Royal Microscopical Society. 243 



works up to the present time, which state generally that when a 

 ray of light proceeds from a denser to a rarer medium, such as 

 glass to air, beyond angles given, total internal reflexion will occur. 

 But the fact is, that air has nothing to do with the phenomenon ; 

 its refractive power is so very small that it can only be discerned 

 by dealing with enormous masses, and has been accurately deter- 

 mined by the difference between the real and apj^arent position of 

 celestial bodies seen through the whole stratum of our atmosphere. 

 Any portion of the back of a total reflecting prism that has the air 

 withdrawn, by being set on a vacuum tube, will not show any 

 difference at the spot where the air is absent ; and if the whole 

 prism is placed under the receiver of an air-pump, total reflexion 

 will continue precisely as before : and as also bearing on the question 

 before us, I may add that " Newton's rings," or the coloured bands 

 seen between contiguous surfaces of glass are not caused by films of 

 air, they are the mere effect of interval upon the colour undulations 

 of light, considered upon the wave theory. When all air is with- 

 drawn the colours are still there. I allude to this as not being 

 altogether foreign to the present question, and to show that in what 

 are deemed total reflecting surfaces there is some influence on light 

 extending beyond that surface, and that external objects placed 

 thereon do not require to he in absolute contact in order to abstract 

 light. For very oblique incidences they must be much nearer, as 

 I have found that the rays within a few degrees of 90^ will not 

 touch objects such as diatoms, but pass beneath them. At 41° the 

 objects may be seen when they lie quite loose on the surface, and 

 hence the reason why the least oblique rays are the most available. 

 Sir Isaac Newton leaves it to be inferred that his colour rings are 

 caused by air, and also that the same element performs its part in 

 total reflexion. After he ends the experiments in the first part of 

 his Optics, fourth edition, so accurately conducted and clearly 

 expressed, towards the end in Book III. (which consists mainly 

 of queries) there is the following remarkable passage in question 29, 

 pace 346, relating to total reflexion, which I will venture to quote : — 

 " The rays of light in going out of glass into a vacuum are bent 

 towards the glass, and if they fall too obliquely on the vacuum, 

 they are bent backwards into the glass and totally reflected ; and 

 the reflexion cannot be ascribed to the resistance of an absolute 

 vacuum, but must be caused by the power of the glass attracting 

 the rays at their going out of it into the vacuum and bringing them 

 back, for if the further surface of the glass be moistened with water, 

 or clear oil, or liquid and clear honey, the rays which would other- 

 wise be reflected, will go into the water, oil, or honey, and therefore 

 are not reflected before they arrive at the farther surface of the 

 glass, and begin to go out of it. If they go out of it into the 

 water, oil, or honey, they go on becau°'^ ^^f^ attraction of the glass 



