1862.] on Mr, Graham* s Researches on Dialysis, 425 



substance, yet a much better result may be obtained by causing the 

 diffusion to take place, not into free water, but into or through the 

 combined water of a soft solid, which, as before observed, scarcely 

 affects the diffusion of crystalloid, but almost arrests that of a colloid 

 body. When a piece of vellum, or of membrane, or of parchment- 

 paper, or even a layer of mucus, is interposed between a colloid 

 solution and a quantity of water, the colloid, in order to get to the free 

 water, must pass through the membrane, that is to say, must unite 

 diffusively with the combined water of the membrane, which however 

 it is incapable of doing. The crystalloid body can pass through, but 

 the colloid cannot. This constitutes dialysis, which means the separa- 

 tion of crystalloid from colloid bodies, through a membrane that will 

 allow the passage of crystalloid, but will not allow that of colloid 

 particles; because the crystalloid particles, having a highly diffusive 

 power, can unite diffusively with the combined water of the soft solid 

 septum, so as to reach the external free water ; whereas the colloid 

 particles, having a very feebly diffusive power, cannot. The process of 

 dialysis is altogether different from that of filtration. Filtration refers 

 to the passage of masses through appreciable pores, but in a dialytic 

 septum there are no pores, and the movement is not molar but mole- 

 cular. The dialytic septum will allow chemical action to take place, 

 or that low form of chemical action which constitutes diffusion ; but, 

 being quite impervious to the mechanical passage of liquid, it will not 

 allow of filtration. 



Colloid Solutions. — The solution of various colloids has been here- 

 tofore effected by means of crystalloid chemicals, comprising acids, 

 bases, and salts. By dialyzing these liquids, the crystalloid reagents 

 diffuse away, and leave the colloid bodies in simple aqueous solution. 

 Mr. Graham has thus obtained pure colloidal solutions in water of 

 numerous mineral and organic substances, such, for instance, as silica, 

 tin-stone, alumina, haematite, chrome, Prussian blue, prussiate of copper, 

 sucrate of copper and other sucrates, tannin, gum, caramel, albumin, &c. 

 These colloidal solutions are for the most part unstable. Either spon- 

 taneously, or on the addition of a very minute quantity of some or other 

 crystalloid reagent, they pectize, or become converted into solid jellies. 

 Hence Mr. Graham speaks of two colloidal states, the peptous, or dis- 

 solved, and the pectoics or gelatinized. Colloid bodies are characterized 

 by their non-crystalline habit, by their low diffusibility, by their che- 

 mical inertness, by their high atomicity, and above all by their mutabi- 

 lity. All these properties are exceedingly well manifested by colloid 

 silica, or co-silicic acid. 



[W. O.] 



