220 Dr. L. J. Blake on the electrical Neutrality of 
employing the Kohlrausch condenser, the condensing-disk C 
was directly connected with the electrometer-needle, and with 
this needle was put in metallic connexion with the earth. 
After the porcelain dish containing the electrified liquid was 
swung under C, the needle, and therefore C, were insulated. 
Only that electricity which was fixed by the closely lying 
electrified surface of the liquid could now be present upon C. 
While the liquid was under C, a deflection of the needle would 
show any convection of electricity from the surface of the 
liquid, and also any change of potential of that electrified 
surface. The following table gives the deflections, in scale- 
divisions, observed while the sea-water was 3 minutes under 
the disk C. As above, 70 scale-divisions correspond to a 
charge of 1 Daniell. 
Positive charge, 480 elements. 
Negative charge, 480 elements. 
Sea-water at 
room-temp. 
Sea-water 
at 100° C. 
Sea-water at Sea-water 
room-temp. at 100° C. 
+ 6 
+ 16 
+ 12 
+25 
+ 1 < - 9 
-15 -5 
- 9 
These numbers are opposed to convection of electricity by the 
vapour; otherwise no discrepancies in the signs of the deflec- 
tions would occur. If convection by the vapour had occurred, 
a greater deflection would have appeared with a liquid rapidly 
evaporating than when cool. But in two cases the contrary 
takes place. When one considers that the rising vapour had 
condensed in large drops upon C, and that between C and the 
closely adjacent liquid a strong electrical attraction must have 
existed, the slight difference between the deflections for hot 
and cold sea-water loses all value as evidence of a convection 
of electricity by means of vapour. 
The particles of dust which lie upon the surface of the 
liquid, or float in the surrounding air, are doubtless the bearers 
of the electricity upon C; for when the surface of the liquid 
was intentionally strewn with dust-particles, the charge upon 
the electrometer-needle was increased, while it was diminished 
when the surface was carefully cleaned. In another place the 
influence of dust-particles will be more clearly seen. 
In regard to the experiments of Buff, already mentioned, I 
quote the following passage * : — " A glass flask, which con- 
* Liebig'8 Annalen, lxxxis. p. 203, 1854. 
