316 
Fishery Bulletin 116(3-4) 
%F0 
Figure 3 
Prey-specific abundance at a family level plotted against the frequency of 
occurrence (%FO) for families of prey species found in analysis of stomach 
contents of blue sharks (Prionace glauca ) sampled in northern Peru between 
February and December 2015. The explanatory axes for foraging patterns 
are those modified by Amundsen et al. (1996). The 2 diagonal axes represent 
the importance of prey (dominant versus rare) and the contribution to niche 
width (high between-phenotype contribution [BPC] versus high within-phe- 
notype contribution [WPC]). The vertical axis defines the predator feeding 
strategy. The families include Ancistrocheiridae (ANC), Argonautidae (ARG), 
Bolitaenidae (BOL), Chiroteuthidae (CHI), Coryphaenidae (COR), Cranchi- 
idae (CRA), Delphinidae (DEL), Engraulidae (ENG), Enoploteuthidae (ENO), 
Enteroctopodidae (ENT), Galatheidae (GAL), Gonatidae (GON), Histioteuthi- 
dae (HIS), Loliginidae (LOL), Macrouridae (MAC), Mastigoteuthidae (MAS), 
Merlucciidae (MER), Myctophidae (MYC), Octopodidae (OCD), Octopoteuthi- 
dae (OCT), Oeogopsidae (OEO), Ommastrephidae (OMA), Onychoteuthidae 
(ONY), Pholidoteuthidae (PHO), Scombridae (SCO), Sphyraenidae (SPH), 
Thysanoteuthidae (THY), Tremoctopodidae (TRE), Vampyroteuthidae (VAM), 
and Vitreledonellidae (VIT). Other stomach contents include one bird (BIR) 
and flying fish (Exocoetidae) eggs (EXO). 
However, because size class I (110.0-149.9 cm TL) was 
represented by only 2 samples, only size class II (150.0- 
229.9 cm TL, n= 32) and size class III (230.0-309.9 cm 
TL, n =40) were considered for subsequent analyses. 
From all the factors assessed (sex, season, size class, 
and fishing ground [by latitude and longitude]), ANO- 
SIM showed a significant difference in the diet by size 
class and fishing ground longitude. These differences 
were small (Table 2), but indicated that an overlap ex¬ 
ists between the composition of dietary factors (Clarke, 
1993). A small difference in diet composition existed 
between size class II and III (Table 2). The nonmetric 
MDS plot showed that this small difference is explained 
by an overlap of diets (Fig. 5A). According to SIMPER 
analysis, in order of importance, the prey families that 
most contributed to the diet of size class II were Argo¬ 
nautidae and Thysanoteuthidae, whereas in size class 
III they were Argonautidae and Ancistrocheiridae. The 
%PSIRI for size classes showed a similar trend (Fig. 
2). Argonauta spp. represented the main prey taxon 
for both size classes, although a considerable decrease 
in this taxon was observed from size class II (28.5%) 
to size class III (8.6%). Furthermore, larger size indi¬ 
viduals (size class III) exhibited a tendency to forage 
on other prey species such as A. lesueurii (7.8%) and 
Japetella diaphana (4.7%), in addition to an increase 
in the importance of 'unidentified cephalopoda’ in the 
diet. Niche breadth analysis revealed that both groups 
had high degrees of specialization, even though higher 
values were estimated for size class III (Table 2). Blue 
sharks of both groups were considered top predators 
(Table 2). 
