Ill 



were obtained: ill <>n closing circuit, deflection (W) was first 38, then 

 settled at 22, on standing at •_'•_' several seconds, radium placed nearest 

 II gave deflection (E) to 344. On removal deflection was W. 



Since an E deflection indicates a decrease in the resistance of II 

 fciie first results secured seemed fairly definite. To make sure the move- 

 ments were not due to the E. M. F. of the electrolyte, weights were 

 placed on tlie keys by which the battery and galvanometer circuits were 

 both kept closed for a considerable time until the needle had ceased to 

 drift. Four additional readings were taken, the five sets of readings 

 being as follows: in all the lists of readings deflections indicating a 

 decrease of resistance by the presence of the radium are marked +. 

 those indicating an increase are marked — : 



Table I. 



Reading. 



Reading on 



addition of 



radium. 



Reading on 



removal of 



radium. 



Deflection. 



Two drops of ITS 0. t were added to the water with which the tubes 

 were now filled. This of course greatly increased the conductivity. It 

 also made it much more difficult to balance the bridge. In securing the 

 data given in table II the radium was placed alternately upon the two 

 tubes, N and S. 



Table II. 



Reading at 

 beginning. 



Radium on N 



Result. 



Radium on S. 



Result. 



