372 SCIENCE PROGRESS 



both ways, until, after forty-eight hours, the solution and gel 

 were the same colour throughout. 



To determine whether the by-product of the reaction had 

 any appreciable influence on the formation of layers, tubes 



were started, (i) and (2) containing 20 c.c. of — Na«CO? in 



20 



1 per cent, agar gel, and (3) and (4) with the same strength 



N 



Na 2 C0 3 H NaCl. Five c.c. of dilute calcium chloride were 



20 



poured on each. Diffusion proceeded at the same rate in each 



tube throughout the experiment, with the formation of distinct 



though rather faint strata at distances below the top of the gel, 



in mm., as follows : 



N 

 experiment 1 5 c.c. of — K 2 Cr0 4 were poured on top of (5) 15 c.c. 



N 

 Pb(N0 3 ) 3 agar gel, and (6) the same strength agar gel contain- 



. . N 

 ing in addition KNO3. After nine days numerous well-marked 



100 J 



strata had been formed very close together. The average 

 distances of the last twenty layers were 0*70 and 0*67 mm. 

 respectively. Doubling or trebling the amount of the by- 

 product of the reaction therefore appears to have little or no 

 effect on the formation of the strata. 



A convenient method of performing the experiment is to fill 

 a U-tube rather more than half full with pure agar gel, and 

 pour reagents of known strength into each limb. In one case 



N 

 5 c.c. of - - K 2 Cr0 4 were placed in one limb, and an equal measure 



N 

 of — Pb(N0 3 ) 2 in the other. The reagents gradually diffused 



through the gel, the progress of the diffusion being indicated 

 by a slight opalescence of the gel containing the lead salt, and 

 by a yellow colour in the other limb. At the point of meeting, 

 a distinct, though somewhat faint, disc of lead chromate was 

 formed. The ratio of the average distance of the outer and 



