56 Louis Schwendler — On the General [No. 2, 



r" 

 Dividing by q", and putting -?, = v" we have 



A' = V V "^ V ' 



f" v" \/b" + {b" + d" +f" s/a" 



This expression has a maximum* for 

 b" =f° + d" 

 which contradicts the regularity condition f = b + d so long as d is different 

 from zero. 



Thus, in order to fulfil the regularity condition, and the maximum 

 current, for the differential method simultaneously, we must put up 



d = o 



It has, however, been shewn that in order to have immediate balance, 

 when adjusting balance by a variation in the resistances, we have to 

 alter the resistances of the four branches b + d, a + h,f, and w -\- j3 simul- 

 taneously according to a relation already given. Thus it is proved that 

 adjustment of balance by an alteration of the resistances must be rejected, 

 since, as pointed out before, a variation of the resistances of the coil b is 

 impracticable. 



We are obliged, therefore, to adjust balance by moving the coils or 

 their armatures, and the further solution of the problem is only required, 

 when this mode of adjustment is adopted. 



Maximum magnetic moment. 



It has now been proved that d is to be made zero, in order to be able 

 to fulfil the conditions of regularity and maximum current simultaneously ; 

 and that therefore, to obtain immediate balance, readjustment of balance is 

 to be effected by a movement of the two coils a and b or their armatures, 

 and not, as has been generally proposed, by an alteration of the resistance 

 in the branches (a -j. h) and (b + d), 



Hence 7i appearing in the denominator of P only, and h > o not being 

 any more required for adjusting balance^ the best value we can give to h 

 is : — 



h = o 

 which will make P, obviously largest. f 



* In order to keep the balance in Station II rigid when b" varies we must suppose 

 v" simultaneously variable with b". This is perfectly justified, for v" can he altered by 

 an appropriate movement of the coils to keep up the balance in Station 11^ without al- 

 tering the outgoing current A". 



f The resistances d and h, without exerting magnetic force, were originally intro- 

 duced in order to investigate the possibility of adjusting balance by an alteration of the 

 resistances in the branches. But since it has been shown that this mode of adjustment 

 is to be rejected it is of course clear that the dead resistances in these branches shoidd 

 be made zero when P will become largest. 



