Flies and Fly -faking. 319 



water, and stand a spoon in it. When held above the 

 eye and viewed at any angle through the sides, the de- 

 scribed appearance of the surface and the utter invisi- 

 bility of that portion of the spoon uncovered by the 

 water will be noted. But when viewed directly through 

 the bottom, the surface will then appear transparent, and 

 objects beyond it can easily be seen. 



Enveloping my head and the glass with a black cloth, 

 that no light might enter from below (a course invari- 

 ably followed in all my experiments with this tank), I 

 carefully examined the surface of the water. It ap- 

 peared perfectly opaque, until happening to cast my 

 eyes directly upward, I saw a clear and transparent 

 space bounded on one side by the end of the tank, and 

 on the other by a curved line strongly fringed with 

 prismatic color. In this clear space the windows of 

 houses, distant tw r o hundred feet and more, could readily 

 be seen, clothed, as was every object visible within it, 

 with all the colors of the rainbow. This unexpected 

 appearance of color seemed to discredit the directions 

 of the books that the clothing of the angler should be 

 sober in hue, since no matter how quiet his clothing, it 

 appeared he must look to the fish as though arrayed like 

 Solomon in all his glory. But subsequent reflection con- 

 vinced me that this phenomenon was due to a lack of 

 parallelism between the surfaces of the glass and of the 

 water. This gave to the body of water through which 

 the object was viewed the form of a truncated prism, 

 to the well-known action of which upon light I attribute 

 the presence of the colors. 



The transition from that portion of the surface which 

 was transparent to that which was opaque, was quite 

 abrupt. With thirteen inches depth of water, the curved 



