98 Bumstead and Wheeler — Radio-active Gas. 



of a gold-leaf electroscope of C. T. R. Wilson's pattern, in 

 which the leaf is supported by means of a sulphur bead from 

 a rod kept charged to a potential at least as high as that of the 

 leaf : so that any leak of electricity from the leaf must be 

 through the gas and not over the support. A number of con- 

 trol experiments showed that the observed activity was not 

 due to the vessels or drying tubes used, nor to the city supply 

 pipes through which the water reached the laboratory ; and 

 the fact that water, once boiled, did not recover to any appre- 

 ciable extent its power to give off a radio-active gas when left 

 to stand, either stoppered or unstoppered, for two weeks 

 showed that the presence of the gas was not due to contact 

 with the air, nor to a dissolved, or suspended, radio-active 

 solid. The latter conclusion was strengthened by our inability 

 to find any evidence of radio-activity in the solid residue left 

 on evaporating the water. As rain water does not contain a 

 radio-active gas (although it does contain an active solid resi- 

 due), the only hypothesis which seemed to be left to account 

 for the presence of the gas in this surface water was that the 

 water had come in contact with the gas in its passage through 

 the ground and had dissolved some of it. Accordingly a piece 

 of gas-pipe was driven about 1*5 meters into the earth and its 

 top connected by a rubber tube to a flask filled with water ; 

 on allowing the water to run out, the flask became filled with 

 the gas from under ground which was then introduced into 

 the electroscope and tested as before. It proved to be much 

 more active than the gas from water (three or four times for 

 equal volumes) and as it was very easily obtainable in as large 

 quantities as one wished, it was much more convenient to 

 work with. 



In order to compare it with the water-gas the rate of decay 

 of the activity of both gases was observed. For this purpose 

 an electroscope was made as nearly air-tight as possible by 

 the use of solder, sealing-wax and asphaltum varnish, and by 

 regrinding the brass cocks with which it was provided ; after 

 some difficulty it was made so tight that, upon being partially 

 exhausted, it would lose less than one-half per cent of its 

 exhaustion in 24 hours. When either gas was introduced into 

 this electroscope, an initial rise of activity occurred lasting for 

 about O'l day, followed by a regular decay according to an 

 exponential law in which the activity fell to one-half its value 

 in about four days. The curves for the water-gas and for the 

 earth-gas were indistinguishable when reduced to the same 

 scale ; one of them is given in ^. 2. During the initial rise, 

 observations of the leak were made as rapidly as possible, and 

 afterward once or twice a day for about two weeks. As the 

 sensitiveness of the gold leaf varied slightly from day to day, 



