424 An Essay on the J\eftectiov, Refraction, and htjiection 



evident demonstration of a fact which is as palpable as the pen 

 I am now holding in my hand. 



That this fluid has originally some connexion with the air, is 

 extremely probable ; because, if it existed beforehand independent 

 of air, it would not require the presence of a medium in order to 

 reflect light. It is reasonable then to suppose that it originally 

 makes a part of the atmospheric air, but that there is some sub- 

 stance in the composition of glass which attracts it more strongly 

 than the air does of which it makes a part ; so that whenever 

 glass is exposed to the air for the first time, a decomposition of 

 the particles that are nearest to it immediately takes place (in 

 the same manner as chemists procure the decomposition of other 

 compoiiiui.s bv applying to them some substance that attracts to 

 it one i)art<of the compound, and repels the rest) ; and, in con- 

 sequence of this separation, the fluid that attaches itself to the 

 glass, recovers an original property in its nature, that of reflect- 

 ing light, which it had lost while it made apart of the composi- 

 tion of air. Whether I am right or wrong in this conjecture, it 

 is evident that there is a substance of some sort that attaches 

 itself to glass and other mediums, and that it only does reflect 

 light'when it is so attached ; and, having proved this, I now 

 come to consider the cause of the refraction of light. 



The refraction of light, I conceive, may be explained much in 

 the same way as the reflection of light, with this difference, that 

 one substance repels the rays, and the other refracts them. Glass 

 is a coinponiid of diflferent substances, and the fluid that reflects 

 light mav lie attracted by one part of the compound, while the 

 refracting substance may be attached to another part. These 

 fluids cannot penetrate the pores of the glass, and therefore ne-.. 

 cessarily remain upon its surface; and as the substances of which 

 glass is composed are blended together, they in like manner 

 blend themselves on its surface, each fluid being immediately 

 over that particular sul)stance by which it is atttacted ; and the 

 ravs of light, according as they strike upon either fluid, are .re-, 

 fleeted back or refracted onwards towards the further surface," 

 where the same phaenomena are repeated. 



Now, as it has already been proved that there is a substance 

 adhering to the srrface of glass that has the property of reflect- 

 ing light, it is more than probable that the refraction of light is 

 produced bv a similar cause ; and if no other proof of it could 

 be adduce;!, the analogy of the two cases would be sufficient to 

 estaiilish its authority. Fortunately however there is no neces- 

 sity to beg the question, for the proofs 1 have to advance in fa- 

 vour of this hypothesis are sufficiently strong to establish it with- 

 out having recourse to a mere probability. 



In the first place, if the fluid that reflects light covered the 



whols 



