118 Prof. C. .Gr. Knott on Swans Prism Photometer. 



principle that solids are regarded as rigid, strings as inex- 

 tensible, and so on. And it is npon the recognition of such 

 constraints that Lagrange's method is founded. But the law 

 of equal partition disregards potential energy. However 

 great may be the energy required to alter the distance of the 

 two atoms in a diatomic molecule, practical rigidity is never 

 secured, and the kinetic energy of the relative motion in the 

 line of junction is the same as if the tie were of the feeblest. 

 The two atoms, however related, remain two atoms, and the 

 degrees of freedom remain six in number. 



What would appear to be wanted is some escape from the 

 destructive simplicity of the general conclusion relating to 

 partition of kinetic energy, whereby the energy of motions 

 involving larger amounts of potential energy should be 

 allowed to be diminished in consequence. If the argument, 

 as above set forth after Maxwell, be valid, such escape must 

 involve a repudiation of Maxwell's fundamental postulate as 

 practically applicable to systems with an immense number of 

 degrees of freedom. 



VII. On Sivans Prism Photometer, commonly called Lummer 

 and Brodhuns Photometer. By Prof. C. G .Knott, J).Se* m 



IN 1849 William Swan, subsequently Professor of Natural 

 Philosophy in the University of St. Andrews, read a. 

 paper " On the Gradual production of Luminous Impressions 

 on the Eye and other Phenomena of Vision " before the Royal 

 Society of Edinburgh (see Transactions, vol. xvi.) . This paper 

 contains some results of high interest, but I have no recollection 

 of ever having seen it referred to in modern literature on the 

 subject. 



On April 4, 1859, Professor Swan gave a second paper on 

 the same subject, much briefer than the first, and entirely 

 occupied with descriptions of greatly improved forms of 

 apparatus (see Transactions, vol. xxii.). Among the forms 

 of apparatus described is his " Prism Photometer." This is 

 simply and solely the form of photometer described in 1889, 

 exactly thirty years later, by Lummer and Brodhun, and 

 named after them in all recent literature (see Zeitschrift fur 

 Instrumentenkunde, Bd. ix.). I cannot do better than give 

 Swan's description in full, and reproduce his own diagram. 



He writes : — "An arrangement which, from an imperfect 



* Communicated by the Author: read before the Royal Society of 

 Edinburgh, Dec. 17th, 1899. 



