THE LUMINIFEROUS ETHEE. 97 



applied it to explain the laws of double refraction. In 

 their theory the conclusions arrived at were rigorously 

 derived from the premises ; but the results did not alto- 

 gether agree with observation; that is to say, although 

 they could by the adoption of certain suppositions be forced 

 into a near accordance with the observed laws of double 

 refraction, yet they pointed out the necessity of the existence 

 of otiier phenomena which were belied by observation. Our 

 own countryman Green Avas the first to deduce Fresnel'slaws 

 from a rigorous dynamical theory, although neai'ly simulta- 

 neously MacCullagh arrived at a theory in some respects 

 similar, though on the whole I think less satisfactory. 



Still all these theories followed pretty closely the analogy 

 of ponderable matter ; and at least in the first three 

 mentioned the ether was even imagined to consist of discrete 

 molecules, acting on one another, like the bodies of the 

 solar system regarded as points, by forces in the direction 

 of the joining line, and varying as some function of the 

 distance. I have already quoted the very strong language 

 in which Newton rejected the idea of the heavenly bodies 

 acting on one another across intervening spaces which were 

 absolutely void. But the conception has nothing to do with 

 the magnitude of the intervening spaces ; and the con- 

 ception of action at a distance across an intervening space 

 which is absolutely A^oid, is not a bit easier when the space in 

 question is merely that separating two adjacent molecules, 

 when the ether is thought of as consisting of discrete mole- 

 cules, than it is when the space is that separating two bodies 

 of the solar system, though in this latter case it may amount 

 to many millions of miles. If the ether be in some unknown 

 manner the link of connection whereby two heavenly bodies 

 are enabled to exert on one another the attraction of 

 gravitation, then according to the hypothetical constitution 

 of the ether that we have been considering, we seem 

 compelled to invent an ether of the second order, so to 

 speak, to form a link of connection between two separate 

 molecules of the luminiferous ether. But since the nature 

 of the ether is so very different as it must be from that of 

 ponderable matter, it may be that the true theory must 

 proceed upon lines in which our previous conceptions derived 

 from the study of ponderable matter are in great measure 

 departed from. 



If we think of the ether as a sort of gigantic jelly, we can 

 hardly imagine but that it would more or less resist the 



H 



