618 
Fishery Bulletin 96(3), 1998 
Hudson River stock. However, genetic variation may 
have been pared further, i.e. the subsequent history 
of the Coos River population suggests that some of 
the demographic and life history factors that con- 
tribute to low levels of genetic variation among na- 
tive populations of striped bass were pronounced in 
the Coos River population. 
The Coos River population has experienced its own 
bottleneck; recent estimates suggest a reduction in 
its order of magnitude from 10 4 * to 10 3 or 10 2 . Fluctu- 
ating levels of annual spawning success also reduce 
the effective population size (N e ). Over generations 
N is approximated by the harmonic mean of each 
generation and strongly reflects periods of low abun- 
dance (Crow and Kimura, 1970). Many striped bass 
populations are sustained by occasional, extremely 
successful or “dominant” year classes (Raney, 1952). 
Dominant year classes have been rare but important 
for the Coos River population, occurring in 1940 and 
1958 (McGie and Mullen 4 ). 
Other factors that may have contributed to a re- 
duced N for the Coos River population are intrinsic 
to all populations of the species. Among these is 
skewed sex ratios. Males greatly outnumbered fe- 
males on the Coos River spawning grounds (Morgan 
and Gerlach 1 ). Estimates of male to female ratios on 
spawning grounds of other systems range from about 
10:1 to 100:1 (Chapman, 1990). Another factor is 
variance in progeny production among females, i.e. 
nonrandom family size (Gall, 1987). Large female 
striped bass can produce on the order of 10 6 eggs per 
year. Because of variable environmental conditions 
within a spawning season, some cohorts of eggs may 
show low or no survival whereas other cohorts may 
flourish (Dey, 1981; Secor and Houde, 1995). Thus, in- 
ordinate success by a few females would cause particu- 
lar genotypes to be overrepresented (Chapman, 1990). 
The significant difference in allele frequencies for 
nDNA locus 22 between the San Francisco Bay and 
Coos River collections indicates either a founder ef- 
fect or subsequent genetic drift. Also, the deviation 
from Hardy-Weinberg equilibrium of one of the three 
loci in the Coos River population is suggestive of 
small N e or some other violation of the assumptions 
that lead to Hardy-Weinberg frequencies (Weir, 
1990). In general, however, it appears that nDNA 
diversity was not strongly affected by the multiple 
bottlenecks that the Oregon populations experienced. 
In contrast, we have shown a stark decrease in 
mtDNA diversity. Our findings concerning nDNA and 
mtDNA diversity in Oregon striped bass are congru- 
4 McGie, A. M., and R. E. Mullen. 1979. Age, growth, and popu- 
lation trends of striped bass, Morone saxatilis, in Oregon. Or- 
egon Department of Fish and Wildlife Information Report Se- 
ries, Fisheries 79-8, 57 p. 
ent with their having experienced a recent popula- 
tion bottleneck. A single breeding pair of diploid ani- 
mals contains four nuclear genomes and one trans- 
missible mtDNA; thus, a population that goes 
through an extreme bottleneck can lose all of its 
mtDNA; variability while still retaining a significant 
fraction of its nuclear variability (Wilson et al., 1985). 
A pattern of highly reduced mtDNA diversity and 
little altered nDNA diversity is consistent with a re- 
cent and unprolonged population bottleneck, as we 
hypothesize to have occurred during the establishment 
and history of the Coos River striped bass population. 
There is an intriguing inverse relationship between 
genetic diversity (reflected in mtDNA) of the striped 
bass populations investigated and the frequency of 
hermaphroditism. Inbreeding depression has been 
firmly associated with detrimental effects in captive 
vertebrate populations (e.g. Kincaid, 1976, Laikre and 
Ryman, 1991), and recently, strong evidence of reduced 
fitness and reproductive impairments due to inbreed- 
ing depression in wild vertebrates has emerged (e.g. 
Jimenez et al., 1994; Keller et al., 1994; O’Brien, 1994). 
Moser et al. (1983) investigated the phenomenon 
of hermaphroditism in Coos River striped bass in 
some detail. Protandry was suspected because young 
hermaphrodites had ripe, motile sperm and imma- 
ture eggs, whereas older hermaphrodites (ages 7 to 
10) had normal appearing eggs and only small 
patches of testes. Reproductive impairment of older 
hermaphrodites was evident. Hermaphrodites with 
small testes and large ovaries showed annual accre- 
tions of eggs and one or more ovarian ducts blocked 
by adhesions. Each egg mass, representing previous 
spawning seasons, became progressively more degen- 
erated toward the interior of the gonad. Moser et al. 
(1983) also found that the oldest hermaphrodites had 
more constricted stomachs and intestines and more 
swollen abdomens than normal prespawning females. 
The four oldest hermaphrodites, 10 years old, had re- 
tained their eggs for up to 5 years. Also, it is possible 
that the absence of hermaphrodites among striped bass 
greater than 10 years of age (n= 7) indicates earlier 
mortality of hermaphroditic individuals, perhaps as a 
consequence of numerous annual egg mass accretions. 
We are aware of only a single observation of her- 
maphroditism among Atlantic coast striped bass. 
Westin (1978) reported one hermaphrodite among 
wild individuals from Chesapeake Bay that were 
sacrificed after being held in captivity for one year. 
However, in addition to surveying Coos River striped 
bass for evidence of hermaphroditism, Moser et al. 
(1983) also examined striped bass from San Fran- 
cisco Bay. Of more than 500 individuals, two were 
hermaphrodites. Thus, hermaphroditism in striped 
bass appears to be exceedingly rare in native, out- 
