual method of manipulation.

The presence of iron whose sulphuret has so great an affinity for that of copper, presents an especial difficulty which it appears exceedingly hard to over

come.

With nitre all the copper can be extracted (from the sulphuret), but with difficulty, and finding by repeated experiments the quantity which produces the maximum result. Sulphuret of copper fused with a mixture of an alkaline carbonate and metallic iron, allows a certain quantity of copper to be set free; but this quantity never surpasses above three-fourths of that contained. This maximum is obtained by using four parts of carbonate of soda, and thirty to forty per cent. of iron filings. The slag is black and homogeneous. Copper pyrites is decomposed by the alkaline carbonates, but without the production of metallic copper. A black crystalline homogeneous slag is formed, which contains an alkaline sulphuret, a sulphuret of iron, sulphuret of copper, and oxide of iron.

With black flux a very fluid homogeneous crystalline slag is produced, the colour of which is black with a metallic lustre, and in which not the smallest trace of metallic copper can be distinguished. The result is the same as the addition of iron, which is perfectly inert, and remains disseminated in the slag.

Copper pyrites is readily attacked by nitre, and either

copper or sulphuret of copper can be separated from it by means of this reagent. But in order to arrive at this result, it is necessary to guess the quantity of nitre; so that it is evidently not a good method of assay.

It is necessary, in order that the slag be fluid, to add a little borax or an alkaline carbonate; so that the reduced metal may collect into one button. With one part of nitre, two of carbonate of soda, and one of borax, pure copper pyrites gives from thirty-six to forty-six per cent. of sulphuret; with double that quantity of nitre it gives thirty per cent. of metallic copper.

The roasting of pure sulphuret of copper, or copper pyrites, (a mixture of the sulphurets of copper and iron), is a very easy operation-so Berthier observes. The only precaution necessary is to manage the heat very carefully for some time, in order to avoid agglomeration; but after a certain quantity of oxide has been produced, a red heat may be employed without inconvenience. The ore must be stirred continually so as to bring necessarily every portion into contact with the atmosphere.

When the roasting is considerably advanced, it is best to push the heat to bright redness from time to time, in order that the sulphates and sulphurets may mutually decompose each other. Lastly, when no more sulphurous acid is evolved, the heat may be raised to whiteness for some minutes to decompose the last traces of sulphates-which decomposition this process very seldom effects. When the roasting has been executed with care, the substance fused with three, or four times its weight of black flux, gives nearly all the

copper it contains, and all the contained iron remains in the slag, partly as oxide and partly as metallic iron; but in a very finely divided state. If the roasting has been imperfect, the slag will contain sulphur, which will abstract a considerable amount of metallic copper from the assay button.

When the ore is poor and contains much iron, the slags resulting from the assay are always pasty; and some care must be taken, otherwise they will retain as much as two to three per cent. of copper. It is preferable, in these cases, to estimate the copper by the wet way by dissolving in aqua regia, and subsequent precipitation by metallic iron. The humid method (Berthier) of assay is preferable for the ores belonging to this class.

Mr. Lewis Thompson has proposed several methods for the assay of copper ores, both by the wet and dry method. The following is an outline of his wet and dry process.

In the dry process, the ore is pulverized finely and heated to redness in a crucible, and nitre thrown upon it till there is an excess which may be determined by the deflagration producible by an excess of that agent, on the introduction of a piece of charcoal. When this happens, charcoal is to be added until all the excess of nitre is destroyed. The pot and its contents is then to be raised to a full red heat for about a quarter of an hour, and the button produced in the operation weighed.

This process is not perfect, for reasons before pointed out. And the amount of copper obtainable from the same sample of ore varies in its quality very much.

The same, however, cannot be said of his humid