160 
IOWA ACADEMY OF SCIENCE 
the electrodes between which the cloud was produced were charged. In all oh-, 
servations with a field the upper electrode was charged negatively and the lower 
positively. As the dp. for the expansions gradually increased the velocities of 
the clouds diminished, as is shown in fig. 2. At dps. from 22 cm. to 24 cm. a 
practically constant velocity was attained which was invariably between 5 and 
6 seconds for a distance of 2 millimeters. It was because of this constant 
velocity that all observations for the determination of e were taken at these 
differences of pressures. It is very probable that at these expansions practically 
all of the droplets carried but one ion. 
It might be interesting to note further that on several occasions while ex- 
panding at as high a difference of pressure as could be attained with apparatus, 
colloidal clouds were distinctly observed. These clouds consisted of very minute 
particles, as was evident from the fact that their falling velocities were very 
slow. In timing them it was found that it required 15 to 18 seconds for the 
surface of the cloud to fall through a distance of two millimeters. This indi- 
cates that the droplets of the colloidal nuclei are three to four times smaller 
than those of the fieeting nuclei. Barus’ pamphlet did not come to my hands 
until the close of work and I was much impressed with the coincidence of 
phenomena observed under such different conditions. 
Below is given a series of preliminary determinations of e taken during the 
first weeks of the work. The mechanism used for producing the expansions 
in the fog chamber and for throwing on the battery terminals to charge the 
field was crudely mechanical, so that the conditions of the successive observa- 
tions could only be duplicated in a rough way. In taking these observations 
Wilson’s method was followed. First, successive expansions were produced to 
get rid of the dust-like persistent nuclei. These were followed by two obser- 
vations with the radium in place, one with the field off and the other with 
the field on. 
Each determination represents the average of a series of eight to ten 
alternate readings taken in the manner described. The calculations were made 
by means of Wilson’s equation. The difference of potential between the elec- 
trodes of the fog chamber was in every case supplied by a storage battery of 
1600 volts. 
3.87X10-^° 
4.65 
3.82 
4.10 
4.20 
3.64 
4.27 
4.43 
3.73 
4.03 
4.55 
4.40 
S. Units 
Mean 4.14X10-^° E. S. Units 
The data which follow were obtained under decidedly improved conditions 
of the apparatus. The mechanical device for producing the expansions was 
discarded and electrical contrivances were substituted so that by the closing 
of a key the expansion in the fog chamber could be produced — instantly fol- 
lowed, when desired, by the connection of the poles of the storage battery with 
the electrodes. The method of observation were also slightly changed. Some 
were taken by what might be termed a group method. First, one or two ex- 
pansions were produced to get rid of the persistent nuclei. These were fol- 
