238 ANNUAL, REPORT SMITHSONIAN INSTITUTION, 1910. 



DIFFERENCES. 



24.60-19.66=4.94 

 29.62-24.60=5.02 

 34.47-29.62=4.85 

 39.38-34.47=4.91 



Meandif.=4.93 



by the constancy during all this time in the value of the velocity 

 under gravity. This constancy was not attained without a consider- 

 able amount of experimenting. It is sufficient here to state that the 

 heating effects of the illuminating arc were eliminated, first by filter- 

 ing the light through about 2 feet of water, and, second, by shutting 

 off the light from the arc altogether except at occasional instants, 

 when the shutter was opened to see that the star was in place or to 

 make an observation of the instant of its transit across a cross hair. 

 Further evidence of the complete stagnancy of the air is furnished 

 by the fact that for an hour or more at a time the drop would not 

 drift more than 2 or 3 millimeters to one side or the other of the 

 point at which it entered the field. 



The observations in Table I are far less accurate than many of 

 those which follow, the timing being done in this case with a stop 

 watch, while many of the later timings were taken with a chrono- 

 graph. Nevertheless this series is presented because of the unusual 

 length of time over which the drop was observed and because of the 

 rather unusual variety of phenomena which it presents. 



The column headed G shows the successive times in seconds taken 

 by the droplet to fall, under gravity, the distance between the cross 

 hairs. It will be seen that, in the course of the four and one-half 

 hours, the value of this time increases very slightly, thereby showing 

 that the drop is very slowly evaporating. Furthermore, there are 

 rather marked fluctuations recorded in the first 10 observations, 

 Avhich are probably due to the fact that, in this part of the observa- 

 tion, the shutter was open so much as to produce very slight convec- 

 tion currents. 



The column headed F is the time of ascent of the drop between 

 the cross hairs under the action of the field. The column headed 

 e n is the value of the charge carried by the drop as computed from 

 (4). The column headed n gives the number by which the values 

 of the preceding column must be divided to obtain the numbers 

 in the last column. The numbers in the e n column are in general 

 averages of all the observations of the table which are designated 

 by the same numeral in the n column. If a given observation is 

 not included in the average in the e n column, a blank appears oppo- 

 site that observation in the last two columns. On account of the slow 

 change in the value of G, the observations are arranged in groups 

 and the average value of G for each group is placed opposite that 

 group in the first column. The reading of the voltmeter, taken at 



