﻿58 . Prof. J. H. Jeans on the 



radiant energy must be of the form 



x-^/cvryx, (i) 



and that the total energy of radiation per unit volume 

 must be of the form 



0T 4 (2) 



This method, it is seen, arises essentially from the con- 

 templation of a conservative dynamical system. 



2. The second method consists in regarding the radiator 

 and the aether surrounding it as a non-conservative system. 

 This method is not concerned with the equilibrium between 

 matter and aether, but with the rate of transfer of energy 

 from the former to the latter. Assuming the source of 

 radiation to be the motion of electric charges in the radiator, 

 it can be shown by this second method also that the radiation 

 must be given by formulae of the type (1) and f 2) *. 



3. In the opinion of the present writer neither method has 

 so far led to a valid determination of the form of the 

 function /of formula (1). The two theories can, however, 

 be differentiated by their treatment of the constant, a of 

 formula (2). 



4. In the second theory the quantity a enters as a physical 

 entity, dependent on the magnitude of the electric charges 

 only (as soon as the temperature scale has been determined). 

 From the observed value of a it is found that the electric 

 charges producing the radiation must, if the theory is valid, 

 be of the order of magnitude of 1*8 x 10" 10 C.G.S. electro- 

 static units f — a circumstance which points very clearly to 

 electrons as the source of radiation. 



5. In the first theory, on the other hand, the quantity ar 

 enters, mathematically as a constant of integration, and the 

 theory itself does not supply the means of evaluating it. Since, 

 however, crT 4 is to be the amount of energy per unit volume, 

 the physical dimensions of a are known. 



The thermodynamic^ argument by which, in this first 

 theory, the formula o-T 4 is reached, is concerned only with 

 phenonema taking place in the aether. Thus we should 

 naturally expect that it would be possible to evaluate a in 

 terms of quantities which measure the properties of the aether. 

 The properties of the aether, as contemplated by the argument 

 which leads to the formula <rT 4 , are, however, all contained 

 in the one constant which measures the velocity of wave- 

 propagation, and obviously a quantity of the physical dimen- 

 sions appropriate for cr cannot be constructed from this alone. 

 Even if we attribute a further property, e.g. material density 



* u On the Laws of Radiation," Roy. Soc. Froc. vol. 70 A. p. 545. 

 t L. c. ante, § 9. 



