Mr. B. Stewart on the Radiation and Absorption of Gases. 219 



and consequently d, e, f disappear also, And even were this 

 not the case, in a perfectly symmetrical arrangement these latter 

 quantities would be made up of terms which would mutually de- 

 stroy each other. Circular polarization therefore presupposes a 

 want of symmetry, which in the calculation is the more general, 

 in nature the rarer case. 



The conclusion to which the present investigation leads is, that 

 simply by the further development of the formal side of the laws of 

 light , that is, the laws of refraction and reflexion by isotropic trans- 

 parent bodies, without doubtful hypotheses, and in fact by keep^ 

 ing to just the most general form, it is possible to arrive at a com- 

 plete theory of double refraction, of chromatic dispersion, and of 

 circular polarization. 



A step further may still be made in the direction we have here 

 entered upon, and I will briefly point out how. The velocity of 

 light is regarded as a function of x, y, and z ; it may, however, 

 be taken still more generally as a function of the time /, for it is 

 plainly a limitation to suppose the particles of the body origi- 

 nally at rest. This can be easily introduced into the calculation 

 by giving to p p the value 



k p t + a P x + b p y + CpZ + dp 



Up 



Since, however, the differential equations (A) were not formed 

 on this supposition, they cannot in this case be taken as a safe 

 basis for the calculation, and the more generally valid differen- 

 tial equations would have to be deduced in another way. 



XXX. On the Radiation and Absorption of Gases. 

 By Balfour Stewart, M.A., F.R.S* 



AS the relation of radiant heat to gases and vapours is at 

 present engaging the attention of two very eminent expe- 

 rimentalists, and is also a subject of much importance in meteo- 

 rology, the attempt, however imperfect, to discuss it from a 

 different point of view may not be without interest. 



It would appear to be established that the spectra of gases pre- 

 serve to a great extent the same character or type through a 

 considerable range of temperature. We know that this is the 

 case with the vapour of sodium, which consistently absorbs the 

 double line D ; and we know also that the characteristic line of 

 thallium is always prominent in its spectrum, although, when 

 the temperature becomes very high, other but weaker lines are 

 added. Coupling this with the proposition which asserts that 

 the absorptive power of a body for any kind of heat is propor- 



* Communicated by the Author. 



