fishing continued throughout the summer. A 

 second major migration took place on June 28, 

 and smaller migrations during the last of 

 July and the middle of September. A few sock- 

 eye salmon migrated from the lake through 

 the summer. Future programs requiring an 

 estimate of total migration should take into 

 consideration an extended migration period 

 throughout the summer and fall. It is not 

 known whether these summer migrants re- 

 mained in Naknek Lake until the following 

 spring, or whether they migrated immediately 

 to the sea. 



In addition to juvenile sockeye salmon seven 

 other species of fish were caught in the nets; 

 881 sculpins (Cotius spp.); 57 coho salmon 

 (Oncorhynchus kisutch); 23 rainbow trout 

 (Salmo gairdneri); 9 Dolly Varden (Salvelinus 

 malma); 4 brook lamprey (Entosphenus lamottei); 

 15 blackfish (DalUapectoralisj; and 22 nine- 

 spine stickleback (Pungitius pungitius). Almost 

 all of the catches of these species were made 

 at night. The coho salmon were caught from 

 June 2 through 29, and some daytime migration 

 of this species was observed by personnel 

 counting adult sockeye salmon at the weir 

 through June and July. Rainbow trout migrated 

 sporadically from June 4 until October, and 

 Dolly Varden first appeared at the end of 

 September. A few lampreys were evident 

 during May and early June, and blackfish 

 appeared late in the summer. A few stickle- 

 backs and large numbers of sculpins were 

 caught intermittently throughout the sum- 

 mer. 



By the very nature of Brooks River this 

 is not likely to happen. In the first place, 

 juveniles attempting to swim from lower 

 Brooks River or Naknek Lake would encounter 

 Brooks Falls, which is about 8 feet high. An 

 efficient gravity flow fish ladder enables adult 

 salmon to bypass the falls, but it would be 

 extremely difficult for juveniles to do so. 

 Furthermore, several stretches of Brooks 

 River have rapids so swift that it is difficult 

 for a man to keep his footing. 



Although it was deemed unlikely that small 

 salmon, particularly fry, could migrate up- 

 stream, a stationary trap was constructed to 

 intercept a portion of small fish that might 

 do so and so provide a measure of the serious- 

 ness of the problem. The trap was located on 

 a small point of land about 100 yards below the 

 outlet of Brooks River on the right bank 

 (fig. 17). It was simply constructed with two 

 rigid leads of 1/4-inch-mesh wire cloth trap 

 and a box trap of 3/8-inch-mesh wire cloth 

 lined with 1/8-inch-mesh nylon bobbinet. One 

 lead intersected the shore at an angle, and 

 the other extended toward midstream where 

 the current was too swift for fry to swim 

 upstream. We assumed that small fish attempt- 

 ing an upstream migration would seek a path 

 of least resistance along the shore. The trap 

 was usually emptied and cleaned twice daily. 

 Fish entering the trap eventually escaped, but 

 a relative index of migration was obtained 

 from fish remaining in the trap at the time 

 of checking. The trap was installed June 2 and 

 removed September 14. The catch by month 

 is shown in table 8. 



MIGRATION OF JUVENILE 

 SALMON UPSTREAM 



The question had been raised as to whether 

 juvenile sockeye salmon originating in Brooks 

 River or Naknek Lake might migrate upstream 

 into Brooks Lake. Such a phenomenon has 

 been noted at some lakes, notably Karluk 

 Lake on Kodiak Island, Alaska, and Babine 

 Lake in British Columbia (Roy Jackson, per- 

 sonal communication). An upstream migration 

 on a large scale at Brooks Lake would seri- 

 ously complicate studies of survival and pro- 

 duction of the indigenous lake population. 



Three features of the trapping study are of 

 particular interest: (1) The maximum catch of 

 sockeye fry (244) was in July but was insigni- 

 ficant compared with the total probable down- 

 stream migration from Brooks Lake; (2) the 

 catch of fry in June was small despite the fact 

 that maximum downstream migration was 

 observed during this period; and (3) threespine 

 stickleback (Gastemsteus aculeatus) and nine- 

 spine stickleback, which do not usually occupy 

 the same waters, were captured together. 



Observations of Brooks River incidental to 

 other activities supported the conclusion that 

 the upstream migration of sockeye salmon 



24 



