342 Mr. Macdonald's Observations of the diurnal Variation 
diurnal of the variation, increased from about 7 in the morn- 
ing till 5 in the afternoon, and that it decreased till 7 in 
the morning. The irregularities which are apparent do not 
destroy the general evident inference of result, as it is well 
known that the common increasing or decreasing variation is 
subject, in the same place, and on the same day, to certain de- 
viations, from latent causes. It appears in general, that such 
diurnal variation of the variation as had been observed during 
thunder, is greater than it ought to have been, cceteris paribus. 
It has been remarked, that heat weakens the magnetic virtue, 
and that cold strengthens it. Supposing, with the great Hal- 
ley, the existence of 4 magnetic poles, by blending this sup- 
position with the above principle, well ascertained, attempts 
have been made to account for the diurnal variation of the 
variation. The south-east magnetic pole being less heated in 
the morning, either by the sun or by subterraneous fire, than 
towards noon, and in the afternoon, and being at the same 
time, by passing through the meridian of Celebes, nearer Su- 
matra than the south-west magnetic pole, it draws to it in the 
morning the south end of the magnetic needle more powerfully 
than the other attracts ; and, consequently, the variation diur- 
nal of the variation ought to be, and actually is, less in the 
morning than in the afternoon. In the progress of the day, 
the south-east magnetic pole having become heated, and the 
south-west pole being at the same time less heated, attracts 
the south end of the magnetic needle more powerfully than the 
other does ; and hence the east diurnal variation of the varia- 
tion appears greater in the afternoon than in the morning. It 
is found in Europe, that this diurnal variation of the variation 
is greater in summer than in winter. This seems to point out 
