Gall et al : Geographic variation in population genetics of Oncorhynchus tshawytscha 



83 



S Oregon/ 

 N. Cahl 



Omagar Creek 



Eel River/ 

 Calif Coast 



KlafTialh/ 

 Trinity 



Mid Oregon 



Sacramento/ 

 San Joaauin 



1 000 996 992 98B 



Genetic Identity 



Figure 2 



Dendrogram based on UPGMA clustering of genetic identi- 

 ty indices (Nei 1972). Identification numbers are defined in 

 Table 1. Brackets on left side indicate geographic grouping, 

 with Blue Creek and Omagar Creek as outliers (collection #6, 

 indicated as 6*, was from mid-Oregon). 



and comparable to the middle Oregon area which 

 shared the highest total gene diversity (Table 3). 



Based on an overall estimate of 0.106 for Gst (Table 

 3), the number of immigrant individuals contributing 

 genes to an average population, Nm, was estimated to 

 be 2.11 individuals per generation. Estimates of gene 

 flow within each geographic cluster were highest in the 

 Sacramento-San Joaquin system (Nm 15.57) and low- 

 est in the Klamath-Trinity drainage (Nm 3.97). 



Discussion 



The genetic structure of chinook salmon populations 

 reported here appears similar to that reported pre- 

 viously. Distributions of variant alleles at Mdh-4, AH-1 

 Pgdh, Pgm-2, GPI-H, and Gpi-2 were similar to those 

 reported by Bartley and Gall (1990). However, average 

 heterozygosity estimates for the Klamath-Trinity 



Table 3 



Hierarchical gene diversity analyses of 37 samples of chinook 

 salmon from Oregon and California.* HgQ = average gene 

 diversity of samples within areas; Hp and Gjq = total gene 

 diversity and relative gene diversity for a given area, respec- 

 tively; Nm = average number of migrants exchanging genes 

 per generation; H,, Hf, and Gg^ = within-sample, total, and 

 relative gene diversity, respectively. 



0.106, Nm 2.11 



drainage were somewhat higher than reported by Utter 

 et al. (1989) and Bartley and Gall (1990). Bartley and 

 Gall (1990) observed a range of 0.008-0.016 for this 

 drainage, compared with the range of 0.028 for the 

 Shasta River sample to 0.064 for the sample from 

 Omagar Creek found in the present study. One reason 

 for the higher estimates in the present study was the 

 inclusion of the Mdhp-2 locus, which is highly polymor- 

 phic in the Klamath-Trinity drainage (Appendix A); 

 Bartley and Gall (1990) and Utter et al. (1989) did not 

 report data for this locus. Generally, comparisons of 

 heterozygosity estimates between this study and earlier 

 studies are difficult to interpret due to the improved 

 laboratory procedures that have greatly increased the 

 number of isozyme loci available for analysis. 



Two samples from the Klamath-Trinity drainage. 

 Blue and Omagar Creeks, were genetically differen- 

 tiated from other samples from within the basin. For 

 example, Mdhp-2(78) had an average frequency of 0.32 

 in eight other samples from the drainage, whereas the 

 allele occurred at a frequency of 0.14 in Blue Creek and 

 was not found in the Omagar Creek sample. Further- 

 more, Omagar and Blue Creeks had higher frequencies 

 of the Tapep-l(130) and mMdh-l(-900) alleles than did 

 other Klamath-Trinity samples. These frequencies in- 

 dicated that fish from Omagar and Blue Creeks are 

 genetically closer to southern Oregon populations than 

 to Klamath-Trinity populations. This result was unex- 

 pected given the pattern of geographic clustering foimd 

 by Utter et al. (1989) and Bartley and Gall (1990). 

 However, earlier studies did not sample populations 

 near or below the confluence of the Trinity and 

 Klamath Rivers, as was done in the present study. 



