490 
Fishery Bulletin 99(3) 
allopatric 
allopatric 
sympatric sympatric 
Barnacle larvae 
Large calanoids 
Small calanoids 
Larvaceans 
Euphausiids 
□ 
M 
Decapod larvae 
Bryozoan larvae 
Cladocera 
Chaelognaths 
Gammarld amphipods 
□ 
Other 
Malacostraca 
Cnidarians and 
ctenophorans 
Hyperiid amphipods 
Gastropods 
Figure 3 
Percent total number of prey consumed by YOY walleye pollock and Pacific her- 
ring from sympatric and allopatric aggregations in Prince William Sound, Alaska, 
in July-August 1995, October 1995, and November 1994. 
in terms of prey numbers and weights. Numeric overlap 
for the November 1994 sympatric species was observed at 
both stations, and the mean was approximately twice that 
for the October 1995 sympatric species (i? 0 = 0.97 versus 
0.43). Biomass overlap between sympatric fish occurred at 
three out of four autumn stations when based on princi- 
pal prey taxa, but only at one station (in November) when 
based on prey species. Overall, diets of sympatric pollock 
and herring overlapped more in terms of biomass in No- 
vember (i? 0 =0.95) than in October (J? 0 =0.69). We did not 
have autumn samples from the same year to compare al- 
lopatric diet overlap between species. 
Prey selection 
We noted selection from zooplankton prey resources by 
pollock in summer and by pollock and herring in autumn, 
1995 (Fig. 5). Selection patterns were almost identical 
whether calculated from shallow or deep zooplankton 
abundances; values were always in the same direction 
(selection or avoidance), and their magnitude was within 
10 points. Neither species selected small calanoids, the 
most abundant zooplankton taxon in both seasons. In 
summer, even though small calanoids made up >50% of 
YOY pollock and herring diets (Figs. 3 and 4), this taxon 
was avoided by pollock and was consumed in close propor- 
tion to its availability by herring (Fig. 5). Summer pollock 
moderately selected for large calanoids, gastropods, and 
larvaceans, but summer herring did not strongly select for 
any prey category. 
Changes in zooplankton composition from summer to 
autumn were reflected in fish diets and prey selection. The 
percent density of small calanoids in the zooplankton de- 
clined by nearly 30% from summer to autumn, the per- 
centage consumed by fish was likewise much reduced, and 
selection values were more strongly negative. In contrast, 
both large calanoids and larvaceans were more abundant 
in zooplankton samples in autumn than in summer and, 
