440 Captain Lefroy's Preliminarij Report 07i 



diate stations. These dates are Jan. li, Feb. 19, March 17, 

 April 24, July 3l, Aug. 12, Sept. 12 and 18, Oct. 7 and SO, 

 Nov. 28. None of the stations singly give quite so many ap- 

 pearances as the previous year. The five Canadian stations, 

 which gave 121 appearances in 1848, give but 99 in 1849, the 

 remainder being made up from the other commands. 



These observations having been continued throughout the 

 night, may be referred to for testing an apparent law which 

 was noticed in the observations made by Serjeant Henry and 

 myself at Lake Athabasca in the winter of 1843-44, and which 

 is fully confirmed by the series at Toronto, namely, that the 

 aurora borealis does not appear with equal frequency at all 

 the hours of darkness, but is subject, like most other phaeno- 

 mena in meteorology, to influences having a diurnal period as 

 well as an annual one. The present series places the hour of 

 maximum frequency at 10 or 1 1 p.m. ; probably a longer con- 

 tinuance will be necessary to fix it accurately. 



Table showing the number of the times on which Aurora is 

 reported at each hour of the night. 



Any observation before 8 p.m. is here set down at 5, and so on. 



The aurora appears in Canada in every month of the year. 

 The greatest number of observations is in April ; and there is 

 a very marked excess in February, March, and April of each 

 year over any other period. Taking them by the seasons, 

 there are in the — 



Spring, March, April, May, 1848 40 1849 41 



Summer, June, July, August, ... 21 ... 29 



Autumn, September, October, November, ... 31 ... 34 



Wir.ter, December, January, February, 1848-49 37 1849-50 (about) 20. 



