82 
MR. C. T. R. WILSON: INVESTIGATIONS ON LIGHTNING DISCHARGES 
18h. 36m. ; the sphere was exposed. During the period covered by the portion of the 
record reproduced the sphere was momentarily lowered at 18h. and at 18h. 5m ; it was 
also lowered at 18h. 10m. and kept in its case till 18h. 11m. when it was again raised. 
The potential gradient at the times of lowering the sphere amounted to +90, +60 
and +40 volts per metre. It is plain from the record that the gradient remained 
positive throughout : the principal sudden changes of gradient were positive, and 
amounted to about 150 volts per metre ; two, however, at about 18h. lm. and 
18h. 9m. 30s. were negative and equal to about 60 volts per metre. Positive discharges 
evidently also occurred during both the short periods for which the sphere was lowered. 
The characteristic recovery of the field after both positive and negative discharges 
is well shown. The two peals of thunder recorded probably belong not to the 
discharges immediately preceding them but to the previous discharges. The 
discharges were probably at a distance of 20 km. or more. 
Fig. 10 (August 15, 14h. 18m. 20s. to 14h. 30m. 15s.). 
At 14h. 18m. 30s., when the cover was removed, the potential gradient was nega¬ 
tive ( = —3600 volts per metre). This negative potential gradient had increased to 
about —5000 at 14h. 19m. 6s. ; at this moment the negative field was nearly 
destroyed by the passage of a lightning flash at a distance of 4T km., the sudden 
change in potential gradient being +4800. The five subsequent flashes also produced 
positive changes in the potential gradient ; the beginning and ending of the thunder 
is marked on the record in each case. The magnitudes of the sudden changes of 
potential gradient vary from 3900 (the third shown in the fig.) to 14,600 volts per 
metre (the last); the distances of these two discharges were practically the same, 3 '7 
and 3’8 km. 
The striking feature of this record is the abnormal character of the curve of 
recovery of the field after the passage of every discharge except the last; instead of 
the rate of recovery of the field being most rapid immediately after the discharge, it 
is at first zero or very small and gradually increases to a maximum, falling off again 
with the increasing field as in the normal type. The last discharge shown in fig. 10 
as well as all the subsequent discharges of the record of which this is a part were 
followed by a recovery of the field of the normal type. The recovery curves following 
the discharges of the immediately preceding record of the same storm were also 
normal in character. 
Rain began about 14h. 20m., became heavy about 14h. 25m., and ceased about 
14h. 31m. 30s. 
The potential gradient was negative throughout the period covered by fig. 10 until 
reversed by the last discharge shown. At 14h. 28m., when the cover was 
momentarily replaced, the potential gradient was - 4800 volts per metre. The small 
hump in the curve at 14h. 24m. 45s. is due to the shielding effect of a horse and cart 
which passed within a few yards of the test-plate. 
In spite of the negative potential gradient, which would tend to produce an 
