106 
Fishery Bulletin 99(1 ) 
Table 5 
Table of pairwise F st values among major regions in the 
Fraser River drainage. 
Lower Fraser 
Mid Fraser 
Upper Fraser 
Lower Fraser 
0.0000 
Mid Fraser 
0.0115 
0.0000 
Upper Fraser 
0.0147 
0.0062 
0.0000 
Thompson 
0.0146 
0.0125 
0.0161 
rather reflect patterns of historical colonization and pres- 
ent day gene flow (or lack thereof). Concordance of both 
allozyme and microsatellite data indicates that the simple 
method used in our study to designate alleles does not 
greatly bias or skew the results. 
During the retreat of the Wisconsin glacial ice sheet, 
the headwaters of the Fraser and Thompson Rivers were 
ice free before the lower Fraser River channel was ice 
free, and therefore drained through the Columbia River. 
McPhail and Lindsey (1986) suggested that freshwater 
fish colonized the upper Fraser River and the Thompson 
River by means of the Columbia River during this time. Al- 
lozyme analysis of chinook salmon show that the Thomp- 
son River populations are distinct from the population of 
the other Fraser River tributaries (Utter et al., 1989; Teel 
et al., 2000). Similarly, coho salmon from the Thompson 
River are genetically distinct from coho salmon in the low- 
er Fraser River (Small et al., 1998a). In our analysis, the 
Thompson River populations formed a distinct group, con- 
sistent with the hypothesis that the Thompson River wa- 
tershed may have been colonized by a different founder 
group than other regions of the Fraser River. In addition, 
strong genetic substructuring within the Thompson River 
watershed was observed. This structuring, also observed 
by Teel et al. (2000), suggests that there may be sufficient 
genetic isolation within the Thompson River watershed to 
allow for the persistence of locally adapted populations. 
If the upper Fraser and Thompson Rivers were both col- 
onized by means of the Columbia River, then tests of the 
genetic relationship between these groups might show the 
upper Fraser and Thompson Rivers more closely related to 
each other than to other regions. However, based upon our 
study, they are the most distantly related, suggesting that 
either the upper Fraser was not colonized by the same 
population that founded the Thompson River populations, 
or that migration may have obscured the origins of fish 
inhabiting this region. This hypothesis is currently being 
tested by analyzing chinook salmon populations from the 
Skeena and Nass Rivers (possible source populations) and 
by increasing the resolution of the genetic structure of 
coho salmon in the Fraser River watershed by analyzing 
more microsatellite loci. 
The high bootstrap support for the lower Fraser group 
suggests that this region was colonized by a single founder 
population. This hypothesis seems likely because the riv- 
er mouths are separated by approximately 15 kilometers. 
The close genetic association between the Harrison and 
Chilliwack Rivers (F s/ =0.003) indicates either that stray- 
ing occurs routinely between them or that colonization 
was so recent that the populations have not diverged. 
The red-flesh Chehalis and Chilliwack populations were 
introduced in the 1980s from broodstocks originating in 
the upper and middle Fraser River. Sources of the brood 
stocks included the Bowron, Chilko, and Quesnel Rivers 
and Slim Creek 2 These populations are artificially main- 
tained by selecting red-flesh fish for broodstock. Our anal- 
ysis places these populations with the upper Fraser River 
populations, reflecting their origins. 
The regional groupings and patterns of genetic relation- 
ships within each region provide a starting point for dis- 
cussion of the events that led to the repopulating of these 
regions by chinook salmon and the degree of isolation of dif- 
ferent populations. Although only in the early stages, this 
information forms a base upon which to begin assigning 
management and fishery enhancement priorities such that 
genetic diversity present in wild populations is preserved. 
This information, along with life history and ecological da- 
ta, will also be useful for the determination of whether a 
regional grouping of populations can be considered an evo- 
lutionarily significant unit (Waples, 1991). 
Acknowledgments 
Samples were collected with the assistance of J. Candy 
and the staff of the Habitat and Assessment Branch of the 
Department of Fisheries and Oceans. Funding was provided 
by the Department of Fisheries and Oceans. C. Wood and 
R. E. Withler provided helpful discussion. We thank five 
anonymous reviewers for improvements to the manuscript. 
Literature cited 
Angers, B., L. Bernatchez, A. Angers, and L. Desgroseillers. 
1995. Specific microsatellite loci for brook charr reveal 
strong population subdivision on a microgeographic scale. 
J. Fish. Biol, (suppl A) 47:177-185. 
Banks, M. A., B. A. Baldwin, and D. Hedgecock. 
1996. Research on chinook salmon ( Oncorhynchus tshaw- 
ytscha) stock structure using microsatellite DNA. Bull. 
Natl. Inst. Aquacult. suppl. 2:5-9. 
Bartley, D., B. Bentley, J. Brodziak, R. Gomulkiewicz, M. Mangel, 
and G. A. E. Gall. 
1992. Geographic variation in population genetic structure 
of chinook salmon from California and Oregon. Fish. Bull. 
90:77-100. 
Beacham, T. D., C. B. Murray, and R. E. Withler. 
1989. Age, morphology and biochemical genetic variation of 
Yukon River chinook salmon. Trans. Am. Fish. Soc. 118: 
46-63. 
Beacham, T. D., R. E. Withler, and T. A. Stevens. 
1996. Stock identification of chinook salmon ( Oncorhynchus 
tshawytscha) using minisatellite DNA variation. Can. J. 
Fish. Aquat. Sci. 53:380-394. 
2 Cross, C. 1998. Personal commun. Department of Fisheries 
and Oceans, 555 West Hastings St. Vancouver, B.C., V6B 5G3, 
Canada. 
