The Food of the Smaller Frexh- Water Fishes. 67 



the retreating' overflow of streams, others from permanent hikes, 

 and still otliers from creeks and rivers. The food from the difi"er- 

 ent localities varies but little, on the whole, and it is scarcely 

 worth while to discuss the separate collections. That of these 

 nineteen specimens was almost purely animal, traces of a minute 

 flowering plant (Wolffia), ancl small quantities of filamentous 

 AlgaB only being taken by two of the specimens. Fishes were 

 eaten ])y but two, and were reckoned at two per cent, of tlie 

 food of the whole. One of these found was recognizable as a 

 Cyprinoid, but the other could not be determined. Insects 

 amounted to more than ninety per cent., all of them aquatic, with 

 the exception of a few gnats (Culicid;\i) taken b}'^ eight of the 

 fishes. Nearly half of the food consisted of larvie of Chirono- 

 mus and Corethra. Aquatic coleopterous larv.e were reckoned at 

 eleven per cent., and specimens of Corixa, taken by three of the 

 fishes, at two. A single fish had also eaten Galgulus. A fourth 

 of the food consisted of neuropterous larvae (Ephemeridi'o and 

 Libellulidjie). Crustaceans, though captured by more than half 

 the fishes, made but four per cent, of the food. As far as recog- 

 nized, this element consisted chiefly of the amphipod, AUorchestes 

 dentata., and the common isopod, Asellus. A few specimens of 

 Cypridid;e were noticed in two of the fishes, and Cyclops and 

 other Copepoda were taken by five. One fish had eaten a 

 Lumbriculus, a species closely allied to tlie common eaithworm. 

 A careful comparison was made of the food of specimens of 

 various ages — those, consequently, in which the situation of the 

 vent was widely different — ^ but no differences of food whatever 

 were distinguishable. It is highly probable, consequently, that the 

 explanation of this peculiar character must be sought elsewhere 

 than in the food. With respect to the othei- relations of food to 

 structure, we have at present only to note the coincidence of fishes 

 and aquatic insects as the principal elements of the food with the 

 large mouth and inferior development of the gill and pharyngeal 

 apparatus, and short and simple intestine. 



FAMILY COTTID^. 



This curious family, chiefly marine, is represented in tlie State 

 by several species from I.,ake Michigan, mostly from its deeper 

 waters, and by a single one recently discovered in our streams. 



