Electricity by Evaporation. 215 
III. Saturated Solution of Sodium Chloride. 
At the 
beginning. 
After After 
5 minutes. 10 minutes. 
(a) Room tem- 
perature 
(6) 100° 0. ... 
1 scale-dir. 

— 10 scale-div. —1-5 seale-div. 
-1-5 „ -1-5 „ 
Nearly all the deflections are less than those from the elec- 
trical action already described as peculiar to B. The two 
exceptions exceed the limit of B's action by an amount so 
small that they need not be considered. The difference of 
average values in the Tables I. A and I. B for the experiments 
lasting 15 minutes equals — - 74millim. scale-division. This 
corresponds to a negative charge upon B of 0*016 Daniell 
(0*0145 volt). This charge is too small in proportion to the 
sea-water evaporated, to be used as a basis for mathematical 
calculations concerning the electricity resident in the clouds*. 
Nor is it a sufficient ground for the assertion that the simple 
change of a liquid into a vapour produces electricity. 
(b) A second method of experiment, in which the insulation 
does not come at all into consideration, was as follows : — A 
brass disk of 125 millim. diameter was fastened horizontally 
by its centre to the end of a stout wire which was connected, 
without the condenser, directly with the electrometer-needle. 
Both needle and disk C could be together either insulated or 
connected with the earth. The porcelain dish containing the 
liquid and resting in the insulated sand-bath was brought 
close under the disk 0, and the liquid connected with the 
earth. The sand-bath having been heated and the flame em- 
ployed afterwards extinguished, the vapour from the liquid 
was condensed very copiously upon the disk C. The reading 
of the electrometer-needle connected with the earth was first 
taken. Then the needle, and consequently the condensing- 
disk C, was insulated, and any deflection was observed. After- 
wards the electrometer-needle was connected with the earth 
and its reading again taken. Thus a change in the needle's 
position of rest during the experiment, which always occurs 
with variations of room-temperature, was observed. 
* Such calculations have been carried out by Freeman, Phil. Mag. 
[5] vol. xiii. p. 898 (June 1882). 
