Active Deposit of Radium in Electric Fields. 551 



decay o£ the matter itself. The second product radium B 

 which results from this transformation is also positively 

 charged, so that any radium B which is produced in the 

 space is attracted back to the cathode by the electric held. 

 In the case of actinium A the decay is so rapid that none o£ 

 the actinium A ever reaches the anode by the process of 

 partition, and the same probably holds in the case of thorium 

 A. Such a partition accounts for the fact that the distribu- 

 tion is independent of the quantity of emanation used, and 

 also causes the anode and cathode activities to be a function 

 of the dimensions of the particular apparatus employed. 

 Further discussion of the consequences of this hypothesis at 

 the present stage, however, seems rather premature. 



In conclusion, it might be pointed out that the pheno- 

 menon of the distribution of active deposits in electric fields 

 is merely a problem in the theory of ionization. The deposit 

 atom behaves apparentlv like any other positive ion until it 

 loses its charge. In consequence, the activity-voltage curves 

 for the active deposit are identical with the ionization 

 curves, but may be modified owing to the superposition of 

 effects which are physical characteristics of the substances 

 employed. If the distribution is determined in various 

 gases, the activity-voltage curves should change with the 

 ionization curves. In the particular cases of vapours great 

 modifications might be introduced, especially if these were 

 electrified or near their condensation points. 



Summary, 



(i.) The distribution of the active deposit of radium 

 between the electrodes in an electric field is independent of 

 the concentration of the emanation for small quantities but 

 depends upon the dimensions of the apparatus. 



(ii.) No negatively charged active deposit exists. Badium A 

 is initially positively charged and is then subjected to re- 

 combination like any other positive ion. Such neutralized 

 active matter then diffuses to the metal surfaces irrespective 

 of their potentials. 



(iii.) From a comparison with the distribution in the case 

 of actinium, it has been shown that the anode activity con- 

 sists of two parts; first the active matter neutralized by 

 recombination, and secondly a constant activity due to un- 

 charged deposit which is independent of the field and only 

 varies with the dimensions of the apparatus. 



(iv.) The constant anode activity is not due to absorbed 

 emanation or to a surface condensation of emanation on the 



