the seine was open soutlieast and zero wlien it was 

 open northwest/ A time-space extrapolation of the 

 average catch (in 30 minutes in a 14-mile-wide 

 [0.46-km.] band) indicated that at least 750,000 

 juveniles migrated daily past any given line of 

 latitude off southeastern Alaska in 1964. Thus, 

 for tiie 30 to 60 days of strong migration, it is evi- 

 dent that major stocks of fisli were involved. 



The fork length of the pink salmon at this stage 

 varies from 10 to 25 cm. The average size was 

 significantly larger in the northern samples, pre- 

 suniabl}- because of the presence of lish which had 

 already migrated a considerable distance from 

 southern production areas (fig. 2). The northern 

 samples also included small fish that presumably 

 had just entered the sea from nearby channels and 

 bays. Mixed with the juvenile pink sahiKin wore 

 juvenile sockeye, chum, coho, and chinook salmons 

 and steelhead trout {Salmo gairdneri), which sug- 

 gests that the migratory cue at this stage is simihxr 

 for all species. 



CANADA 



ISO-W. 



Figure 2.— Fork lengths of juvenile pink snlnion s.uiiplcd 

 August 4-14, 1964, in eastern Gulf of Alaska. 



The northward migration of juvenile salmon is 

 indicated by the purse-seine catches and tag re- 

 turns. Six pink salmon were recovered a year after 

 tagging, all in southeastern Alaska (fig. ?,). All 

 were recovered south of the i)oint of release. If il 

 is assumed that they entered the ocean near the 

 point where they were recaptured as maturing 

 fish, they must have migrated northwestward be- 

 fore they were tagged. The specimen tagged near 

 lat. 59° N. by long. 138° W. had traveletl about 

 350 miles (648 km.) by Sei)tember 22, 1961, when 

 it was tagged. Figure ;J also illustrates the loca- 

 tions of release and recovery for 1 sockeye salmon, 

 1 steelhead trout, and 58 coho salmon that were 

 tagged as juveniles, along with the pink salmon. 



*For a discussion of the validity of purse seine gear for deter- 

 mlnlDK directional migrations of salmon in areas of strong ocean 

 currents, see Hartt (1960 : .S-10). 



Figure .3. — Release and recover.v diagram of .salmon :ui(l 

 steelhead trout tagged as juveniles and recovered 1 or li 

 years later tall rohn and ])ink salmon were recovorod 1 

 year after tagging I . 



These lish were also recovered south of the points 

 of release; some traveled northwestward ovei' 

 1,000 miles (1,852 km.) before they were tagged. 

 The rate of travel of juvenile pink salmon dur- 

 ing their first few months at sea is diflicult to esti- 

 mate on the basis of the few tag returns received 

 to date, because the distances involved are rela- 

 tively short and the date of ocean embarkation can 

 only be approximated. If we assumed that the 

 specimen tagged near Yakutat (fig. 3) on Septem- 

 ber 22 had left Dixon Entrance on August 1, and 

 then had followed the cfjastline (.■(50 niiles or 

 ()48 km.), its rate of travel would be 6.6 miles 

 (12.2 km.) per day. By the same method, if we as- 



444 



U.S. FISH AND WILDLIFE SKKXK'l-: 



