SALMON" AND TROUT TF ALASKA. 39 



headfirst from the quiet waters of the lake into the current of the 

 outlet. Seton Lake is 17 miles in length and about 1 mile in width. 

 In this body of water there can be no percepti])le current setting 

 down the lake, since the winds would produce more tide than gravity; 

 temperatures, while higher at the head than at the foot, will exhibit 

 no sensible gradual increase after a short distance from the inflow of 

 the cold streams. Yet these schools, apparently guided by instinct 

 alone, approach the outlet directly. Were not the same phenomenon 

 exhibited b}^ migrating fry, as humpback, for example, it might be 

 supposed that the ^^ear's residence in the lake has familiarized them 

 with its geography. 



Upon reaching a pronounced riffle they "turn en masse and head 

 upstream, circling and moving more or less rapidly in the more quiet 

 stretches before venturing to approach the dam," i. e., the main fall. 

 Their timidit}'^ in approaching swift water was notable. Onl}^ with 

 the waning light would the first few allow the current to carry them 

 down, the movement, or rather passivity, gradually becoming more 

 and more general. It will be remarked that in heavy runs the aver- 

 age size is slightly smaller. This would seem to indicate that fishes 

 not otherwise quite ready to migrate are drawn out by the general 

 movement. The hesitancy to encounter swift water is noticeable in 

 the adult as well. Spaulding at Karluk remarked it as a prominent 

 feature of the movement of spawners in the small streams, and schools 

 often are noted standing in a swift current, seeming to hesitate to 

 ascend or descend as the case may be. It would appear that this 

 trait ma}' have had no small influence in producing a resident form, 

 though there is no reason to believe that the dwarf sockeyes as now 

 known (0. nerka liennerlyi) are the product of the regular form, i. e., 

 only tard}" young that have failed to migrate with their fellows, and 

 thereb}^ remained to reach sexual maturity in fresh waters. Their 

 distinctive size, varying -with waters inhabited, and the small number 

 of eggs present in the ovary when yet little developed, mark them as 

 at least a distinct race. 



It has been reported " that hj the damming of streams for pro- 

 ducing reservoirs to supplj^ San Francisco and Oakland with water, 

 the salmon then in those streams were landlocked. As a result they 

 remained in the reservoirs and reproduced. Ultimately, by reason of 

 the confinement and its effects, thc}^ became dwarfed, decreasing 

 from their original weight of 12 to 14 pounds to less than 1 pound at 

 maturity.'' The continual breeding of this species (0. tschawytscha) , 

 in confinement in fresh water seems to produce dwarfing, even with 

 abundance of food. At the Trocadero Aquarium, where they have 



o Report Commissioner of Fisheries, California, 187(5-77, p. 5 and (i. 

 & See report by W. N. Lockington, in Report Commissioner of Fisheries, California, 

 1878-79, p. 50. 



