62 VOLTAIC BATTERY CONTROLS CAPILLARY ATTRACTION. 



water does not wet the surface of pure quicksilver, but stands upon it in drops of a 

 more or less rounded form : if, however, the electrical relation of these substances be 

 changed if the mercury be in contact with the negative pole of a voltaic battery, and 

 the water with the positive, a remarkable phenomenon ensues the water now wets the 

 mercury. To this important fact, and its applications, I shall hereafter return. 



201. The best method of showing that the voltaic battery has entire control over 

 capillary attraction is to take a shallow vessel containing a quantity of mercury, as A A, 

 fig. 23, and place upon it, in the position marked a, a drop of water ; on making this 

 drop communicate with the positive electrode of a battery, and the mercury with the 

 negative, in a moment the drop loses its rounded form and spreads out in a thin sheet 

 on the metallic surface, completely wetting it, and according as the tension of the bat- 

 tery increases, the drop expands more and more. Thus, if the current from 5, 10, 20, 

 40, 80, &c., plates be successively passed through it, the diameter of the circular space 

 it occupies closely follows the increase, and appears to continue to do so until the drop 

 becomes so thin that the electricity, in the shape of a spark, can pass through : then, 

 of course, the experiment cannot be continued. / 



202. If, therefore, in the arrangement of (198), the electrical relations of the saline 

 solution and the mercury be changed, by the process here indicated, we should expect 

 that the passage through the chink would take place ; and it is so. This experiment 

 affords a very elegant illustration of a result obtained by Porrett many years ago, which 

 was applied by Dutrochet to the explanation of endosmose. He observed that if two 

 quantities of water were separated from each other by a membranous partition, and 

 one of them made positive and the other negative, all the water in contact with the 

 positive pole would escape through the membrane into this negative partition. When, 

 in the arrangement of (198), the water escapes through the chink on being electrified, 

 it does not move slowly, by the action of its own weight/but is also impelled downward 

 in the way described by Porrett. For, take a tube, c c,fig. 24, the diameter of which is 

 about one tenth or one twelfth of an inch, and insert in its axis a platina wire, a, then 

 "let the lower extremity touch a surface of water, and a volume of that fluid will rise in 

 it to a certain height by common capillary attraction. . If, now, the tube, charged with 

 its water and wire, be placed, as in the figure, .on the surface of some mercury in a 

 watch-glass, so that the extremity of the tube shall just touch the metallic surface, and 

 the wire a be then connected with the positive electrode, and the wire b with the 

 negative, in an instant the water will begin to flow out of the tube, and spread over the 

 mercury, and will continue to do so until its level has sunk to the end of the platina 

 wire. With a wider tube, such as that described in (198), this passage might be impu- 

 ted to the mere gravitation of the parts of the fluid urging them downward ; but, in this 

 instance, 'owing to the narrowness of the tube, that force is nullified by capillary attrac- 

 tion. : The water is therefore driven out of the tube by an active force; and that this 

 is really the case, is abundantly proved by breaking the battery connexion and rais- 

 ing the tube slightly above the mercurial surface ; the water then precipitately re- 

 turns back into the tube again. 



203. In Chapter V. it was shown how the common phenomena of capillary attrac- 



