Waldman et a!.: Population bottlenecks and DNA diversity in Morone saxatilis 
615 
By the mid-1920s, striped bass were being commer- 
cially harvested in Coos Bay, and in 1945, annual 
landings from Coos Bay reached a high of 231,000 lb. 
The adult population also appeared to peak in 1945 
at about 69,000 individuals (>age-3), based on catch- 
per-unit-of-effort sampling. Pathological hermaph- 
roditism was noted, but it was rare among Coos River 
striped bass during this period; Morgan and Gerlach 1 
reported a 3% (n=124) incidence in 1950. 
Since 1945, the Coos River striped bass popula- 
tion has crashed, whereas the incidence of hermaph- 
roditism has increased dramatically. Between 1950 
and 1975, population estimates of adults ranged from 
as many as 43,000 in 1963 to as few as 7800 in 1973. 
No adult population estimate is available for 1980, 
but in that year Moser et al. (1983) found 11 of 42 
(26%) wild fish to be hermaphrodites. Population size 
of Coos River adult striped bass was not evaluated 
again until 1988 and 1989, when estimates of be- 
tween 1000 and 3000 for both years were obtained. 
Estimates to date for the 1990s are of an adult popu- 
lation size under 1000. Furthermore, virtually no 
natural recruitment appears to be occurring (but 
supplemention is occurring through stocking of 
hatchery-produced offspring of San Francisco Bay 
broodstock). A standardized seine-haul survey of ju- 
venile production begun in 1978 showed a decline in 
catch-per-unit-of-effort from 2.9 in 1978 to between 
0.3 and 0.1 until 1986, and then only infinitesimal 
levels or zero through 1995. Recent estimates are that 
hermaphrodites make up 30% of the naturally pro- 
duced adult population in the Coos River system 
(Reimers et al. 2 ). Additionally, during 1993, an angling 
2 Reimers, P. E., R. E. Bender, J. A. Johnson, T. Rumreich, D. J. 
Van Dyke, T. A. Confer, R. C. Smith, J. A. Hutado, and R. S. 
Boots. 1990. Tenmile-Coos-Coquille Fish District: a review 
of stocks of concern. State conservation department report, 
Oregon Department of Fish and Wildlife, 41 p. 
guide captured a hermaphroditic striped bass from the 
Umpqua River, the first reported from that system. 
We hypothesized that Coos River striped bass 
would show reduced genetic diversity because both 
the history of the population’s establishment and its 
subsequent demographics favored inbreeding. To test 
this hypothesis, we surveyed both nuclear DNA 
(nBNA) and mitochondrial DNA (mtBNA) variation 
in the Coos River population, the similarly non-na- 
tive San Francisco Bay population (the source of the 
Oregon populations), and the source for the San Fran- 
cisco Bay population, the Hudson River, New York. 
We also examined mtBNA variation in a second Or- 
egon population (Umpqua River). 
Methods 
Coos River striped bass (both wild fish and fish that 
were originally hatchery-cultured) were captured in 
gill nets during spring 1993. These fish were distin- 
guished on the basis of size; significant stocking of 
hatchery-cultured striped bass (age-0 only) did not 
begin until 1989. Only wild Oregon striped bass were 
used as broodstock until 1991, when broodstock from 
California were used. Umpqua River specimens were 
collected in 1992 by angling. San Francisco Bay 
samples were collected by means of gillnetting in the 
lower Sacramento and San Joaquin rivers during 1991 
and 1992. Collections of striped bass from the Hudson 
River are described in Wirgin et al. (1990, 1993). 
Total BNA was isolated from livers or blood by the 
CTAB method (Saghai-Maroof et al., 1984; Wirgin et 
al., 1990), phenol-chloroform extractions, and etha- 
nol precipitations. To determine mtBNA haplotypes, 
BNAs were digested with Acc I, Hind III, and Rsa I, 
electrophoretically separated in 1.2% agarose gels, 
and visualized in Southern blot analyses by using 32 P 
