Astronomical and Nautical Collections, 209 



not produce any sensible light in the direction D/, as the 

 calculation founded on the law of interferences demon- 

 strates. It is therefore sufficient that the projection of 

 the elevations, or the depth of the depressions, should be 

 very small in proportion to the magnitude of an undulation 

 of light, in order that the surface of the mirror should reflect 

 no sensible light, except at an angle equal to the angle of 

 incidence ; and when the greatest inequalities do not exceed 

 the hundredth of an undulation, for example, that is, the four 

 or five millionth of an inch for the yellow rays, the mirror must 

 necessarily exhibit a perfect polish ; [and there is reason to 

 suspect that, when the scratches are tolerably regular, it will 

 have the appearance of a bright polish, even if they are in- 

 comparably larger than this. Tr.] 



There is a consequence of this explanation which requires 

 to be remarked. Since the lengths of the undulations are 

 different for the different kinds of coloured rays which con- 

 stitute white light, it may be imagined that the asperities of 

 the surface may be of such a magnitude, as to afford a pretty 

 regular reflection of the longest, those of the red rays, and 

 yet to dissipate a considerable portion of the violet rays, the 

 length of undulation of which is shorter by a third ; so that 

 in the regularly reflected image of a white object, the red 

 and orange rays may predominate, while the green, and espe- 

 cially the blue and violet rays, may be in a smaller propor- 

 tion, so that the tint would become reddish : and this result 

 is confirmed by experiment. Instead of carrying the polish to 

 a certain point only, which it might be difficult to ascer- 

 tain, employ a mirror merely ground plane, and worked 

 with fine emery, and incline it to the incident rays until you 

 begin to distinguish a sufficiently distinct image of a white 

 object seen by reflection ; this image will appear tawny, and 

 even of an orange red colour, like that of the setting sun, if 

 the object is so bright as to be visible without too much incli- 

 nation of the mirror. The tint of the image is the same, 

 whatever the nature of the reflecting substance may be, whe- 

 ther of steel, for example, or of a greenish crown gla.^^s. In 

 proportion as the obliquity of the mirror increases, the image 

 becomes brighter and more brilliant ; and when it become^ 



JAN.— MARCH, 1828. P 



