150 Prof. A. LI. Hughes on Characteristic 



could be measured, it certainly being less than '000001 mm. 

 After the first few hours' initial heating, liquid air was 

 placed around the trap between the pump and apparatus.. 

 By the time the tempera! ure had been reduced to room- 

 temperature (the pumps working continuously), one may be 

 confident that ihe vacuum was very good. (Outgassing at a 

 series of gradually decreasing temperature is said to be 

 particularly effective*.) The whole procedure usually 

 occupied two days. Liquid air was placed around the char- 

 coal tube in apparatus II. two or three hours before 

 observations were made. 



In apparatus II. a wide-mesh gauze P was introduced 

 between Q and Z, to increase the steadiness of the electro- 

 meter needle. Slight fluctuations in the cells giving ihe 

 15 volts had much less inductive effect on Z and the asso- 

 ciated electrometer than when Z was faced directly by Q at 

 600 volts, when the same relative fluctuations in the cells 

 would produce an eflect forty times as big. 



Experimental Results. 



Boron f . — A set of observations obtained with boron on 

 one occasion (using apparatus I.) is shown in fig. 2. Hori- 

 zontally we have the voltage applied between T and F, 

 measuring the energy of the impinging electrons, and 

 vertically we have the photo-electric current from Z per 

 unit electron current between T and F. It will be seen that 

 a^ the observations were repeated (a run with decreasing 

 potentials following one with increasing potentials and vice 

 versa), there was a steady increase in the magnitude of the 

 effect. To eliminate the time factor to some extent, the 

 mean of the four runs was found and plotted, as shown in 

 the upper part of the figure. It will be seen that there is 

 an abrupt change in the slope of the curve at 150 volts. 

 On attempting to repeat this some days later, it was found 

 that a practically continuous line, with only a slight break 

 such as that shown in fig. 3, was obtained. Many attempts 

 were made to find out the conditions which gave rise to the 

 type of curve shown in fig. 2. The effects were tried of 

 subjecting the boron to an intense electronic bombardment, 

 of taking it out and grinding a new surface, and of using a 

 newly-fractured unground surface. Only on one other 

 occasion was a set of curves similar to that shown in fig. 2 



* Dushman, Gen. Elec. Rev. xxiv. p. 251 (1921). 

 t The author is indebted to Dr. W. R. Whitney, of the General 

 Electric Company, for the boron. 



