Flies and Sooks. 95 



HOWS, however, were obtainable in the form of " white- 

 bait," believed to be identical with the English fish of 

 that name. They were silvery in color, except upon the 

 back, which was light olive-green. 



The sky was entirely overcast and leaden in color, 

 with rain. The water in the tank was in what would 

 generally be considered first-class fishing condition, i.e., 

 very slightly turbid, the tank having just been filled. A 

 whitebait was selected of the same length as the wings 

 of the flies, and suspended in like manner from the cop- 

 per wire, with two flies on each side. Throughout these 

 experiments the flies were manipulated two or three 

 inches below the surface of the water, as .is customary in 

 salmon-fishing. 



A marked apparent increase in the size of both fish 

 and flies when submerged, over that in air, was first no- 

 ticed — an apparent increase of one half, if not somewhat 

 more. 



At the extreme further end of the tank the white- 

 bait was unmistakably a fish, and nothing else. As long 

 as the files were sufiiciently near the observer to enable 

 him to distinguish their details of construction, they 

 bore not the faintest resemblance to the fish. But when 

 moved to the farther end of the tank, the bodies of all, 

 except those of the Silver Gray and Silver Doctor, dis- 

 appeared altogether. Then, when those files were moved 

 which were provided with mixed wings of which the crest 

 of the golden-pheasant formed part, the wings seemed 

 to flash with reflected light in a manner and with an ap- 

 pearance not unlike that of the fish. The effect was 

 extremely beautiful to ray eye. It was like the inter- 

 mittent flash of a flrefly, lighting up the closely con- 



