S. H. Freeman— Electrification by Evaporation. 431 
In the above observations the rate of evaporation was not 
measure 
The earlier observations on aa and salt water gave defle ec- 
tions about 0:1 that given a Daniell’s cell, the deflections 
with the dry dish being uty the same (see table I). Though 
the dish was insulated for five minutes or more, the position ‘ot 
the spot of light was read every 30 seconds. Far the greater 
part of the deflection was attained in the first minute or two, 
and after five minutes scarcely any further deflection could be 
discovered. 
The observations thus indicated that the small deflections 
obtained were mainly due to leakage from electrified parts of 
the apparatus. The sources of this leakage were carefully in- 
vestigated and three principal ones were discovered, viz: the 
charged needle of the electrometer, the vulcanite Alnstn support- 
ing the electrodes of the elect oinevek and the glass cup of the 
commutator. In the later ex perime Sie care was taken to elim- 
inate these sources of leakage as far as possible. The connec- 
tions were so made, that the electrodes and commutator were 
not touched for days before the experiments, and the charge 
on the needle was not replenished within about 24 hours of the 
ex periment. hen these precautions were taken, the deflec- 
tions became less than 0°01. and generally less than 0-005 of a 
Daniell’s a (see table IT). 
n contrast with these deflections, Tait and yoy ob- 
tained by dashing liquids on red hot platinum, 
Solution CuSO, a potential +300 times that of a Daniell’s cell. 
Solution NaCl a potential +120 f - “ 
eRe ome d water a potential : 24 i“ ‘i 
hol 3 ae “ te 
ure the deflections obtained n almost eve 
as pt set state ceased, ee nol of light et beyond th the 
limits of the scale. 
To show the pani gt of the apparatus to very slight fric- 
tion it was so arranged that the drops of liquid could fall u upon 
the evaporating dish from an insulated ea: in metallic 
connection with it. A few drops of liquid, falling 3 or 
were sufficient to give deflections of nearly the size obaihad 
from a Daniell’s cell and much larger than any obtained in 
evaporation. 
In order to compare the results obtained with the phe- 
nomena of atmospheric electricity the writer computed the 
virtual depth of water which would be necessary to produce 
one flash of lightning, if the vapor in a thunder-cloud had 
been electrified as indicated by the experiments. By virtual 
