two inches loiiy. liad iikkU' niiicty-six per cent, of tlieir food dI' Enlonios- 

 traca and the small larva' of jj^nats (Chlronomus) already mentioned, seventy 

 per cent, of the first and twenty-six of the second. This ^roup comprised 

 five specimens of black bass under three-quarters of an inch in length, 

 two rock bass of similar size, two of the large-moutlied sunflsh (ChaMio- 

 bryttus) from seven-eighths of an incli to an inch long, nineteen of the 

 commoner suntlshes (Lepomis) ranging in length from an inch to two 

 inches, live of the genus Centrarchus. one inch and under, four croppies 

 (Pomoxys) from three-quarters of an inch to an inch and a half, and six 

 indeteriuinable si)ecimens, probably Lepomis, from seven-sixteenths to five- 

 eighths of an inch long. A single sheepshead an incli and an eighth in 

 length had eaten Chlronomus larvjv (seventy-five per cent.) and larva- of 

 the "river fly" (Ilexagenia). A single grass pickerel about an inch and a 

 quarter long had taken about sixty per cent, of its food from Entomos- 

 traca and young Ampliipoda, the remainder consisting of little fishes. 



The first food of the common white-fish was determined experimentally, 

 the breeding habits of this species making direct observation impossible. 

 Three hundred and forty very \oung fi-y fed with fragments of the lirook 

 shrimp. Gammarus. in a hatching house, were examined in January. 1881, 

 and thlrty-tive of them, which had apparently taken food, were dissected. 

 Minute fragments of Gammarus were found in but eighteen of these, 

 while five contained minute insect larvie. four. Entomostraca. and eight, 

 small particles of vegetation.— objects accidentally conveyed to them in 

 the water of the hatching house. In two hundred and forty-two others, 

 confined in spring water, only eight were found to have eaten anything, 

 and these had taken only AlgcT and vegetable fragments. In February of 

 the same year, fourteen specimens, confined in a small aquarium and sup- 

 plied with" living objects, plant and animal, from stagnant pools, were 

 proven to feed freely upon the smallest Entomostraca presented to them, — 

 chiefly Cyclops and" Canthocamptus, ten of the fourteen eating Cyclops. 

 three Canthocamptus. and one a specimen of each. 



A little later, a more extensive experiment was conducted by means of 

 a large aquarium, in which there were placed several hundred fry, kept 

 constantly supplied with all the living objects which a fine gauze net 

 would separate from the waters of Lake Michigan. Of one hundred and 

 six of these, dissected within the following fortnight, sixty-three had taken 

 food consisting almost wholly of the smallest Entomostraca occurring in 

 the lake (a minute Cvclops and a slender Diaptomus). The other objects 

 encountered were rotifers, and diatoms and other unicelkilar Alga?, ap- 

 pearing, however, in such trivial quantity as to contribute nothing of Im- 

 portance to the support of the fry. 



A dozen specimens of small gizzard shad, ranging in length from four- 

 fifths of an inch to nearly two inches, had eaten about ninety per cent, 

 of Entomostraca, two per "cent, of Chlronomus larva\ and, for the remainder, 

 Alga>. 



Tiie true minnows iCyprinida') seem to agree with the suckers in the 

 more minute character "of their early food. Six examples— three-eighths to 

 three-fourths of an inch long — too small to determine, tiut apparently be- 

 longing to the genera Minnilus. had eaten Entomostraca. Chlronomus 

 larva\ many Protozoa, and unicellular Alga', a few filamentous Alga^ and 

 minute fungi and fungus spores, a water mite, and a few accidental in- 

 sects. In several specimens of the common chub minnow (Semotilus), 

 from five-eighths of an inch to an incli in length, seven per cent, of the 

 food was Entomostraca. and the remainder consisted of filamentous Alga?. 

 It should be noted, however, that twenty per cent, of that of the smallest 

 specimen, which was five-eighths of an inch long, was Cyclops, and it may 

 be that Semotilus lives wholly on Entomostraca at first, merely changing 

 its habit earlier than most of its allies. Two other minnows of the genus 

 IS'otropis. an inch and a half in length, had eaten nothing but Ento- 

 mostraca. The Cyprinida\ like the sucker family, are toothless when 

 young. 



