the most important of all, and that which is undertaken the most often. It has also been the most studied, because the results it affords are of great importance in the purchase of the precious metals. This class of cupellation is effected without difficulty, because the oxide of copper forms so slowly, that the litharge is always enabled to pass it into the body of the cupel. After having weighed the lead and placed it in the cupel, as soon as it is perfectly fused, place it in the alloy to be assayed, wrapped either in blotting-paper or thin leaf-lead. It is essential, in this class of assay, to employ a sufficient quantity of lead to carry away all the copper. We may always be sure of succeeding whatever the alloy may be, by employing the maximum proportion of lead, that is to say, the quantity necessary to pass pure copper; but as the loss which the silver undergoes increases with the length of the operation and with the mass of the oxidized matters, it is indispensable to reduce this loss as much as possible by reducing the proportion of lead to that which is strictly necessary. Long experience has proved, that silver opposes the oxidation of copper by its affinity, so that it is necessary to add a larger amount of lead in proportion to the quantity of silver present.

M. d'Arcet has obtained the following results by the most accurate experiments.

It is remarkable that below the standard of 500, the same proportion of lead must be employed, whatever that of copper. This fact is repeatedly verified by experiment. Whenever fine silver is fused in a cupel, it is always necessary to add lead, in order to cause the button to unite and form well. If less than ths of lead be employed, the button will be badly formed; the litharge could not separate but by the action of a very strong heat, and a considerable loss of silver would ensue. If, on the contrary, roths of lead is exceeded, the cupellation goes on well, but the loss would be greater, on account of the duration of the process. These proportions also ought to vary with the temperature. M. Chaudet has found, that to cupel an alloy containing 900 ths of silver, 5 parts of lead are required in the middle of the muffle, 10 in the front, and only 3 at the back.

1000

The proportion of copper carried off by litharge varies not only with the temperature, but even for the same temperature in relation to the amount of copper and lead the alloy contains. By cupelling 100 parts of copper with different proportions of lead in the same furnace, M. Karsten obtained the following results:

From which we see, that the lead carried away from th

15

tooth of its weight of copper. Much less lead yet can be employed in a cupellation by making the alloy maintain its richness of copper throughout the operation. This can be accomplished by adding to the alloy in the cupel small doses of lead, in proportion as that first added disappears by oxidation. If, for example, an alloy composed of 4 parts of copper and 1 of silver, be fused with 10 of lead, and by adding successive small doses of the latter, as already pointed out, but 7 parts will be consumed, although in the regular way, from 16 to 17 would be employed.

The proportion of oxide of copper contained in the litharge increases each instant, and goes on incessantly increasing when an alloy of copper and lead is cupelled which it contains. According to M. Karsten, this proportion is always about 13 per cent. at the commencement, and 36, or more than a third, at the end of the operation.

In the assay of the coined alloys of copper and silver,

the loss of silver may even amount to

loss is variable, and is proportionally standard of the alloy is lower.

5ths; but the

1000

greater as the

The following table contains the results of many experiments made on this subject.

These numbers, however, are not constant, and vary with the circumstances under which the assays are made; two assays made from the same ingot, by the same assayer, can differ as much as four or five thousandths. Tillet has remarked that the cupels can retain double as much silver as is lost; which proves, as has already been mentioned, that the silver obtained by cupellation is not perfectly pure, but may retain as much as 1 per cent. of lead.

Sulphuret of Silver may be cupelled directly, without the addition of lead; but as it passes readily into the ordinary cupels, Schlutter remarks that the cupels for this purpose must be made of 2 parts of pulverized brick, and 1 of glass. Moreover, this means cannot be considered exact; it is much better, therefore, to make the assay in an ordinary cupel, and use lead.

Argentiferous Galenas may be cupelled directly, whenever they are nearly pure, or contain but 2 to 3 per cent. of foreign matters, such as sulphate of baryta,