170 DR. S. CHAPMAN ON THE LUNAR DIURNAL MAGNETIC VARIATION, 



amending paper ('Met. Zeitschr.,' 12, p. 589, 1913) withdrew his earlier suggestion of 

 a primary internal magnetic field concerned in the lunar variations; it may be 

 remarked incidentally that such a field is rendered unlikely by the magnitude of the 

 components other than semi-diurnal in the magnetic variations.* These appear to be 

 excited by the semi-diurnal atmospheric tide in conjunction with a variable electrical 

 conductivity of the atmosphere, depending on the solar hour angle ; their phases 

 change by multiples of 2?r in the course of a lunation, according to ascertained laws. 

 While only secondary phenomena, they are comparable in magnitude with the 

 main (semi-diurnal) component of the lunar variation, so much so, in fact, that it 

 is hardly possible for more than a very small fraction of the latter to be due to 

 internal causes, since this portion would not account for any part of the secondary 

 components. 



Since perigee and apogee occur at all phases of the moon during a sufficiently long 

 period of time, no explanation of the perigee-apogee phase change in the magnetic 

 variations can be looked for in any direct solar action. 



Tims far, therefore, the attempt to assign a known cause to this remarkable 

 phenomena has been unsuccessful. It remains to make one more suggestion, which is 

 at once very tentative and far from definite. We should naturally suppose that the 

 tidal effect of the moon is to produce a lunar magnetic variation A cos (2t + a.), 

 the amplitude (A) of which undergoes a regular variation of the kind observed, 

 but which shows 110 change of phase ; the observed phase change might be produced 

 if in some other way the moon produced a semi-diurnal magnetic variation of type 

 B cos (2t + /3), where ft exceeds a by about 90 degrees, provided that B/A is fairly small 

 and that B increases more quickly than A as the moon's distance diminishes. The 

 combination would slightly increase the theoretical variation of amplitude, but this 

 need only be by a small amount. 



Not much importance can be attached to this suggestion, however, since it is 

 very difficult to conceive of any way, other than tidal, in which the moon could 

 produce a semi-diurnal variation of terrestrial magnetism. Certainly no appreciable 

 direct magnetic effect of the moon is at all likely, nor would it, in any case, supply the 

 desired variation in the present instance, since its effect would be diurnal and not 

 semi-diurnal. The matter must therefore remain a mystery for the present, but it is 

 a mystery whose solution may be the clue to some important new fact relating to 

 magnetism of the atmospheric tides ; it is with this hope that I venture to publish the 

 preceding very inconclusive discussion. 



In conclusion it is a pleasure to acknowledge the assistance which has been placed 

 at my disposal, in the execution of the computations involved in this paper, by the 

 Government Grant Committee of the Royal Society. 



* Of. 'Phil. Trans.,' A, vol. 213, p. 279, 1913. 



