OSMOSE; DIALYSIS. The earlier discoveries of Dutrochet and Graham have been briefly described in the article on Diffusion (q.v.). The subject has, however, been much extended recently, principally by the investigations of Graham; and as the whole phenonama are exceedingly interesting and important, since secretion, absorption, and various other organic processes are to a great extent dependent on them, some further detail, especially of these later facts, may here be given.
When two differkt liquids are separated by a bladder or other membrane, or a piece of calico coated with coagulated albumen, there is always a more or less rapid transfer ence of the two liquids in opposite directions through the diaphragm. In certain cases, the explanation given in the article referred to is compiete, but in others it appears be insufficient. Graham has made an extensive series of experiments upon osmose, where distilled water was on one side of the diaphragm, and various liquids and solu tions on the other, and has arrived at many general results, of which the following are the more important. The osmose is considered as positive when more of the water passes. through the diaphragm than of the other liquid. Such substances as gum, gelatine, etc.; produce scarcely any effect. Solutions of neutral salts, such as common salt, Epsom salts, etc., follow the ordinary law of diffusion, as if no diaphragm had been interposed. Acid salts in solution, and dilute acids, pass rapidly into the water—or the osmose is negative; while alkaline solutions give, in general, a strong positive effect.
In all the cases in which an osmotic action occurs which cannot be explained by capillary forces, there is chemical action on the diaphragm; and conversely, such osmose cannot be produced if the material of the diaphragm be not acted on by the liquids in_ contact with it.
But the most remarkable results of Graham's later investigations are those relating to dialysis—i.e., to the separation of the constituents of mixtures, and even the decomposi
tion of chemical compounds, by osmose. The results of his earlier investigations, above given, show a remarkable difference between two classes of bodies; gum, gelatine, etc., which form viscous solutions, on the one hand; and salts, acids, and alkalies, on the. other. The first class he has called colloids; the second, crystalloids. The former are extremely sluggish, the latter comparatively rapid in their action. Thus, of common salt and albumen, under precisely similar circumstances, there pass through the dia phragm in a given time quantities which are as 25 to 1 by weight. Hence, if a solution containing both classes of substances be opposed to pure water, the crystalloids will rapidly through the diaphragm, and the colloids slowly. This process promises to be of very great value in medical jurisprudence, as, without introducing any new substance (except the diaphragm and distilled water), we have the means of separating from the general Ly colloidal contents of animal viscera such poisonous crystalloids as white arsenic, vegetable alkaloids, etc., which by the old methods was in general attended with great 'difficulty, and often uncertainty. These methods are still in their infancy, but enough is already known to show how valuable they must soon become to the chem ist and• the toxicologist. One economical application has been proposed, and shown to be practicable. When a bladder is filled with the brine of salt beef, and suspended in freth water, the salt after a time nearly all disappears, and there remains in the a rich extract of meat fit for making soup.
For a brief notice of the speculations which Graham's researches have led him to form as to the nature of mutter, we refer to the article on that