74 
MR. C. T. R. AVILSON: INVESTIGATIONS ON LIGHTNING DISCHARGES 
The observations discussed in this paper were made at the Solar Physics 
Observatory, Cambridge, mainly during the summer months of 1917. 
I. Methods of Measurement. 
The method and apparatus used in the measurements are substantially those 
described in a paper “ On Some Determinations of the Sign and Magnitude of Electric 
Discharges in Lightning Flashes.The induced charge on an exposed earthed 
conductor (test-plate or sphere) is used as a measure of the electric field. The test- 
plate virtually forms part of a flat portion of the earth’s surface, and the vertical 
electric force or potential gradient at ground level is equal (in electrostatic measure) 
to 4-rQ/A, where Q is the charge on its exposed surface and A is its area. The 
charge Q on the earth-connected sphere of radius It, when exposed at a height h> 
great compared with It, is a measure of the potential at that height; the zero 
potential of the sphere being the resultant of the undisturbed atmospheric potential V 
at the height k and of the potential Q/Pi, due to the charge on the sphere, so that 
Q/R = — V. The earthed conductors can be shielded from the earth’s field : the 
test-plate by means of an earth-connected cover, the sphere by lowering it into a 
conducting case resting on the ground. The quantity of electricity which flows to 
earth through the connecting wire on exposing or shielding the test-plate or sphere, is 
measured by a special type of capillary electrometer in which the readings indicate the 
total quantity of electricity which has traversed the instrument ; the sign and 
magnitude of the charge on the exposed conductor, and thus of the potential 
gradient, at the beginning and end of an exposure are thus determined. The sign 
and magnitude of sudden changes of potential gradient which occur while the 
conductor is exposed are indicated by the direction and magnitude of the resulting 
displacements of the electrometer meniscus. The total flow of electricity between 
the atmosphere and the test-plate or sphere during an exposure is also measured 
—being given by the difference between the electrometer readings before and 
after the exposure. The principal improvement introduced has been the provision 
of apparatus for giving a photographic trace of the electrometer readings ; rapid 
changes in the field occupying less than one-tenth of a second are in this way 
recorded. 
In the observations described in the previous paper the sphere was supported in a 
manner which did not admit of absolute measurements being made, as the charge 
measured included that on the upper part of the support as well as that on the 
sphere itself; in these earlier measurements therefore the sphere was standardised by 
comparison with the test-plate. The method of supporting the sphere is now such 
that the charge on the sphere alone is measured, while the disturbing effect of the 
* ‘Roy. Soc. Proc.,’ A, vol. 92, p. 555, 1916. 
