MIRAGE. 



569 



FIG. 7. SPECTERS OF BEN LOMOND. 



relieves it from the surface of the apparent 

 water, and according to its modification of 

 form gives the appearance of islands, architect- 

 ure, or foliage. S;ind being silica, the prime 

 constituent of glass, is therefore a good reflect- 

 ing material, and when the angle of light is 

 low, as in this case, it assumes the brilliance 

 of a surface of water. 



Every phase of the mirage we have been re- 

 viewing has been referred by Sir David Brew- 

 ster, the great expounder of optical phenomena, 



FIQ. 8. REFRACTION AND REFLECTION. 



to what he calls " fits of extraordinary refrac- 

 tion in the earth's atmosphere." That is to 

 say, the earth's atmosphere is subject to fits 

 of changing density, which cause objects to 

 appear in places where they are not. The na- 



ture of the reflection of light is sufficiently il- 

 lustrated by a boy playing with a bit of look- 

 ing-glass in the sunshine. By placing a glass 

 prism in the path of a sunbeam, we have the 

 illustration of both refraction and reflection, 

 as in the figure. The prism is nothing more 

 than a triangular piece of glass, from the surface 

 of which a portion of the sunlight is reflected, 

 but the greater quantity of it passes through 

 the prism as shown by the arrow-marked re- 

 fracted light. If the prism were replaced by 

 a transparent medium, such as the atmosphere, 

 which decreased in density from its base up- 

 ward, the refraction would be in the same direc- 

 tion. Therefore atmospheric refraction could 

 only deflect the image of an object downward 

 instead of upward, and consequently images 

 of objects on the earth's surface could not be 

 produced in the air by refraction. If they 

 were, the science of civil engineering, astron- 

 omy, and navigation would be impossibilities, 

 for their accuracy depends entirely upon the 

 telescope's seeing in straight lines whenever 

 and wherever that instrument is brought into 

 use, and so unchangeable is the minute quan- 

 tity of refraction due to the earth's atmosphere 

 that in all astronomical observations where pre- 

 cision is required it is dealt with as a constant 

 factor. If it were in the least subject to change, 

 exactness would be unattainable, for no astron- 

 omer could tell what condition of refraction ex- 

 isted in the atmosphere at the time of his ob- 

 servations. The same would be true of a navi- 

 gator finding the position of his ship on the 

 ocean by his quadrant, and equally true of all 



