320 Prof. E. "Rutherford and Mr. H. Robinson on tlte 



suddenly increased so that the heating effect of the current 

 changed to 71 per cent, of the initial amount. 



In this way the variation of the deflexion of the galvano- 

 meter to be expected for the assumed distribution of the 

 heating effect was directly determined. The values obtained 

 are marked by crosses in the main curve fig. 3. It is seen 

 that the observations lie close to the curve throughout the 

 wbole range. The lag of the resistance-coil was slightly 

 greater than that due to the emanation-tube, and in con- 

 sequence the initial points lie slightly above the curve. The 

 agreement between the two curves shows clearly that the 

 assumed distribution of heating effect between the emanation 

 and its products is in close accord with direct experiment. 



It is seen from the curve fig. 3 that the heating effects due 

 to the emanation and radium A have practically vanished 

 after 20 minutes, and the remaining heating effect is due to 

 radium B and radium C together. 



A number of experiments were made to test whether 

 radium B provided an appreciable part of the heating effect 

 observed. For this purpose the decay of the heating effect 

 was carefully followed for about three hours after removal of 

 the emanation and the results compared with those to be 

 expected theoretically for any assumed distribution between 

 the heating effects of radium B and radium G. The decay 

 curve observed was found to agree closely with that to be 

 expected if all the heating effect arose from radium C alone. 

 The experiments were rendered difficult by the fact that a 

 small fraction of the emanation adhered to the walls of the 

 emanation-tube, and was gradually released during the time 

 of the experiment. The effect of this became appreciable 

 after two hours, when the heating effect due to radium C had 

 decayed to about 14 per cent, of its initial value. 



In addition, the method is not very sensitive, for the decay 

 curve over the region examined is not much affected even if 

 radium B provides 5 per cent, of the heating effect of 

 radium C. It was concluded from the observations that 

 radium B could not provide more than 5 per cent, of the 

 heating effect due to radium C; but for the reasons men- 

 tioned the actual percentage could not be deduced with any 

 confidence. 



Agreement of Experiment with (Calculation. 



The relative heating effects of the emanation, radium A 

 and radium C in equilibrium can be readily calculated if it 

 be assumed that the heating effect is a measure of the kinetic 

 energy of the expelled a particles. Since the expulsion of 



