GOLD.— Assaying.

121

small glass flask, called a parting flask, and this planis still adopted when assaying is not practised on alarge scale; but when, as in the Loyal Mint, thereare large numbers of assays constantly required tobe made, an apparatus formed of platinum, andintroduced by Messrs. Johnson, Matthet, & Co., isnow very generally employed. It may be useful,A Fig. 87.

Fig. 38.

however, to describe both these methods. Fig. 36represents a convenient apparatus for boiling sixcornets separately. The large tube, a, is connectedwith the gas supply, and from it rise a number ofsmall rose Bunsen burners. On these rest the part-ing flasks, C, into each of whioh is introduced from2 to 3 ozs. of nitric acid of spec. grav. l - 26. Thelong tubes, D, dipping into the mouths of the flasks,

Fig. 39.

Fig. 40.

are intended to condense as far as possible the acidvapours, the condensed liquid running back into theflasks. In cases where a large number are in actionat once they should be connected with a flue tocarry off these vapours. The acid should continueto boil for from three to five minutes after the glasseshave cleared of fumes ; the acid is then poured off,the cornets washed with hot distilled water, and again

VOL. II.

boiled for from fifteen to twenty minutes in nitricacid of about spec. grav. 1'30. As the ebullitionis at times very violent, and the cornet extremelyfragile, owing to its spongy nature, it is generallyrecommended that a piece of charcoal be introducedinto each flask. This, however, is objectionable, forit may cause the formation of nitrous acid, which byabsorption in the acid will even dis-solve portions of the metal. Thissource of error is completelyeliminated by using instead smallballs of porous earthenware, asrecommended by Field.

The last aeid having beenpoured off, the flask is completelyfilled with distilled water, and asmall, smoothly - finished, porousclay crucible is placed over themouth of it. The two vessels,thus adjusted, are then inverted,so as to allow the cornet to fallgently through the water into thecrucible, and by a dexterous move-ment of the hand the flask is with-drawn, in such a manner as toprevent the overflow of any liquidfrom the little crucible. The wateris now carefully decanted and the,silver which it contains is subse-quently extracted; the cruciblecontaining the cornet is heated toredness in the muffle. Under thisfinal heating it is not fused, butshrinks in bulk, loses its brownappearance, and assumes thepeculiar colour and lustre of gold;while at the same time it is ren-dered more compact, less brittle and fragile, andwhen cold, can be removed by a pair of forceps tothe scales, in whioh it is weighed with the sameprecision and care as the original alloy.

The second method, however, of parting abovereferred to is far preferable where large numbers ofassays are frequently made. The apparatus requiredis formed entirely of platinum, the joints being madewith gold. It is represented to scale in front eleva-tion and section in Figs. 37 and 38, and consistsof two boilers furnished with movable covers andcondensing tubes, which rest in water troughs whenthe apparatus is in action. The aeid vapour is partlycondensed, and passes into a suitable receptacle, theremaining fumes passing into a flue. The two boil-ings are performed one in each boiler. Each stripof metal, after being coiled into a cornet, is placedin a small platinum cup, as shown in Fig. 39 ; andthe requisite number of these is introduced into aplatinum tray, shown in Fig. 10, the bottom ofwhich is perforated, and which is divided into anumber of compartments. The whole is then placedin the first acid, and boiled for about fifteen minutes.The tray is then removed, dipped into distilledwater, and placed in the second stronger acid forabout twenty-five minutes. On its removal the16