ISOLATION OF AN ION MILLIKAN. 239 



the mean time corresponding* to each group, is labeled V and placed 

 just below or just above the mean G corresponding to that group. 

 The volts were in this case read with a 10,000-volt Braun electrom- 

 eter which had been previously calibrated, but which may in these 

 readings be in error by as much as 1 per cent, though the error in 

 the relative values of the volts will be exceedingly slight. The PD 

 was applied by means of a storage battery. It will be seen from the 

 readings that the potential fell somewhat during the time of observa- 

 tion, the rate of fall being more rapid at first than it was later on. 



MULTIPLE RELATIONS SHOWN BY THE CHARGES ON A GI\T:N DROP. 



Since the original drop in this case was negative, it is evident 

 that a sudden increase in the speed due to the field — that is, a de- 

 crease in the time given in column F — means that the drop has caught 

 a negative ion from the air, while a decrease in the speed means that 

 it has caught a positive ion. 



If attention be directed, first, to the latter part of the table, where 

 the observations are most accurate, it will be seen that, beginning with 

 the group for which G=23.4'3, the time of the drop in the field 

 changed suddenly from 71 to 380 seconds, then back to 71, then down 

 to 39, then up again to 71, and then up again to 380. These numbers 

 show conclusively that the positive ion caught in the first change — 

 i. e., from 71 to 380 — carried exactly the same charge as the negative 

 ion caught in the change from 380 to 71. Or again, that the negative 

 ion caught in the change from 71 to 39 had exactly the same charge as 

 the positive ion caught in the change from 39 to 71. 



Furthermore, the exact value of the charge caught in each of the 

 above cases is obtained in terms of Trig from the difference in the 

 values of e„, given by equation (1), and if it be assumed that the 

 value of m is approximately known through Stokes's law, then the 

 approximately correct value of the charge on the captured ion is 

 given by the difference between the values of Cn obtained through 

 equation (4). The mean value of this difference obtained from all 

 the changes in the latter half of Table I (see Differences), is 

 4.93X10-". 



Now it will be seen from the first observation given in the table 

 that the charge which was originally upon this drop and which was 

 obtained, not from the ions in the air, but from the frictional process 

 involved in blowing the spray, was 34.47X10-^°- This number comes 

 within one-seventh of 1 per cent of being exactly seven times the 

 charge on the positive, or on the negative, ion caught in the obser- 

 vations under consideration. In the interval between December, 

 1909, and May, 1910, Mr. Harvey Fletcher and myself took observa- 

 tions in this way upon hundreds of drops which had initial charges 



