5IO Roosi-vclt Wild Life .liinals 



times jumping at the same time and forming a verj- attractive spectacle." Hubbs 

 ('21, p. 263) found a remarkable difference between the habitat of the young and 

 the adult of this species. He found that soon or immediately after hatching on 

 the shoals of the inland lakes he studied in southern Michigan, the young fish 

 moved outward "to assume a pelagic habitat over deep water." and only rarely 

 during the first month of their existence were they seen to return to the shallow 

 waters, although "they were repeatedly and carefully sought for there." Often they 

 were most abundant over the deepest part of Portage Lake where the water was 

 near a hundred feet deep. In streams Hubbs found the young would keep over 

 the deepest part of the quieter stretches, away from the shores. It was evident 

 that during August and September the young return to shallow water, "at first 

 intermittently but soon permanently." The adults showed a "practically exclusive 



selection of the shoal community Not once was an adult seen over the 



deep lake water associated with the young." Cahn ('27, p. 64) makes similar 

 observations to those of Hubbs concerning the habitat of this fish. The adult 

 was found to live entirely in shallow water and the ynung, u]i to two-thirds adult 

 size, live over deep water, but both young ami adult livt- at all times near the sur- 

 face. He says: "Both young and adults are surface species, living normally within 

 less than a meter of the surface of the water, and spending most of their time 

 within ten or twelve centimeters of the surface. . . . The silversides never 

 under any conditions descend below the upper meter of water, this being the maxi- 

 mum depth sought by the adults, while nothing can drive the immature individuals 

 more than a few centimeters below the surface. Hence the statement that the 

 silversides is the most characteristic of our surface fishes." Cahn notes (p. 69) 

 that the small size of the young fish together with their inconspicuous coloring and 

 transparent bodies afford a protection against enemies Ijoth in the water and in 

 the air. 



Food. Baker ('16, p. 180) gives the results of the exaniinatinn of the fond 

 of four examples of this species from Oneida Lake, taken in the \icinity of Ladd 

 Point, September 3, 1915. More than 90% of the food was adult midges; the 

 remainder was amphipods, entomostracans, water mites and bryozoans. Forbes 

 ('83, p. 70) in his examination of 25 Silversides taken in the northern and central 

 parts of Illinois found the food to be purely animal matter, a little over half con- 

 sisting of insects and less than half of crustaceans. About a third of the entire 

 amount constituted larvae of Chirojwinus; the crustaceans were all Entomostraca. 

 Spiders and terrestrial insects, accidentally washed or fallen into the water. 

 amounted to 12% of the food. Evermann and Clark ('20, \'ol. i. p. ^^jS) at Lake 

 Maxinkuckee found the food to be insects and entomostracans. Forbes and Rich- 

 ardson ('09, p. 228) say: "It seems to live wholly on animal plankton, apparently 

 catching its minute prey one liy one. as a jiike captures fish. Its mouth, though 

 small, is well equipped witli tcetli. ;nid its gill-rakers are unusually well dcveln|H'd. 

 being nuiuerous, slender, armed with minute denticles, and longer than the gill- 

 filaments. Corresponding to its predacious habit, its intestine is uncommonly short, 

 the whole alimentary canal being considerably shorter than the body without the 

 head." Thev mention the finding of a small unrecognizable minnow among the 



