Disappearance of Gas in the Electric Discharge. 243 



the walls were doubtless saturated wiih C0 2 . If the vessel 

 were baked to remove the G0 2 , and then filled to a pressure 

 less than 1 mm. and the discharge passed, higher values of 

 oc could be obtained at the lowest pressures, although they 

 always fell off on approaching zero more rapidly than would 

 be expected from the rest of the curve. 



12. In the light of these experiments a new vessel (No. 4) 

 was prepared. It was similar to vessel 2, except that the 

 cylindrical anode was replaced by a silver coating on the 

 walls and that a tube 5 cm. long and 1*5 cm. in diameter, 

 dipping into liquid air, projected from the bottom. In this 

 vessel the values of a did not fall off until a pressure of 

 0*03 mm. was reached. 



The relation between i g Ji e for this vessel was measured 

 with i e about '005 m.a., this current being so small that re- 

 combination was inappreciable and no change in the ratio 

 occurred with further reduction of i e . The results are shown 

 in fig. 6. The curve with its sharp maximum (in the neigh- 

 bourhood of wdiich measurements were irregular) is interest- 

 ing. The rapid fall after the maximum cannot be due mainly 

 to recombination, for, as has been stated, a further reduction 

 of the current did not affect the values obtained in this 

 region ; it is perhaps due to the fact that the voltage-drop 

 along the free path of an electron is becoming comparable 

 with the ionizing potential. But this is a matter requiring 

 further investigation. . 



13. Table I. summarizes the results of measurements of a 

 in this vessel. The first three columns give V, the mean value 

 of i e , and the range of pressure ; the fourth gives the value 

 of a ; the fifth the value corrected for recombination, where 

 data are available, as explained in par. 10. 



Table I. 



V (volts). i e (m.a.) 

 400 -05 



•19 

 100 -006 



•006 

 •001 

 400 -003 



006 

 •011 

 •018 



», -oio 



•008 

 •012 



p (mm.) 



a 



a (corrected) 



085- -022 



2-3 



2 - 2 



124- -030 



24 5 



1-9 



108-075 



2-1 



24 



125- -074 



1% 



19, 



108--075 



2-1 



24 



•232--175 



1-9 



1-9 



•317 — -215 



2-4 



? 



•169- -114 



2*2 



20 



•152- -080 



2-6 



2-3 



•075- -051 



2-0 



20 



•177--078 



2-3 -, 



.>.._> 



•254- -190 



2-4 



? 



