98 The American Sdlmon-fisherman. 



the centre of wMch lies directly over the fish, and the 

 diameter of which is to the depth of water above the fish, 

 as twenty is to thirteen. Within this area surrounding 

 objects were clearly visible to me, while other portions of 

 the surface v/ere always as opaque as a stone wall. When 

 the flies were displayed within the transparent space, it 

 was almost impossible, with one ^ception, to distinguish 

 one variety from another, no matter how widely tbeir 

 characteristics might differ. When a comparatively dull- 

 colored object is viewed against a bright light, the vivid 

 impression of the background swallows up that produced 

 upon the retina by the more feebly illuminated interven- 

 ing object; and to this was the result attributed. The 

 Fiery Brown here showed decided superiority. Its pecu- 

 liar-colored hackle, wound the whole length of the body, 

 fairly burned with transmitted light. 



When the flies were examined looking toward the sun, 

 somewhat the same effect was produced, and by the same 

 cause. No matter how clear the water might be, at four 

 feet, their characteristics became obscure. If the ob- 

 server then changed his position to the other end of the 

 tank, so that the sun was at his back, every detail of con- 

 struction, even to the hook itself, was plainly visible the 

 full length of the tank. In the one case the illuminated 

 side was viewed against a dark background; in the other 

 case the shaded side was seen against a bright back- 

 ground. The difference was very marked when the sun 

 was on the meridian of the tank, diminishing gradually 

 as it bore more or less to one side. 



The jungle-cock neck-feather, employed so frequently 

 upon the cheeks of flies, was a very conspicuous feature 

 at all times when anything beyond the flash of the crest- 



