02 Dr. W. F. Gr. Swann on 



of the volume density of electrification necessary to account 

 for the earth's Held is 1<»-" B.M.U., which corresponds to a 

 deficit of one electron in about 10 ls pairs of positive and 

 negative electrons in neutral matter. 



The electrostatic field at the surface of the sphere amounts, 

 when expressed in electrostatic units, to 10 6 . The difficulty 

 in this respect may be overcome by imagining that, corre- 

 pponding to the volume distribution p, there is a surface 

 distribution of 5 units per sq. cm., the total amount of which 

 is just equal to that in the volume distribution. There will 

 then be no electrostatic field, but there will be a magnetic 

 field, which, in virtue of the absence of the electrostatic field, 

 will be the same for a moving as for a fixed observer*. 

 This magnetic field will be composed of two parts, which, at 



the equator, comprise [r) 7rpcoa 2 due to the rotation of the 

 volume distribution and ^irsaoa (see appendix, Problem I.) 

 due to the rotation of the surface charge. Hence, remem- 

 bering that 4:7rcrs=— ^7ra z p, we have H = 4 - irpo)a 2 , which 



gives H of the same order as before. 



With regard to the difficulty of explaining how the surface 

 charge is prevented from mixing with the volume charge, it 

 may be remarked, that the action of the centrifugal force of 

 the earth's rotation can easily be shown to result in distri- 

 bution somewhat of the kind described, and it would give 

 the field in the right direction, but on calculating the effect, 

 even for the most favourable case, where gravity doc- nor 

 pull the electrons, we find an effect of far too small an order 

 to account for the earth's field, so that unless other pheno- 

 mena, such as intense corpuscular pressure in the atoms of 

 some part of the earth's substance, be brought to our aid, 

 there seems little evidence in support of the distribution 

 required. 



TJie effect <>/' <t r<uii<tl electrostatic polarization resulting 

 from (he rotation. 



If we imagine that the rotation of the earth results in an 

 electrostatic polarization in a sense perpendicular 10 the axis 



• Of course it is not necessary thai the distribution shall be of the 

 simple nature described ; we might, for example, have thick concentric 

 sheila of positively electrified matter separated by thin shells oi 

 lively electrified matter. 



