142 



On the Electrolytic Diffusion of Liquids. [Dec. 8. 



diameter than the meniscus tube, and about 15 centims. long. To the 

 upper end of E is attached an india-rubber tube F, provided with a 

 pinch-tap G. H and I are two sheet platinum electrodes, each about 

 7 centims. long, and 18 millims. wide, for connexion with a voltaic 

 battery. The connecting wire of H is hermetically sealed in a glass 

 tube, which fits air-tight into the bung. 



In using this apparatus, the vessel A is shifted from its place, and 

 the heavier liquid poured into it. The tube B is then filled by means 

 of suction at J with the lighter liquid up to a level in E, a little above 

 the bung. The tube P is then closed by means of the pinch -tap G,, 

 the vessel A replaced, and B, &c, lowered by means of the rack and 

 pinion until the pressure of liquid in A just balances that in B, the 

 difference of level being approximately determined beforehand, by 

 taking the specific gravities of the two liquids. A definite meniscus is 

 then easily formed in the tube D, by opening the pinch-tap and 

 raising B until a drop of liquid issues below, and then lowering it a 

 minute distance. It is particular that no air bubble exists in B, and 

 in order to facilitate the escape of any, the interior of the upper bung- 

 is made of a funnel shape, and coated very smoothly with sealing 

 wax. 



In an experiment I made with this apparatus, the heavier liquid 

 was a solution of nitrate of mercury of specific gravity 1'30, and the 

 lighter one a solution of cupric nitrate, specific gravity 1*22. With 

 an upward current from 18 Grove's elements in single series, a colour- 

 less horizontal line soon appeared below the meniscus in D, advanced 

 downward, and underflowed the end of the meniscus tube. Neither- 

 the meniscus in the lower tube D, nor that in the upper one E, shifted 

 in position during the passage of the current. These results were 

 repeatedly verified with the meniscus at different distances, varying 

 from one-sixteenth to one-eighth of an inch above the bottom of the 

 tube. 



Bemarlcs. — These results show first, and most conclusively, that 

 liquid diffused downwards continuously through the meniscus during 

 the passage of the upward Current ; and second, that during the con- 

 tinuance of the current, either no manifest expansion occurred in the 

 bulk of the liquid in B, and that equal volumes of liquid diffused in 

 two opposite directions through the lower meniscus ; or, that any 

 expansion of the bulk of liquid in tube B was compensated for by 

 downward diffusion of an equal bulk of liquid. Another possibility is 

 that the united volumes of the metallic deposited copper, and of the- 

 acid element from which it had been separated by electrolysis, was 

 greater than before such separation, and that this was compensated for 

 by the volume of liquid diffused downwards through the meniscus. 



