AURORA BOREALIS. 



We had frequent opportunities of observing the Aurora 

 Borealis in the autumns of 1826 and of 1827. From the 

 25th of August until the 9th October, about the time of the 

 departure of the Blossom from the northern regions in both 

 years, this beautiful meteor was visible on every night that 

 was clear, or when the clouds were thin and elevated.* It is 

 remarkable that, in both years, its first appearance was on the 

 25tli August. The season of 1826 was distinguished by an 

 almost uninterrupted succession of fine weather and easterly 

 winds, and that of the following year by continued boisterous 

 weather and winds from the westward. In the former year, 

 the weather being fine, the Aurora was more frequently seen 

 than in the latter; but in 1827 the displays were brighter, 

 and the light more frequently passed to the southward of the 

 zenith. It never appeared in wet weather. 



In 1826, when, as before mentioned, the weather was settled, 

 the Aurora generally began in the W.N.W. and passed over 

 to the N.E., until a certain period, after which it as regu- 

 larly commenced in the N.E. and passed to the N.W. ; whilst 

 in 1827 the appearance of the meteor was as uncertain as the 

 season was boisterous and changeable. The period when 

 this change in the course of the light took place coincided very 

 nearly with that of the equinox ; and as the Aurora Borealis 

 has been supposed to be affected by that occurrence, we ima- 

 gined that tlie change might be in some way owing thereto, 

 but the irregularity of the meteor in this respect in 1827 gave 

 a contradiction to diis hypothesis. It was, however, uniform 

 in making its appearance always in the northern hemisphere, 

 and generally in the form of elliptical arches from 3o to 7" of 

 altitude, nearly parallel with the magnetic equator. These 



* In 1826 it was visible on twenty-one nights; in 1827 only eleven. 



