366 De la Rive on the Aurora Borealis. 
ter, Farquharson, and by all voyagers, establish the following 
conclusions :— 
1. During the day preceding the night on which an aurora ap- 
pears, the declination of the magnetic needle to the west is al- 
ways augmented 10, 20 or 30 minutes, or more. 
2. On the contrary, at the middle, and at the end of the ex- 
hibition, the needle deviates from its normal state to the east. 
3. Finally, the needle often undergoes irregular perturbations 
during an aurora, amounting to several minutes 
It happens ordinarily that the maximum deviation of the nee- 
dle during the day preceding the night of the aurora, is at noon, 
or half an hour after noon; and the deviation due to the disturb- 
ance may be 5 to 30 minutes or more, beyond that of the days 
before or following. Sometimes the maximum western deviation 
is at other hours in the morning, and it is probable that in such 
cases there is an aurora during the day. Arago cites several ca- 
ses of this kind. Thus, on the 17th of August, 1828, the decli- 
nation from 84h. a.m. till noon was 5’ above the mean of the 
month for the same hours; and on the same day, at 10h. P.™., 
Messrs. Coldstream and Foggo perceived feeble traces of an au- 
rora which was probably the end of a day aurora, During the 
evening the needle was in its ordinary position. 
The magnetic observations made in the regions near the pole 
confirm the influence on the needle. Thus at Reykinwik (64° 
8’ 15’ N.) MM. Lottin and Bravais, having made numerous ob- 
servations on the diurnal variation of the needle parallel with sim- 
ilar observations at Paris and Cherbourg, were struck with the 
Lowenérn made in 1786, 50 years before, they satisfied them- 
selves that the effect was due to auroras invisible to them because 
without hesitation to the aurora. ‘This conclusion is confirmed 
by the very numerous and excellent observations of MM. Lottn 
and Bravais. 
We thus see, that for a long period observations near the pole 
have shown that auroras must be more frequent than was SUP- 
posed, and this is confirmed by the facts observed in Canada and 
the United States. : 
We therefore conclude, that the production of auroras, north- 
ern and southern, is the normal mode of neutralising the positive 
electricity of the atmosphere with the negative of the earth. 
