238 



Fishery Bulletin 101(2) 



intraregional variation among populations, and spawning 

 sites within a tributary system may or may not support 

 genetically discrete populations. Time of adult migration 

 also appears to promote genetic differentiation. 



The Birkenhead River population, the most distinctive 

 of the 52 individual chinook salmon spawning aggregates 

 surveyed, occupies the upper portion of the Harrison River- 

 Lillooet River drainage and is characterized by a very early 

 adult migration time. Historical recreational fisheries on 

 this population occurred in the Birkenhead River during 

 April and May (Fraser et al., 1982). Conversely, the Har- 

 rison River population in the lower reaches of the same 

 drainage is a late run, entering the Harrison River just 

 prior to spawning in October and November. The Birken- 

 head River population has low levels of polymorphism and 

 heterozygosity at both allozyme loci (Teel et al., 2000) and 

 microsatellite loci, likely reflecting small population size at 

 least in recent history and low levels of gene flow because 

 of its spatial and temporal isolation from other populations. 

 Another example of a spawning population isolated tempo- 

 rally from neighboring sites is the Louis Creek population 

 of the north Thompson region. The migration time of Louis 

 Creek chinook salmon is early or "spring run," whereas 

 other north Thompson populations are later "summer run" 

 migrants. The distant clustering of the Louis Creek chinook 

 salmon with the lower Thompson populations may reflect 

 either a common origin or more recent gene flow due to 

 common migration times. The low level of allelic diversity 

 observed in the Louis Crook population and its distinction 

 from the lower Thompson populations indicate that current 

 levels of gene flow are low, and the genetic similarity may 

 be due to common ancestry. 



Significant linkage disequilibrium was detected in 

 samples from four populations: Harrison River, Tete Jaune 

 (main stem Fraser River), Fontoniko Creek, and Bessette 

 Creek. Linkage disequilibrium may reflect sample admix- 

 ture (Waples and Smouse, 1990) in the Harrison River and 

 Tete Jaune samples. The Harrison River samples were 

 obtained from broodstock collections at a hatchery on the 

 Chehalis River, a tributary of the Harrison River Initial 

 broodstock for the hatchery was derived from chinook 

 salmon collected from the Harrison River, and over time 

 broodstock has been developed from fish returning to the 

 hatchery. Chinook salmon returning to the Chehalis hatch- 

 ery were also used to found the Chilliwack River popula- 

 tion, which is maintained by production in the Chilliwack 

 hatchery and spawning in the Chilliwack River During the 

 1990s, chinook salmon were transplanted back from the 

 Chilliwack hatchery to the Chehalis hatchery Thus, the 

 samples examined in our study, collected between 1988 and 

 1994, may reflect some mixing of genetically related but 

 heterogeneous groups offish from the Harrison, Chehalis, 

 and Chilliwack rivers in the Chehalis hatchery broodstock. 

 The Tete Jaune samples were obtained at Tete Jaune Cache 

 in the extreme headwaters of the Fraser River Because the 

 samples were collected from the mainstem Fraser River, 

 there is potential for admixture of populations, although it 

 is thought that there are few chinook salmon spawning sites 

 upstream from this location. Significant linkage disequilib- 

 rium was detected in single-year samples from Fontoniko 

 Creek (a tributary of the McGregor River) and Bessette 

 Creek (a tributary of the Shuswap River). Population ad- 

 mixtures would not typically be expected in such terminal 

 locations, and the cause of the disequilibrium is unknown. 



