236 



Fishery Bulletin 101(2) 



Portage s 



Harrison o 

 — Chilllwack o 



Raft s 



I — Bndge s 



Salmon ST s 



 Eagle so 

 Lower Shuswap 

 " " ' 'le Shuswap o 

 Bessette S 

 South Thompson 

 - Lower Adams o 

 Little 



— Finn S 

 Mahood s 



Cleanwater s 



I — Westroad s 



I Endako s 



A I Elkin s 



H I Chiiko s 



I Taseko S 



I Upper Chilcotin s 



— I Chilcotin s 

 (— Lower Chilcotin s 



M| Cariboo s 



4 I — Quesnel S 

 Hj— Stuart s 

 ^ Nechako s 



I Cottonwood s 



L Horsefly S 

 — Salmon UF s 

 Willow s 



Dome S 

 Goat s 



Holmes s 



Spius s 

 — Nicola s 

 Coldwater s 



Deadman s 



Bonaparte s 



Louis s 



I Lower Fraser 



South Thompson 



North Thompson 



Lower Thompson 



Birkenhead 

 Birkenhead I'^S 





Middle Fraser 



Horsey s 

 Tete Jbune s 



I — Fontoniko s 

 I McGregor s 



i- Bowron s 



l]i- Indian s 



1i- Chillrwack red s 



Hj— Chehalis red s 

 T- Slim s 



Swift s 



Upper Fraser 



0.1 



Figure 2 



Neighbor-joining dendrogram of Fgj- differences for 52 spawning groups of Fraser River chinook salmon surveyed at 

 12 microsatellite loci in Hardy- Weinberg equilibrium. Bootstrap values at the tree nodes indicate the percentage of 

 100 trees where populations beyond the node occurred together. Populations have been designated as ocean type (O) 

 or stream type (S) after the name of the population. 



to be more similar to each other but that the degree of 

 differentiation among populations within tributaries 

 varied among tributary systems (Table 61. For example, 

 in the Nicola River drainage, temporal variation in allele 

 frequencies within the Nicola and Coldwater samples was 

 about three times larger than differences between them 

 (Table 6). There was no significant difference in allele fre- 

 quencies at any locus between the two locations in relation 

 to annual variation within location (all 13 P>0.10). Thus 

 it seems likely that there is sufficient gene flow between 

 these two sites (and likely the Spius Creek site) that the 

 Nicola River drainage can be considered to contain a single 

 chinook salmon population. Significant differentiation was 

 observed between the lower and middle Shuswap River 

 samples at Og(>2 (F=7.82, P<0.05), indicating that there 

 has been some restriction of gene flow between the two 

 spawning locations. Structure was observed among the 

 three sites sampled from the Nechako River system, with 

 significant differences observed at Oki'4. Of^o'2. and Ots9 

 (all F>6.0, P<0.05l. Similarly, afti-r accounting for annual 

 variation, significant differentiation between the Quesnel 



and Horsefly sites in the Quesnel River drainage was 

 observed at Ogo2 (F=6.63, P<0.05), and among four Chiiko 

 River sites at Otsl07, Ogo4, Ots9, and Omy325 (all >6.6, all 

 P<0.05). The most pronounced allele frequency differentia- 

 tion was observed between the Harrison and Birkenhead 

 River populations in the Harrison River-Lillooet River 

 drainage. Interpopulation differentiation was 24 times 

 greater than intrapopulation variation; the two popula- 

 tions differed in allele frequencies at all loci except Ssa 297 

 and OtslOl (all F>6.6. all P<0.05). 



Discussion 



Population structure 



The genetic structure of Fraser drainage chinook salmon 

 populations has a strong geographic basis, indicating that 

 isolation by distance is a major component of population 

 structure. Regional differentiation accounts for approxi- 

 mately twice the variation in allele frequencies as does 



