324 Prof. A. Schuster on Magnetic Precession. 



a possible case if the conductivity is of an electrolytic 

 character, w ^ 



n = -r — T, ft) = X T-TTT-r, CO. 



2a 2 8xl0 17 

 Hence the time of revolution of the magnetic system would 

 be 2*5 X 10 17 days or about 7 x 10 u years. jSTo admissible 

 value of yLt could reduce this number very materially. If the 

 currents could be imagined to be confined to a sheet of 

 thickness t, the angular velocity would be increased to 



0= ■— ^CO. 



07r at 

 Taking //,/£ = 1 this would give for the period of revolution 

 Sira/3 or as 27ra = 4x 10 9 the periodic time would be 5 x 10 9 

 days, or 14 x 10 6 years, which is still of a much bigger order 

 of magnitude than the period of the secular variation. To 

 produce a revolution of the magnetic system of e.g. 500 

 years t/ji would have to be equal to 36 X 10~ 6 , so that for 

 possible values of fi, t would have to be reduced to molecular 

 dimensions. 



10. The fact that a current sheet of molecular dimensions 

 would show a magnetic precession of angular velocity com- 

 parable with that of the secular variation suggests the possi- 

 bility of the phenomenon being rather of a molecular than 

 a molar character. If terrestrial magnetism is due to the 

 rotation of the electron round the atom, a precession of the 

 required amount might be produced. But the subject is not 

 capable of theoretical treatment without making some assump- 

 tions of too speculative a character to be introduced here. 



An experimental investigation would be more likely to 

 help, for it might give an answer to the question : — " Does 

 the magnetic axis of a transversely magnetized sphere or of a 

 steel disk magnetized along a diameter, lag behind when the 

 sphere or disk is rotating rapidly ? " Some trials which 

 were made, under not very suitable conditions, have given 

 me a negative answer to the question, but I hope to be able 

 to arrange for more decisive experiments. 



11. In the above investigation it has been assumed that 

 electric currents are conveyed equally by the two opposite 

 electricities, so that the current as a whole has no moment of 

 momentum, but as far as I can see without detailed calcu- 

 lation the general result of the investigation would not be 

 altered, even assuming the existence of angular momentum. 

 As far as the results of this investigation are concerned, we 

 must conclude that the inertia of electric currents, which 

 doubtless does exist, is too small to account for the observed 

 secular variation of terrestrial magnetism. 



1 



