Repetitions of this experiment have shewn, that tho’ the separation succeeds in some cases, it does not perfectly in all: that if there is any particle of the platina imperfectly dissolved in the gold (which will generally be the case, unless the quantity of gold is three or four times greater than that of the platina), this part will be retained, after long trituration, undissolved by the mercury, uncomminuted by the pestle, and too ponderous to be washed off in its gross form. A variety of mixtures of platina and gold were treated in the manner above described; and the gold, recovered from the amalgams, submitted to further examinations. Where the proportion of platina was large, the microscope almost always discovered still some granules of it on the fracture of the ingot: where the proportion was small, the recovered gold was frequently, but not constantly, found to be pure. From these experiments it appears, that mercury has a greater affinity with gold than platina, and that platina is capable of being totally separated by elutriation; but that the process is too vague and undetermined to be applicable in the way of assay, as we have no mark of the precise time for discontinuing it, and as we can never be certain, without making another assay, whether the whole of the platina is separated or not. As a preparatory examination, where the quantities of platina and gold to be separated are large, it is nevertheless of good use, as greatest part of the platina may by this means be washed over with little trouble, and the gold brought into a less compass, so as to be commodiously submitted to a perfect purification by the means hereafter pointed out. This process has a similar effect on platina and gold to that of stamping and washing on metallic ores; which could not be reduced into pure metal in the furnace to advantage, without the previous separation of great part of the earthy and stony matter by water. 2. _By Precipitation with Alkalies._ Gold is precipitated totally by fixed alkaline salts, but platina only in part. When solutions of the two metals are mixed together, so much of the platina remains suspended, after saturation with the alkali, as to be readily distinguishable by the yellow colour, which it communicates to the liquor. It has been objected, that tho’ the platina was discoverable, when thus mingled superficially with the gold, it may nevertheless, when combined more intimately by fusion, elude this method of trial. 1. Mixtures of gold with small proportions of platina were therefore kept in fusion, by a very strong fire, for several hours, and afterwards dissolved in aqua-regis. The solutions being diluted with water, and a pure fixed alkaline salt gradually added, so long as any effervescence of precipitation ensued, the liquors remained manifestly coloured, tho’ apparently paler than when the two metals had been dissolved by themselves.