Nelson et al.: Population structure of Oncorhynchus tshawytscha of the Fraser River 
101 
' 
Table 1 (continued) 
j 
r 
Chehalis Red 
30 
Bowron 
50 
Holmes 
44 
Indianpoint 
40 
Slim 
66 
UFr 
521 
L Shuswap 
185 
M Shuswap 
167 
Eagle 
37 
Coldwater 
54 
Nicola 
231 
Thompson 
674 
0.033 
0.02 
0 
0.125 
0.038 
0.047 
0.016 
0.006 
0 
0.019 
0.006 
0.01 
0.2 
0.01 
0.136 
0.013 
0.053 
0.102 
0.005 
0 
0 
0 
0.009 
0.004 
0.017 
0.01 
0.034 
0.025 
0.008 
0.046 
0.014 
0.003 
0 
0 
0.006 
0.007 
0.083 
0 
0.034 
0.025 
0.03 
0.039 
0.008 
0.009 
0 
0 
0.004 
0.006 
0.017 
0.01 
0.023 
0.025 
0.03 
0.073 
0.016 
0.009 
0 
0 
0 
0.007 
0 
0.03 
0.023 
0 
0.008 
0.016 
0.016 
0.006 
0.014 
0 
0.002 
0.007 
0 
0 
0 
0 
0 
0.002 
0.019 
0.003 
0 
0 
0 
0.006 
0.017 
0 
0 
0 
0 
0.001 
0.003 
0.006 
0.014 
0 
0.004 
0.004 
0 
0 
0 
0 
0 
0 
0.016 
0.015 
0 
0 
0 
0.008 
0 
0 
0 
0 
0 
0 
0.011 
0.021 
0.081 
0.009 
0 
0.013 
0 
0 
0 
0.025 
0.023 
0.007 
0.103 
0.066 
0.068 
0 
0 
0.048 
0 
0.01 
0 
0.013 
0.038 
0.009 
0.116 
0.096 
0.095 
0 
0.011 
0.065 
0 
0 
0 
0 
0.008 
0.003 
0.057 
0.093 
0.041 
0.111 
0.089 
0.08 
0 
0.01 
0 
0 
0 
0.001 
0.041 
0.027 
0 
0.046 
0.065 
0.044 
0 
0 
0 
0 
0 
0 
0.016 
0.003 
0.027 
0 
0 
0.007 
0 
0 
0 
0 
0 
0 
0.038 
0.003 
0.041 
0 
0 
0.013 
0 
0 
0 
0 
0 
0 
0.038 
0.003 
0 
0 
0 
0.011 
0.9 
0.8 
0.75 
0.875 
0.829 
0.808 
0.535* 
0.575* 
0.297* 
0.796 
0.68* 
0.602 
0.9 
0.88 
0.886 
0.925 
0.857 
0.896 
0.941 
0.898 
0.92 
0.889 
0.883 
0.933 
dard deviations were calculated for each locus and for all 
loci combined according to Weir and Cockerham ( 1984) by 
using FSTAT (Goudet, 1995). FSTAT also provided pair- 
wise values for populations and regions. We used the 
notations F„F ,andF for Weir and Cockerham’s ( 1984) 6, f, 
and F, respectively. In all analyses, probability values were 
adjusted to correct for the number of simultaneous tests as 
discussed by Lessios (1992). Thus, the significance level is 
taken to be 0.05/number of simultaneous comparisons. 
Graphical analysis of genetic relationship between pop- 
ulations was performed. A neighbor-joining (NJ) (Saitou 
and Nei, 1987) dendrogram was constructed by using PHY- 
LIP 3.5c (Felsenstein 1 ). The allele frequency matrix was 
resampled 1000 times in a bootstrap resampling proce- 
dure and Cavalli-Sforza and Edwards (1967) chord dis- 
tances among populations were estimated for each matrix. 
A consensus NJ dendrogram of chord distances was gener- 
ated to determine the stability of nodes within the dendro- 
gram. For presentation, a NJ dendrogram with the origi- 
nal branch lengths was constructed and bootstrap values 
over 50% were plotted onto the nodes of the dendrogram. 
In several populations we suspected a null allele at 
Ots 102. These populations had moderate heterozygote de- 
ficiencies and a large number of individuals whose other 
two loci amplified but that failed to amplify at Ots 102, 
even after PGR stringency was relaxed. We generated an 
estimate of the frequency of the null allele by recoding, as 
null allele homozygotes, two out of every three of the indi- 
viduals failing to amplify at Ots 102. Corrected allele fre- 
Table 3 
Observed precision of allele size determination (bp) for 
repeated analysis of heterozygous fish at loci Ots 101, 
OtslOO, and Otsl02. “n” is the number of times each fish 
was analyzed. Standard deviation (SD) was calculated 
and the range of observed measurements is shown. 
Locus 
n 
Mean 
SD 
Range 
OtslOl 
44 
221.77 
1.08 
220-224 
103 
199.47 
0.96 
196-201 
44 
167.93 
0.79 
166-169 
103 
153.49 
0.94 
151-155 
Ots 100 
18 
367.17 
1.86 
360-367 
103 
322.07 
1.73 
318-325 
18 
281.28 
1.23 
279-283 
103 
251.81 
1.13 
249-254 
Ots 102 
74 
270.35 
1 34 
268-273 
70 
225.54 
1.10 
223-228 
70 
188.30 
1.08 
186-190 
74 
181.15 
0.86 
180-183 
1 Felsenstein, J. 1993. PHYLIP (Phylogeny Inference Package), 
version 3.4. Univ. Washington, Seattle, WA. (Available from 
author, Department of Genetics, Box 357360, Univ. Washington, 
Seattle, WA 98195-7360.] 
