by a Charged Condenser moving through Space. 



159 



the galvanometer in the main and subsidiary experiment 

 respectively. 



In view of some uncertainty as to the effective resistance 

 of the commutator when rotating *, the precaution was taken 

 of placing in series with the rotating coil in the subsidiary 

 experiment, such a resistance as to make the resistance of 

 the circuit containing the shunted galvanometer and the 

 rotating coil the same as the resistance of the circuit com- 

 posed of the galvanometer and the two rotating coils in the 

 main experiment. Thus the quantities R x and R 2 , which 

 contain the resistance of the commutator, cancelled out. 



The value of 8 2 was obtained by reversing the current 

 through the galvanometer, and taking half the deflexion 



observed. 



L 



The accompanying table shows the values of — obtained 



in three successive experiments; the constancy of the numbers 

 in the last column gives an idea of the degree of consistency 

 of the readings. 



Speed revs., 

 per inin. 



Deflexion on 

 reversal. 



5 2 



n 2 



735 

 800 

 800 



230 

 253 



248 



0-156 

 0-158 

 0-155 



Mean -* =0-156 

 n 9 



ant 



The time of swing of the galvanometer was 11*0 seconds, 



1 the value of ^ was 280, so that in view of 



the fact that W = 0*44, & = 2'3, a = 31*3, l = S cms., wc find 

 for the deflexion to be expected in the main experiment, for 

 a case where y = 18 miles per second (2*9 x 10 G cm./sec) f 

 V7+V 2 = 1200 volts, %=800, and ^=11 seconds, 



5, = 10-2. 



* The commutator was kept well wetted with paraffin oil while 

 running-, to keep its effective resistance constant. 



t The fact that the resultant velocity of the aether drift happened to 

 coincide with the velocity (18 miles per second) of the earth's orbital 

 motion, is a mere coincidence ; the directions were of course not the 

 same. 



