Theory of Refraction in Gases. 



481 



(8) Experimental Verification, 

 We shall suppose that the density of the gas is given b^ 



P r. XoP 



pIpo= 



(H-a076 



■d 



1 + 



76(1 -faO 



v 



where P is the pressure in cms. o£ mercury, t is the tempe- 

 rature centigrade, a the ordinary coefficient o£ expansion, 

 and Xq ^ constant. The factor in brackets I shall call the 

 compressibility term. 



In my paper on refractive indices *_, to which I must refer 

 to avoid undue repetition, I reduced the observations by 

 supposing that the compressibility term did not vary with 

 temperature, being guided thereto by Mascart''s conclusion 

 that the temperature deviation was too great to be accounted 

 for by the variation of the compressibility term. My experi- 

 ments do not give such a large deviation as Mascart obtained, 

 and in the case of SO2 and NH3 I find that it can be exactly 

 accounted for in this way. 



Sidjoliur Dioxide. 



My experiments on the refractive index may be reduced 

 as in the following table. 



Temp. 



Katio. 



Pl+Vo- 



Eatio 

 -0319(;,,+^,)- 

 ^^ 76(1 + -00390 



Eatio 

 1 + 



X(l+-0003950 

 76(1 4- -00390 



81 



11-535 



137 



11-050 





14-585 



37-4 



13-305 



130 



12-700 





14-575 



14-9 



14-465 



126 



13-775 





14-585 



14-2 



14-490 



121 



13-825 





14-600 



14-2 



14-270 



81 



13-825 





14-600 



Bote. — Eatio means 

 displaced for 1cm. c 



the number of tands ivroor. 

 lifference of pressure. ^^^^^^ ' ' ' 



14-585 



The compressibility term was calculated from the last two 

 observations, v/hich give 



\o=-0319 per 76 cms. at 0° C. 



In order to get the value of //, the numbers in the last 

 column have to be multiplied by 76, by the wave-length 



* Phil. Trans, vol. cci. p. 435 (1903). 

 Fhil. Mag. S. 6. Vol. 6. No. 34. Oct. 1903. 



21 



