Johnson et al.: Reproductive success of Clupea pallasi after the Exxon Valdez oil spill 
755 
1 989 year class 
Yolk-sac edema P=0.348 
Figure 4 
For the 1989 year class, mean (±SE) per- 
cent yolk-sac abnormalities (edema) and 
yolk volume for larval Pacific herring by 
site and region in Alaska, 1995. Sample 
size is shown in each bar. S = Sitka, M = 
St. Mathews Bay, K= Ketchikan, F = Fish 
Bay, C = Port Chalmers, R = Rocky Bay, Y 
= Seymour Canal. No significant differ- 
ences existed between regions for either 
parameter (P>0.348). 
one region (SE). Therefore, the chances of detecting 
any oil-related effects against the natural background 
variation were negligible when herring were com- 
pared between regions. Although responses among 
some age classes within Port Chalmers and Rocky 
Bay were occasionally significant, these differences 
were highly variable, did not indicate reproductive 
impairment of the 1989 year class, and were incon- 
sistent between sites. 
Of the four key reproductive parameters we ex- 
amined, spinal defects were particularly important 
because exposure of herring eggs to oil frequently 
causes spinal defects (Linden, 1978; Kocan et ah, 
1987; Rice et al., 1987; Pearson et al. 6 ), that could 
6 Pearson, W. H., D. L. Woodruff, S. L. Kiesser, G. W. Fellingham, 
and R. A. Elston. 1985. Oil effects on spawning behavior and 
reproduction in Pacific herring ( Clupea harengus pallasi). Fi- 
nal Report OF-1742 to American Petroleum Inst., Battelle Ma- 
rine Res. Lab., Sequim, WA, 108 p. [API publication 4412.] 
result in reduced swimming ability and long-term 
survival. Spinal defects, however, can also occur natu- 
rally as a result of other environmental factors. In 
our study, herring from an uncontaminated site, 
Ketchikan, had the highest percentage of spinal de- 
fects (7%). Ketchikan samples were collected at least 
40 km from any urban area, and it is unlikely that 
these fish were exposed to industrial or other urban 
pollutants. Whether the incidence of spinal defects 
at Ketchikan was just random noise or a response to 
some underlying environmental factor is impossible 
to determine, but it is evidence that similar results 
could occur in PWS without implicating oil as a cause. 
In fact, a 10% incidence of gross abnormalities was 
observed in PWS herring 23 years prior to the spill 
(Smith and Cameron, 1979). 
Reproductive success of herring in PWS was con- 
sistently better in 1995 than that reported in earlier 
studies. For example, we observed a mean hatching 
success of 78-86% compared with 53% in 1976 (Smith 
and Cameron, 1979), 62% 7 in 1989 (McGurk et al. 8 ), 
85% in 1990 (McGurk et al. 9 ), 59-79% in 1991 (Kocan 
et al., 1996a), and 19-56% in 1992 (Kocan et al., 
1996b). The viable hatching 10 that we observed in 
PWS (79%) also exceeded previously reported per- 
centages; 53% n in 1989 (McGurk et al. 8 ), 57% in 1990 
(McGurk et al. 9 ), 35-37% 12 in 1991 (Kocan et al., 
1996a), and 13-33% 12 in 1992 (Kocan et al., 1996b). 
Incidence of spinal abnormalities in PWS was about 
5% in our study compared with 7% in 1989 (McGurk 
7 To avoid desiccation effects, and because egg survival was sig- 
nificantly less in the +1.5-m collections in the McGurk et al. 8 
data set, these data were not included in this comparison. Es- 
timated egg survival was 59% when the +1.5-m data were in- 
cluded. 
8 McGurk, M., D. Warburton, T. Parker, and M. Litke. 1990. 
Early life history of Pacific herring: 1989 Prince William Sound 
herring egg incubation experiment. Final report, contract 
number 50ABNC-7-00141, Triton Environmental Consultants 
LTD., No. 120-13511 Commerce Parkway, Richmond, British 
Columbia, Canada V6V 2L1. 
9 McGurk, M., T. Watson, D. Tesch, B. Mattock, and S. 
Northrup. 1991. Viable hatch of Pacific herring eggs from 
Prince William Sound and Sitka Sound, Alaska, in 1990. Re- 
port number 2060/WP 4269, Triton Environmental Consult- 
ants LTD., No. 120-13511 Commerce Parkway, Richmond, Brit- 
ish Columbia, Canada V6V 2L1. 
10 To conform with McGurk et al. 8,9 , % viable hatch was defined 
as 100 [(no. live larvae - no. abnormal larvae )/(no. hatched 
eggs)] x (no. eggs hatched/no. eggs total). The value defined by 
Kocan et al. (1996, a and b) as % viable larvae is nearly syn- 
onymous with % viable hatch. Our % live larvae (Table 1) in- 
cluded abnormal larvae, but McGurk et al. 8 9 excluded abnor- 
mal larvae in their definition of % viable larvae (% viable = 
100 (no. live larvae - no. abnormal larvae )/no. hatched). 
11 As previously, +1.5-m data were not included; estimated % vi- 
able hatch was 50% when these data were included. 
12 Percent viable larvae values reported by Kocan et al. (1996, a 
and b) should be increased by 2% to approximate percent vi- 
able hatch. 
