The Food of Fishes. 21 



erly be made the nucleus about which all the facts of its 

 natural history are gathered. 



In a paper on the food of Illinois fishes published in the 

 second bulletin of this Laboratory, the subject was treated 

 in a general and cursory way, the amount of material upon 

 which that paper was based being insufficient for exact or 

 detailed description. The favor with which that prelim- 

 inary notice was received, has made it possible to under- 

 take a more serious investigation ; and this paper contains 

 an account of the food of the Acanthopteri of the state 

 which I believe to be nearly or quite sufficient for the stu- 

 dent of science and for the practical fish-culturist. It is 

 still necessary only to study the food of specimens under a 

 half-inch in length, and to test the value of the general 

 conclusions here reached, by occasional examinations of 

 fishes taken from other waters at other seasons of the 

 year. Among the results of this study, those relating to 

 the food of the young are especially worthy of attention, 

 and these have therefore been summed up separately. 



The explanation of certain structural conditions about 

 the mouth, throat, and gills, has proceeded so far as to make 

 it very likely that a number of definite general corre- 

 spondences between structure and food will be made out, 

 which will enable us to tell with considerable accuracy 

 and detail what the food of an unknown fish must be, by a 

 mere inspection of the fish itself; provided, of course, 

 that we know what food is accessible to it in its habitat. 

 It seems likely to prove to be a general rule that a fish 

 makes scarcely more than a mechanical selection from the 

 articles of food accessible to it, taking almost indifferently 

 whatever edible things the water contains which its ha- 

 bitual range and its peculiar alimentary apparatus enable 

 it to appropriate, and eating of these in about the ratio 

 of their relative abundance and the ease with which they 

 can be appropriated at any time and place. If this is so, 

 knowing the structure of a fish and the contents of a body 

 of water, we shall be able to tell, a priori., what the fish 

 will eat if placed therein. 



This is, in fact, the objective j)oint of the present inves- 

 tigation, — to arrive at a knowledge of the correlations of 



