FISHERY BULLETIN: VOL. 70, NO. 1 



In October, 1970. when fertilization was dis- 

 continued, the underyearhng sockeye salmon 

 were about 30 "^r heavier than the fish caught in 

 October of 1969. 



DISCUSSION 



The period of freshwater growth is a very im- 

 portant stage in the natural i)roduction of ju- 

 venile sockeye. It is one in which mortality may 

 be high, particularly in the very early stages of 

 their growth. An abundant food of the optimum 

 size readily available for the fry to prey upon 

 is one factor which increases their survival rate 

 (LeBrasseur et al., 1969) . LeBrasseur and Ken- 

 nedy (1972) determined that the biomass of 

 zooplankton in May, 1970 before fertilization 

 was greater than at the same time in 1969. At 

 this time the numbers of fry caught in the mid- 

 water trawls began to increase and the catches 

 were greatest, reaching a maximum in June at 

 all stations shortly after fertilization com- 

 menced. LeBrasseur and Kennedy (1972) have 

 demonstrated that the phenomenal increase in 

 standing stocks of zooplankton in 1970 over 1969 

 was attributable to the sustained additions of 

 nutrients. It is evident that the young sockeye 

 fed very heavily (Figures 6 and 7) upon the in- 

 creased number of Epischura in the lake (Le- 

 Brasseur and Kennedy, 1972). Epischura 

 reached their greatest abundance in the upper 

 10 m in June when the average surface temper- 

 atures was about 15°C which was ca. 5° lower 

 than that in July. In their diel migration, the 

 young sockeye are able to take advantage of the 

 most favorable temperature for food conversion 

 for sockeye, which lies between 5° and 17°C, 

 with a general physiological optimum at 15°C 

 (Foerster, 1968; Brett et al., 1969). This may 

 account for the rapid rate of growth in June com- 

 pared with July. 



Although the increase in length was greater 

 in any one month in 1970 than it was in the same 

 month in 1969, the increase in weight of about 

 30% in October, 1970 was less than expected 

 from the 10-fold increase in zooplankton abun- 

 dance reported by LeBrasseur and Kennedy 

 (1972). The underyearling sockeye in the lake 

 in 1970 were the progeny of a large escapement 



of 72,000 adults in 1969 and a large number of 

 new recruits could be expected. Since most of 

 the spawning occurred along the lakeshore it was 

 not possible to determine the number of fry that 

 entered the lake, particularly when it was dis- 

 covered that fry were always caught in the trawls 

 from March to August. An estimate of the 

 total numbers of sockeye in the lake was cal- 

 culated from the fish targets recorded on the 

 high-frequency echo sounder during surveys of 

 of the lake conducted once every 3 weeks 

 throughout the year. From this analysis, 

 coupled with pertinent catch data from each tow, 

 a maximum estimate of 1 x 10'' juvenile sockeye 

 was determined. Johnson (1961) suggests that 

 above a threshold concentration of 5,000 fish per 

 hectare crowding may affect growth. LeBras- 

 seur and Kennedy (1972) have demonstrated 

 that the lake is sufficiently large that density 

 of even a maximum estimate of 1 x 10^ juvenile 

 fish could not have been a limiting factor in their 

 growth. However, the movement over a distance 

 of 8 km in 1 day of large layered masses of 

 juvenile fish in August (observed with high-fre- 

 quency echo sounder) from one locality where 

 the food source became depleted (O.D. Kennedy, 

 personal communication) to another where food 

 was abundant suggests that fish density may 

 have been a seasonal factor. 



The size of the yearling smolts leaving the lake 

 from April to June each year was always larger 

 than the same year class of juveniles caught at 

 the six regular trawling stations during the same 

 period because the migrating smolts school near 

 the outlet of the lake. In 1969 the 781 yearling 

 smolts caught in the Robertson Creek weir aver- 

 aged 67.5 mm in length and 2.6 g in weight. In 

 1970 the average length of 1,423 yearling smolts 

 was 71 mm and 3.5 g. In 1970 the average length 

 of the yearling smolts decreased from 72 mm in 

 April and May to 68 mm in June. The largest 

 yearling smolts migrated first in April and May 

 together with the 2- and 3-year-old smolts, 

 and in June, almost all the smolts were smaller 

 yearlings. 



In April, 1971, 589 yearling smolts were sam- 

 pled from the weir at Robertson Creek. Their 

 average length and weight was 79 mm and 4.8 g. 

 During this period the largest yearling smolts 



46 



