I 



ON COMPARING AND EEDUCING MAGNETIC OBSEEVATIONS. 83 



Secondly, the moon would produce tides in those regions, which would 

 be the cause of a lunar semi-diurnal magnetic effect. 



Thirdly, the sun, acting as the moon does, would likewise produce 

 tides which would be the cause of a solar semi-diurnal magnetic effect. 



Fourthly, these various effects would be increased during those hours 

 when the sun is powerful, inasmuch as the upper atmospheric regions 

 become better conductors at high temperatures. 



19. If we now leave the regular variations, and turn to magnetic dis- 

 turbances, there seems reason to suppose that the earth, like any other 

 magnet, may be subject to small and abrupt changes of magnetism, and 

 it is quite conceivable that such changes may produce secondary currents 

 in the moist conducting strata of the earth, and likewise in the upper 

 atmospheric regions. We know, as a matter of fact, that there are such 

 earth-currents, and the observations made at Greenwich show that they 

 are intimately associated with the disturbances registered by the self- 

 recording magnetographs. 



20. The late Dr. Lloyd was the first to remark that ' the rapid changes 

 of the earth currents are much greater in proportion to the regular daily 

 changes than the corresponding movements of the magnetometers.' We 

 may perhaps interpret this to mean that a smn U but abrupt magnetic 

 change is associated with a larger earth-curijnt manifestation than 

 another change of the same size, but of a more gradual nature. This 

 would appear to be in favour of the view that such earth currents are 

 secondary currents due to small but abrupt changes which take place in 

 the magnetism of the earth. In conformity, too, with this hypothesis, 

 •cases may be pointed out where the magnetic disturbance, while rapidly 

 varying, is yet altogether on one side of the normal, and where the cor- 

 responding earth currents pass alternately from strong positive to strong 

 negative. 



21. Quite recently (see Appendix V. to this report) the Rev. Professor 

 Perry and Professor Stewart have brought before the Royal Society the 

 results of a preliminary comparison between the fluctuations of the 

 declination at Kew and at Stonyhurst (neighbouring stations), and have 

 derived the following conclusions : — 



(1) In the very great majority of cases the angular value of the 

 declination disturbance is greater for Stonyhurst than for Kew. 



(2) The ratio "- ^ is certainly greater for disturbances 



of short than for those of long duration. 

 If we add to these conclusions the fact noticed by these observers that all 

 the disturbances occur in couplets, we may be disposed to agree with them 

 that in the case of disturbance as exhibited by a suspended magnet there 

 are probably two causes at work, the first of these being a change in the 

 magnetism of the earth, and the second an induced current due to this 

 change. 



22. It would thus appear that in this provisional working hypothesis 

 the principle of current induction is brought forward with the object of 

 explaining both the regular and the irregular magnetic fluctuations. It 

 is sought to explain the former by the hypothesis that in the upper 

 atmospheric regions we have conductors moving across lines of magnetic 

 force, and hence animated by a current. It is sought to explain the latter 

 on the supposition that small but abrupt changes of the magnetism of the 

 earth by a method similar to that in a Ruhmkorff's coil cause secondary 



G 2 



