822 .Professor Sir James Dewar 



42 hours : 37 c.c. diffused in ; bright gas discharge. Helium : yellow 

 brilliant, green feeble. Neon : orange set quite distinct. Hydrogen : 

 very feeble F line only. 



72 hours : 63 c.c. condensed. Same appearance as shown at 42 hours 

 intensified, with F line of hydrogen quite plain. 



90 hours : 78 c.c. condensed. Helium, neon, hydrogen and nitrogen all 

 visible. Neon orange and yellow about equally intense with the 

 complete set of 5 helium lines ; F line subdued comparatively, so 

 also blue, green and violet nitrogen. Full red gas glow discharge 

 with still some phosphorescence. 



The day after -this the neon greatly intensified, while the helium 

 faded out in the blue and green. The hydrogen then disappeared ; 

 then the nitrogen intensified, and the remaining lines greatly 

 diminished. 



When, instead of platinum wire poles, B has external electrodes, 

 conveniently formed hy small deposits of silver top and bottom, then 

 the green line of helium is the first to show, and all the helium 

 spectrum develops before anything else. 



The sequence of the experiment was as follows :— 



Large double film of 98 sq. cm. membrane surface. Eate of diffusion 

 for air at 15° C, 1-15 c.c. per sq. cm. per day, or 4*7 c.c. per hour 

 through the whole surface. Weight of charcoal 1| grams. 



4 hours : 19 cc. diffused in ; faint shadow of the green helium line, with 

 much phosphorescent discharge. 



9 hours : 42 c.c. in. Helium green line well seen ; no other. 



11 ,, 52 ,, Same as in 9 hours, but more definite. 



22 J ,, 106 ,, All the five principal helium lines only. 



23J ,, 111 ,, Neon orange set visible in addition to the helium 



lines previously showing. No sign of hydrogen. 



The total gas extracted from the charcoal had 40 per cent of 

 oxygen in this case. 



Relative Rates of Helium and Xeox Diffusion. 



The three gases helium, neon and hydrogen are not of course 

 affected to the same extent by charcoal at liquid air temperature. 

 Helium is hardly absorbed at all ; neon is only very slightly taken 

 np, Avhereas the condensation of hydrogen is quite appreciable. 

 Therefore, in an equal mixture of the three gases over cooled char- 

 coal, the helium and most of the neon would be free, while the 

 pressure of the hydrogen would be sensibly reduced. Xow, the rela- 

 tive proportions of the three gases present in air may be taken as 

 20 of neon, 5 of helium, and from | to 2^ of hydrogen (London 

 air), expressed as parts per million volumes of an*. If, therefore, 

 they diffused through the indiarubber at the same rate, the neon 

 should be apparent in one quarter the time of the helium, while the 

 hydrogen would only appear in about twenty times the period 

 necessary for the neon. 



The second experiment just detailed showed that when the hehum 

 was visible spectroscopically in 18 hours, not until 42 hours could 



