Water Radioactivity, 567 



indirect. \i a flask of tap-water is closed up air-tiglit, and 

 allowed to stand for several days, it is found that the rate of 

 decay of its radioactivity is considerably less than that of the 

 (:jas ilriven out of it ; in tact, on one or two occasions the 

 radioactivity has actually increased on standing. This can 

 be explained only by assuming that there is a continuous 

 production of a radioactive emanation in the water. Further, 

 tap-water is found to partially recover its radioactivity on 

 standing after it has been thoroughly boiled, but never to its 

 original value. In general, the recovery has been to about 

 one-tenth of its original value. 



Induced Radioactivity. 



Just as radium emanation possesses the power of inducing 

 radioactivity on bodies immersed in it, so the radioactive gas 

 from water possesses the same property. To this is due the 

 fact that when the gas is introduced into the electroscope 

 the rate of leak increases for a time before it begins to 

 decrease. The rate of decay of this induced radioactivity is 

 the same when produced by the gas from tap-water as when 

 produced by radium emanation. That is, it decays according 

 to an exponential law, such that it falls to half its value in 

 about 35 minutes. 



This induced radioactivity is greater on a negatively 

 charged conductor than on a positively charged one. When 

 a conductor is not charged, there appears to be no induced 

 radioactivity. When charged positively or negatively, the 

 amount of induced radioactivity is at first proportional to 

 the absolute value of the potential, and approaches a limiting 

 value with high potentials. The rate of decay is the same 

 for the induced radioactivity, whether produced by positive 

 or negative electrification. 



Experiments on induced radioactivity were made with air 

 that bad been circulated through water several times by 

 passing through a water-pump. In this way the conduc- 

 tivity of the air in a large tank could be increased to about 

 80 times its normal value. Metal rods were suspended in 

 this tank, and these could be charged to any desired potential. 

 They were taken out after a definite time and suspended in 

 another tank filled with ordinary air, and the current through 

 this second tank, when the rod was connected to a quadrant 

 electrometer, and the tank to a battery, compared with a 

 similar rod which had not been made radioactive. In every 

 case the voltage applied to the testing-tank was sufficient to 

 give the saturation current. After each test the brass rod 



2 P2 



