FEEDING HABITS OF KINGFISHERS. 217 



that millions of cjprinoids or minnows were found there, 

 as though they soitght in this canal an asylum from the* 

 attacks of carnivorous fishes, which are quite rare, as 

 compared with the number in the river near by. 



During the season of 1874 I took notes on such king- 

 fishers as were seen about two creeks, a mill-pond, and 

 the Delaware River. In each of these localities large 

 fishes of many kinds are more or less abundant, and the 

 percentage of small cyprinoids from two to four inches 

 long being much smaller than in the canal, it would 

 evidently be irksome to so voracious a bird as the king- 

 fisher to wait until fish of the proper size for swallowing 

 without preliminary butchering should come within reach. 

 It therefore, in some measure, seems to depend upon the 

 size of the captured fish whether or not it is killed by the 

 kingfisher before it is swallowed. 



Both habits having been found to be true of this bird, 

 it is desirable to know why the habit of killing the fish 

 before eating it should be the invariable practice of this 

 bird in some localities, as stated. I can only suggest 

 that this may depend upon the anatomical characteristics 

 of the fishes caught. When an abundance of cyprinoids, 

 which are fishes with soft-rayed fins, are to be obtained, 

 then little or no preparation is necessary to make them 

 fit for food ; but if young perch with their spiny fins, 

 or tough, hard-scaled fish of any family, have to be de- 

 pended upon, then the kingfisher is forced to prepare the 

 food before it can be safely swallowed. 



10 



