in the Velocity of Propagation of Light in Water. 353 



molecules which finally act upon a given aether molecule, a num- 

 ber which must appear in the coefficients of the differential 

 equations of the motion, must lead to a relation between the 

 velocity of luminous propagation and the density of bodies. 

 The fundamental conception is certainly at present wanting con- 

 cerning the method to be adopted. But not only the possibility, 

 but also the necessity of a connexion appears from the nature 

 of the differential equations supported by the views we have 

 given. It may be permitted here to return to a point already 

 touched upon. On page 265 we arrived at the surprising result 

 that the velocity of propagation of light in liquids and gases in- 

 creases universally with the increase of their temperature, while 

 with most solid bodies it diminishes. Such an opposite beha- 

 viour may be explained by assuming that there occurs in aether, 

 on heating it, an increase of density or a diminution of elasti- 

 city. The enormous alteration in density of liquids and gases, 

 brought about by their increase of volume, causes such a dimi- 

 nution in the density of the circumposed aether, as entirely to 

 mask the alteration specific to the aether itself, because the ma- 

 terial molecules in their separation from one another are accom- 

 panied by the enveloping aether. But the variation in the den- 

 sity of solid bodies being small, their influence upon the aether 

 may be so completely overbalanced by the increase of density or 

 diminution of elasticity which the aether undergoes with increase 

 of temperature, that an increase of the index of refraction is 

 brought about. We thus see that these phenomena are not in- 

 explicable. And although the hypothesis adopted for their ex- 

 planation is at present only brought forward to explain facts, it 

 is nevertheless simple and acceptable. We see, however, that 

 here also the alteration in the density of a body is in uninterrupted 

 connexion with the optical relations. 



The opinion of Jamin, that the vibration hypothesis does not 

 require any such definite connexion, must be rejected as un- 

 founded, since, although hitherto no such relation has been 

 established from theory, while the necessity for the existence 

 of one has been felt, although vaguely, by most physicists, at- 

 tempts have been made to establish divers empirical relations 

 which have not, however, as yet proved satisfactory. These 

 were, for the most part, obtained accidentally and without any 

 substantial foundation. It was tried whether any function of the 

 index of refraction, or of Cauchy's constants A and B, gave a 

 plausible relation to the density. 



For a long time after the undulatory theory had gained sup- 

 port, it was still supposed that the refractive power n 2 — 1 was a 

 constant. This view was caused by the proof given by Biot and 

 Arago of its constancy for gases, and by the extension of the 



