i;i;>i n;i.i> TO PRODUCE DlsciiAKcK THi;m <;n v.\!;l<r> OASES 



885 



In order to check this result a sample of hydrogen was used which had been 

 occluded l>y palladium foil. This gave results in close agreement with the above, the 

 minimum sj>ark potential being 308 volts. 



Diagram No. 2. Hydrogen. 



TOO 



600 



aoo 



'too 



iOO 



ISO 



to 



4O AO 



Pressure, mm. of Mercury. 



60 



7Q 



to 



Nitrogen. 



This gas is extremely troublesome to deal with. Although each sample tried 

 gave smooth curves for the relation between the spark potential and the pressure, yet 

 no agreement could be obtained between different samples of the gas, even though 

 they were prepared iu the same way and with the greatest care I was able to 

 bestow. 



Thus, for example, a specimen of nitrogen prepared from air by absorption of the 

 oxygen with alkaline pyrogallol gave for the minimum 347 volts. 



A sample from ammonium nitrite purified by passage through caustic potash and 

 sulphuric acid and over phosphoric anhydride gave 351 volts. 



Another sample prepared in the same way and dried with especial care, by being 

 allowed to stand in contact with phosphoric anhydride all night, gave 3G9 volts. 



A sample prepared by the removal of oxygen from air by means of metallic copper 

 gave 388 volts. 



Although these values for the minimum spark potential vary so widely, yet the rate 

 of change of the spark potential with the pressure, at pressures well above that 

 o>nes]>< uuling to the minimum. \\as approximately the same in all the samples. 



My rxperienre \\itli the nitrogen j s entirely in accord \\ith that of WAHIH'KC. He 



VOL. CXC1II. A. 3 D 



