ON THE LIESEGANG PHENOMENON IN GELS 373 



inner edges and centre of the disc, from the surface of the gel 

 impregnated with lead nitrate, to that of the corresponding 

 measurements from the other surface, was 1*335 ; while the ratio 

 of the square roots of the molecular weights of the ions is i'34- 

 At the moment of formation of the first disc, the yellow colour 

 of the chromated gel was apparently uniform throughout, 

 except within few mm. of the disc, where it was paler. The lead 

 nitrate, being hypertonic, diffused further in the direction of the 

 potassium chromate with the formation of eight successive discs, 

 approximately ris mm. apart. The ninth and following discs 

 were of basic chromate approximately 0*9 mm. apart. The 

 striking fact was observed that, during the formation of discs 

 in that part of the tube where the colour had previously been 

 uniformly yellow, the layer of gel next to the precipitate was 

 always colourless, affording occular proof of the depletion of 

 chromate from that portion of the gel. 



An additional feature was noticed in the case of a test-tube 



prepared with 20 c.c. — K 2 Cr0 4 agar on which 15 c.c. of 



r 20 



— Pb(N0 3 ) 2 were poured. This gel had been strained through 



muslin instead of filtering, so that it was slightly cloudy. After 



twenty-four hours a clear space appeared below the precipitate, 



as if the cloudiness had begun to subside. This space grew to 



a depth of 4-75 mm. in eight days. At the same time the 



yellow colour of the gel adjacent to the precipitate gradually 



became paler. That the gel was still solid was evinced by its 



power of supporting the growing precipitate. And it appears 



that the clarification was caused by the breaking up of the 



agglomerated portions of the gel by the rapidly moving chromate 



particles. 



Since NaCl is negatively adsorbed, U-tubes were prepared con- 



N N N N 



taining, respectively, — NaCH AgN0 3 and— AgN0 3 +— NaCl, 



separated by 1% agar. At first only a diffuse precipitate was 



formed in the latter case, while two clear discs resulted in the 



N 

 former. But later, owing to adsorption of AgN0 3 , the — NaCl 



N 

 began to diffuse into the — AgN0 2 with the production of 



distinct strata. 



