﻿EQUIPMENT AND WORK OF AN AERO-PHYSICAL OBSERVATORY. 25 



These values were obtained, however, with a burning match-collector, 

 and we propose to apply a correction to them for that reason. Sir Wil- 

 liam Thomson, in a paper read before the meeting of the British Associa- 

 tion in 1889, gives observations made by McLean and Goto for that year, 

 showing " that an enclosed mass of air is electrified negatively by the 

 burning of a paraffin lamp, of coal gas, of sulphur, magnesium, and sev- 

 eral other substances, while, on the other hand, the burning of charcoal 

 electrified a room positively." In some experiments made by us in 

 1890 it was found that the flame in the dark room where the electrom- 

 eter was installed electrified the air of the room and materially affected 

 the readings. The electrification amounted to as much as 19 or 20 volts 

 negative, while an average value for the air outside (the nozzle of the 

 collector was about a metre from the wall and 12 metres from the ground) 

 was 50 volts positive. Ventilating the room thoroughly, we found, caused 

 a disappearance of this negative electrification. We would reduce the 

 value 6.8 volts found by Exner to something like 34 volts, following 

 Pellat's relative weights of 1,5, and 10 for match, water, and flame. 



Another set of observations made upon an exposed mountain side 

 gave Exner the following values : 



Metres. 



Volts. 



Metres. 



Volts. 



3 



110 



140-150 



210 



230-250 



280-405 



480 



18 



520-550 



5 



19 



20 



25 



550 



6 



660 



7 



820 



12 



30 



970 



14 











which would make the linear potential gradient of much higher value' 

 Valuable observations were made by the United States Signal Service 

 , under the supervision of Dr. Mendenhall. A resume of the observations 

 can be found in Third Memoir, vol. v, National Academy of Sciences. 

 The influence of varying temperature and humidity were in part elim- 

 nated by a long series of observations. The instruments used were modi- 

 fied Mascart electrometers and large similar water-droppers. The methods 

 and adjunct apparatus were alike at all stations. 



For the particular question which we are discussing — the potential 

 gradient — Ave are obliged to refer to a table not found in the published 

 report, although it was one of the most important. This table gives the 

 most extensive and comparable values yet obtained for determining the 

 potential gradient in the lower layers of the air. The values are some- 

 what smaller than might be anticipated, but this may be explained by 

 the proximity of buildings. 



