212 ELECTRICAL CURRENTS OF THE EARTH. 



tliem repeatedly with a clotb, and renew the water in wliicli they are 

 immersed. To procure homoiioiieous vases of terracotta for containing 

 the water in whi(;h the porous cylinders are to be immersed is more 

 difficult, and in order to succeed we must leave them to imbibe water 

 for scA'eral days, and then pro^'e them ; but even then there are scarcely 

 to be found, among many, two between which signs of heterogeneity do 

 not present themselves. 



In some experiments I have been accustomed to excavate a hole of 

 moderate dimensions at tlie two extremities of the circuit, and to fill' 

 each hole with tlie same earth, into which the terra-cotta vase was then in- 

 troduced. Most frequently I have satisfied myself, before commencing 

 the experiment, that there was no current between the electrodes, by 

 sinking the two vases in two holes of moderate size made in contiguity 

 with one another. I have also, whenever it v/as practicable, reversed 

 the position of tlie two earthen vases and noted the difference, if thei-e 

 were any, of the deviations obtained in the two cases, in order to dis- 

 cover whether the heterogeneity of the vases was noticeable, and to 

 what extent in the current detected. 



I have sought finally to ascertain whether, in any case, it would be 

 possible to substitute for the electrodes vdiich 1 have descaibed two 

 plates of copper sunk in the ground, which v>'ould be much more simple 

 and convenient ; and I have found that, whatever might be the state of 

 these plates before submitting them to experiment, that is to say before 

 using them, either in a different condition of purity or oxidation as they 

 most frequently occur, there was always realized, from the first, Vv^ith 

 the galvanometer of 2,000 coils, a very strong deviation, of which it was 

 impossible to foresee the direction. It constantly occurred, however, that, 

 on keeping the circuit closed and leaving the plates of copper buried and 

 undisturbed in the earth, this deviation slowlj" diminished, and aftei 

 eighteen or twenty hours became comparatively insignificant. At this 

 juncture, it was onh^ necessary to stir slightly one of the plates, or to 

 press the contiguous earth, or to throw a little Avater on the spot where 

 the two plates were sunk, in order to excite a deviation, which would 

 afterwards very gradually disappear. It was also found that on with- 

 drawing the plates from the ground, when the deviation had ceased, the 

 latter reappeared, if the plates were replaced in the earth either at the 

 same or at any other point. It is scarcely necessary to say that when 

 a current was made to pass, with electrodes of copper, across the mixed 

 circuit, the effects of secondary polarity were realized. 



In conclusion : there would be no security in the results if, in these 

 experiments on the electric current of the earth, electrodes of copper 

 were used without the proper precaution ; but by employing those elec- 

 trodes only after they have been left buried for twenty-four hours in the 

 earth and with a closed circuit, the proper currents of the earth are ob- 

 tained with the same deviation and the same constancy as with elec- 

 trodes of zinc, and even with greater intensity; and this probably 

 through the greater extension and depth of the electrodes of copper in 

 comparison with those of zinc. 



Galvanomete)'. — Unfortunately I have not been able always to use, in 

 these protracted experiments, the same galvanometer ; nevertheless, in 

 two of the most important series I have constantly used a galvanometer 

 of 2,000 coils, with a distinctly astatic system, and which underwent no 

 variation in the whole series of experiments. 



When I have wished to ascertain the electric state of the atmo- 

 sphere, I have used a thin wooden rod, G or 7 metres, (20 or 23 feet) in 

 height. At the upi)er extremity of this rod a porcelain insulator was 



