Notice of the late Aurora Borealis. 105 



and remarkably comoid — the temperature warm. The 31st, 

 also warm. In the forenoon, clouds were visible, consis- 

 ting of bars of whitish vapor, very regularly disposed, form- 

 ing the mackerel back sky. About noon, a smart shower. 



On Sunday evening, September 9th, at 8 o'clock, I walk- 

 ed out, and saw a considerable light in the north, rounded 

 into a semi-eliptical form. It extended probably through 

 ninety degrees of the horizon. It was not very well defined, 

 but was rather darker in the centre near the horizon. In 

 about five minutes, I saw, almost directly overhead, a light 

 cloud hanging in the clear sky, apparently lower than clouds 

 generally are, which I immediately recognized, as a frag- 

 ment of the aurora. It moved gently to the west, and as it 

 moved became more and more distinct, and better defined, 

 putting on gradually the appearance of beams of light, hav- 

 ing a southern inclination. Suddenly they became very 

 straight, and well defined on their north side, as if I then saw 

 all their northern sides in a line. Proceeding still west, and 

 probably a little south, they seemed to lengthen, and de- 

 scend towards the western horizon. Looking at them in 

 this direction, I judged their deviation from a perpendicular, 

 to the horizon to be from fifteen to eighteen degress. Pro- 

 ceeding still farther west, they grew gradually shorter, and be- 

 gan to open, and soon had the appearance of five or six col- 

 umns, of about the same height, the northernmost standing 

 near the horizon, and the rest a little higher and higher, 



forming a part of an arch ; thus \\m\l 



About the time, this disappeared, some scattered frag- 

 ments of a more considerable arch, appeared a little north 

 of the zenith. It soon became more distinct, and the part 

 east of the meridian, although rather faint, was well formed. 

 It seemed to consist of bars of light lying in this manner : 

 s 



N 



The movement westward, and the wheeling motion were 

 distinct, but not sufficiently so to enable me to estimate the 

 rate of motion. It gradually approached and passed the 

 zenith, becoming narrower, and of a more uniform breadth. 

 When it had passed the zenith a little, it seemed to have col- 



Vol. XIV.— No. 1. 14 



