Although as many as 25 species of fish were 

 found in a system, the abundance of most was 

 low. Species of salmon other than sockeye were 

 not sufficiently abundant to compete seriously 

 for rearing area. Juvenile sockeye salmon and 

 one other species of fish constituted 75 to 95 

 percent of the catches. 



Of the species that may be competitors with 

 juvenile sockeye salmon, the threespine stickle- 

 back (Gasterosteus aciileatus) is most im- 

 portant in the Igushik, Snake, Wood, Kvichak, 

 Nuyakuk, Chignik, and Karluk systems and part 

 of the Naknek system. In other parts of the Nak- 

 nek system the ninespine stickleback (Piingitiiis 

 pungitius) , pond smelt (Hijpomesus olidus), or 

 pygmy whitefish (Prosopimn coulteri) seem to 

 be of greatest significance. Species of fish in most 

 systems that have varying degrees of impor- 

 tance as predators on juvenile sockeye salmon 

 include Arctic char (Salvelinns alpinus), lake 

 trout (S. 7iamaycush), northern pike (Esox 

 luciiis), juvenile coho salmon (O. Icisntch). and 

 Dolly Varden (S. rnahna). 



The data on the species in the other systems 

 of southwestern Alaska are too few to permit 

 generalizations. 



MIGRATION OF PRESMOLT SOCKEYE SALMON 



To study the relation between density of par- 

 ent spawners (numbers per unit lake area), 

 growth and abundance of progeny, and capacity 

 of the rearing areas, we had to know how long 

 the young sockeye salmon remain in the lake that 

 they first enter as fry. Fyke nets were fished, 

 therefore, in some interconnecting rivers during 

 the summer to detect possible migrations from 

 various lakes. Changes in length frequencies of 

 juveniles and their abundance in tow net catches 

 supplied indirect information on their move- 

 ments. 



Wood S\stem 



Catches of fyke nets fished in the connecting 

 rivers of the Wood system indicated no appre- 

 ciable migration of young sockeye salmon during 

 the summer between Little Togiak Lake and 

 Lake Nerka in 1961 or between Lake Nerka and 

 Lake Aleknagik in 1962." Two other kinds of 

 observations in the Wood system support the 



conclusion that these fish do not migrate during 

 the summer : ( 1 ) the abundance of young-of-the- 

 year in tow net catches after mid-August is gen- 

 erally proportional to the number of parent 

 spawners for each lake (the density of spawners 

 varies widely between lakes within years) ; and 

 (2) the growth of young-of-the-year is often re- 

 duced in lakes that had high densities of spawn- 

 ers the previous fall. We, therefore, assumed 

 that in the main lakes of the Wood system, 

 juvenile sockeye salmon remain within their lake 

 of origin at least during their first growing 

 season. 



K\ ichak S\ stem 



Most of the sockeye salmon in the escapement 

 to the Kvichak system spawn in tributaries or 

 on beaches in Iliamna Lake. The only possible 

 significant presmolt interlake migration would 

 be from Lake Clark to Iliamna Lake, and we 

 have no indication that this movement occurs. 

 We investigated this problem directly only at the 

 time of normal smolt migrations in 1959, 1962, 

 and 1963 when we found that the juveniles leav- 

 ing Lake Clark are mostly age II.'-' Because no 

 age III smolts have been captured at the outlet 

 of Iliamna Lake and no adults with three fresh- 

 water annuli have been captured in the system, 

 we assume that the age II juveniles migrating in 

 June from Lake Clark are smolts that pass di- 

 rectly to the outlet of Iliamna Lake. As indicated 

 elsewhere in this paper, the abundance of spawn- 

 ers within Iliamna Lake is greatest in the east 

 end. This concentration produces a greater 

 abundance of fry in that area. Entire age groups 

 move in the lake toward the outlet in the west 

 end. however; this movement is comparable to 

 interlake migrations in other systems. 



Naknek S\steni 



Interlake migrations of presmolt (age and 

 age I) sockeye salmon are known in the Naknek 

 system.'  In one movement fish left an area of 

 high density of juveniles to enter areas of low 

 densities (Coville Lake to downlake areas), and 

 in another the migration was from a low- to a 

 high-density area (Brooks Lake to Naknek 



^^ Burgner. R. L., and .T. M. Green. 1963. Study of interlake migra- 

 tion of retl salmon fry, Agulowak River. Univ. Wash.. Fish. Res. Inst., 

 Circ. 182. 13 pp. 



'- Orrell. R. F. 19ri3. Atiundance and age of Lal<e Clark red salmon 

 smolts, 1!)(;2. Univ. Wash. Fish. Res. Inst.. Circ. ISO, 13 pp. 



'•' Ellis, Robei't J. 19G3. The abundance and distribution of juvenile 

 red salmon and associated species in lakes of the Naknek River system 

 and Karluk I.nke. Bur. Commer. Fish. Biol. Lab., Auke Bay. Alaska. 

 [Manuscript. SO pp.] 



428 



U.S. FISH AND WILDLIFE SERVICE 



