8 . DISTRIBUTION OF THE ACTIVE DEPOSIT OF 



plates, A. The guard plates were kept 3.0 cm. apart by wood- 

 en blocks C and D at top, and bottom. The latter were 

 supported on glass legs, E. The apparatus was set in a 

 shallow glass dish and was covered by a rect- 

 angular glass jar. Control of the atmospheric conditions o^ 

 the exposure was obtained by sealing the bell jar with a layer 

 of liquid in the shallow glass dish. To secure dry air, con- 

 centrated sulphuric acid was the liquid used. In experi- 

 ments on the effect of water vapor, water replaced the acid. 

 For other gases a mercurj- seal was used to keep the desired 

 vapor in the vessel. The thorium hj'-droxide was placed in a 

 shallow wooden box G, under the plates and supported by 

 the glass legs E. Connections with the plates were made by 

 wires passing through glass tubes, F F, bent to pass under the 

 bell jar through the liquid. 



A number of advantages can be claimed for this type of 

 vessel in studying the distribution of the active deposits. 



The field is practically uniform over the plates A, any 

 distortion of the field at the edges being eliminated bj^ the 

 guard plate. 



No insulating material is near the electrodes. 



Equal areas of electrodes are presented to the active 

 deposit. 



The activities of the two plates are measured under iden- 

 tically the same conditions, in the most simple manner, by a 

 gold leaf electroscope. 



Good control of gaseous conditions can be secured. 



A series of observations was made with applied potentials 

 varying from 6 to 12,000 volts. The highest voltage was 

 obtained by the use of a Wimshurst machine run by an electric 

 motor. The length of the spark gap was .35 cm. and the 

 diameters of the knobs 3.0 cm. and 1.5 cm., giving an estimated 

 voltage of 12,000. The air was dried by standing over con- 

 centrated sulphuric acid. The results obtained are shown in 

 Table 1 and plotted in curve A, Fig. 4. 



