132 PROCEEDINGS OF THE AMERICAN ACADEMY 



of the needle, and about as quickly as they could be seen from 

 without. 



On June 18, at 2 h. 30 m., p.m., these experiments were repeated. 

 As before, no difference could be detected between the potential of the 

 air a few feet from the ground, and that of the ground. It is to be 

 remembered, however, that iu the present adjustment of the electrom- 

 eter, a difference equivalent to two volts was the minimum difference 

 that could be detected. With the same instrument, adjusted for 

 greater sensitiveness, employing the same methods, I have found in 

 another locality a difference equivalent to a volt and more between a 

 point in air 10 feet from the ground and the ground itself. The sky 

 had been perfectly cloudless for many hours, and was now without 

 clouds, except one or two very small rounded cumuli in the east. 

 These also in the course of an hour disappeared, and the sky was 

 again cloudless. The height of the kite was determined experiment- 

 ally, by sending up a carrier on the kite-string, to which was attached 

 thread, with markers at certain distances, made of folded wrapping 

 paper, and of just sufficient weight to keep the thread perpendicular. 

 The distance between the last marker and a level line of sight from 

 the summit was estimated, and added to the known length of thread. 

 My own carrier device having failed, Mr. Willard Gerrish, of the 

 Observatory, suggested a conical-shaped carrier, which answered well. 

 When the kite was about 350 feet high, the needle was deflected off 

 the scale, but did not press against the side of the case of the instru- 

 ment. When the kite was about 200 feet high, the deflections were 

 in the neighborhood of 10, and variable in character. At this time, 

 then, it was easily possible to notice and record the fluctuations in the 

 value of the potential of the air. At times the needle would start 

 suddenly and swing off the scale ; sometimes remaining off, sometimes 

 immediately returning. At other times it would fall to 5, 7, or 8, 

 and either remain at these figures for a while, or vary greatly 

 therefrom. 



At 4 h. 30 m., with a cloudless sky, the kite being at a height of 

 300 feet, the deflection was 23-|-. The kite was then pulled down 

 until only about half as high as before, and the average deflection was 

 15-|-. At 4 h. 50 m., the kite being about 400 feet high, the length of 

 kite-string being about 700 feet, and the distance between the kite 

 and the ground beneath it about 1,000 feet, there being a deep glen 

 between this and the next hill, the deflection was off the scale, and 

 apparently much greater than when at 300 feet. The sparks obtained 

 by presenting the ground-wire to the kite-wire were larger than before, 



