312 ON THE ELECTRICAL CURRENTS OF THE EARTH. 



remaineil at the same angle during' tlie whole experiment, which sometimes con- 

 timied for an hour; but I have observed also, without any change having occur- 

 red in the state of the sky, a movement in the needle almost periodic. Twice I 

 have seen the needle deviate at lirst by an ascending current, and after some 

 minutes descend to zero, then pass into the opposite quadrant and return after- 

 wards to the previous deviation, becoming eventually fixed under the action of 

 the current ascending in the wire. It has seemed to me that this phenomenon 

 was presented when the water which filled the cavities of the electrodes was in 

 movement and flowed rapidly aAvay around the porous vessels. Reflection on 

 the conditions under which we are compelled to operate in tliis sort of experi- 

 ments, will suffice to evince the difiiculty of solving all the doubts which may 

 present themselves in the prosecution of our inquiries. 



Notwithstanding the diificulties inherent in such researches, and which impose 

 on the physicist the greatest reserve in his conclusions, we may regard, I think, 

 the following results as founded on a large number of facts conformable with one 

 another and obtained under different circumstances : 



When a metallic line is stretched upon the earth, but isolated from it, while 

 the extremities of the wire communicate with the earth at two points having a 

 different elevation, an electric current circulates constantly in the wire, the cause 

 of which current can be attributed neither to the chemical action of the electrodes, 

 nor to that (jf the terrestrial strata in which they are sunk. 



This current is constantly directed in the wire from the lowest towards the 

 highest point, and its intensity is greater in the longer lines and as the difference 

 of level of the extremities is more considerable. 



The intensity of this current does not vary sensibly with the depth of the 

 cavities in which the electrodes are sunk, and is the same in the wire suspended 

 at some metres from the ground as in tliat which is in contact with it. 



Two circumstances present themselves as constantly associated with this phe- 

 nomenon, circumstances which, by their analogies, may assist in explaining it ; 

 I mean the difi'erence of temperature of the two extreme points and the difference 

 of electric tension of these points. I shall only remark here that I could cite 

 results in which the influence of difference of temperature could not be consid- 

 ered as cause of this phenomenon, which to me appears to be due to terrestrial 

 electricity. 



