MEASUREMENT OF DISCHARGES. 531 



The time of oscillation of the bar was 30 seconds in the experi- 

 ments of Weber and Zollner, and 55 to 65 in those of Wiedemann ; 

 the swing was determined by the recoil method, or by the mixed 

 method. 



1115. F. Kohlrausch* has introduced a modification into the 

 experiment which had been pointed out by Weber. Instead of 

 taking as multiplier a frame in which the value of g might be 

 directly calculated, he uses an ordinary galvanometer of great sen- 

 sitiveness, with astatic needles, and even with correcting magnets, 

 and calculates the value of g by the difference of dampings, 

 according as the circuit is open or closed. 



Neglecting the coefficient of self-induction of the system, and 

 the small difference between the times of oscillation of the magnet 

 in the two cases, we have sensibly 



2RK = r TT 



2 _ 2 RKA-A 

 M 2 T 



and the expression of the resistance becomes 



S 2 H 2 A-A T 

 ~K **~~^' 



Besides determining the surface S of the inductor, and the 

 time of oscillation, both the moment of inertia of the needle, and 

 the horizontal component of the terrestrial magnetism, must be 

 determined. 



As the square of this latter quantity comes into the final formula, 

 the relative error due to its determination is doubled; and this is 

 also the case with that which is due to an inclination of the axis 

 in the plane of the meridian. 



It is simpler to determine the constant g of the galvanometer, 

 by comparison with an absolute galvanometer.! If, further, the 

 inducing frame is itself transformed into a tangent galvanometer, 

 the ratio of the intensities of the two fields is at once eliminated. 



* F. KOHLRAUSCH. Ann. Pogg. Ergdnzungs Band vi., p. i. 1874. 



! MASCART, DE NERVILLE, and BENO!T. Resume d' Experiences sur la 

 Determination de I' Ohm. 1884. Ann. de Chim. et de Phys. [6], Vol. vi., 

 p. 5- 1885. 



M M 2 



