0-3 DAYS OUT 

 August 11 WE-69 



ISl CS I FS 'SP 



FS I SPl is' CS I FS • SPI 



MILES FROM TRAP SITE 



CS I FS ISPl 



Figure 1 1 . — Tag recoveries per trap per time period by 20-inile intervals measured from the point of release in middle Chatham Strait, 



1942. 



August had moved into Frederick Sound during the 

 latter part of the first week and during the second 

 week continued to drift eastward. Thus, in 1941 the 

 migratory behavior of fish tagged during August in 

 lower Chatham Strait was quite different from that of 

 those tagged during August in 1938 and in 1939, when 

 more of the fish remained near the tagging station in- 

 stead of moving into Frederick Sound. 



The infrequent recoveries in lower Chatham Strait 

 from releases in Icy Strait and in upper and middle 

 Chatham Strait in 1941, coupled with the sparse 

 recoveries in the northern areas from releases in lower 

 Chatham Strait, indicate that most of the fish moving 

 through lower Chatham Strait in mid-August of 1941 

 were en route to streams in lower Chatham Strait, 

 Frederick Sound, and Stephens Passage. 



The results from the single release in lower 

 Chatham Strait in 1942 are given in Figure 15 and 

 Table 10. For 68 of the tags recovered from this experi- 

 ment, only the date of recovery was given. The dis- 

 tribution of the recoveries for which the locations were 

 known was similar to the distributions observed for 

 the 1941 experiments except that relatively more 

 tagged fish were recaptured in middle Chatham 

 Strait. However, unless it is assumed that the 89 tags 

 with known recovery locations constituted a represen- 

 tative sample of the recoveries, the movements in- 

 dicated cannot be accepted without serious reserva- 

 tion. 



Summary and Discussion of General 

 Migration Patterns as Determined 

 from Tag Recaptures 



From each of the pink salmon tagging experiments 

 a sequence of spatial distributions of tag recaptures 

 per unit of effort during successive recovery periods 

 was determined. The apparent migratory pattern 

 generated by the recaptures from a single release may 

 be described as the result of an interaction of two 

 basic types of movement — one directional, the other 

 random. The directional component is defined as a 

 movement with complete spatial orientation, whereas 

 the random component is a nondirected movement in 

 which the fish disperse in all possible directions about 

 a center of density. It is obvious, of course, that any 

 particular geographical distribution of recoveries is 

 affected by the distribution of recovery effort and also 

 by the particular mixture of racial stocks represented 

 in the population of tagged fish. 



Verhoeven (see footnote 6) concluded that pink 

 salmon migration consisted largely of a random to- 

 and-fro movement. He was influenced by multiple 

 recaptures of several fish tagged in the Ketchikan area 

 of southeastern Alaska, and it is not possible to deter- 

 mine if the successive recapturing altered the natural 

 migratory behavior of the fish. The experiments 

 reported here neither confirm nor definitely refute 

 Verhoeven's hypothesis of random to-and-fro move- 

 ment. However, the results of these experiments do 



21 



