202 



MKSSKS. A. SCHUSTEE AND G. HEMSALECH ON 



the quantity x being larger a small error does not affect the result so largely as one 

 in the difference x, x,. 



TABLE I. Zinc, X = 48107 

 Date, Feb. 15, 1898. No. XV. v = 102. Distance between poles : 1 centim. 



X = 4924-8 (a). 





The above experiment was only considered to be of a preliminary nature, and the 

 photographs obtained were used chiefly to find the best method of reducing the 

 observations, but it gives a good idea of the general nature of the results. It is seen 

 that the velocity near the pole is a very uncertain quantity, owing to the sources of 

 error which will l)e discuased (p. 210), but the velocity gradually diminishes, and about 



