HAVE FISHES MEMOEY. 387 



it disappeared instantly at sig-lit of a pike hook. Later it was found 

 that it bore the marks of having been pricked on a former occasion. 

 If a net is dragged through a swarm of carp, kept and fed for breed- 

 ing- purposes, various observers report that for a week or two they 

 remain away at feeding- time, and nothing- can induce them to appear. 

 Mr. Mullert, of New York, whose numerous observations I have liad 

 to quote frequently, mentions that the iish bred by him in large 

 quantities, which as a rule swim close up to him, llee if he holds a 

 little net in his hand when he opens the aquarium. The same is 

 reported by Von Quasowski, in Constance. 



According to various observations, swarms of fish absent themselves 

 if one of their number has been caught with the hook, Semon, for 

 instance, reports this cx)ncerning- a shoal of Echenek^ observed by him 

 in the Torres Strait. The same is told me of the bream {Ahxim.ls 

 hrama) and the dace {Idus Tnelmiotus). 



The above experiences in connection with the taming of fish may 

 suffice to prove that impressions once received can be retained. The 

 same conclusion is forced upon us with still greater clearness by the 

 letters, in the neighborhood of 150, that concern themselves with the 

 behavior of fish at feeding times, whether in ponds or streams or aqua- 

 riums. The statements in all are so absolutely identical that the 

 reported facts may well be accepted as the result of the aggregate 

 experiences of all observers of fish. 



Goldfish, long accustomed to be fed, grow so tame that they always 

 come close to the spot at which the keeper appears to feed them. An 

 interruption of months in the practice of feeding does not cause loss 

 of the habit. Herr M. Schmidt-Metzler, in Frankfort on the Main, 

 reports that his goldfish, which pass the winter in a hothouse, swim 

 up to him the very day they are put back into the pond in the spring. 

 Mr, H. Mullert (Brooklyn) writes that various sorts of goldfish and 

 tench, kept in difi'erent aquariums, take food from his hand, and some 

 allow him to caress them. For four months every year they are 

 released in a pond. While there they are timid, and do not come up 

 to the owner or attendant, but as soon as they are back in the aquarium 

 they feel at home, and without fear swim up to the attendant that 

 feeds them as though sixteen weeks liad not intervened. The same is 

 told of the perch, the AV:Y//>///;'A?/;?t'A?6S' [shovelnose sturgeon], minnows, 

 bitterlings, tench, sheath-fish, trout, and various species of carp. In 

 many cases fish follow the keeper for some distance as he walks along 

 the side of the pond. 



Certain marks of an optical character seem to bind the fish to the 

 attendant, although the opinion advanced by one observer, W. von.Der- 

 schau, that they do not know the attendant as an individual, must be 

 indorsed. Ives (Crookston) tells that in his trout hatchery the attend- 

 ant wore a scarlet coat; anyone that put on this coat could lure the fish 



