202 ANNUAL OF SCIENTIFIC DISCOVERY. 



ter by three inches in depth, and one side is covered by a disk of 

 parchment-paper, so as to form a vessel like a sieve. A mixed solu- 

 tion, which may be supposed to contain sugar and gum, is placed upon 

 the septum to the depth of half an inch, and the instrument is then 

 floated upon a considerable volume of water contained in a basin. 

 Three-fourths of the sugar diffuses out in twenty-four hours so free 

 from gum as to be scarcely affected by subacetate of lead, and ready 

 to crystallize on the evaporation of the external water by the heat of 

 a water bath. 



In this case, then, the effect of the septum is to produce an unequal 

 action of diffusion, and to cause the separation above described. The 

 explanation of this change depends, according to Professor Graham, 

 not upon any degree of capillarity, but upon the effect of the septum, 

 which is a true colloid. The crystalloid sugar is capable of taking 

 water from the hydrated colloidal septum, and thus obtains a medium 

 for diffusion ; but the colloid gum has little or no power to separate 

 the combined water of the same septum, and does not, therefore, open 

 the door for its escape by diffusion, as the sugar does. 



This separate inaction of the colloidal septum is defined by Profes- 

 sor Graham under the term dialysis. In cases where bodies belong- 

 ino 1 to the crystalloid series have unequal diffusive properties, they 

 may be separated from each other without any intervening septum ; 

 the natural diffusive power of each crystalloid being sufficient to de- 

 termine its diffusibility. To effect separation of different crystalloids, 

 therefore, mere diffusion in a column of water is necessary. The 

 mixed solution of crystalloids is conveyed by means of a pipette to the 

 bottom of a column of water contained in a cylindrical glass jar. A 

 kind of cohobation takes place ; a portion of the most diffusing sub- 

 stance rising and separating from the less diffusive substances, more 

 and more completing as it ascends. 



We are thus led by these experimental deductions to give to the 

 phenomena of osmosis a new and distinct reading. The consideration 

 of the properties of gelatinous bodies or colloids appears, Prof. Gra- 

 ham thinks, to show that osmosis is an affair of the dehydration (that 

 is to say, of the removal of water) from the membranous gelatinous 

 septum, under influences having a catalyptic character. The colloid- 

 al septum is capable of hydrating itself to a higher degree, in contact 

 with pure water, than in contact with a saline solution; while the 

 saline solution, on its part, removes the water from the membrane and 

 establishes the effect. 



In addition to this explanation, we obtain by these new experi- 

 mental deductions another method of conducting analytical inquiries. 

 For instance, soluble albumen may be obtained in a state of purity by 

 exposing it to the separate action of the colloidal septum, with an ad- 

 dition of acetic acid ; or, again, if blood, milk, or other organic fluids, 

 are charged with a small quantity of arsenious acid, and the mixtures 

 so formed are placed upon a membranous septum or dialyser, the op- 

 posite surface of the membrane being exposed to water, the greater 

 portion of the arsenious acid will pass through the membrane to the 

 external water in the course of twenty-four hours. In an experiment 

 given by Prof. Graham, the arsenic, thus separated and diffused 

 through the external water, was so free from the organic matter that 



