Diffusion of Neon through Hot Quartz, 705 



a faint " blue argon "spectrum with the yellow helium line. 

 After about three hours 1 heating the helium spectrum was 

 fully developed, and the strong neon line at 5852 was well 

 marked. The tube was then left hot all night. Next morning 

 the helium and neon spectra were both quite strong, all the 

 prominent neon lines being easily recognized. We could 

 not be certain of any increase in the intensity of the blue 

 argon spectrum which was present, either in this or in other 

 similar experiments. This is important since it proves that 

 the helium and neon had not leaked into the apparatus through 

 the sealing-wax joints which connected to the two quartz 

 tubes. In that case the argon spectrum should have increased 

 proportionately with the others. This possibility is also 

 excluded by the fact that the helium and neon present in the 

 air cannot be detected spectroscopically in presence of the 

 very large excess of argon which accompanies them. We 

 have tested this question several times very carefully by 

 examining the spectrum of the inert gases of the atmosphere 

 under conditions similar to those which prevailed during the 

 other experiments. We were never able to detect even the 

 strongest helium and neon lines. The faint argon spectrum 

 which was observed arose in all probability from the trace 

 of air absorbed by the walls of the tubes before the heating 

 commenced. 



It is clear from these experiments that neon diffuses through 

 quartz at about 1000° C, but its coefficient of diffusion is 

 smaller than that of helium. This follows since the helium 

 is visible first, although there is about ten times as much 

 neon as helium present in the atmosphere. Of course helium is 

 more sensitive to the spectroscopic test than neon, but it does 

 not seem likely that this will account for the facts, as the strong 

 neon line appears at about the same time as the weak helium 

 lines, in spite of the disparity in the amounts of the two 

 gases present in the atmosphere. Arguing from the general 

 similarity in the properties of the inert gases, it would seem 

 likely that argon also would diffuse through hot quartz. If 

 it does, the coefficient of diffusion must be very small, otherwise 

 accumulation of argon would certainly have been observed in 

 these experiments. This question is by no means easy to test 

 experimentally, as there is always the possibility that a minute 

 accumulation of this gas might arise from defective air- 

 tightness of the apparatus. 



It seems likely that an easy method of measuring the 

 amount of helium and neon in the atmosphere might be bnsed 

 on this property of diffusion through heated quartz. By 

 attaching a quartz tube heated to a constant temperature to 



