THE object of these notes is to call attention to the difficulty of making correct comparison between the results obtained from the examination of small parcels of ore and the results which may be expected from the continuous working of large plants on. The same ore; to describe certain apparatus and tests used by the writer to achieve accuracy, having this end in view, and to offer some general advice to the young consulting metallurgist in the conduct of metallurgical investigation work generally, more especially that work which is preliminary to new plant construction or the improvement of existing plants. Criticism is offered as to the relative value of the results obtained, studying only their true commercial sense, and emphasis is laid on the importance of solving each problem entirely by itself, taking into consideration all the local factors bearing on the case at issue. The subject is a large one, and the writer must admit that he has hovered just on the fringe. He would explain at the beginning that his treatment is confined to that branch of metallurgy dealing with the extraction of gold from its ores by cyanide, but offers no apology that he has not dealt in detail with the numerous tests which form a necessary portion of the metallurgical investigator's programme. Most of these tests are well known, simple, and leave but small loop-hole for error.