SUPPLEMENT. 
25 
scarcely pretend to ascertain the absolute heat of any particular 
month or day, so we must consider the influence of heat upon 
magnetism to operate in the like manner ; viz. that for a short 
time it scarcely manifests itself, yet in the course of a century 
the constancy and regularity thereof become sufficiently apparent. 
It would therefore be idle to suppose that such an influence could 
be derived from an uncertain or fortuitous cause. But if it be 
allowed to depend upon the constancy of the sun’s motion, and 
this appears to be a cause sufficient to explain the phenomena, 
we should, agreeably to Newton’s first law of philosophising, 
look no further. 
As we therefore consider the magnetic powers of the earth to 
be concentrated in the magnetic poles, and that there is a diurnal 
variation of the magnetic needle, these poles must perform a 
small diurnal revolution proportional to such variation, and 
return again nearly to the same point. Suppose then that the 
sun in his diurnal revolution passes along the northern tropic, 
or along any parallel of latitude between it and the equator, 
when he comes to that meridian on which the magnetic pole is 
situated, he will be much nearer to it, than in any other, and in 
the opposite meridian he will of course be the farthest from it. 
As the influence of the sun’s heat will, therefore, act most power¬ 
fully at the least, and less forcibly at the greatest distance, the 
magnetic pole will consequently describe a figure something of 
the elliptical kind ; and as it is well known that the greatest 
heat of the day is some time after the sun has passed the meridian, 
the longest axis of this elliptical figure will be north-easterly in 
the northern, and south-easterly in the southern hemisphere. 
Again, as the influence of the sun’s heat will not from those 
quarters have so much power, the magnetic poles cannot be 
moved back to the very same point, from which they set out, 
but to one, which will be a little more northerly and easterly, or 
more southerly and easterly, according to the hemispheres in 
which they are situated. The figures, therefore, which they 
describe, may more properly be termed elliptoidal spirals. The 
north magnetic pole may by this means, be carried with a slow 
but constant motion, more and more to the north.-eastward, till it 
82. <P 
