The final weight of the gold cornet in dicates the absolute quantity of this metal in the assayed sample. The difference between the weight of the button after cupellation (deducting the silver added) and the first sample, is the weight of the copper, or other base metal in the gold ; and the difference between the gold cor net, together with the silver added, and the button after cupellation, is the quan tity of silver with which the gold was al loyed.
The silver is usually recovered from the solution left after parting, by immer sing in it plates of bright copper, which dissolve and precipitate the silver in its metallic form.
Touch-needles for gold are formed in the same manner as for silver, but more of them are required, as the various com binations of three metals are to be exami ned by them in this case. Four sets of them are usually employed ; one in which pure silver is used for the alloy, another in which the alloy is two parts silver and one of copper, a third with two parts cop per and one of silver, and a fourth of copper only. In trials with these nee dles nitric acid is of singular service; a drop of it is let fall on the streak of metal on the touch-stone ; in eight or ten se conds it is washed off and the effect ob served. If the streak preservesits gold en brilliance unaltered, the metal is judged to be of a certain degree of fine ness. If it looks red, dull, and coppery, it is less fine ; but if the streak is almost entirely effaced, it contains very little gold.
A peculiar set of weights are used for assaying.
The quantity of metal taken for an as say is always very small ; in this country generally from 18 to 36 grains troy for silver, and from 6 to 12 grains for gold. This is the integer, and whatever be its real weight, it is denominated the assay pound. This imaginary pound is then subdivided into aliquot parts, but differ ing according to the metal. The silver assay pound is subdivided into 12 imagi nary ounces, each ounce into 20 penny weights ; and for assaying, these again into halves.
The following is the table of the pro portions of lead required to different al loys of copper. In the three first columns is shown the absolute increase of the quantity of lead in alloys of decreasing fineness. In the three last columns will be seen the gradual diminution of the protecting power of fine metal against ty of lead, required for the same weight of copper, under different mixtures.
scorificati on, in proportion to the increase of alloy, shown by the decreasing quanti It should be remarked, however, that many assayers, of good authority, use proportions of lead to alloy, considerably different from the above table ; and that the whole numbers here given may be considered as rather high in proportion to the quantity of lead.
The proportions of lead for gold assay ing are nearly the same as for silver.
Assays of alloys with platina are con ducted nearly in the same manner as for the mixtures of silver and gold. Silver is seldom alloyed with it ; but gold is more frequently ; and is known by the much greater heat it requires in the fusion ; by the edges of the button appearing thicker and rounder than in common assays of Fold ; by its colour being duller, and tend ing to yellow ; and its being entirely crystallised on its surface.
The action of nitrous acids on the alloys of platina is very remarkable. By itself, platina is as insoluble in this acid as gold, and a mixture of these two metals equally resists its action ; but when silver enters into the mixture, in the proportion of 2i, or three times the weight of the gold and platina, and when the platina is not above a tenth of the gold, the platina is totally soluble in nitrous acid, together with the silver, and the gold alone remains un touched.
when the gold mixed with platina is to be freed from it in the above manner, it must be laminated very thin ; a weak acid is first added, and boiled for some time. If the platina is above two per cent. of the gold, the acid assumes a straw colour, which deepens in proportion to the platina, and at the same time the cor nets assume a brownish green. A strong er acid is then added, and boiled three times successively, to detach the last por tions of Dlatina, which are separated with difficulty. By laminating very fine, u-ing the acid liberally, and long boiling, all the platina may be separated in one ope ration, when it does not exceed a tenth of the gold : and above that proportion, the colour of the gold is so much debased, and the appearances on the cupel so striking, that fraud can hardly escape an experienced eye. Parting might be used, even when the platina was more than a tenth of the gold, but then more silver must be added, which would render the cornet so very thin, after the action of the acid, that it could hardly be annealed with out breaking.
When alloys of silver alone with p latina are treated with nitrous acid, the silver dissolves as usual, but the liquor soon becomes muddy, with a very fine bulky black precipitate, which continues in creasing till all the silver is dissolved, and which is found to be entirely platina when collected. A part of the platina, how ever, remains in the solution ; for, on add ing muriatic acid to the liquor separated from the black precipitate, white Luna cornea falls down, after which carbonate of potash will throw down a green coagu lum, which is oxide of platina. The above effects of nitrous acid will therefore de tect an alloy of silver and platina.