A Sportsman 447 



particular lost fly and parted leader. This did not 

 show much memory, and would shade the theory of 

 Dr. Dunn that the primitive memory of anadromous 

 fishes directed them to the particular stream of 

 their early life, directed by the dermal sense in its 

 magnet polarization as the needle guides the mariner 

 to point of destination. 



A. H. Gouraud, in one of his articles concerning 

 the movements of shad to the Connecticut River 

 from Long Island Sound, said they approached the 

 mouth of the river from the east, but did not reach 

 the mouth by direct course from the waters of the 

 Sound, as the discharge from the river pursued a west- 

 erly course in its current, making a loop, so to say, and 

 the shad took their course up the current of the river 

 waters, thereby occasioning a distance of a score of miles 

 more than would be required if the shad had gone direct. 

 This loop course in the sound had been clearly defined 

 by the net fishermen. This would indicate that the 

 current of the river was the guide for the shad to follow. 

 He adds: "This fluvial characteristic may be due to 

 distinctive mineral particles which, retained in solution 

 far out at sea, may be revealed to the delicate percep- 

 tiveness of the fish, and so guide it from the deep to 

 its bourn." 



I would add to the mineral particles also the vege- 

 table, an important element, distinct, and noticeable 

 by fish. It is the following of these elements that 

 guides the fish, which may to an extent have an edu- 

 cated perception as well as taste. I have frequently 

 found whole fresh-water clams up to two inches and 

 over in length in the stomachs of trout, which are 



