500 Prof. E. Edlund on the Magnitude of 



equal to 0*1326 metre, and r = 2*57r= 31*416. From these 

 determinations we obtain, by means of equation (7), 

 x = 0*02325 D.Ar. 

 The same determination was repeated on a subsequent occasion 

 with another magnet, of the same length as the first, but having 

 a diameter of 15 millim. The same cylinder w T as employed, 

 and the velocity of rotation was the same as in the first expe- 



Cot 77 



riment. These determinations gave — 3— =56*0. The current 



due to unipolar induction gave a deviation of 87'0 scale-divi- 

 sions; and one Daniell cell, with a resistance of 100,000 ohms 

 in the current, gave 357 scale-divisions. The numerical infe- 

 riority of these deviations in proportion to those first obtained 

 is due to the smaller distance between the mirror and the scale. 



5*2151 

 These observations give A= .. ~ 5 D. With these values of 



the unknowns we obtain %= 0*02297 D Ar; and, consequently, 

 for the mean of the two determinations 

 ^ = 0*0231 DAr. 

 This equation refers to points of the terrestrial surface in the 

 neighbourhood of the equator, where the vertical component of 

 the force of unipolar induction of the earth is at its maximum* 

 and whence it diminishes as we approach the magnetic poles, 

 where its value becomes zero. The horizontal component of 

 the same force, on the other hand, is zero at the equator, but 

 augments as we recede from the equator, acting, as I have ex- 

 plained in the memoir cited, in the northern hemisphere towards 

 the north, and in the southern hemisphere towards the south. 

 Although r and co of equation (1) are modified when we recede 

 from the terrestrial surface in a vertical direction, either above 

 or below the surface, we may consider these values as nearly 

 constant, if the distance does not exceed a few thousand metres. 

 For distances much greater from the surface of the earth, the 

 induction gradually approximates to zero, because the co of 

 the numerator only increases in proportion to r } whereas the 

 denominator increases nearly in the ratio of the cube of the 

 same quantity. For great distances below the surface of the 

 earth, equation (1) can scarcely be considered as applicable, 

 seeing there is room for doubt whether the electromagnetic 

 force there can be supposed to be approximated by a magnet 

 situated in the axis of the earth. 



§5. 



In each horizontal layer of air of 1 metre thickness (Ar=l) 

 there exists, as explained above, an electromotive force of the 



