240 



Fishery Bulletin 101(2) 



five genetically distinct groups of chinook salmon in the 

 Sacramento and San Joaquin rivers of California, 8.2% of 

 genetic variation determined from a microsatellite survey 

 was not contained within populations (Banks et al., 2000). 

 In our present study of Fraser River chinook salmon, 95% 

 of genetic variation was present within populations, where 

 3.0% was due to differences among regions (including the 

 lower Fraser, putatively colonized independently) and 1.3% 

 was due to differences among populations within regions. 



Pairwise Fgj< values between geographic regions of the 

 Fraser drainage (Birkenhead excluded) did not exceed 

 0.080. Thus, if indeed Fraser River chinook salmon are 

 descendants of genetically distinct chinook salmon "races" 

 from two (or more) glacial refugia, the observed relatively 

 low levels of differentiation suggest that introgression has 

 occurred among the races before or since their colonization 

 of the Fraser drainage. Support for the idea of the Fraser 

 drainage as a chinook salmon "melting pot" comes from al- 

 lozyme data (Fig. 19 in Myers et al., 1998) which indicate 

 that chinook salmon of the both the lower and interior 

 Fraser drainages are genetically intermediate to chinook 

 salmon sampled from three postulated glacial refugia (the 

 Columbia drainage, the Pacific coast and Beringia, the 

 northern refuge). Minisatellite DNA variation has also in- 

 dicated that Fraser River chinook salmon are genetically 

 intermediate to populations in southern and northern Brit- 

 ish Columbia, groups putatively derived from southern and 

 northern refugial areas (Beacham et al., 1996). 



Teel et al. (2000) suggested that the presence of popula- 

 tions in the south Thompson region that have high propor- 

 tions of ocean-type juveniles, and the genetic distinctive- 

 ness of the region, indicate that hybridization of ocean- and 

 stream-type chinook salmon occurred within this region of 

 the Fraser drainage. However, our study revealed that chi- 

 nook salmon of the lower Thompson and south Thompson 

 regions are the most genetically distinct of the interior chi- 

 nook salmon groups, apparently strongly isolated from each 

 other as well as from chinook salmon from all other regions. 

 The south Thompson populations were not more closely re- 

 lated to the ocean-type chinook salmon of the lower Fraser 

 than were the predominantly stream-type populations in 

 other interior regions. Moreover, whereas the high level of 

 allelic diversity observed in the south Thompson region 

 may support the idea of historical hybridization, the pres- 

 ence in the south Thompson samples of a relatively large 

 number of unique alleles (i.e. alleles not observed in other 

 Fraser regions) does not support the idea that the hybrid- 

 ization occurred between the ocean-type chinook salmon 

 that colonized the lower Fraser and the stream-type popu- 

 lations elsewhere in the drainage. The genetic distinction 

 of the south Thompson chinook salmon seems more likely 

 due to a unique colonization history than to recent gene 

 flow within the Fraser drainage. South Thompson chinook 

 salmon may have resulted from a historical admixture of 

 refugial ocean- and stream-type races, but their genetic 

 similarity to other interior Fraser chinook salmon groups, 

 and the intermediate genetic position of all Fraser River 

 chinook populations, indicate that all groups of Fraser 

 chinook salmon may possess similar racial admixture. 

 The ocean-type chinook salmon populations of the south 



Thompson may have originated from stream-type fish and 

 thus reflect adaptation to environmental conditions condu- 

 cive to the production of large juveniles capable of smelting 

 in their first year of life. 



Our data provide some evidence of gene flow among 

 chinook salmon in different regions of the Fraser Mid 

 Fraser chinook salmon were characterized by a high level 

 of allelic diversity but were notable in their low level of dif- 

 ferentiation from chinook salmon of both the upper Fraser 

 and north Thompson regions. Mid Fraser fish were also 

 the most closely related of all the interior groups to lower 

 Fraser chinook salmon. As for the south Thompson region, 

 a high number of unique alleles in the mid Fraser region 

 indicated that the high level of diversity was unlikely to 

 result solely from hybridization of postglacial founding 

 populations from the lower and upper Fraser regions. 



Further genetic analysis of chinook salmon in British Co- 

 lumbia and refugial areas will reveal the nature of genetic 

 variation in the Fraser River and other regions of southern 

 British Columbia that likely represent "contact zones" of 

 postglacial recolonization. Although extensive introgres- 

 sion may complicate the identification of conservation units 

 based on "important phylogeographic subdivisions within 

 species, those based on historical separations or fluctua- 

 tions that are still evident in the gene pool" (Moritz, 1994), 

 it may also have endowed the chinook salmon of southern 

 British Columbia with a high level of adaptive diversity and 

 evolutionary potential. It is thus important to determine if 

 the regional structure evident within the Fraser drainage 

 represents an intermediate step in the erosion of genetic dif- 

 ferentiation among refugial groups by ongoing gene flow, or 

 in the differentiation of introgressed founding groups adapt- 

 ing to environmental variability. Comparison of the distribu- 

 tion of variation at neutral and adaptive loci, such as that 

 conducted for sockeye salmon of the Fraser River drainage 

 (Miller et al., 2001), may help determine the geographic 

 scale of adaptation for Fraser River chinook salmon. 



Acknowledgments 



We would like to acknowledge all those people involved 

 in spawning ground sample collections. Numerous people 

 from various groups were involved, including regional staff 

 R. Bailey and B. Rosenberger, and many staff of salmon 

 enhancement facilities throughout the drainage. Scale 

 samples from selected populations were provided to us by 

 D. Gillespie of the Ageing Laboratory at the Pacific Biologi- 

 cal Station. The manuscript was improved by constructive 

 comments from two referees. Funding was provided by 

 several sources from the Fisheries and Oceans Canada. 



Literature cited 



Banks, M. A., M S. Hlouin, B. A. Baldwin, V. K. Ra.shbrook. 

 H. A. Kitzf;('ral(i, S. M. Blanken.ship, and D. HedRecock. 



1999. Isiilation and inlicrilanco of novel niicrosatollitt's in 

 chinook .salmon iOncorhymliiiK tshaiiytachnl J. Horcd. 

 90: '281-288. 



