The mixture ia fused in a larger crucible of the sanie kind. The metallic bead obtained, taining both the metals, is poured out in the ordinary way and weighed, when cold. If iron be not present in the ore, about 5 grains of fine iron ivire must be added as an indispensable eondition.
For commercial purposes, the following method is by far the best in use ; after a little practice, it yields very accurate results. Take 100 grains of the powdered ore and dissolve them in agua regia.. Dilute the solution with distilled water, and filter. Now pass a stream of sulphuretted hydrogen through the solution for half an hour. Filter again, and wash the filter with the sulphuretted hydrogen water. Eoil the filtrate and peroxidize the solution with a few drops of nitric acid. If the filtrate be not oxidized, a drop of nitric acid will produce a dark colour, which appears! only instantaneously, and must be closely watched. Add barium carbonate until alkaline ; boil, filter, and wash the precipitate ; then dry and ignite it iu a crucible. The re,sidue so obtained must now he fused with an equal weight of arsenic, as in the operation previously described. The button of metal reduced, however, need not be weighed. If it be not much larger than a pea, the whole may be taken for the next operation, but if it be much larger, it must be broken up aud a weighed quantity taken. During this time, the smallest and strongest muffle furnace proculable must have been urged to its utmost heat ; tbe back of the muffle aliould be difficult to SLC on account of the high temperature. Some small clay cups similar to those shown in Fig. 281, are now placed in the muffle to heat, and the operator proceeds as follows : Remove the door of the muffle, and place a little borax glass in one of the cups, and push it back to the far end of the muffle, and shut the door again. When the clay and borax are at the same beat as the furnace, wrap up the weighed piece of speiss in a piece of paper, and with the tongs drop it into the melted borax. The furnace should be hot enough to instantly melt the button, this being one of the in dispensable conditions of the assay. There must only be sufficient borax to allow the melted button to revolve in, and not enough to drown the button.
The button immediately begins to rotate, like a silver-lead bead in a eupel, and after a few moments the cup must be seized with the tongs and drawn quickly out, and plunged, bead and all, as quickly as possible, into a basin of water. When cold, examine the borax, and if iron be still coming off, only the characteristic tinge of iron will be observed ; but if the very smallest trace of cobalt has come off, the borax will be tinged blue. Directly this occurs, the operator knows that only cobalt, nickel, and perhaps copper remain. Sup posing that no blue tinge appears in the borax, a fresh little cup is placed at the back of the muffle, the button again dropped in and cooled as before, until the blue tinge appears. The operation is proceeded with till all the iron is gone, when the button is taken and placed in a small crucible and covered over with charcoal ; it is melted till no more arsenic fumes are given off, and is then treated exactly as before, occasionally testing the button to see when the cobalt is exhausted. When this point is gained, the button will stop revolving for a time, and the operator must be on the watch for this sign. The cup is now drawn out again, cooled, and borax examined for the characteristic reddish-brown colour of nickel. Any copper which may be present will come off last. In conducting this process the requisite heat is the only difficulty, but by adding some extra iron pipiug to the chimney of a Plumbago Crucible Company's muffle furnace, placing some anthracites at the bottom of the fire, and carefully choosing the coke, a very high temperature may be got, as long as the muffle itself is a small one. The estimation of the metals by this process is as follows : Product, after removal of the iron and excess of arsenic = Ni,As Co,As, in varying proportions. Remove the Co,As, and Ni,As is left. But when the weight of each is known, it is easy to calculate the relative amounts of nickel and cobalt. For instance, suppose the weight of the speiss of Ni,As Co,As = 20 grains, and the weight of the after the removal of the Co,As = 10. The amount of Co,As = 10 grains by difference. Then Ni, 193 : 10 grains • • .. 118 : x.