In this paper is shown a practical possibility to perform flotation in a non-aqueous medium (petroleum, xylol, benzol), wherewith it is pointed out that a series of flotation phenomena, as compared with flotation in water, go in a quite contrary direction. Thus, hydrophilic minerals, as sodium chloride, quartz, are floated well, but the more hydrophobic pyrite can’t be floated. It has also been found that usually employed collecting reagents are unfit for flotation in non-aqueous medium. In the present experiments have been employed alcohols of the saturated series as collecting reagents, other collectors having not been found. It is stated that the quantity of alcohols necessary for the experiment is very great and that the consumption of them depends on the magnitude of the radical of the alcohol molecule and on the degree of humidity of the alcohol. As to other factors – character of froth-formation, influence of pulp density, time of agitation, etc. – there exists an analogy with flotation in aqueous medium.

By means of numerous examples taken from experimental work the author shows the falsity of the view which considers organic protective colloids as flotation poisons or in any case as substances possessing only the function of depressing minerals in flotation. It has been found that such colloid as casein possesses good collecting properties with respect to quartz, feldspar, apatite, and does not collect sulphides. Gelatine has analogous properties. Tannin has proved to be a selectively acting depressor with respect to galena, but not depressing zinc-blende, chalcopyrite, and other sulphides under the same conditions. By means of protective organic colloids it has been found possible to satisfactorily divide many mixtures of minerals, both sulphides and non-sulphides. These so interesting properties of organic colloids in flotation, discovered when studying them more closely, bear witness to the necessity of undertaking systematical investigations of this question.