

362 Wm. A, Norton on the Variations 



to 8 a. m.. the line moves towards the south and the needle east- 

 wardly. When the changes are both increments, if the first is 

 less than the second, as from 10 a. m. to 1 p. m., the line rises and 

 the needle moves toward the west ; but if it is greater than the 

 second, as from 1 p. m. to 4 p.m., the line falls back, or toward 

 the south, and the needle moves towards the east. When the 

 first change is a decrement and the second an increment, as at 

 about 10 a. m. and toward midnight, the line rises and the motion 

 of the needle is toward the w r est; and when the first is an incre- 

 ment and the second a decrement, as near 3 p. m. and 5 a. m., it 

 falls and the needle moves eastward. It appears therefore that 

 the needle should move toward the west when the curve of the 

 horizontal force is concave upward, and toward the east when 

 the same curve is convex upward. The westerly movement 

 should then be from 8 a. m. to 1 p. m., and from 7 or 8 p. m. to 2 

 or 3 a.m.; and the easterly movement from 1 p. m. to 7 or 8 p. m., 

 and from 2 or 3 a. m. to 8 a. m. These results accord with obser- 

 vation, with the single exception, that the time of minimum de- 

 clination is generally about two hours later than 8 p. m. 



If we compare the curves of horizontal force for the different 

 quarters of the year, we find that, while the points of maxima and 

 minima, as well as the points of inflexion, are pretty nearly the 

 same for all, the curvatures are in general greater for the two mid- 

 dle than for the first and last quarters of the year, and therefore the 

 daily changes of declination should be greater toward the middle 

 than toward the beginning or end of the year — a result which 

 accords with fact. There appears, however, to be generally 

 a more rapid variation of the horizontal force toward midnight, 

 during the cold than during the warm months ; which must be 

 attended with corresponding differences in the small nocturnal 

 increase of declination. This result seems also to be in accord- 

 ance with fact ; but it would be premature to attempt the detail- 



ed explanation of such minute differences among the variations 

 from a limited series of observations made only at the station of 

 the needle. A similar remark may be made with respect to 

 certain small discrepancies which may be observed, between 

 theory and fact, in relation to the relative amounts of the varia- 

 tions at the same hour in the different quarters of any one indi- 

 vidual year. A theory which furnishes a sufficient explanation 

 of all tlie laws deducible from the observations, cannot reasona- 

 bly be rejected on the score of small discrepancies in quantity, 

 when the observations are much less extended than the theory 

 calls for. The precise movements of the line of equal magnetic 

 action of the surface stratum upon which the motion ot the 

 needle depends, can only be ascertained with certainty by institu- 

 ting special observations at a variety of places in every direction 

 and at various distances from the station of the needle. The 





