314 



FISHERY BULLETIN OF THE FISH AND WILDLIFE SERVICE 



The slightly increased percentage return of 

 smolts that had spent two winters in fresh 

 water probably could be expected because of 

 their larger size and better chance of survival 

 in the sea. This conclusion follows that of 

 Foerster (1954), who showed a direct relation 

 between marine survival and size of smolts. 

 The discrepancy in age composition between 

 the 1957 smolts and the adults returning in 

 1959 after two winters in the ocean could be 

 due to inaccurate smolt sampling or to good 

 survival conditions in the ocean. Good survival 

 of fish spending two winters in the ocean was 

 evident in nearly all districts of Bristol Bay in 

 1959. 



DISCUSSION 



The methods used in this study do not lead 

 to a finite evaluation of smolt abundance and 

 age composition. However, their reliability has 

 been supported by most tests conducted during 

 the past 5 years. The tests involved the simul- 

 taneous fishing of two nets side by side and 

 two nets 2 miles apart up and down the 

 Kvichak River, and the relating the age of the 

 smolts to the fresh-water age of returning 

 adults. It is important to note that the great 

 differences in index values from year to year 

 permit a large error in the accuracy of the 

 method but the index method still enables de- 

 tection of important changes in smolt abun- 

 dance. Tremendous differences in abundance 

 from year to year do occur, as is evident from 

 the striking changes in the size of the adult 

 population of Kvichak River red salmon (table 

 11). The large adult escapements have pro- 

 duced large smolt migrations and small adult 

 escapements have produced small smolt migra- 

 tions. Two extremes in the smolt index values 

 are apparent for the past 5 years, the mini- 

 mum represented by 1956 and 1957, and the 

 maximum represented by 1958 and 1959 (table 

 4). The index of 1955 is neai'ly four times 

 larger than the next smaller value (1956) but 

 eight times smaller than the next higher value 

 (1959). 



The methods described in this paper have 

 been shown to be reasonably accurate to date. 



Final evaluation of these methods, however, 

 must wait until data are available on the return 

 of adult salmon from a longer series of smolt 

 migrations. Only after such an evaluation 

 period can the e.stimation of the population 

 variables and the prediction of subsequent adult 

 returns be achieved with accuracy. 



SUMMARY 



The purpose of the smolt study, conducted 

 from 1955 through 1959, was to provide an 

 annual index of abundance and to determine 

 the age and size composition of Kvichak River 

 red salmon. 



The relative index method used to assess 

 smolt abundance was based on the season-long 

 catch of a single fyke net set at the index site 

 throughout the same daily fishing period (2200 

 to 0100 hours) and under similar river con- 

 ditions each year. It was designed to deter- 

 mine major fluctuations in smolt abundance 

 from year to year. 



The annual indices of Kvichak smolt abun- 

 dance were for 1955, 11.2; 1956, 3.3; 1957, 1.3; 

 1958, 100.00; 1959, 85.9. These index values 

 are relative to the combined smolt catches of 

 1958, which were assigned a value of 100.0. 

 The tremendous differences in smolt index 

 values permit a large error in the accuracy of 

 the method, but the index method enables the 

 detection of important changes in abundance. 



Large parent escapements have produced 

 large smolt migrations and small parent escape- 

 ments have produced small smolt migrations. 



Weighted length-frequency samples, verified 

 by scale samples, were used in age determina- 

 tion. Two predominant age groups of smolts 

 occur in Kvichak River: those that have spent 

 one winter in fresh water and those that have 

 spent two winters in fresh water. 



The smolt age composition for the 5 years 

 indicated a trend toward cyclic variation and a 

 relation of size to population density, smaller 

 fish occurring during years of large abundance. 

 The similarity of smolt ages with the fresh- 

 water ages of returning adults supports the 

 methods and results of age determination. 



