Colloidal Suspensions.
Methods of Preparation.—The mixtures of this class have been, for the most part, prepared artificially. The principles of some of the methods which have been employed for this purpose may, therefore, be first described.
Of these principles the most important one is, that when an insoluble substance is produced in the absence of electrolytes by a reaction between two chemical compounds, it almost invariably separates in the state of a colloidal suspension. By the term electrolyte is here meant any dissolved substance which is a good conductor of electricity, one, therefore, whose molecules are, according to the ionic theory, largely dissociated into electrically charged atoms or atom groups called ions. Most salts and strong acids or bases are such electrolytes; but water, neutral organic substances like alcohol or sugar, and very weak acids or bases, are not. Electrolytes must not be present in considerable quantity, for the reason that ions coagulate these suspensions. Thus, when a saturated solution of hydrogen sulphide, a slightly ionized substance, is added to one of arsenious oxide, also slightly ionized, no coagulated precipitate of arsenious sulphide results, but only a turbid yellow liquid, which, when poured through filter-paper, leaves nothing behind. It will be noted that in this case the other product of the reaction is water, an un-ionized compound. If this reaction be carried out with a solution of arsenious chloride, instead of with one of the oxide, the ordinary precipitate of arsenious sulphide is obtained; for, in this case, the hydrochloric acid produced by the reaction, being largely dissociated into hydrogen and chlorine ions, coagulates the colloidal suspension. So, also, upon adding hydrochloric acid to the colloidal mixture resulting from the former experiment, a large precipitate is immediately produced. As a second illustration of this method, hydrogen sulphide water may be added to a solution of mercuric cyanide. In this case also a black opaque colloidal suspension of the sulphide results; for the three substances involved in the reaction, hydrogen sulphide, mercuric cyanide, and hydrocyanic acid, are non-electrolytes; but, upon the addition of hydrochloric acid, or, still better, of magnesium chloride, to this solution, the precipitate immediately coagulates. It is not necessary, of course, that electrolytes be entirely excluded, but only that they be not present at any point at such a concentration as will produce coagulation. The method is, therefore, of fairly general applicability. Thus, a colloidal suspension of Prussian blue can be prepared by mixing dilute solutions of nearly equivalent quantities of ferric chloride and potassium ferrocyanide; for the other product of the reaction, potassium chloride, has a coagulating effect only at higher concentrations.
A second method which has until recently been even more com-
