ROSA, DORSEY AND MILLER: INTERNATIONAL AMPERE 213 



ments. The limitations of simplicity of form are, however, imposed 



upon a second apparatus with which the instrument itself is compared. 



The method (c) allows the use of multiple layer coils and the resulting 



high sensibility. The ratios which enter into the constanl of the instru- 

 ment may be determined experimentally by methods which are compara- 

 tive and allow an extreme precision. The constant can be redetermined 



at any time. 



The Rayleigh type of current balance as used in this work embodies 

 the advantages of the methods (b) and (c). It consists essentially of 

 three multiple-layer coils of square cross-section; two large fixed coils 

 and a smaller moving coil. All of the coils are coaxial and their planes 

 are horizontal. In this work three pairs of fixed coils (50, 50 and 40 cm. 

 in diameter) and three moving coils (25, 25 and 20 cm.) were used. The 

 moving coil was suspended midway between the other two from one pan 

 of a precision balance. The current was led in and out of the moving 

 coil at the balance pan by means of fifty silver wires. These wires were 

 slack and, on account of their fineness, did not interfere sensibly with 

 the oscillations of the balance. 



The force exerted upon the moving coil by the current in the fixed 

 coils was weighed by means of the balance, as follows : The strength and 

 the direction of the current in the coils were arranged so that the moving 

 coil was pulled dowm by a force of about 3 g. This force and the weight 

 of the moving coil were counter-balanced. Then the rest point a of 

 the balance was determined. The current was then reversed in the 

 fixed coils alone. The force then became a pull up instead of down, 

 giving a change of about 6 g. A weight of 6 g. was put on one pan of 

 the balance from which the moving coil was suspended, and very nearly 

 compensated for the change in the force. Another rest point b was 

 taken, the difference between a and b gave the difference between 6 g. 

 and twice the electrical force upon the moving coil. The drift of the 

 balance was eliminated by taking a series of the rest points a and b 

 and plotting these against the time. 



The determination of the force in absolute units required the knowl- 

 edge of the absolute value of the acceleration of gravity. The value 

 at the Bureau of Standards was determined in terms of the value at 

 the United States Coast and Geodetic Survey by Mr. Wm. II Bei 

 of the Survey. Mr. G. R. Putnam of the Survey has connected the 

 latter value with Potsdam by relative pendulum observations. Thus, 

 the value at the Bureau of Standards has been referred to Potsdam, 

 where, as the result of a long series of observations, the absolute value is 



