18 



CHANGES OP THE MAGNETIC DECLINATION 



number of hourly readings (equal to 17), and m the mean disturbance, then m = 



I 



t 



V A 2 



This quantity is analogous to the mean error of an observation. In 



^ t 



the following comparisons we must always bear in mind that the observations for 

 the present discussion are rather limited, and that the comparisons with results at 

 Lake Athabasca and Fort Simpson are of a date nearly ten years earlier. This 

 interval is perhaps favorable to the comparison. 



At Van Rensselaer Harbor the mean disturbance force is greater than at either 

 place just named, and pretty regular during two well-marked periods, as shown by 

 the following table : 



TABLE OF THE MEAN DISTURBANCE OF THE DECLINATION AT VAN RENSSELAER HARBOR, TAKEN 

 WITHOUT REGARD TO DIRECTION, FOR EACH OF THE OBSERVATION HOURS, AND EXPRESSED IN MlNUTES 



OF ARC. 



Local Mean Time. 









."I 



The disturbing force is least during the day (if such an expression is admissible 

 in this case), from 10 A. M. to 7 P. M., and greater and equally regular during the 

 hours of the night (?), from 8 P. M. to 8 or 9 A. M. At Lake Athabasca the hours 

 ^\ / of least disturbance are between 9 A. M. and 7 P. M., and at Fort Simpson from 

 10 A. M. to 7 P. M. Captain Lefroy, in his discussion of the disturbances of the 

 declination remarks : " There are indications in each of the three curves (for Lake 

 Athabasca, Toronto and Sitka) of a small increase in the mean disturbance about 

 noon." At Van Rensselaer Harbor we find the maximum disturbance at this very 

 hour preceded and followed by quite small values; this circumstance certainly 

 deserves our particular attention. Further coincidences of the disturbing force can 

 be noticed at 5 P. M., at which hour at Van Rensselaer, Lake Athabasca, and Sitka 

 the minimum disturbance has been observed. At Fort Simpson, in April and May, 

 1844, the mean disturbance was but one-fourth of that observed in January, Feb- 

 ruary and March at Van Rensselaer, and the ratio of the minimum to the maximum 

 value was 5.6 and 2.0 at the two places respectively. 



By adding the squares of the differences for each hour of the day and month, we 



find the mean monthly disturbance by the formula f-t _J. The 



\N 24 



ance for each month is as follows : 



mean disturb- 



In January, 1854 

 In February, " 

 In March, " 



30' 

 65 

 40 



1 Principally clue to a very large disturbance. 



