Table 21. — Age composition and calculated^ mean fork 

 lengths'^ on September 1 of age and age I sockeye salmon 

 captured in tow nets in lakes of the Naknek and Karluk 

 systems, weighted by surface area and catch per tote, 1961-62 



Age fish Age I fish 



Mean Mean 



fork fork 



System, lake, and year Frequency length Frequency length 



Percent Mm. Percent Mm. 



Naknek system 



Coville Lake 



1961 100.0 54 



1962 100.0 60 



Grosvenor Lake 



1961 100-0 46 



1962 100.0 52 



Naknek Lake 

 North Arm 



1961. ._ 



1962 62 5 56 37 5 98 



Iliuk Arm 



1961 84 1 57 15 9 78 



1962 69 2 61 30 8 91 



South Bay 



1961 64 3 47 35 7 88 



1962 26 63 74 85 



West End 



1961 75 1 65 24 9 8S> 



1962 91 8 69 82 



Brooks Lake 



1961 96 3 51 3 7 



1962 97.1 60 2 9 



Karluk system 



Karluk Lake 



1961 88.7 40 11 3 83 



1962 81.9 49 18 1 76 



^ Curves describing the average length were calculated for each lake 

 each summer: only those data involving 10 or more fish were used. 

 - After at least 48 hours, preservation in 10 percent Formalin. 



nek system (table 22) seems to reflect the results 

 of a combination of at least three factors: (1) 

 movement of the large fish first in the interlake 

 migrations, (2) midseason recruitment of re- 

 cently emerged fry (especially evident in Gros- 

 venor Lake), and (3) differences in productivity 

 of waters. The average size of juveniles taken in 

 tow nets in Coville Lake increased from the up- 

 per to the lower end and increased again in 

 juveniles taken in fyke nets as they migrated 

 from Coville Lake to Grosvenor Lake. The juve- 

 niles captured in fyke nets in Grosvenor River 

 were larger than those taken in tow nets in the 

 lake. In the next basin downsystem, Iliuk Arm, 

 smaller fish were more abundant at the inlet 

 (upper third) even though the larger juveniles 

 that migrated from Grosvenor Lake enter this 

 end. The average size was the same in the upper 

 third of South Bay and the adjacent area of Iliuk 

 Arm but increased toward the outlet of the lake. 

 Chignik system. — The number of parent 

 spawners per unit lake area in the Chignik sys- 

 tem was greater in 1960 than in 1961, and, cor- 



Table 22. — Calculated mean fork length of age sockeye 

 salmon from the upper end of Naknek system to outlet on 

 September 1, 1962 



Sampling area Fork length 



Mm. 

 Coville Lake 



Upper half 56 



Lower half 60 



Coville River 61 



Grosvenor Lake 52 



Grosvenor River .. 63 



Naknek Lake 

 Iliuk .\rm 



Upper third 57 



Middle third 63 



Lower third 61 



South Bay 



Upper third 61 



Middle third 66 



Lower third 



West End > 69 



' Does nut include samples from the small basin in the northwest 

 corner of Naknek Lake. 



respondingly, age progeny were more abundant 

 in 1961 than in 1962. The mean lengths of the 

 age progeny on September 1 were less in the 

 year of greater abundance (table 23) . The rela- 

 tions are similar to those in the Wood system. 

 In the Chignik lakes, however, the apparent 

 growth of progeny at similar densities of spawn- 

 ers and progeny is better than in the lakes of 

 the Wood system ; this difference may be due to 

 a combination of richer nursery areas and fewer 

 competitors in the Chignik lakes. 



Karluk system. — The escapement of sockeye 

 salmon in Karluk Lake was greater in 1960 than 

 in 1961, and, correspondingly, age progeny 

 were more abundant in 1961 than in 1962. In 



1960 about 349,000 adults entered the system 

 and the average catch per tow of age fish in 



1961 was 8.9 from August 16 to September 15. 

 The escapement in 1961 was about 297,000, and 

 the average catch per tow in 1962 was 4.3. 



The average length of age sockeye salmon 

 in Karluk Lake was greater in 1962 than in 1961 

 (table 21 ) . The juveniles were larger in the year 

 of lesser abundance (1962), so it might be as- 

 sumed that the size of age fish is density- 

 dependent. But because of the long period of 

 recruitment of recentl.v emerged fry into the lake 

 — caused by the extended spawning period (4-5 

 months) — the mean size of young-of-the-year 

 at the end of the summer is based on fish that 

 may differ in age by many weeks. Differences in 

 the abundance of early- and late-spawning stocks 

 might result, therefore, in apparent but non- 



SOCKEYE SALMON IN MAJOR RIVER SYSTEMS IN SOUTHWESTERN ALASKA 



435 



