FISHERY BULLETIN: VOL. 82. NO. 1 



TABLE 6.— Fork lengths and apparent growth rates of juvenile 

 spring chinook salmon recaptured at the Dalles Dam after release 

 into the Deschutes River. 1977. Fork lengths are means ± standard 

 errors for the number of samples shown in parentheses. 



DISCUSSION 



Determination of the migratory characteristics of 

 juvenile chinook salmon during smolting has been 

 complicated by the variety of migratory behaviors 

 displayed by the juveniles. Some fry migrate from 

 tributaries shortly after emergence from the gravel 

 (Reimers 1973; Ewing et al. 1980), but there is little 

 evidence that the fry move into the estuary at that 

 time (Schluchter and Lichatowich 1977). In some 

 stocks, a general movement of fish through the river 

 occurs during the fall of the first year (Reimers 1973) 

 with a majority of the fish entering the ocean during 

 the fall of the first year (Reimers 1973; Schluchter 

 and Lichatowich 1977; Buckman and Ewing 1982). 

 In other stocks, seaward movement occurs primarily 

 in the following spring when the fish are more than 1 

 yr old (Mains and Smith 1964; Diamond and Pribble 

 1978: Raymond 1979). Krcma and Raleigh (1970) 

 reported migration of juvenile chinook salmon into 

 Brownlee Reservoir (Snake River, Idaho) in fall and 

 spring for 2 consecutive years. The migration pattern 

 seems to depend upon stock, size, and rearing con- 

 ditions and may be highly variable. It is therefore 

 important in the culture of various stocks of juvenile 

 chinook salmon to determine the timing of maximum 

 migration tendency. 



In the present study, the major migration of fish 

 released early into Pelton ladder occurred in mid- 

 May. Fish from the same brood released into the 

 Deschutes River at about this time were found to 

 migrate 213 km to the Dalles Dam within 7 d, sug- 

 gesting that the migrational behavior was seaward 

 directed (Hart et al. 1981). It is difficult to confirm in 

 the Deschutes River that the release of fish into 

 Pelton ladder 1 mo before the time of maximal migra- 

 tion tends to increase the time during which the fish 

 will migrate. Release of the fish 1 mo later than the 

 time of maximal migration tends to decrease the time 

 for migration. It is important to note that it is not 

 necessary to release the fish early to insure that all 



migrate to sea. Releases late in the migration period 

 were recovered to the same extent as those released 

 earlier. Migration tendency seems to be retained for 

 some time, even though the fish are not permitted to 

 begin migration. These results suggest that late re- 

 leases hasten the seaward migration, thus removing 

 the populations of hatchery fish quickly from the 

 river system and affording maximum protection to 

 the wild stocks. 



Those groups released later than July were recap- 

 tured in the trap in decreasing numbers (Tables 1,2). 

 In 1977, nonmigrant fish were recaptured in increas- 

 ing numbers from releases after 12 July (Table 1). 

 This result indicates that the decrease in numbers of 

 fish recaptured at the trap was due to decreased 

 migration tendency and not due to increased mor- 

 talities at the higher water temperatures. 



A major advantage of utilizing a closed system such 

 as the Pelton ladder for studies of migration was that 

 fish populations and flows could be effectively con- 

 trolled. Variables which remained uncontrolled in- 

 cluded photoperiod, lunar periodicity, temperature, 

 and food supply. Of these, photoperiod seems the 

 most important in stimulating seaward migration. 

 Previous studies utilizing a closed system for study- 

 ing seaward migration of steelhead trout, Salmo 

 gairdnvri, (Zaugg and Wagner 1973; Wagner 1974) 

 and coho salmon, Ocorhynchus kisutch, (Lorz and 

 McPherson 1976) also concluded that photoperiod 

 was an important factor affecting the timing of sea- 

 ward migration. 



Lunar phase has been suggested to affect the onset 

 of migration, based on the correlation between peaks 

 in plasma thyroxine levels and lunar phase (Grau et 

 al. 1981). Assuming maximal migration occurred on 

 22 May in both 1977 and 1978, this date correspond- 

 ed to the time of a new moon in 1977 and that of a full 

 moon in 1978. These brief data do not support the 

 hypothesis that the migration is influenced by the 

 lunar phase. 



Temperature may have had a dual influence on 

 migration. Temperature has been suggested as a 

 releasing factor for salmon migration (Hoar 1958; 

 Baggerman 1960), but we were unable to show a 

 statistical relationship between daily migration and 

 average daily temperature (Fig. 4). 



Temperature also serves to increase growth rates in 

 salmonids in the presence of abundant food supplies. 

 Wagner (1974) suggested that a critical size was 

 required in steelhead if migration were to take place. 

 The importance of size on migration of spring 

 chinook salmon can be seen by comparing the extent 

 of migration of the slow- and fast-reared fish in 1978 

 (Fig 3). The slow-reared fish may have failed to mi- 



162 



