THE ARCTIC SEAS. 119 



of one which I observed in Lower Canada, in 

 Februaiy, 1837, will give a notion of tiie appear- 

 ance of this meteor in its more usual state. " I first 

 obsei'ved it about half-past eight o'clock ; a long, low, 

 irregular arch of bright yellow light extended from 

 the north-east to the north-west, the lower edse 

 of which %vas well defined ; the sky beneath this 

 arch was clear, and appeared black, but it was only 

 by contrast with the light, for on examination I 

 could not find that it was really darker than the 

 other parts of the clear sky. The upper edge of the 

 arch was not defined, shooting cut rays of light 

 towards the zenith : one or two points in the arch 

 were very brilliant, which were varpng in their 

 position. Over head, and towards the south, east, 

 and west, flashings of light were darting from side 

 to side : sometimes the sky was dark, then instantly 

 lighted up with these fitful flashes, vanishing and 

 changing as rapidly; sometimes a kind of crown 

 would form around a point south of the zenith, con- 

 sisting of short converging pencils. At nine o'clock, 

 the upper and southern sky was filled with clouds or 

 undefined patches of light, nearly stationary ; the 

 eastern part, near the top, being bright crimson, 

 which speedily spread over the upper part of the 

 northern sky. A series of long converging pencils 

 was now arranged round a blank space about 15° 

 south of the zenith, the northern and eastern rays 

 blood-red, the southern pale yellow ; the redness 

 would flash about, as did the white light before, still 

 not breaking the general form of the corona. In a 

 few minutes all the red hue had vanished, leaving 



