ELECTRIC PHZNOMENA OF THE ATMOSPHERE. 403 
that it charges itself more powerfully with resinous electricity ; 
that each transformation leaves behind a quantity which the 
electricity repelled by the superior induction had received in 
deposit, and that the ascending portions have a tension which 
increases with the number of the transformations, which tension 
they retain on account of their perfect insulation, and the elec- 
tric energy of which augments to a degree that nothing here 
below can reproduce, too much surrounded with moist air and 
with inducing and conducting terrestrial bodies. 
47. There are days very favourable to make these observa- 
tions, it is when small cwmuli, sufficiently thin to allow their in- 
ternal motions to be distinguished, are spread over the sky. 
On examining attentively what takes place, the following is ob- 
served. Independently of the motion of translation of the 
entire mass, each part of the cloud changes its position with 
respect to the other parts, whilst the evaporation is taking place : 
these motions are more extensive, as the evaporation proceeds 
more rapidly; but they are not the same throughout the mass. 
Towards the edge which receives the direct rays of the sun, the 
evaporation being there greater, the last opake vapours become 
strongly vitreous, and they are seen to descend towards the 
earth, to pass even below the mass of the cloud, and to keep 
itself there ; whilst on the opposite side the vapour extends and 
disperses itself until all is transformed into elastic vapour, but 
without that great agitation and strong repulsion from above to 
below. When, by a particular disposition in the upper part of 
the cloud, the solar rays are received and readily absorbed, 
and there thence results a prompt evaporation, a vaporous 
eruption is seen to proceed from below the cumulus and to ex- 
tend itself before the mass. These last portions are not vapor- 
- ized so quick as the first, and if we compare the rapidity with 
which the cwmulus at first diminished, with that of the last thin 
and semi-transparent portions, we soon recognise the whole in- 
fluence of the vitreous electricity of the latter, which far from 
assisting the temperature to vaporize these flakes, opposes it by 
the repulsion of similar tensions. 
48. We may imitate one of these transformations, and judge 
by the result obtained of the powerful tension that the insulated 
clouds may acquire, when the vapours which constitute them 
have undergone several such changes. A watch glass a (fig. 3.), or 
a glass capsule, is pierced at the bottom by a small hole through 
