440 M. De la Rive on Chemico- capillary Actions. 



is observed in the interior of the crack on the inner surface of 

 the tube ; but if the copper wire be removed, which suppresses any 

 metallic contact, the fissure fills with small crystals of metallic 

 copper, as well as the inside of the tube. In this case, according 

 to M. Becquerel, the liquid lodged in the fissure acts as the cop- 

 per wire which has been suppressed ; and there are two currents, 

 which add on to each other, for they are in the same direction — 

 one due to the reaction of the liquids, the other arising from the 

 oxidation of the reduced metal. When the fissure is filled with 

 metal, it may happen that there is no direct contact between the 

 two liquids, and the couple then ceases to act. The same results 

 are obtained with other metallic solutions, such as nitrate of 

 silver. It may be remarked that, capillary spaces of the same 

 section being far more imperfect conductors of electricity than 

 the metallic wires whose functions they perform, the electroche- 

 mical effects should also be feebler. 



M. Becquerel assumes that liquids enclosed in capillary spaces 

 play the part of solid conductors. This he finds supported by 

 the circumstance that, from M. E. BecquereFs experiments, the 

 conductivity of liquids in capillary tubes is greater than is due to 

 the section of the liquid column. This important fact would tend 

 to prove that the increase in density of the liquid layer retained 

 on the surface of the glass by molecular attraction comes into 

 play to form a voltaic circuit consisting solely of liquids. 



I shall not dwell on the various improvements successively 

 made by M. Becquerel in the construction of these apparatus, 

 nor on the numerous experiments made to replace monosulphide 

 of sodium by other solutions. Before concluding I will restrict 

 myself to remarking upon a few points which appear to me more 

 important. 



I will first call attention to a series of facts which would seem 

 to prove the production of electric currents in the fissures of 

 cracked glass vessels without the apparent intervention of che- 

 mical actions. In all the phenomena previously described, the 

 fissures of cracked vessels, though very narrow, nevertheless 

 enable the two liquids which moisten the sides to react chemi- 

 cally upon each other in the capillary spaces and to give rise to 

 electrochemical effects. Now observe what takes place if the 

 magnitude of the fissures be varied. If the size is relatively 

 great, the solutions mix more or less rapidly, giving rise to the 

 ordinary precipitates which result from a double decomposition. 

 With a smaller size the electro-capillary phenomena of metallic 

 reduction which we have described are obtained. At a lower 

 degree there is an electrical current without any apparent che- 

 mical action — a current which has the same direction as if the 

 reaction took place. The cracks may finally be so narrow that 



