Prof. Challis on the Principles of Theoretical Physics. 317 



in the application of partial differential equations to the deter- 

 mination of fluid motion ; and I venture to assert that, till these 

 were known, the science of hydrodynamics was in so imperfect 

 a state that an undulatory theory of light was not possible. 



There will appear to be the more reason for making the above 

 request when the nature of the task which I have, to a con- 

 siderable extent, accomplished is stated. It is known that there 

 are certain phenomena of light which are dependent only on 

 properties of the medium in which it is generated, and through 

 which it is transmitted. The following is a list of them : 

 (1) the uniformity of the rate of propagation ; (2) the identity 

 of rate for rays of different intensity ; (3) the difference of in- 

 tensity of different rays ; (4) the variation of the intensity, ac- 

 cording to a certain law, with variation of distance from a centre ; 

 (5) the co-existence at the same instant of different portions of 

 light in the same portion of space ; (6) the interference of diffe- 

 rent rays according to given circumstances ; (7) the composite 

 character of light; (8) its colour; (9) results of compounding 

 colours; (10) the different kinds of polarized light; (11) the 

 circumstances of the interference and non-interference of polar- 

 ized light. To account for all these facts, the single hypothesis 

 is made that the medium which is the vehicle of light is a 

 perfect fluid pressing proportionally to its density. Out of this 

 one hypothesis the explanations of all these different phenomena 

 are to be evolved by mathematics alone ; and if any well-ascer- 

 tained mathematical result be contradicted by any one of the 

 facts, the whole theory must be abandoned. Now the hydro- 

 dynamical propositions that I have spoken of, if my demonstra- 

 tions of them be true, do, in fact, give these explanations. Since, 

 however, the principles on which the demonstrations rest are in 

 some respects in advance of any previously proposed, additional 

 elucidation and confirmation may reasonably be required. It is 

 on this account that I intend, as already intimated, to revise 

 both the propositions and their demonstrations. 



We have now to consider what hypotheses of the second 

 class — those relating to the qualities of the constituents of 

 bodies — may be allowable and sufficient for the foundation of a 

 general physical theory. First, it may be assumed that the 

 constitution of material substances is atomic, the adoption hypo- 

 thetically of this constitution being justified by the facts of 

 chemical combination and analysis. As the known inertia of 

 masses must be due to the inertia of the constituent parts, it 

 may also be assumed that the atoms are inert. It is necessary 

 to make a supposition respecting the forms of the atoms, other- 

 wise the mutual action between them and the aether cannot be 

 brought within the province of mathematical calculation. I have 



