392 PELTIER ON THE CAUSES OF THE 
by vapours, but a contrary effect is produced in them; they are 
completely inactive under a clear and dry sky, and they give 
signs of electricity when the air is become moist, when the va- 
pours produced in the day-time begin to be condensed, when in 
short the air becomes a better conductor and facilitates the elec- 
tric radiation. Humidity of the air is for the fixed apparatus 
what the flame of Volta is for electroscopes—it is the means of 
losing the induced electricity accumulated at the extremities. 
These instruments would in fact be well adapted for ascertain-_ 
ing the quantity of water contained in the air, if the supports 
could remain perfectly insulated ; if, becoming moist, they did 
not in every part allow the escape of the electric quantities which 
the radiation leaves free. But why is not aqueous vapour indif- 
ferent as the air is? Why does it modify the induction of the 
globe? Why does it alter the energy of this inductive action 
even sometimes so far as to change the signs of our instruments ? 
These are the questions we are about to consider. 
32. Since the experiments of Volta, of Laplace, and of Lavoi- 
sier in 1781, and according to much later experiments, it”has 
been said that the electricity of vapours arose from chemical 
segregation; that the vapours of saline solutions carried off 
vitreous electricity and left to the liquids resinous electricity. 
We have repeated these experiments, we have analysed them, 
and we have already announced that their results are inappli- 
cable and contrary to the natural phenomena*. The experi- 
ment, it is well known, consists in making red-hot a capsule of 
metal (platina is preferable), into which is poured a few drops o 
a saline solution. During the time that the liquid remains in a 
globular form, the electroscope, whether provided or not with 
condensing plates, gives no sign of electricity, although the drop 
of water be reduced to half. The same thing happens when the 
drop of water suddenly moistens the capsule of platina and be- 
comes a mass of vapour. In one, as in the other case, there is 
no sign of electricity, notwithstanding the great quantity of 
vapour which rises. Between these two phases of the pheeno- 
menon, when the vessel has its temperature lowered between 
140 and 110 degrees, and when some small parts of the satu- 
rated drop begin to touch the metal, the vapour is there pro- 
* Comptes Rendus, Ac. Sc. 1841, t. 12, p. 307, and that of the 30th N : 
vember 1840. 
