106 
MR. C. T. R. WILSON: INVESTIGATIONS ON LIGHTNING DISCHARGES 
electrically polarised (the induced charges of the original conducting filament 
remaining when the filament freezes), and will thus tend to remain with their long 
axes parallel to the direction of the electric force. A study of the optical phenomena 
of “ false cirrus ” would be of interest in this connexion, as would also an experimental 
investigation of the effects of an electric field on super-cooled drops. 
XV. Pressure Within a Charged Portion of a Cloud. 
The pressure within a charged cloud—like that within a charged soap bubble—must 
be less than the pressure outside. If the whole charge be supposed to lie near the 
surface of a sphere the analogy with the soap bubble is complete, and this case may 
be considered in finding the order of magnitude of the effect. The reduction of 
pressure within the cloud by the charge is 27rP = F 2 /8 7 r where a- is the charge per 
unit area of the surface of the sphere and F is the radial electric force immediately 
outside. Just before the passage of a discharge F = F 0 = about 100 in electrostatic 
measure, so that F 2 / 8 ?r is about 400 dynes per square centimetre, i.e., about ^sVo °f an 
atmosphere. 
If we consider the charge to be distributed uniformly in a horizontal layer of 
thickness which is small compared with its horizontal dimensions, the diminution of 
pressure midway between the top and bottom of the charged layer, due to mutual 
repulsion of the charged drops, is again F 0 2 / 87 r dynes per square centimetre. 
XVI. Thunder Resulting f rom Sudden Contraction due to Loss of Charge. 
Thunder is generally regarded as entirely due to the sudden expansion of the air 
along the track of a lightning flash. It is evident however that the sudden 
contraction of a large volume of air (the contraction corresponding to an increase of 
pressure of some tenths of a millimetre of mercury) must furnish a by no means 
negligible contribution to the thunder which follows the discharge. 
XVII. Energy Dissipated in Lightning Discharges. 
If we take the estimates arrived at above (V = 10 9 volts, Q = 20 coulombs) for 
the order of magnitude of the potential in the charged portions of a thunder-cloud 
immediately before the passage of a flash, and of -the quantity discharged in the flash, 
we obtain for the order of magnitude of the energy dissipated in an average discharge, 
|QV = 10 10 joules = 10 17 ergs. 
We may also arrive at an upper limit for the energy if we assume that the 
distribution of the charges is stratiform and that the vertical electric force is 
uniform and equal to F 0 throughout the height H through which the discharge 
extends. From the value found for the average electric moment, 2 QH, since V must 
