AND ON THE ELECTRIC FIELD OF THUNDERSTORMS. 
83 
ionization current from the ground to the atmosphere, the total charge received by 
the test-plate from the atmosphere between 14h. 18m. 30s. and 14h. 28m. has 
been positive and equal to the charge which a potential gradient of 17,000 volts per 
metre would have induced on the earth-connected plate. The charge carried by the 
rain must thus have been positive and must have exceeded to the above extent the 
negative charge carried by the ionization current. The greater part of this charged 
rain has evidently fallen between 14h. 26m. and 14h. 28m. 
Fig. 11 (June 12, 1917, 16h. 38m. 40s. to 16h. 50m.). 
This is a portion of the second record taken on a sultry afternoon with towering 
cumulus in all directions. The first record ran from I5h. 55m. to 16b. 19m. A cap 
was seen to form on the summit of a large cumulus cloud in the E.N.E. at 15h. 59m., 
and another on one of the lower heads of the same cloud now in the N.E., about 
16h. 16m. The potential gradient throughout this first record was positive and about 
50 volts per metre. No thunder was heard and no sudden changes of the field are 
shown on the record. 
The second record, of which fig. 11 is a portion enlarged, extended from 16h. 27m. 
to 17h. 4m. The large cumulus cloud was N. by E. with its edge at an elevation of 
about 60 degrees at 16h. 30m. and due N. about 16h. 50m. The potential gradient 
diminished from +44 volts per metre at 16h. 30m. to 15 at 16h. 35m. and then 
became negative, being —29 at 16h. 40m., —175 at 16h. 45m. (all the above being 
occasions of momentary lowering of the sphere), and —106 volts per metre at 
16h. 49m. when the sphere was finally lowered. The field was zero at 17h. 14m., 
and had become positive at 17h. 16m., when the observations ceased. 
All the sudden changes of field observed were positive; two of 18 and 14 volts per 
metre, both due to flashes at a distance of about 7 km., occurred before the field had 
become negative. The other two (equal to 120 and 320 volts per metre respectively) 
are shown in the figure ; they both show characteristic recovery curves ; in both cases 
the potential gradient was reversed, in the first by a discharge at a distance of 8'2, 
in the second by one at a distance of about 7 km. 
The discharge at 16h. 45m. 50s. is an interesting one. The negative potential 
gradient had reached a steady value—about 170 volts per metre—before the passage 
of the discharge. The discharge—at a distance of about 7 km.—caused the gradient 
to become positive ( = 150 volts per metre); the negative field was again re-established, 
practically exponentially, a steady value being finally reached equal to about 
105 volts per metre. The sphere was brought down at 16h. 49m. No thunder was 
heard after this time. 
In Plate 5 are enlargements of small portions of some of the traces, showing 
details in the changes of potential gradient associated with lightning discharges. 
In each case the time in seconds is shown, reckoned from the moment at which the 
discharge, as indicated by the record, began. 
n 2 
