AND ON THE ELECTRIC FIELD OF THUNDERSTORMS. 
95 
There can be little doubt that it is by the agency of precipitation that the 
separation of positive and negative charges in a thunder-cloud and consequent 
production of an electric field is effected, the larger raindrops or hailstones carrying 
down a charge of one sign while the charge of opposite sign is attached to small 
drops or cloud particles carried up in the ascending air stream. It is not proposed to 
discuss here how the large and small particles may acquire charges of opposite sign : 
whether for example the thunder-cloud is essentially a frictional electrical machine 
(disruption of drops, which Simpson* regards as the important factor, being included 
under this head) or an influence machine as Elster and GEiTELf contend. 
It is obvious that any view that places the seat of electro-motive force of a thunder¬ 
storm within the thunder-cloud implies that the cloud is essentially bipolar, equal and 
opposite charges being in any given time transferred from within the cloud to its 
upper and lower portions. The actual charges residing at any moment in the positive 
and negative portions of the cloud will in general be quite unequal, since the 
conditions determining the rates of dissipation of the charges at the top and bottom 
of the cloud will be very different; an important part of the loss of charge from the 
lower part of the cloud is obviously the charge carried down to the ground in rain¬ 
drops. The lower charge may indeed to a large extent reside on rain-drops falling 
from the cloud, and may thus extend all the way to the ground. Rain may not 
however reach the ground, or may lose a large part or the whole of its charge before 
reaching it by processes to be considered later. 
Consider a cloud in which there is an upward stream of charged cloud particles or 
small drops and a downward stream of oppositely charged large drops ; the total vertical 
electric current is the sum of the currents carried by the upward and downward 
streams. If the density of electrification of the two streams were the same and 
uniform throughout the greater part of the vertical thickness of the cloud, then the 
whole of this portion of the cloud would be electrically neutral. Above a certain 
level however the small drops alone will remain, and again it is only the large drops 
which fall below the lower margin of the cloud ; equal and opposite charges will in 
this way be liberated in the upper and lower portions of the cloud. The assumption 
of uniform density of electrification in the two streams is of course an extreme and 
improbable one, and the concentration of the charges in the upper and lower parts 
of the cloud alone is not likely to be so complete as this supposition would imply ; 
it serves, however, to indicate the possibility of the positive and negative charges of 
a cloud being separated by a considerable vertical thickness of electrically neutral 
cloud. 
The factors which determine the rates of dissipation of the upper and lower 
charges and the magnitudes of the maximum charges are considered in a later 
* Simpson, loc . cit . 
t Elster and Geitel, ‘ Wied. Ann.,’ 25, p. 116, 1885; ‘ Physikal. Zeitschr.,’ 14, p. 1287, 1913; 
Geitel, ‘Physikal. Zeitsch.,’ 17, p. 455, 1916. 
