20 SECTIONAL ADDRESSES. 



currents and associated magnetic phenomena. There will also be 

 mechanical reactions, and Schuster showed how these reactions can be 

 calculated. It is certain that the induced currents must tend to destroy 

 the motion of the inducing field, and that one effect must be to reduce the 

 period of rotation. Such a reduction in the period of rotation would result 

 even if the magnetic axis coincided with the axis of rotation, but when the 

 two axes do not coincide there is another retarding couple acting on the 

 magnetic field. A circular movement of the magnetic pole about the axis 

 of rotation may be regarded as produced by two radial movements at 

 right angles operating from that axis. Such motions of the magnetic 

 field will induce currents in the conducting layer, and the reacting 

 forces will tend to destroy the movements which produce them, that 

 is, the tendency will be to make the two axes coincide. The total 

 result is, therefore, to slow down and eventually destroy the rotation of 

 the magnetic axis and to reduce the angle of separation of the two axes 

 and eventually cause them to coincide. A bird's-eye ^saew of the magnetic 

 and geographical poles taken over a long period of time would reveal a 

 spiral path for the magnetic pole, the latter drawing nearer and nearer to 

 the geographical pole. 



Schuster has calculated the value of the retarding couple for many 

 variations of the conductivity of the outer medium. The retarding 

 couple is zero when the conductivity is zero or infinitely great, and is a 

 maximum when the conductivity is about 2.0x10"", which Schuster 

 points out is one which normally would be regarded as being very small, 

 as it is about 2.4x10"" that of mercury. The magnitude of the retarding 

 couple for such a conductiAaty has been calculated to be such that it 

 would take 125 centuries to lengthen the day by one second. Notwith- 

 standing the smallness of this, Schuster points out that its existence would 

 almost certainly have been detected by astronomers, and a value of the 

 conductivity smaller than 5 X 10~" or greater than 1 X 10"'^ is more likely. 



It is, of course, not necessary to assume a large volume of outer space 

 to have uniform conductivity to produce such effects. An outer layer 

 will suffice, and the conductivity may be uniform or patchy, but the 

 reactions will be of the sign indicated. It is certain that the movements 

 of the magnetic field are not simple as outlined above but are very 

 complex, and that unexpected reversals occur, so that it is not possible 

 to predict the conditions even 20 years ahead. The theory advanced 

 is, however, still capable of explaining the variations, for any conducting 

 layer may not only vary greatly over considerable areas, but there may 

 be relative motion between the earth and portions of the layer which 

 also varies. • 



Unfortunately, such an explanation of the cause of secular change 

 does not help very much in explaining the cause of the field itself, but it. 

 is obvious that any complete theory must not only explain such a change 

 but must also account for the present difference between the earth's 

 magnetic axis and its axis of rotation and the variation in intensity. 



Theory of an Iron Core. 



As is well known, the first idea was that the core of the earth is of 

 iron and is magnetised. But, apart from our lack of knowledge of the 



