Relation between Uranium and Radium. 2,79 



that an indicator so excessively fragile as a gold-leaf can be 

 reproduced to give readings so nearly comparable. 



Such an electroscope as that described is properly charged 

 when the leaf forms an angle with the strip of between 30° 

 and 40°, and no one w T ith experience of electroscopes would 

 use the instrument with deflexions outside these limits. It 

 is therefore of interest to note that the values of the readings 

 are sensibly constant over this range, and so long as the leaf 

 is not obviously undercharged or overcharged the readings 

 may be depended upon. In our work we aimed at a charge 

 (about 400 volts) usually obtained from rubbed sealing-wax, 

 such that the deflexion was 35°, and the divergence or hori- 

 zontal distance between the tip of the leaf and the inside of 

 the brass strip was about 11 "5 mm. at the initial charge. 

 The potential, divergence, and sensitiveness of the leaf at 

 various deflexions is given in the following table : — 



Volts. 



Divergence. 



Sensitiveness 



645 



16*2 



•72 



540 



14-3 



•79 



443 



12-4 



•89 



385 



11-0 



1-00 



354 



10-0 



100 



289 



8-0- 



1-02 



260 



7-0 



1-05 



188 



4-9 



1-04 



The practical working range is represented by a bracket. 



For minute leaks where the " saturation current" is easily 

 attained, the table shows that it is better to work with a too 

 low than with a too high charge. It may be noted that the 

 divergence was about *8 mm. greater if the charging-rod 

 was near to the leaf-system. In the above table the charging- 

 rod was as remote from the leaf-system as possible. Of 

 course in all measurements the charging-rod and support of 

 the leaf-system were carefully connected to the case of the 

 instrument and earthed. 



As before (in the determinations of radium) those readings 

 of the instrument were selected which occurred in the interval 

 from 10 to 15 minutes after the introduction of the emanation. 

 Although the rate of leak increases rapidly both before and 

 after this period owing to the progressive disintegration of 

 the emanation, the leak during the interval is always constant 

 and constitutes a temporary maximum which is a far more 

 convenient point to select than the true maximum which is 

 reached after some hours. It is difficult to ensure that an 

 electroscope, which is frequently being dismantled and cleaned, 



