THE RATIO OF THE SPECIFIC HEATS OF GASES. 289 



Chlorine, bromine, iodine and iodine chloride, though 

 diatomic, have y near 1 '3, and hence eight degrees of 

 freedom. This is only one instance of what, as was said 

 above, holds generally. Two halogen atoms together in 

 one molecule invariably give a low value to y, and hence, 

 according to Boltzmann, an abnormally large number of 

 degrees of freedom. What this means it is impossible at 

 present to say. 



With regard to the polyatomic gases it is sufficient to 

 say that, with the exception of those with more than one 

 halogen atom in the molecule, nearly all have a value of y 



that makes /3 approximately equal to — n, where n is the 



•5 



number of atoms in the molecule. The relation is only 

 roughly approximate, the divergences being in many cases 

 too great to be accounted for by errors of observation. 

 Calculating the number of degrees of freedom from this 

 result we find that we have, in addition to the three 

 required by the motion of translation, approximately one 

 for each atom in the molecule. 



It is an interestinor task to devise modes of vibration 

 that satisfy this relation, but the possibilities with gases of 

 high atomicity are too numerous to give much scientific 

 value to the work. 



There is one very striking feature in the table on p. 

 279, and that is the absence of any gases with values of y 

 between 1*67 and 1*41. This is a strong argument in 

 favour of the dynamical theory of gases in general, and 

 Boltzmann's theorem in particular. The former value of y 

 gives q = 3, and the latter gives q = 5, both of which, as 

 we have seen, are quite intelligible ; but what would be 

 done with a gas whose y was 1*5, and whose molecule had 

 consequently four degrees of freedom ? Such a gas would 

 be very hard to reconcile with our present notions, and if 

 ever discovered would bring about a profound change in 

 the hypothesis of molecular physics. 



J. W. Capstick. 



