Royal Society. 499 



subject to periodical latos, connecting them with the seasons of the 

 year and the hours of the day at the particular stations. 



In preparing for the press the Jiourly observations of the declina- 

 tion in the years ISiS, 1844 and ISiS, at the same two stations, the 

 author found his previous opinions strongly confirmed ; and be- 

 lieving that the evidence thus obtained of periodical laws is far too 

 systematic, and rests on a series of too long duration to make it pro- 

 bable that it will be otherwise than confirmed by the continuation of 

 the observations in subsequent years, he has been induced to make it 

 the subject of a communication to the Royal Society ; although it is 

 probable that the exact periods, and the mean numerical values of 

 the effects produced, or their proportions to each other in the dif- 

 ferent seasons and at the different hours, may hereafter receive 

 modifications. 



Tiie disturbances which are the subject of this investigation 

 have two leading characteristics, — 1st, the irregularity of their 

 occurrence, many days together frequently passing without any 

 trace of them being discoverable; and 2nd, the large amount of 

 deviation from a mean or normal position, to which the needle is 

 subject during their continuance. It is this last feature which has 

 led to their general recognition, and to the establishment of the fact, 

 that when they take place their influence usually extends simul- 

 taneously, or nearly so, over all parts of the globe at which observa- 

 tions have been made. The same feature still aflTords their best and 

 most convenient distinguishing mark. 



Availing himself of this characteristic, the author separated from 

 the whole mass of hourly observations in the three years at both 

 stations a sufficient portion of disturbed observations to form an 

 adequate basis for investigation. The portion thus separated con- 

 sisted at Toronto of the 1630 largest deviations of the declination 

 magnet from its mean position ; the whole number of hourly observa- 

 tions in the same period being 22,376, and the proportion therefore 

 being 1 in 13-6; and at Hobarton of the 1479 largest deviations, 

 the whole number being 21,436, and the proportion 1 in 14'5. Of 

 the 1650 disturbed observations at Toronto, 472 were in 1843, 612 

 in 1844, and 566 in 1845. Of the 1479 disturbed observations at 

 Hobarton, 415 were in 1843, 562 in 1 844, and 502 in 1845. Whence 

 it appears, that at both stations, situated as they are in opposite 

 hemispheres and nearly at opposite points of the globe, 1843 was 

 the least disturbed year of the three, and 1844 the most so. Taking 

 the number in 1845 aa unity, the numerical proportions at each 

 station are as follows: — 



Toronto. Hobarton. 



1843 0-84 0-83 



1844 1-08 1-12 



1845 1-00 1-00 



This accordance, and the fact that the separated disturbed observa- 

 tions in these years occurred for the most part on tiie same days at 

 the two stations, are noticed as corroborating the conclusions, de- 

 rived from former investigations, of the character of these disturb- 



