440 Prof. Balfour Stewart on the Cause of the 



The argument as I have now advanced it is of a general 

 nature and is twofold. 



(1) In the first place it brings forward objections to the 

 localization of the cause of the diurnal variations anywhere 

 else than in the higher regions of the Earth's atmosphere ; 

 then points out that if this cause exist in these regions it 

 must be in the shape of a system of electric currents ] and 

 finally endeavours to show that there are no valid objections 

 to such a system of currents. 



(2) It endeavours to show that, if we take the diurnal 

 variation of the declination-needle, which is of a remarkably 

 simple type, we have great difficulty in imagining that it can 

 be brought about by any temporarily induced displacement of 

 the magnetic system of the earth, while we can with the 

 greatest ease account for it by means of a system of currents 

 in the upper regions of the Earth's atmosphere. 



At this point Dr. Schuster Jias endeavoured to apply mathe- 

 matical analysis to the subject, and he has obtained results 

 which are very satisfactory and encouraging *. He had pre- 

 viously, in the Keport of the Magnetic Committee of the 

 British Association for the year 1885, pointed out the existence 

 of a test f which will enable us to decide whether or not the 

 diurnal variation can be caused by a series of electric currents 

 which embrace both the upper atmospheric regions and the 

 Earth, and complete themselves by electric discharges through 

 the lower regions of the Earth's atmosphere into the ground ; 

 and he now endeavoured to ascertain whether the supposed 

 vertical currents at the equator, to which allusion has been 

 made, could possibly be of this nature. If this were the case, 

 he argues, and if a unit pole were carried from a position A 

 at the Earth's surface in a path which should return again to 

 A, and in the centre of which there should exist such a ver- 

 tical current, then through all its course work would be done 

 either by or against this pole, inasmuch as there would be a 

 continued effort of the vertical current to turn the pole round 

 it in one direction. In other words, the pole and the point A 

 being immersed in the system of currents, there would be a 

 potential having multiple values. On the other hand, if the 

 system which produces the daily variation be outside of A, 

 whether above or below it, there will be at A a true potential, 

 and if the unit pole be carried as before, in any path from A 

 and back to A again, as much work will be done against the 

 pole as will be done by it. Let us consider, now, a little path 

 embraced between two parallels of latitude near each other, 



* Proc. Lit. & Phil. Society of Manchester, January 12, 1886 : Phil. 

 Mag. April 1886, p. 349. 



t The reader will find a complete discussion of this subject in Maxwell's 

 ' Electricity and Magnetism.' 



