102 Dr. C. V. Burton on a 



conductors moving between one level and another, and 

 making contact intermittently with the extremities of an 

 elongated vertical conductor. It would be easy to devise a 

 machine for performing continually a cycle of operations, 

 whereby any electromotive influence due to gravity would be 

 rendered effective in gradually imparting contrary charges 

 to a pair of insulated hollow conductors. But apart from the 

 excessive minuteness of the effect to be looked for, and the 

 thermoelectric and other disturbing influences almost neces- 

 sarily encountered, another consideration steps in to render 

 any such attempt nugatory. To guard against stray electric 

 influences, our whole apparatus would require to be enclosed 

 in a conducting envelope, the charge induced on which by 

 gravitational influence would exactly neutralize, throughout 

 the interior of the envelope, the quasi-electromotive intensity 

 due to the direct action of gravity. Thus, even if we suppose 

 positive and negative electrons to be oppositely acted upon 

 by a gravitational field (in which case, for example, the 

 aggregate attractive force on the negative electrons in a body 

 might greatly exceed the weight of the body), it would still 

 be impossible, by means of laboratory experiments, to detect 

 any electromotive effects due to the earth's gravity. 



57. The case of the earth moving in its orbit under the 

 sun's attraction may be regarded as that of a body moving 

 freely under a sensibly uniform field of gravity. Making 

 any assumptions that suggest themselves as to the number 

 of electrons in a gram of matter, the relative masses of 

 positive and negative electrons, and the forces experienced 

 by these under gravitational influence, (64) may be used to 

 estimate the resulting electrical distribution on the earth's 

 surface (or in the upper strata of the atmosphere) arising 

 directly from the gravitational attraction of the sun. But 

 with such assumptions as I have tried, it appears that 

 the effect to be expected is very minute. For example, 

 making the same assumptions as in § 52, and in addition 

 assuming that nearly all the inertia of neutral matter is the 

 inertia of positive electrons, the extreme difference of 

 potential between opposite poles of the earth comes out as 

 about 80 microvolts. Moreover, the shifting of this feeble 

 electrical distribution, owing to the earth's (solar) diurnal 

 rotation, could not have any sensible influence on the 

 phenomena of terrestrial magnetism. 



