458 



Messrs. Rowland and Hutchinson on the 



The last column gives the values of V. This work is not 

 intended as a determination of" Y, but the calculation is made 

 merely to show to what degree of approximation the effect 

 follows the assumed law. 



The deflexions are about the same as those obtained in the 

 Berlin experiments — 5 to 8 millim. on reversing. The force 

 measured then was 1/50000 H ; now it is 1/125000 H. The 

 sensitiveness of the needle in the two cases w^as almost the 

 same. In the former experiment a force of 3 X 10" ^ deflected 

 the needle 1' of arc ; the corresponding number now is 

 2*7 X 10~7, slightly more sensitive. The scale distances were 

 110 and 200 centim. respectively. So this experiment gives 

 about the same scale-deflexion at twice the distance with a force 

 1*0 ^^ great. The agreement between the two is seen to 

 be quite good. 



The observations, except Nos. 1, 2, 15, and 18 given above, 

 were taken in pairs — first one direction of rotation and the 

 other immediately afterwards, everything except the rotation 

 being kept constant. 



The table shows that, in every case except one, the de- 

 flexion for negative rotation is appreciably smaller than the 

 corresponding positive. 



The difference is too great to be due to accidental errors 

 in the readings, as the following table, giving the successive 

 deflexions in the case of #13 and #14 will show. 



Table II. 



#13. 



#14. 



mm. 





67 



5-1 



51 



4-9 



6-6 



39 



76 



5-2 



8-0 



5-0 



5-8 



5-2 



6-3 



4-9 



8-0 



5-0 



80 



50 



4-3 



4-4 



5-9 



6-6 



60 



50 



6-5 



50 



There is but one deflexion in #13 as small as the mean of 

 #14; and but one in #14 as large as the mean of #13. 



