616 
Fishery Bulletin 96(3), 1 998 
Table 1 
Genotypic frequencies for three single copy, nuclear DNA loci. Heterozygosity values in parentheses. 
Locus 25 Dra I Locus 27 EcoR V Locus 22 Hinf I 
Population 
N 
AA 
AB 
BB 
N 
AA 
AB 
BB 
N 
AA 
AB 
BB 
Hudson River 
35 
0 
11 
24 
123 
6 
42 
75 
80 
47 
24 
9 
(0.314) 
(0.342) 
(0.300) 
San Francisco Bay 
64 
6 
23 
35 
49 
3 
22 
24 
65 
41 
20 
4 
(0.359) 
(0.449) 
(0.308) 
Coos River (wild) 
27 
1 
12 
14 
32 
3 
14 
15 
34 
11 
9 
14 
(0.444) 
(0.438) 
(0.265) 
Coos River (hatchery) 
23 
1 
11 
11 
25 
6 
7 
12 
19 
6 
5 
8 
(0.478) (0.280) (0.263) 
radiolabelled DNA probes (Feinberg and Vogelstein, 
1983). Each of the three enzymes generates a diag- 
nostic fragment which was used to characterize 
mtDNA major length-variant haplotypes (defined as 
differences of more than 100 base pairs; Wirgin et 
al., 1990). In addition, Acc I digestion revealed an in- 
formative, single base substitution. Probes were either 
highly purified mtDNA isolated from a single striped 
bass liver (Wirgin et al., 1990) or a gel-purified 1.7 kb 
PCR product containing the striped bass mtDNA con- 
trol region (Wirgin et al., 1995). To determine nDNA 
genotypes, DNAs (10 pg) were digested with Dra I, 
EcoRV, and Hinf I, and analyzed in Southern blot 
analysis with the single copy probes developed from a 
striped bass genomic DNA library: DSB 25, DSB 27, 
and DSB 22, respectively (Wirgin and Maceda, 1991). 
Each of these enzyme-probe combinations revealed a 
single restriction site polymorphism with two alleles. 
Analysis of controlled laboratory matings demonstrated 
the Mendelian inheritance and nonlinkage of loci. 
Genotypic frequencies derived from nDNA analy- 
sis were tested for deviations from Hardy-Weinberg 
equilibrium with the disequilibrium coefficient ap- 
proach (Weir, 1990). Mitochondrial DNA haplotype 
diversity was calculated with the formula of Nei and 
Tajima ( 1981). Chi-square significance (P<0.05) of the 
differences between striped bass populations in 
nDNA allele frequencies and mtDNA haplotype fre- 
quencies was tested by using the randomization ap- 
proach of Roff and Bentzen ( 1989). 
Results 
Nuclear DNA 
River (wild and hatchery) did not deviate significantly 
(P>0.05) from Hardy-Weinberg equilibrium; however, 
locus 22 did differ significantly from Hardy-Weinberg 
equilibrium for the Coos River wild (P<0.01) and hatch- 
ery (P<0.05) samples. Allelic frequencies for the three 
nDNA loci did not differ significantly (P>0.05) between 
the Hudson River and San Francisco Bay collections. 
Of the three nDNA loci, only locus 22 differed signifi- 
cantly in allelic frequencies between the San Francisco 
Bay and Coos River collections (% 2 =19.21; P<0.0001). 
Mitochondrial DNA 
Mitochondrial DNA haplotypic diversity (based on 
mtDNA length variants) showed a clear pattern of re- 
duction (0.810 to 0.0) among the striped bass popula- 
tions along the historical path that led to and includes 
the wild Coos River population (Table 2). The number 
of mtDNA haplotypes revealed decreased from 8 among 
Hudson River specimens to 5 in the San Francisco Bay 
collection, to 2 in the Umpqua, and 1 in the wild and 
hatchery-cultured Coos River samples. The third most 
common haplotype (C-l) found in the Hudson River 
collection (17%) was observed in the great majority of 
San Francisco Bay specimens (81%) and in all Coos 
River specimens. Also, the three least common haplotypes 
in Hudson River striped bass were those absent in striped 
bass from San Francisco Bay. Mitochondrial DNAhap- 
lotype frequencies were significantly different between 
the Hudson River and San Francisco Bay samples 
(^ 2 =71.47; P<0.0001) and between the San Francisco 
Bay and Coos River samples (% 2 =8.21; P<0.05). 
Discussion 
Genotypic frequencies (Table 1 ) at two loci of samples Extensive allelic surveys of Atlantic coast striped 
from the Hudson River, San Francisco Bay, and Coos bass have shown extreme monomorphism at the pro- 
