SALMON AND TEOUT IN ALASKA. 73 



movements of a school, if they occur, can be regarded only as explora- 

 tion either for food or spawning streams. 



Other interesting temperature questions are raised by the sockeyes' 

 selection for spawning of such streams as the one at Bartlett Bay.'' 

 It may be postulated that streams with lakes at levels accessible 

 to salmon possess a higher summer temperature than streams of 

 similar volume without lakes. There is a class of streams, however, 

 such as that at Bartlett Bay, where the lake outlet furnishes only 

 a small part of the volume of the main stream. In the Bartlett Bay 

 stream a temperature of 46° at the mouth June 26 decreased to 39° 

 before the lake in which the sockeyes spawn was reached. The 

 main volume of the stream is glacial water, and there is nothing at 

 the mouth to intimate to human intelligence the existence of lakes 

 and suitable spawning beds in its course. With even tliis temper- 

 ature, however, the river was probably warmer than the surrounding 

 salt water, in which ice was then drifting. 



CURRENTS. 



The influence of tide or current in salt water apparently varies, 

 for reports are contradictory. The Fraser River sockeyes are said 

 to travel on the flood, i. e., with the current. Salmon enter San 

 Francisco Bay on the ebb. At Karluk they come in on the first of 

 the ebb, or preferably on flood just before the current from the 

 lagoon changes. In the Kvichak the last of the flood and the whole 

 of the ebb are said to be the best fishing. By this they must approach 

 on the flood and take the river on the ebb. It is also claimed that a 

 southwest wind blows the fish inshore in Bristol Bay, whereas a 

 northeast wind at Karluk is alleged to prevent their approach. The 

 value of these observations regarding winds is lessened by the fact 

 that certain winds prevent the operation of gear and perhaps influence 

 the salmon not to expose their presence by jumping. It is believed 

 that many fish, in 1903 estimated at some 2,000,000, reached 

 Karluk Lake in this way by passing the bay and estuary unobserved. 

 A similar instance was noted by Mr. Cobb on the Ugaguk in 1906, 

 when an offshore wind had not prevented the schools from reaching 

 the fishing ground. It seems probable that inshore winds, besides 

 obscuring the presence of fish, will accelerate their movement by 

 assisting the current shoreward, while an opposite wind will retard 

 this. 



In such streams as Dolomi it is inconceivable that a current influence 

 can be felt beyond the small bay at its mouth. Tliis may explain 

 why only small fishes reach it, such small fishes perhaps traveling 

 closer to the shore. On the other hand, none of the small salmon 

 streams of that regrion can maintain the identitv of their currents 



"Moser, Bulletin U. S. Fish Commission, vol. xxi, 1901, p. 374. 



