7 lieori/ of Refraction in Gases. 4:83 

 Accordino- to the theory we deduce 



,_ -001377? , -OIU'.) ? If (S-iog, / ,;\^ 



^ 7i;(i+-oo3yi!)'' 7(5(i+-ob39<)) I "^ d V a,vr 



AVe can now determine ^^ for the Xa hne at 0° C. from 



the measurements of dispersion. 

 Ketteler gives 



'^^l^^ =1-0064 from Li red to Na. 



o 



Hence if p refer to the Na Hne, -t, =1*29 7. 

 If -^ is hircre we have 



;-ic 



P'- =1-0064. 



1-6-10^' 



l-e-io'^^'x 1-297 



^\ 

 Therefore ,. .^ w^ ^^^ 



b-10 — „ =-021, 



^'^ - ■ =-0034 for the Na line at 0° 0. 



The supposition that -^ is small is justified. We must 



— 2 



note that this estimate of —^ cannot be considered accurate 



to more than 10 per cent. We may now apply a test by cal- 

 culating the theoretical value of jjl for the Na line at 76 cms. 

 and 0° C. from Baedecker's measurements. We obtain 

 ^>,^^ = 1-000695. 



From my measurements on refraction we deduced 

 ^^^^= 1-000674. 



I think the agreement must be considered very satisfac- 

 tory, especially when we remember the uncertainty of some 

 of the measurements, and the fact that the value of K — 1 is 

 about 7 times the value o^ fju^—1 in the visible spectrum. 



Ammonia Gas, 



In reducing my observations on the refractive index I have 

 used Mascart's value for the compressibility term, viz. 



j -OlBo^P I 



I ^76(1 + -00380 J ' 

 which agrees well with the deviation from Boyle's law. 



212 



