3 - 



2 - 



0-3 DAYS OUT 

 JULY 7 EA-'45 



a.i 

 < 



H 



UI 

 0. 



5° 



LU 



u 

 m 

 q: 

 q:3 



LU 

 DQ 



5 

 :3 2 



? I I 



lioo 



200 



AUGUST 6 -EA H5 



I h" 



.^, 



12.00 / AUGUST 9 EA-45 



I I 



100 



1113 DAYS OUT 



? I I I 



100 



200 



200 



|100 



200 



I 1''°° 



AUGUST 12 EA-15 



I I llOO 



1 FS r rs <;pl 



CS > FS LCS SPl 



200 Oil 100 200 Oil jlOO 



CS IPSLCSSPI CS 'FSLCSSP' 



MILES FROM TRAP SITE 



200 Oil 100 



CS IFS LCS SPI 



Figure 12. — Tag recoveries per trap per time period by 20-inile intervals measured from the point of release in lower Chatham Strait, 



1938. 



indicate that the movement of pink salmon can be 

 more adequately described as consisting of a random 

 and a directional component with the relative impor- 

 tance of the two components varying from year to 

 year. In 1942 fish tagged in Icy Strait, upper Chatham 

 Strait, and middle Chatham Strait distributed them- 

 selves as might have been predicted on the basis of the 

 random to-and-fro hypothesis. In 1941 the move- 

 ments of fish tagged at the same locations showed a 

 much stronger directional component. These tagged 

 fish migrated in large numbers to the Frederick 

 Sound-Stephens Passage region. Thus, it appears that 

 the particular mixture of racial stocks in a given sec- 

 tor of the fishing grounds varies greatly from year to 



year. There was some indication from these ex- 

 periments that the random component of movement 

 was relatively stronger in even years. 



The apparent random component of the migration 

 of pink salmon may be an inherent characteristic of 

 this species, which spawns in a large number of widely 

 dispersed streams. A different behavior might be ex- 

 pected for sockeye salmon, Oncorhynchus nerka, 

 seeking one major spawning ground. The results of 

 these experiments do not provide information on the 

 cause of the random dispersion of tagged fish. This 

 effect could be caused either by the erratic wanderings 

 of tagged fish or could be the result of the superim- 

 position of a great number of directional migrations 



22 



