The Foorl of Fishes. 



67 



exist, they can probably be determined only by taking at 

 one time and place a number of specimens of unlike char- 

 acter in this particular. 



While I believe that the generalizations made above 

 will hold good, at least for fishes of similar form and in- 

 ternal structure among the Acanthopteri, I do not wish to 

 be understood as extending them at present beyound this 

 order. Doubtless, while the characters mentioned must 

 assist greatly in determining the food of a species a jyrhv'r^ 

 they are not by any means sufficient for this purpose when 

 taken by themselves. The discussion of other features, 

 external and internal, bearing upon this subject, must be 

 postx3oned to a later i^eriod of the investigation. 



Table of Food of Young Centrarchid^e. ( One inch and under.) 



