APPENDIX. 33 



These figures seem to indicate the existence of a period of frequency, probably of eleven years as 

 conjectured by Prof. Wolf, the least number probably occurred in 1843, if we make an allowance for 

 invisibility of the phenomenon either by daylight or by cloudy weather. 



Between June, 1840, and July, 1845 (incl.), there were seen, according to the Toronto record, 109 

 auroras. The disturbances at Philadelphia on the dates of their appearance have been classified as 

 follows : The numbers give the relative proportion to the total number, which latter is expressed by 

 100; the average numbers are given resulting from the examination of the disturbances of the decli- 

 nation, the horizontal and the vertical force. 



I^umber of cases. 

 No record at Philadelphia , . . . ' . . . .19 



None of the elements disturbed 30 



But very few disturbances 20 



An ordinary number of disturbances 14 



An unusual number of disturbances 17 



The number of unusual disturbances is therefore less than one-fifth of the total amount, and in 

 fully one-half of the cases the magnetic elements were either not at all or but very slightly affected. 



