206 ATMOSPHERIC PHENOMENA. 
During whole days it appears null, and again, for 
some hours, the pith-balls of the electrometer diverge 
from three to four lines. The atmosphere, which 
in the torrid as in the temperate zone is generally 
in a state of positive electricity, passes alternately, 
in the course of eight or ten minutes, to the nega- 
tive state. The rainy season is that of thunder- 
storms ; and yet I have found, from numerous ex- 
periments made during three years, that at this sea- 
son the electric tension is less in the lower regions 
of the atmosphere. Are thunder-storms the effect 
of this unequal change of the different superimposed 
strata of the air? What prevents the electricity 
from descending towards the earth in a stratum of 
air which has become more humid since the month 
of March? At this period the electricity, in place 
of being diffused through the whole atmosphere, 
would seem to be accumulated on the outer envelope 
at the surface of the clouds. According to M. Gay 
Lussac, it is the formation of the cloud itseif that 
carries the fluid toward the surface. The storm 
rises in the plains two hours after the sun passes 
through the meridian, and therefore shortly after 
the period of the maximum of the diurnal heat in 
the tropics. In the inland districts it is exceedingly 
rare to hear thunder at night or in the morning, 
nocturnal] thunder-storms being peculiar to certain 
valleys of rivers which have a particular climate.” 
It may be interesting to present a very brief state- 
mentof Humboldt’s explanation of these phenomena: 
— The season of rains and thunder in the northern 
equinoctial zone coincides with the passage of the sun 
through the zenith of the place, the cessation of the 
breezes or north-east winds, and the frequency of 
