286 Fortieth Annual Meeting 



several instances were noted of the biter bitten. In every 

 case the victim was seized by the tail and a fish burdened 

 with his prey was so handicapped as to become himself an 

 easy mark. Cases were noted where the second victim 

 had let go his morsel and part of it floated still attached 

 by a thread of skin and so towed about by the captor, who 

 seemed expecting his turn next. Although active and free- 

 swimming when hatched and apparently searching for food, 

 they did not begin to eat one another until at least ten days 

 old and, with none of them taking any other food (so far 

 as known) the last survived two weeks longer, that is, at 

 least 24 days from hatching. And, contrary to all expecta- 

 tions, the last survivor is believed to have been, not the most 

 successful cannibal, but a fish which had not fed at all. 



Their behavior seems to warrant the suggestion that from 

 the first the water should contain something which the young 

 fish can swallow. It seems likely that they may learn to eat 

 before they are able to digest. Other creatures swallow 

 first substances which serve rather to open the organs and 

 be discharged than as food. Perhaps young fish learn to 

 feed by the involuntary swallowing of particles brought into 

 their mouths in respiration, and the voluntary selection and 

 seizing of food comes later. Supposing this to be the case 

 it would seem that in rearing young fish in confinement the 

 effort should be to furnish them water supplied with some- 

 thing to be swallowed, almost anything being better than 

 nothing. It may well be that microscopic creatures sifted 

 out in the movements of respiration may form a large part 

 of the food of fishes of various kinds and sizes. But, if 

 pure water be insufficient for fish reared in confinement, it 

 is not less, but more, important to see that planted fish be 

 released where food not only may be found, but where it is 

 so abundant as to be forced on the attention of the young 

 fish, literally crammed down their throats. Small wonder 

 young fish become cannibals when their comrades are abso- 

 lutely the only thing in sight. From this a next inference 

 will be that at planting young fish should be scattered. 



