43-2 DK. C. CIIKEE; AXALYSIS OF RESULTS FROM THE KEW MAC4XETOnRArHS 
^ (L). A rather striking illnstratiou of the points of agreement and ditference just 
discussed, was afforded ])y the case of V at Kew. As already stated, all the tables 
liad' to he re-calculated, making allowance for the non-application of a temperature 
correction in earlier years of the period. 1 was puzzled for a time by the, at fir.st 
sinht remarkable fact that whilst the new values of differed verv appreciably from 
the old, there was exceedingly little difference between the new and old values of the 
other c coefficients, or of any of the phase angles. The explanation simply was that 
the external diurnal temperature inequality was pretty exactly represented in the 
ma'^netomanh room—of course, on a very much reduced scale—Avith the hours of 
maximum and minimum so retarded as to bring them much closer to the corresponding 
times for the diurnal variation of V. 
§ 70. There are other important conclusions to l)e draAvn from the comparison of 
tenq)erature and inagnetics. TTie source of the solar diurnal inequality in magnetism 
is, of course, the sun, acting in some way or other; tlie only question is Iioaa'. The 
researches of Gauss, Schmidt, Adams, and others all point to the conclusion that 
magnetic force at the Earth’s surface is derived, if not exclusively, at least in a wholly 
])reponderating degree, from centres of force within that surface. Accepting this 
conclusion, one would most naturally perhaps look for the action of the sun on the 
Earth itself as the cause of the diurnal inequality. Maxwell and others have 
pointed ont that tlie tidal effect of the sun by modifying internal strains must exert 
some influence on magnetic strata; Imt the trifling size of the lunar magnetic 
ine([uality indicates that any such effect must 1)6 very small. 
Tlie only other obvious direct solar effect on the Earth itself is the diurnal heating 
and cooling of the surface. It lias been pointed out, however, by Principal Puckee 
that, owing to the great preponderance of ocean, the portion of the Earth’s surface 
exposed to appreciable temperature variation is comparatively small. What I Avish 
to point out in addition to tliis is that any diurnal Aurriation of temperature in the 
Earth diminishes very rapidly as the depth increases, and the rate of decrease is 
more rapid for the part of the surflice heating represented liy the 12-honr and 8-hour 
Fourier terms than for the part ansAveiing to the 24-hour term. Also any seasonal 
underground change of temperature lags liehind the surface change. Noav aa'b haA'e 
seen that the ratios borne by the amplitudes of the higher order Fourier terms to the 
amplitude of the 24-li()ur term are not smaller, but much larger, in the magnetic 
elements than in tem])erature ; whilst magnetic sea.soiial changes seem earlier flian 
those of surface temperature, and much earlier tlian fliose presented hy an element 
such as A-apour pressure knoAA'n to he directly dependent on temperature. Tliere is, 
of course, the further obvious })oint that surflice temperature .sIioaa-s no large 
difference from years of sun-spot maximnm to years of minimum, and still less aatII 
underground temperature. 
This reasoning may appear quite unnecessary to those Avho accept Professor 
Schuster’s able paper in the ‘ Phil. Trans.’ for 1889, p. 467, as provnig the chief if 
