ELECTRICAL CURRENTS OF THE EARTH. 211 



was used ; which wire was in some cases stretched npnii the ground, 

 iu others sunk at a slight depth beneath the surface of the earth, 

 but oftenest suspended on slender rods of wood, like those employed 

 for fii-ld-telegraphs. 



Electrodes. — This is naturally the part of a mixed circuit requiring 

 the greatest attention, and I have been able to avoid causes of error in 

 this respect from the assurance, which I had obtained in the experi- 

 ments of electro-physiology, that electrodes of annilgamated zinc, im- 

 mersed iu a saturated and neutral solution of sulphate of zinc, do not 

 excite between them an electric current, and do not acquire secondary 

 polarities by the passage of a voltaic current. 



The electrodes which I have employed are rectangular j^lates of lami- 

 nated zinc, from G to 8 centimetres {2.^ to 3 inches) in width, and from 

 12 to IG (5 to G inches) in height, perfectly amalgamated and joined to 

 the metallic line by means of a circuit-breaker with two holes, into one 

 of which enters the wire and into the other the extremity of a projec- 

 tion on the i)late of zinc. This plate is immersed in a saturated snid 

 neutral solution of sulphate of zinc, contained iu a iiorcelain cylinder 

 like that of Grove's battery. In the selection of these cylinders care 

 must be taken to reject those which are so porous as to admit too readily 

 of the percolation of the liquid. 



The porcelaLu cylinder thus prepared is immersed in well or spring 

 vrater, which should be the same at both extremities. For the reception 

 of the water in which the cylinders are simk, I have used different con- 

 trivances. Sometimes, after having formed in the ground a sort of ])it, 

 varying in depth from a half metre to two metres, I have made in the 

 bottom of this pit a cavity, shaped like a capsule, from 10 to lii 

 centimetres (4 to 5 inches) Avide, and of such a depth that the por- 

 celain cylinder, when introduced, should reach with its rim the level 

 of the bottom of the pit. Then, in order that the v\'ater poured into 

 this kind of capsule may not be too speedily absorbed by the ground, I 

 line the capsule with a stratimi of tempered potter's earth, such as is 

 used in earthenware manufactures. At other times 1 have used tlov.er- 

 pots, which are sunk iu the ground, the earth being compressed around 

 the vessel. In some cases, finally, the porcelain cylinder was inserted 

 and lixed in a large piece of cork, so that the cylinder might remain 

 floating on the water of a well, but almost entirely immersed therein 5 

 the copper wire cov-ered with gutta-percha, which is joined to the plate 

 of zinc, is wound aronnd a small cord, by means of which the floating- 

 body is made to descend in the well. 



I have made many experiments to assure myself of the homogeneity 

 of the electrodes thus prepared. It is very easy to obtain this homo- 

 geneity and to preserve it with porous cylinders. We begin by having 

 a certain number of such cylinders, quite new, and by filling them to 

 the same height with a solution of sulphate of zinc ; we select those 

 which do not readily imbibe the liquid or allow it to escape, and im- 

 merse simultaneously two of those thus selected, after they have been 

 well dried with a clean towel, in a vessel filled with the same Avater. In 

 this way, even with a galvanometer of 20,000 coils, we i)romptly find 

 the two electrodes perfectly homogeneous. It Imppens, however, n(;t ini- 

 frequently, that, if the cylinders are left in the water for some time or 

 withdrawn and again immersed, a certain current Avill be observed to arise 

 between them. If the cause of the difference which has thus originated be 

 attributable to the plates of zinc, which is the rarer case, it is necessary 

 to reamalgamate them ; if, on the other hand, the heterogeneity be due to 

 the ijorceluin cylinders, we must withdraw these ii'om the 'water, di'y 



