230 



Fishery Bulletin 101(2) 



several interior populations of the South Thompson River 

 system, which produce substantial proportions of both 

 stream- and ocean-type juveniles (DFO'). 



Variability in chinook salmon flesh pigmentation levels 

 exceeds that of other Pacific salmonids, resulting from a 

 phenotypic dichotomy of "red-fleshed" and "white-fleshed" 

 forms that is under relatively simple genetic control (With- 

 ler, 1986; McCallum et al., 1987). Chinook salmon that 

 return to spawn in the Fraser River are mixtures of red- 

 and white-fleshed fish (Godfrey, 1975). The large Harrison 

 River population is considered to be entirely white-fleshed; 

 whereas red-fleshed fish predominate in populations of the 

 upper Fraser and Thompson drainages. Red-fleshed chi- 

 nook salmon are also abundant in mid-Fraser tributaries, 

 but a number of populations are polymorphic for flesh color 

 (e.g. the Quesnel River population consists of approximate- 

 ly equal proportions of the two flesh colors). 



Similar variation in migration timing and juvenile fresh 

 water residence times is observed in chinook salmon of 

 the Columbia River drainage (Myers et al., 1998). In that 

 system, adult migration time is not simply correlated with 

 juvenile life history type; spring-, summer-, and fall-run 

 populations in the lower and mid Columbia regions can all 

 be characterized by ocean-type juveniles. Genetic analysis 

 has indicated that, in the Columbia drainage, juvenile life 

 history type is a better indicator of genetic relationships 

 among populations than is adult migration time and sup- 

 ports the suggestion that the juvenile life history types 

 might represent separate "races" in that area (Myers et al., 

 1998). In British Columbia, which has been recolonized by 

 chinook and other salmon species in the last 10,000 years 

 since the Wisconsin glacial period from as many as four 

 possible refugial areas (Gharrett et al., 1987; Wilson et al., 

 1987; Utter et al., 1989; Cronin et al., 1993), the phyloge- 

 netic relationships of ocean- and stream-type populations 

 have yet to be determined and may vary on a geographic 

 basis dependent on both adaptation and the history of 

 colonization. 



In the Fraser River, a genetic demarcation between fish 

 of the lower and interior Fraser watersheds observed in 

 coho and sockeye salmon and attributed to independent 

 postglacial colonization of the two regions (Wehrhahn and 

 Powell, 1987; Wood et al., 1994; Small et al, 1998; Withler 

 et al., 2000), is also evident in chinook salmon (Beacham 

 et al., 1996; Teel et al., 2000; Nelson et al., 2001). Factors 

 underlying the substantial genetic differentiation among 

 chinook salmon occupying different areas of the interior 

 Fraser watershed, a region postulated to have been colo- 

 nized from the Columbia River refuge (Utter et al., 1989), 

 are less clear (Teel et al., 2000; Nelson et al., 2001 ). 



Differentiation among areas within the Thompson drain- 

 age, and between the interior Fraser and Thompson drain- 

 ages, has also been observed in coho and sockeye salmon. 

 Of the three species, the chinook salmon is the most 



' DFO (Department of Ki.sheries and Occan.sl. 199.5. Kraser 

 River chinook salmon. Kraser River Action Plan. Fishery Man- 

 afiement (jroup. Vancouver, British Columhia. Canada, 24 p. 

 jAvailahle from Fisheries and Oceans, .'i.'j.'') West Hasting St., 

 Suite 1220, Vancouver, British Columbia, Canada V6B 5G3.1 



extensively and continuously distributed throughout the 

 interior Fraser region. The variability in age of maturity, 

 both within and among populations, is greater in chinook 

 salmon than in coho or sockeye salmon — a factor that may 

 increase the effective number of spawning fish each year 

 and reduce genetic variability due to drift both within 

 and among populations of chinook salmon. On the other 

 hand, the development of population-specific anatomical 

 structures and life history traits indicate that recent gene 

 flow has been sufficiently restricted among populations to 

 enable strong adaptation (Taylor, 1991). 



The presence of ocean-type juveniles in the Eagle and 

 Shuswap River populations of the relatively warm and pro- 

 ductive South Thompson River drainage may represent a 

 relatively recent adaptive response to environmental con- 

 ditions enabling attainment of sufficient size in the first 

 spring of freshwater residence for juveniles to undertake 

 seaward migration. Conversely, Teel et al. (2000) suggested 

 that chinook salmon of the south Thompson region were 

 genetically intermediate to those of the mid-upper and 

 lower Fraser areas and might represent hybridization of 

 stream- and ocean-type races that independently colonized 

 the lower and interior portions of the Fraser drainage. 



Conservation of chinook salmon genetic diversity within 

 the Fraser River requires delineation of the phylogeneti- 

 cally and adaptively distinct groups within the drainage, 

 an understanding of their origins and the evolutionary 

 processes promoting and maintaining their differentia- 

 tion, and the ability to manage them on an independent 

 basis. Similar zoogeographic factors will have influenced 

 the coho, chinook, and sockeye populations recolonizing 

 the Fraser drainage, although the differing utilization of 

 glacial refugia, environmental requirements, and propen- 

 sities for homing and straying may have resulted in quite 

 different evolutionary responses among them. 



The objective of the present study was to analyze 

 variation at 13 microsatellite loci in 52 samples of chinook 

 salmon from the Fraser and Thompson River drainages to 

 determine population structure. By detecting differentia- 

 tion in allelic frequencies, levels of allelic diversity, and the 

 presence of unique alleles, we were able to use the high 

 levels of polymorphism and heterozygosity at the micro- 

 satellite loci to indicate the relationships among chinook 

 salmon occupying different regions of the Fraser drainage. 

 The distribution of genetic diversity in the Fraser River 

 drainage among regions, populations, and sampling years 

 is estimated, as well as the stability of population structure 

 within major tributaries of the Fraser Riven We also exam- 

 ined the patterns of divergence within the drainage with 

 respect to ocean- and stream-type life histories. 



Materials and methods 



Collection of baseline DNA samples 

 and laboratory analysis 



Genomic DNA was extracted from either liver, scales, oper- 

 culimi punches, or fin clips from chinook salmon sampled 

 between 1987 and 1998 by using the phenol-chloroform 



