LIGHT. 139 



matter in a so-called solid are as distant as the stars in heaven, 

 still a certain depth or thickness of such solid would present 

 at every point of space a particle or rock in the successive 

 progress of a wave, which particles, to carry on the move- 

 ment, must vibrate in unison with it. 



At the utmost, our assumption, on the one hand, is that 

 wherever light, heat, &c., exist, ordinary matter exists, though 

 it may be so attenuated that we cannot recognise it by the 

 tests of other forces, such as gravitation, and that to the ex- 

 pansibility of matter no limit can be assigned. On the other 

 hand, a specific matter without weight must be assumed, of 

 the existence of which there is no evidence, but in the phe- 

 nomena for the explanation of which its existence is supposed. 

 To account for the phenomena the ether is assumed, and to 

 prove the existence of the ether the phenomena are cited. 

 For these reasons, and others above given, I think that the 

 assumption of the universality of ordinary matter is the least 

 gratuitous. 



Outizv ri TOV iravros Kevov ire\ft ou5e trepiffcrov. 



A question has often occurred to me and possibly to oth- 

 ers : Is the continuance of a luminous impulse in the inter- 

 planetary spaces perpetual, or does it after a certain distance 

 dissipate itself and become lost as light I do not mean by 

 mere divergence directly as the squares of the distances it 

 travels, but does the physical impulse itself lose force as it 

 proceeds? Upon the view I have advocated, and indeed 

 upon any undulatory hypothesis, there must be some resist- 

 ance to its progress ; and unless the matter or ether in the 

 interplanetary spaces be infinitely elastic, and there be no 

 lateral action of a ray of light, there must be some loss. 

 That it is exceedingly minute is proved by the distance light 

 travels. Stars whose parallax is ascertained are at such a 

 distance from the earth that their light, travelling at the rate 

 of 192,500 miles in a second, takes more than ten years to 



