754 
Fishery Bulletin 95(4), 1 997 
1989 year class 
Hatching Effective swimmers 
100-i 1 -i 
i i i i i I i i i i i I r 
SMKFCRY SMKFCRY 
Site 
Figure 3 
For the 1989 year class, mean (±SE) percent hatching, live, effective swim- 
mers, and spinal abnormalities of larval Pacific herring by site and region in 
Alaska, 1995. The 1989 year class, sampled in Prince William Sound in 1995, 
was more likely exposed to oil as eggs or larvae than were other year classes. 
S = Sitka, M = St. Mathews Bay, K = Ketchikan, F = Fish Bay, C = Port 
Chalmers, R = Rocky Bay, Y = Seymour Canal. Sample size is shown in hatch- 
ing graph. Progeny of the 1989 year class did not differ significantly between 
regions for any reproductive parameter (P>0.50). 
(<16%), and differences among sites 
and between regions were not signifi- 
cant (P>0.348) (Fig. 4). 
Yolk volume in larvae from the 
1989 year class did not differ signifi- 
cantly (P=0.486) between regions but 
may have been related to incubation 
temperature. The largest and small- 
est mean yolk volumes were observed 
in PWS but closely overlapped those 
in SE (Fig. 4). Although scatter was 
high (r 2 =0.13), yolk volumes declined 
significantly (P<0.001) as tempera- 
ture increased. It is possible, how- 
ever, that site differences and incu- 
bation temperature were confound- 
ing factors. 
Comparison among age classes 
within sites 
Reproductive success differed signifi- 
cantly among some age classes at 
Sitka, Ketchikan, Port Chalmers, 
and Rocky Bay but not among age 
classes at St. Mathews Bay, Fish Bay, 
and Seymour Canal (Figs. 5-8). The 
few significant differences we ob- 
served were highly variable, incon- 
sistent among sites, and no pattern 
existed for the 1989 year class. For 
example, at Rocky Bay, age-3 and 
age-4 fish had a significantly lower 
percentage of live larvae than age-5, 
age-6, and age-7 fish (Fig. 6), whereas 
at Sitka, age-3 and age-4 fish had a 
significantly higher percentage of 
effective swimmers and a signifi- 
cantly lower percentage of spinal abnormalities than 
age-7 fish (Figs. 7 and 8). 
Other parameters 
Hatching times decreased steadily with increasing 
incubation temperature (Fig. 9). For Sitka, the first 
site sampled, peak hatching occurred about 33 d af- 
ter start of incubation at a mean temperature of about 
4.5°C, whereas at Seymour Canal, the last site 
sampled, peak hatching occurred about 26 d after 
start of incubation at a mean temperature of about 
6.0°C. 
Fertility did not differ significantly (P>0.50) be- 
tween regions for all ages combined or when the com- 
parison was restricted to fish of the same age. For 
all ages combined, fertility in SE ranged from 80% 
at Seymour Canal to 96% at Sitka; in PWS, fertility 
ranged from 88% at Fish Bay to 94% at St. Mathews 
Bay. 
Discussion 
Six years after the spill, reproductive impairment 
was not detected in PWS herring. This conclusion 
was reached by comparing reproductive success of 
fish collected in PWS and SE and among age classes 
within specific sites. Specifically, hatching success, 
larval viability, and fertility did not differ signifi- 
cantly between PWS and SE, including response of 
the 1989 year class. In fact, discrimination of re- 
sponses between regions was not possible because 
the best and worst responses were usually found in 
