of Light find its "Ji'heory. ^ ^ 



As in this case, cos. r = sin. ^, we have the equation (II.) page 

 110 last Number, which is directly derived from the experiments, 

 whence those experiments confirm the theory. In another me-> 

 dium than air I denote the exponents of refraction, for instance, 

 for the ray C with (C 7i), for the ray D with (D n), &c. That 



for such a medium in the above equation, instead of (C ''^OrclT^' 



is to be put, instead of (D w) r^-^' &c. maybe concluded from 

 the experiments with light modified by mutual influence in the 

 refractive medium. * ' 



When the second surface of the glass containing the system 

 of lines is coated with a black resinous varnish, which in its 

 refractive power nearly equals glass, then no light can be re- 

 flected from this coated side, and then only the etched surface 

 reflects light. Then, if, from an apparently very small aper- 

 ture in a shutter, light reflected yrom the etched side falls on 

 the object-glass of the telescope, exactly the same phenome- 

 non appears as when the light under the same angle of incli- 

 nation passes through the system of lines, non symmetric 

 perfect spectra of the second class are seen. The intensity of 

 these spectra is still so great that the distances of the various 

 fixed lines can be determined with great accuracy. I have 

 made an extensive series of experiments on the distances of these 

 spectra produced by reflection under various angles of inci- 

 dence, of which I here omit the details. The equation (V.) 

 represents the aggregate observations as satifactorily as can 

 be expected. In the developement of the general equation 

 for the phenomena produced by reflection after the principles 

 of interference, the same term is obtained as for light passing 

 through, namely, without correction, the e4uation (III.) Con- 

 sequently, the theory is also in this way confirmed by expe- 

 rience. 



It is very remarkable that, under a certain angle of inci- 

 dence, a portion of a spectrum produced by reflection consists 

 of entirely polarised light. This angle of incidence varies 

 greatly for the different spectra, and even still very percepti- 

 bly for the different colours of one and the same spectrum. 



* New Modification of Light, ^c. page 59^ French translation, p. 94, 



