ON LIGHT. 237 



in like manner have to be reversed on interchanging the 

 illuminating and illuminated points. On neither suppo- 

 sition could the same hitiinsic law of communication 

 carry the ray from a through p, to B, and from B, througli 

 P, to A. This, then, is the law of regular reflexion, com- 

 monly expressed by saying that the angle of incidence is 

 equal to that of reflexion and lies in the same plane with if. 

 (20.) If the reflecting surface be a plane, there will be 

 only one point in it which fulfils these conditions. Thus 

 a perfectly polished flat surface of silver, free from 

 scratches, or that of still water, sends no light to the eye 

 from a candle, and is in fact invisible, except at this one 

 point so determined whence the light is reflected to the 

 eye, and in the direction of which from the eye the re- 

 flected candle is seen. With curved surfaces, as well 

 as with those we designate as "rough" or " unpolished," 

 the case is difterent. In all surfaces of this last-men- 

 tioned description the microscope reveals to us such irre- 

 gularities, such innumerable and abruptly broken facets, 

 protuberances and hollows, as to satisfy us that in every, 

 the most minute, visible portion of such a surface, places 

 must occur in which the condition of equal inclination of 

 the two lines in question to the actual surface, as it exists 

 i?i those places, is satisfied so that a ray there reflected 

 may reach an eye liowever situated. By such rays, and 

 by others which have entered into the substance of the 

 object and been there internally reflected or otherwise 

 bent, in a manner presently to be explained, all surfaces 

 not self-luminous become visible as objects, being seen by 

 rays " scattered " from them in every possible direction. 



