292 ILLINOIS STATE ACADEMY OF SCIENCE 



means of a liquid air trap. The pressure was then varied 

 in the apparatus and simultaneous readings were taken on 

 a standard McLeod gauge and the Knudsen gauge. A cali- 

 bration curve was drawn on a large scale from the data thus 

 obtained. This is made possible since the deflection at zero 

 pressure is zero, and the origin thus became an accurate 

 point of the curve. Mercury vapor of course still existed 

 in the McLeod gauge ; this, however, had no effect since this 

 instrument does not read small vapor pressures. The 

 Knudsen gauge consists of a platinum strip rigidly 

 mounted within the bulb of the gauge and it is heated by 

 means of an electric current. A light vane is freely sus- 

 pended in front of the platinum strip. The bombardment 

 of the molecules thrown off from the hot foil causes the 

 vane to turn. The instrument may be used for pressures 

 where the mean free path of the molecules is greater than 

 the distance between the vane and foil. Since this con- 

 dition obtains only when the vacuum is high, it follows 

 that the Knudsen gauge is therefore adapted for the 

 measurement of exceedingly high vacua only. It has been 

 used in measuring pressures as low as 10-8 mm. of mer- 

 cury. 



Having calibrated the gauge the mercury sample was in- 

 troduced into the container, B, and purified by repeated 

 distillations in the vacuum. Later on one sample was also 

 treated with nitric acid before being used. The tube was 

 then heated to a temperature of 250° to 300° C for several 

 hours, while the sample of mercury was protected by warm 

 water in a vacuum flask placed on the container. A Lang- 

 muir pump supported by a Gaede rotary outfit kept the 

 vacuum at the highest point during the period of heating. 

 All vapors were thus driven from the glass and a residual 

 air pressure as low as .00002 mm. was obtained. The tube 

 was then sealed off at E. The whole system was rigidly 

 fastened within a box in order to control the temperature, 

 and the box in turn was placed on a pier to prevent jarring. 

 A fan properly placed kept the temperature uniform. Low 

 temperatures were obtained by cooling the room, while 

 heating coils placed within the box were employed in main- 

 taining higher temperatures. 



