u V T il !•: M A <; N 15 T 1 (' FORCE 



■V.) 



With the exception of the month of November, the values given above for the 

 annual variation are tolerably regular in their progression, and considering the 

 delicacv of the test applied to the observations in deducing the annual variation, 

 this exceptional irregularity in the November value will not affect the general 

 conclusion. We have as the general result: a greater horizontal force in summer 

 (from April to August), and a smaller horizontal force in winter (from September 

 to March) than the average annual value. The maximum occurs in July (at 

 Toronto hi June), and the minimum in January (at Toronto in December). 

 For Toronto we have the expression for the annual variation: — 



3.531 + 0.002 sin (0 + 312°). 

 For Philadelphia (omitting the November value): 



4.176 + 0.001 sin (6 + 30(5°); 

 the angle 6 in both equations counting from January 15th. 



The annual range is 0.0021 (in absolute measure). The transition appears to 

 take place about the time of the equinoxes or a short time before. 



Table XIII. contains the monthly normal values of the horizontal force in abso- 

 lute measure, obtained by adding (algebraically) 4.1730 to the values in the last 

 column of Table XII. These numbers, it will be observed, are corrected for secular 

 change; if we apply the same we obtain the resulting monthly mean values of the 

 horizontal force answering to the epoch January, 1843. The quantity A, mentioned 

 in the explanatory remarks to Table VII., is given in the last column of Table XIII. 



