the Earth's Magnetic Field. 81 



of the effect according to the inverse square law of the 

 distance. 



The manners in which the volume elements of the earth 

 may give rise to a magnetic effect may be classed under 

 three heads. 



(1) The rotation of the earth may give rise to an ordinary 

 current, involving motion of electricity relative to the volume 

 elements themselves. 



(2) Each volume element may act as though it were 

 charged, or the rotation may result in some kind of radial 

 distribution of the charge. 



(3) Each molecule of the earth may be looked upon as 

 equivalent to a very feeble magnet, owing to want of complete 

 compensation of the magnetic fields due to its orbital elec- 

 tronic motions, and the rotation of the earth may bring about 

 an orientation of these molecular magnets in favour of one 

 direction. The only advance of such a theory over an 

 assumption of permanent magnetization would be in its 

 attempt to seek in the rotation itself a reason for the 

 orientation required. 



Any theory which is to explain the earth's magnetism as 

 arising from its rotation must be restricted to some extent 

 by the following conditions : — 



(1) It must provide for the absence of any large effect 

 due to the earth's translatory motion in its orbit, since such 

 an effect would be different along the line of the earth's 

 motion to what it is in a direction perpendicular thereto, so 

 that a large diurnal variation in the magnetic elements would 

 result. 



(2) It must be of such a nature as to predict no appreciable 

 magnetic field in any laboratory experiment in which a 

 sphere of moderate size rotates with a reasonable angular 

 velocity, as such fields would not escape detection. 



(3) It must account for the fact that the sun, rotating 

 with its known angular velocity, produces no appreciable 

 field at the earth. Such a field would, owing to obliquity of 

 the ecliptic, give rise to a large diurnal variation in the 

 magnetic elements. 



(4) It must not be of such a type as to lead to an enormous 

 field at the sun's surface, due to the rotation of that body, 

 since it may be certain that a field for example of the order 

 of 2000 c.G.s. units would have been detected by the Zeeman 

 effect in the sun. Such an effect would be distinguishable 

 from the effect observed in the sun spots in that it would 

 show a regular variation over the sun's disk. 



There is a certain amount of elasticity in the conditions 

 Phil. Mag, S. 6. Vol. 24. No, 139. July 1912, (j 



