308 ON LIGHT. 



tween a lens of light crown, and a plate of heavy flint- 

 glass. In this case the reflexions from the two surfaces 

 are performed either both from a denser medium upon a 

 rarer, or IwtJi from a rarer on a denser according as one 

 or the other glass is uppermost. In the former case 

 two serai-(or one entire) undulation will be gained or 

 lost between tlie reflected ravs at emero^ence, in addition 

 to the entire ones lost l''et7uec}i the glasses : in the latter 

 none. At the central spot, then, the two reflected rays 

 will start on their backward course in exact harmony, 

 and the spot vvill be white, not black ; and a similar re- 

 versal of character will of course pervade the whole 

 series of rings. This result, predicted by theory, has 

 been found confirmed by experiment. 



(92.) It was a favourite idea of Newton that the 

 colours of all natural bodies are in fact the colours of 

 thin pellucid particles of such sizes and thicknesses 

 as to reflect those tints which, in the scale of tints of the 

 coloured rings above described, most nearly correspond 

 to them. This idea we know now to be untenable, if 

 for no other reason than that we are sure the ultimate 

 particles or indivisible atoms of bodies (if any such there 

 be), are at all events many hundreds, thousands, or 

 millions of times smaller than even a single wave-length 

 of any homogeneous ray of light. It will, of course, be 

 asked how we know these wave-lengths. And this we 

 must now explain : in doing which we shall have to de- 

 velop the most astounding facts in the way of numerical 

 statement which physical science has yet revealed. 



(93.) In a series of equal waves running along still 



