• •Alaska Peninsula - 1951-57 



• • Kodiak-l948-5l. 1955- 57 



• •Prince William Sound - 1952- 58 



Moles 



Mean length 

 (MmJ 



Females 



r— 



3 



— p — 

 4 



-I 

 5 



Age in years 



Figure 4.--Nfean lengths of chum salmon In Alaska 

 Peninsula, Kodiak Island, and Prince William Sound 

 areas by age and sex (i>oints are unweighted means). 



points are unweighted means calculated from 

 table 7. The figure shows that despite the 

 inconsistency of the sources of the yearly 

 data, the basic relationships of size with age 

 prevailed in all areas. Age class 3 females 

 from the Kodiak Island area were too few 

 to use. Mean lengths increased from west to 

 east: Alaska Peninsula chum salmon were 

 smallest and Prince William Sound chum 

 salmon largest. 



In table 7 the numbers of fish in each age 

 class, average lengths, and standard deviations 

 for each sample are shown for the three 

 areas for the years 1948-58, Age classes con- 

 taining fewer than 10 fish were not used in 

 the calculations. 



Lengths in each age class varied little from 

 one year to the next, the greatest variation 

 occurring in males in age class 4 from the 

 Kodiak Island area; in 1955 males were 70 

 mm, (almost 3 inches) larger than those in 

 1948, In any year fish of the same age class 

 tended to be relatively the same length. Even 

 though fish of different ages were exposed to 

 different growing conditions during their life 

 in the ocean, they all tended to be relatively 

 large or relatively small at maturity. For 

 example, in 1952 average lengths were greater 

 than the long-term mean, and in 1956, except 

 for females in age class 4 in the Kodiak Island 

 area, they were smaller. 



Age composition. - -The age composition of 

 samples combined by years for each area is 

 shown in table 8. All age data are grouped to 

 give an average value for age composition of 

 chum salmon in the general region under 

 consideration. Age class structure of chum 

 salmon of the central and western part of 

 Alaska averaged about 10 percent 3-year-olds, 

 75 percent 4-year-olds, and 15 percent 5-year- 

 olds. The percentages were fairly consistent 

 from year to year. 



Age composition for the Kodiak Island area 

 varied from that for the Alaska Peninsula 

 and Prince William Sound areas. In the Kodiak 

 Island area about 27 percent of the fish in the 

 samples were in age class 5, which was about 

 twice the percentage of age class 5 fish in the 

 other two areas. Fish in age class 3, on the 

 other hand, formed a greater proportion of the 

 Prince William Sound and Alaska Peninsula 

 runs than they did of the Kodiak Island area 

 runs. When ' year-to-year changes were 

 inspected with regard to the long-term mean 

 age composition, it appeared that all three 

 areas tended to vary in the same direction 

 at the same time. For instance, in the Alaska 

 Peninsula and Prince William Sound areas, 

 low percentages of 4-year-olds were found in 

 1952. In 1956 in all three areas, high percent- 

 ages of 4-year-olds were taken. 



Sex ratio. --Sex ratio by area and by year 

 are given in appendix tables A-3, B-3, C-5, 

 and C-6, Differences in the percentage of 

 males from place to place were negligible, 



RECOMMENDATIONS 



During the analysis we found that because 

 of the inconsistency of collecting data by 

 locality of sampling, season of the fishery, or 

 stage of the spawning migrations, only general 

 conclusions could be drawn about chum salmon 

 in the various areas. More precise estimates 

 of parameters of length, age composition, and 

 sex ratio, as well as other biological statistics, 

 are needed. With reference to the characters 

 measured during this study, we recommend 

 the following for future work: 



1. Since the age composition in the fishery 

 and the spawning migration does change during 

 the season, samples should be taken from both 

 at regular and frequent intervals. 



