3/0 SCIENCE PROGRESS 



recently commenced by the author. And it is suggested that 

 the phenomenon is due to adsorption, by the precipitate, of 

 the substance dissolved in the gel, which serves merely to 

 retain the precipitate in place. 



Considering the surface bounding two phases, such as a 

 solid and a liquid, Gibbs has shown that the concentration, at 

 the surface, of a dissolved substance, will be increased, if thereby 

 the surface tension is reduced, in accordance with the formula 



C rl 



U = — p^ -j~ t where U is the concentration at the surface, C 



that in the bulk of the liquid, and a the surface tension. This 

 effect is well known in the filtering of solutions through fine 

 powders, where the dissolved particles are retained 03^ the 

 solid, as in Dreaper and Davis' Night-blue experiment. In 

 the case under consideration the properties of the gel resemble 

 those of a liquid. The precipitate formed by the interaction of 

 ' reagent " above and the " solute " in the gel will create a solid 

 boundary surface between the liquid and the gel at which the 

 concentration of both dissolved substances will increase. If the 

 reagent is hypertonic to the gel, its surface concentration will 

 be in excess of that of the solute, and the reagent will diffuse into 

 the gel. With an hypotonic reagent the concentration of the 

 solute will be the greater, so that the reagent will be precipit- 

 ated as it collects at the surface, and will be unable to pass into 

 the gel. This is in accordance with Pringheim's observation 

 in 1895. Taking the case of the hypertonic reagent, formation 

 of a precipitate will increase the surface, at the same time des- 

 troying the solute at the surface. Consequently there will be 

 a rush of the particles of the solute towards the growing surface. 

 If the solute is sufficiently dilute, it will presently happen that 

 the layer of the gel next the precipitate will become practically 

 devoid of dissolved substance, when the formation of the first 

 zone will cease. The reagent diffusing through will pass on 

 until it meets with more of the solute, where the production of 

 a fresh layer will commence. From this it follows that there 

 will be none of the solute remaining between the layers. So 

 that if the Liesegang experiment is reversed, and potassium 

 bichromate poured upon a gel impregnated with silver nitrate, 

 the intermediate spaces between the rings will not blacken on 

 exposure to light. Since the soluble products of the reaction 

 will be more concentrated at the solid zones, it would appear 



