176 
Fishery Bulletin 115(2) 
A 1-0 1 
CQ 
X 
<D 
"O 
c 
A SRI 
OSR2 
□ SR3 
OSR4 
ATRl 
OTR2 
□ TR3 
OTR4 
High 
O Moderate 
20 40 60 
Total length (mm) 
Total length (mm) 
Figure 4 
Diet diversity and dietary overlap of winter flounder iPseudo- 
pleuronectes americanus) and summer flounder {Paralichthys 
dentatus) as a function of “preserved” total length (i.e., mea¬ 
sured in the laboratory after specimens were preserved in 70% 
ethanol) and riverine site (for locations of sampling sites, see 
Fig. 1). Diet diversity of (A) winter flounder and (B) summer 
flounder was expressed by the Levins index of niche breadth 
{B; Eq. 4), and horizontal dashed lines differentiate among 
high (B>0.6), moderate (B=0.4-0.6), and low (B<0.4) niche 
breadths. (C) Dietary overlap is expressed by the Schoener 
index (a; Eq. 5), and horizontal dashed lines differentiate high 
(a>0.6), moderate (a=0.4-0.6), and low (a<0.4) overlap. Nonlin¬ 
ear (exponential or logarithmic) regression models were fitted 
to the data and are represented by the solid lines. Winter and 
summer flounder were collected from the Seekonk River (SR) 
and Taunton River (TR) during 2009-2015. 
and bivalves (%/i?/=18.4%) remained important 
food items (Fig. 5A). The largest winter flounder 
also consumed isopods and crabs (e.g., megalope) 
in relatively high proportions 3.1% and 
4.0%, respectively) (Table 2, Fig. 5A). 
Summer flounder within the smallest size cat¬ 
egory (<59 mm TL) had 68.2% dietary similarity 
(SIMPROF: 71=1.88, P=0.69; Fig. 3B), with mysid 
shrimp, copepods, and amphipods representing 
the most dominant prey 59.5%, 22.9%, and 
15.1%, respectively), and sand shrimp and fish 
consigned to secondary importance (%/i?/<1.3%) 
(Fig. 5B). The similarities in diet of summer floun¬ 
der within the other size classes ranged between 
84.2% and 91.3% (SIMPROF: small-medium 
[60-79 mm TL], 7t=0.00, P=1.00; medium-large 
[80-119 mm TL], 71=1.25, P=0.63; large [>120 
mm TL], 7i=0.00, P=1.00; Fig. 3B). Copepods were 
not observed in the stomachs of moderate- and 
large-size summer flounder (%/P/=0.0%; Fig. 5B). 
Moreover, progressive increases in size of summer 
flounder resulted in a decline in the dietary im¬ 
portance of mysid shrimp and amphipods {%IRI: 
mysid shrimp, from 35.6% to 6.4%; amphipods, 
from 43.7% to 17.5%), whereas sand shrimp and 
fish became increasingly more dominant {%IRI: 
shrimp, from 19.2% to 66.8%; fish, from 1.0% to 
6.0%) (Fig. 5B). Of the identifiable fish remains 
in the stomachs of summer flounder, winter floun¬ 
der had the highest %F (2.4%) and %IRI (0.4%) 
(Table 2), verifying predator-prey interactions be¬ 
tween the focal species. 
Spatiotemporal effects on intraspecific diets 
Winter flounder diet in the Seekonk and 
Taunton Rivers varied statistically by site and 
month (2-way PERMANOVA: site, pseudo- 
P=3.04-4.77, P<0.01-0.001; month, pseudo- 
P=4.94-10.07, P<0.001), and the site-month 
interaction effects were not significant (2-way 
PERMANOVA: sitexmonth, pseudo-P=0.73-1.77, 
P=0.06-0.75) (Fig. 6). Principal coordinate analy¬ 
sis revealed that month most closely correspond¬ 
ed with the first PCO axis (PCOl) and accounted 
for 56.7% and 59.3% of the explainable variation 
in diet of winter flounder in the Seekonk and 
Taunton Rivers (Fig. 6, B and D). The second 
PCO axis (PC02), in contrast, was best repre¬ 
sented by riverine sites (SR1-SR4 or TR1-TR4) 
and accounted for 20.5-28.8% of the total varia¬ 
tion in diet. Differences in diet of winter floun¬ 
der across months were attributed mainly to the 
importance of copepods at the onset of this study 
(from May through August-September, %IRI for 
copepods declined from 61.6% to 0.8%), and co¬ 
pepods were steadily replaced by polychaetes 
thereafter (from May through August-Septem¬ 
ber, %IRI for polychaetes increased from 5.1% to 
42.6%) (Fig. 6, A and C). 
