This section is from the book "Leaching Gold and Silver Ores. The Plattner And Kiss Processes: A Practical Treatise", by Charles Howard Aaron. Also available from Amazon: Leaching Gold And Silver Ores.
158. It is well known that many experienced as-sayers, although generally reliable, do not obtain correct results from the assay of gold-bearing sulphides, very rich ores of silver, or cement copper containing gold, and contaminated by basic chlorides. A few facts and suggestions on this subject may not be out of place here. As this is not a treatise on assaying, the reader is supposed to be familiar with the apparatus and manipulations to be employed.
A common practice among assayers, and by some deemed indispensable, is to roast the assay of gold-bearing sulphides. It is better not to do so unless they contain much arsenic, and even in that case, scori-fkation is preferable.
When an accurate assay of an ore which is new to the assayer is desired, it is best to make several assays by different methods. If the assayer knows how to properly clear, weigh, and part the bead, and to make the necessary allowance for silver in the litharge or lead employed, the presumption of correctness is in favor of the highest result, because, although it is easy to lose gold or silver in an assay, it is impossible to extract more than the ore contains. In important assays, such as those on which the purchase of a lot of ore is based, the carefully taken sample should be thoroughly dried at the heat of boiling water, passed through a seive of at least 80 meshes to the running inch, and then very carefully mixed.
It is generally assumed, and so implied in books on the subject, that if two assays, made at the same time, and in the same manner, give correspondiug results, they are correct. This is a great mistake. Two chemical operations conducted under precisely similar conditions should give similar results. Their agreement proves that no accidental loss has occurred, nor any error in weighing, but it does not prove that the method adopted is that by which the highest results can be obtained.
159. There are two principal methods of assaying ores for gold and silver - by crucible, and by scorifica-tion. The crucible assays may be divided into two classes; firstly, that in which a thoroughly oxidized slag is produced; secondly, that in which a matte containing the minimum proportion of sulphur is formed.
The scorified assay is always a completely oxidized assay, and is the most certain in its results, but is subject to certain inconveniences. Either a very small quantity of ore is operated on, or, a very large quantity of lead is obtained, and it usually requires more time than a crucible assay. Yet it is indispensable as a check, for in some cases it gives higher results than can be obtained from the crucible, especially with some arsenical compounds.
Each kind of assay is subject to modifications as to the quantity, kind, and respective proportions of fluxes used, depending on the constituents of the ore; whence it is difficult to give rules to meet the requirements of all cases.
Concentrated gold bearing sulphides, chiefly iron pyrites, with or without an admixture of other sulphides containing silver, are assayed by the different methods as follows: