236 ATMOSPHERIC ELECTRICITY. 



January 12th: Wind S.W., cloud above but not ou Table 

 Mountain ; gradient generally less than 100, but occasionally 

 higher. 



January 13th : Wind S., plenty of clouds, but high, so that 

 the sun was seldom obscured. Potential gradient at sea-level 

 500 volts per metre. 



January 15th: Cloud pouring over Table Mountain, S.E. 

 wind. Observations at Green Point: Potential varied from o to 

 100 volts per metre in the course of an hour. Generally the S.E 

 wind was accompanied by low potential gradients. 



January loth : Cloud over top of Lion Rock ; potential 

 gradient 90. On sun breaking through cloud gradient rose to 

 130. 



From the observations already recorded, it seemed likely that 

 the potential gradient would vary with the altitude, and a number 

 of observations were taken at different heights above sea-level. 

 A very marked difference in the gradient was seen. For 

 example, the electroscope stand was placed in a pool close to the 

 water ,when the potential gradient was found to be over 400. The 

 height of the collecting plate was 150 cm. above the water. On 

 plaicing the electroscope on a rock at a height of about 6 metres" 

 above the sea, the gnadient was reduced to about 300. Repeated 

 observations confirmed this very abnormal reduction, and it may 

 'be due to some difference in the conducting power of the r^ocks 

 and sea-water. The electroscope was then carried up the moun- 

 tains to a height of about 1,000 feet, and observations taken at 

 various points during the descent, the height being taken as in- 

 versely proportional to the pressure indicated by an aneroid 

 barometer. At the highest point the potential gradient was 56, 

 at sea-level it was 260. Many observations showed that the 

 higher the position the smaller the potential gradient, but a more 

 systematic and extended series of obsei"vations is necessary to 

 prove this, and of course the observations at the different points 

 should be conducted simultaneously, as the potential gradient 

 varied from day to day and from hour to hour. 



In April of this year a set of observations wa.s taken between 

 Van Reenen's Pass and Durban, with the object of confirming the 

 inference made at Cape Town, z'i::., that the potential gradient is 

 inversely proportional to the altitude. 



Van Reenen is at an altitude of about 6,000 feet. There is 

 a fairh' steady fall down to Ladysmith at 3,300 feet; from Lady- 

 smith to Durban there are considerable fluctuations as far as 

 Thornville, then a steady fall down to Durban. The observations 

 taken were, however, very incomplete, as the time available was 

 only that due to the stoppage of the train. But the general re- 

 sult was apparently the same ; the potential gradient increased 

 as the altitude diminished. 



During the first of the three days at Van Reenen, the sky 

 was clear up to 3.30. The potential gradient was low, not going 

 above 60 between 7 a.m. and 4 p.m. At 4.30 a violent thunder- 

 storm broke over the Pass, and after this subsided, a very strong 

 degree of positive electrification remained in the air. The poten- 



