678 Prof. 0. W. Richardson on the 



The order from left to right is the order of the observations 

 in time. The slightly higher values with diminishing than 

 with increasing pressures is a spurious effect, caused by the 

 temperature having undergone a slight increase temporarily 

 while the pressure was being reduced. Subsequent experi- 

 ments, made at a higher temperature (1160° C), in such a 

 way that the temperature was kept constant by a continuous 

 regulation of the controlling resistances, as the pressure was 

 altered, showed that there was a slight difference in the 

 position of the ascending and descending curves, but in the 

 other direction. The curve for ascending pressure indicated 

 rather bigger values of the emission at a given pressure than 

 the descending curve. This experiment was carried out with 

 a specimen of the salt which had been heated for a long time. 

 Continued heating only seems to remove the pressure of 

 maximum emission when the test is made at a low tempera- 

 ture. A specimen which had been heated continuously in a 

 vacuum for four days still gave a quite definite maximum in 

 the neighbourhood of 0*16 mm. both on the ascending and 

 descending pressure curves at 1160° C. The maximum 

 pressure was somewhat higher for the ascending than for 

 the descending curve; in all probability this only means that 

 the true equilibrium was not completely reached when the 

 observations were taken. 



The maximum in the pressure curves does not appear to 

 be due to lack of saturation of the thermionic current. I 

 have examined this point most carefully by taking a separate 

 current-E.M.F. curve for each point on the pressure-current 

 curve, It occurs with voltages which are too small for 

 ionization by collisions to be a factor, and, in addition, the 

 shift of the maximum with temperature is against this 

 explanation. 



The following numbers are typical of the results obtained 

 with a specimen of sodium sulphate at 11.60° 0. which has 

 been heated for a long time at that temperature : 



Pressure increasing. 

 Pressure (mm.) ... '01 '084 'IS '18 -26 '31 "46 116 1-84 



Current 67 71 77 82 84 85 76 71 47 



(l = l-25xl0~ 8 amp.) 



In this experiment the heating current was continuously 

 regulated so that practically no fluctuation occurred in the 

 temperature whilst the pressure was being varied. 



Experiments with water vapour.— Immediately after the 



