692 Prof. 0. W. Richardson on the 



vacuum for 14 hours more and a test made at 797° C, which 

 gave the following values for the mean curve : — 



Pressure (mm.).....-... "002 '5123456 8 10 16 25 



Positive Emission . 

 (l=6-3xl0 _9 amp.). } 26 38 46 55 68 60 61*5 62 60 55 35 30 



This curve has a flat maximum at 6 mm. pressure and 

 is intermediate in character between the first curves and 

 those given by barium sulphate prepared chemically. 



Immediately after this experiment air was let in to atmo- 

 spheric pressure, when the emission dropped to 4 divisions on 

 the above scale of current. On reducing the pressure it rose 

 slowly to 15 5 divs. at 15 mm., but part of this may be due 

 to equilibrium not being fully attained at the lowest pressure. 

 On further evacuation the current became steady at 15 divs. 

 at 0*004 mm. A pressure curve was then taken which is 

 shown in fig. 5. The ascending points are shown thus ® , 

 and the descending points thus <•> . The dotted curve repre- 

 sents the mean : it is practically a straight line parallel to 

 the pressure axis. The temperature in this experiment was 

 783° C. 



Several points stand out quite definitely as a result of these 

 experiments with barite. After exposure to air at atmo- 

 spheric pressure the emission is comparatively small and 

 varies very little with the pressure from atmospheric pressure 

 doicn to 0003 mm. One experiment showed an increase and 

 the other a decrease on evacuation from atmospheric pressure, 

 but the change in both cases was small. When the power 

 of emission has been reduced in this way it is recovered again 

 by long continued heating at low pressure. 



The most natural interpretation of these results is that the 

 barium sulphate is decomposed by continued heating in a 

 vacuum and forms a substance which has a greater thermionic 

 activity. I am inclined to suspect that this substance is 

 barium sulphide as the tubes usually smelt of sulphuretted 

 hydrogen when they were taken down. If that is the case, 

 the effect of heating in air at atmospheric pressure w r ould be 

 the reconversion of the sulphide to sulphate. On this view, 

 the true emission from the sulphate is the smaller emission 

 which is obtained after exposing to air at atmospheric pressure 

 and which varies very little with the pressure of the gas in 

 the apparatus. 



The effect of Hydrogen. — If there is anything in the above 

 view a greatly increased emission should be obtained by 

 reducing the sulphate in an atmosphere of hydrogen. A 



