142 Mr. Satyendra Kay on the 



The value of A could be found from the coefficient of Self- 

 Indue lion L, which by definition is the value of \Bd& when 

 i = l. We have the equation 



for finding A in terms of L or vice versa. But L for a circular 

 conductor of one term is given by Fleming for a wire of 

 diameter d in the form 



L = 4™ [log 8 ]™ -2-2} 



The formula makes L infinitely great when d approaches 

 zero, as in our investigation, so that the formula fails us. 

 Of course the value of d to choose is the diameter of the 

 Electron. If d were not involved we could determine A at 

 least experimentally by finding L. If A = V the diameter oE 

 the electron would be of the order l0~ 4 ' 3xl0U cm. 



The idea of Equivalence may help us to form an idea of 

 the order and magnitude of A as follows. 



A circular row of electrons of total charge e moving round 

 with a velocity v in a medium of permeability j& gives us an 

 equivalent magnetic shell on the face of which the pole m is 

 equal to A/juve. The pole density is infinite at the rim, so 

 that we shall imagine this pole to be distributed along the 

 circle itself. 



The error in making the equivalent magnetic ring coincide 

 with the electric circuit is very small as the true radius is 



1 /iy 3 /1.3\ 2 5 '/1.3.5v 2 7 , f 



= a 



r 



Jo 



r a ~"' I /l\ 2 3 /I.3. 2 5 /1.3.5\ 2 7 p 



J o „H . Shrr . dr g + ( 3 ) -J + (j^) ■ g + jjXg) ■ g + *°| 



The numerator in the right-hand expression is less than the 

 denominator but the difference is less than J, so that if A is 

 large the radius of the equivalent ring is equal to a for all 

 practical purposes. 



The magnetic analogue similarly means that a circular 

 line distribution of pole m, moving -with a velocity v in a 

 direction opposite to above, in a medium of inductivity K, is 

 equivalent to a dielectric ring at rest, coinciding with^tlie 

 magnetic current, the charge on its two faces being AKimi. 

 A reciprocal relation exists, a current of one kind being- 



equal to a doublet of the other kind coincident with it. 



To explain a doublet does not, as we know, require postu- 

 lation of two kinds of charges, electric or magnetic, positive 



