ON THE THEORY OF OPTICS. 321 



?-vliibit a single- ray of light, except as the confine between light and dark- 

 nesspor as the lateral limit of a pencil of light. 



When light passes through a space free from all material substances, it 

 moves, with great velocity, in a direction perfectly rectilinear ; when also 

 it passes through a material substance perfectly uniform in its structure, 

 it probably always moves in a similar manner. But in many cases its 

 motions are much interrupted. Those substances through which light 

 passes the most freely, and in straight lines, are called homogeneous trans- 

 parent mediums. Perhaps no medium is, strictly speaking, absolutely 

 transparent ; for even in the air, a considerable portion of light is inter- 

 cepted.* It has been estimated that of the horizontal sunbeams, passing 

 through about 200 miles of air, one two thousandth part only reaches us ; 

 and that no sensible light can penetrate more than 700 feet deep into the 

 sea ; a length of seven feet of water having been found to intercept one 

 half of the light which enters it. 



It is possible that mediums, not in other respects identical, may be 

 homogeneous with respect to the transmission of light ; for example, a 

 glass may be filled with a fluid of such a density, that the light may pass 

 uninterruptedly through their common surface ; but it generally happens, 

 that whenever the nature of the medium is changed, the path of light 

 deviates from a straight line ; thus, the apparent places of the sun and 

 stars are changed by the effect of the atmosphere, because the light, by 

 which we judge of their situations, is deflected, in its passage out of the 

 empty space beyond the atmosphere, first into the rarer and then into the 

 denser air. In the same manner, when we view a distant object over a 

 fire or a chimney, it appears to dance and quiver, because the rays of light, 

 by which it is seen, are perpetually thrown into new situations, by the 

 different changes of the density of the air in consequence of the action 

 of heat. 



When rays of light arrive at a surface which is the boundary of two 

 mediums not homogeneous, they continue their progress without deviating 

 from those planes in which their former paths lay, and which are perpen- 

 dicular to the surface of the mediums ; but they no longer retain the same 

 direction, a part of them, and sometimes nearly the whole, is reflected back 

 from the surface, while the remaining part is transmitted and refracted, 

 or bent. The name refraction is derived from the distortion which it 

 occasions in the appearance of an object viewed in part only by refracted 

 light : thus an oar, partially immersed in water, appears to be bent, on 

 account of the refraction of the light by which its lower part is seen, in its 

 passage out of the water into the air. 



There is no instance of an abrupt change of the density of a medium, 

 without a partial reflection of the light, passing either into the denser or 

 into the rarer medium ; and the more obliquely the light falls on the 

 surface, the greater, in general, is the reflected portion. No body is so 



* Consult Bouguer, Traite d'Optique sur la Gradation de la Lumiere, 4to, Par. 

 170. Herschel's Description of an Actinometer, Ed. Jour, of Sci. iii. 107. 

 Pouillet, Mem. sur la Chaleur Solaire, Comptes Rendus, July 9, 1838. Forbes on 

 the Extinction of the Solar Rays in passing through the Atmosphere, Ph. Tr. 1842, 

 p. 225. 



y 



