Bumstead and Wheeler — Radio-active Gas. 101 



on the connecting rod ; by means of the little electromagnet, 

 M, fastened to the lid of the box, JN" could be lifted off the rod 

 and the quadrants could thus be insulated without the slightest 

 jar. The short-circuiting contact was made of the same piece 

 of brass on both sides, was small in area, and the distance 

 moved was short ; there was thus no trouble from contact 

 potential differences. A small hole through the brass box, 

 opposite the binding screw of the electrometer, enabled one to 

 test the sensitiveness of the electrometer from time to time. 

 When the gaskets were well greased with stiff mutton-tallow, 

 the cylinder retained a pressure in excess of that of the atmos- 

 phere without perceptible diminution in twenty-four hours. 



The gas to be tested was put into the cylinder and readings 

 were made as rapidly as possible during the rise of the induced 

 activity. After the ionization had reached a steady value, the 

 gas was allowed to stand in the cylinder for several days, and 

 readings were taken from time to time to determine the rate 

 of decay of the activity of the gas; as this change was slow, 

 the average of four readings was taken at each observation. 

 Finally the lower plate was removed and the gas allowed to 

 diffuse through the porous plate, readings being taken as rap- 

 idly as possible in the meanwhile. From all these readings the 

 leak due to the so-called spontaneous ionization of the air in 

 the cylinder was always subtracted. The potential applied to 

 the cylinder (90 volts) was more than enough to give the sat- 

 uration current in all cases. The radium emanation used was 

 obtained by bubbling air through a solution containing 0*1 gram 

 of the very impure radium bromide prepared a year or two 

 ago by De Haen ; its activity is about 1000 times that of ura- 

 nium, so that the solution contained probably less than 0*1 

 milligram of radium.. Three small bubbles drawn through 

 this into the cylinder (perhaps a fifth of a cubic centimeter) 

 caused an ionization about six times as great as 1^- liters of the 

 earth-gas ; the chief difficulty with the emanation was to get a 

 small enough quantity into the cylinder. 



Comparison of the Rates of Decay. 



If we assume the exponential law of decay of the radio- 

 activity I=I e~ Kt , the constant k may be determined from any 

 two observations (or two sets of four observations) taken at 

 different times after the excited activity has ceased to rise. 

 Several such determinations were made for the earth-gas and 

 for the radium emanation and are given in the following 

 tables ; t in the formula is measured in hours. 



It will be seen that the agreement is well within the limits of 

 accuracy of the experiments. This constant for the radium 



