Alatorre-Ramirez et al.: Trophic segregation of mixed schools of Thunnus albacares and Katsuwonus pelamis 
253 
Korsmeyer and Dewar, 2001), and consequently pre¬ 
vious authors have suggested that food acquisition is 
one factor that could explain the association of these 2 
tuna species (Sund et ah, 1981; Petit, 1991). The diets 
of yellowfin and skipjack tunas in the ETPO have been 
described individually (Galvan-Magana, 1988; Roman- 
Reyes, 2000; Olson et ah, 2014) with notable intraspe¬ 
cific differences related to size class and time of cap¬ 
ture (yellowfin tuna: Olson et ah, 2014; skipjack tuna: 
lATTC^). The diets of co-occurring yellowfin and skip¬ 
jack tunas in the northern ETPO were evaluated by Al- 
verson (1963), and he found only a minor diet overlap 
owing to the consumption of pelagic red crabs (Pleu- 
roncodes planipes) by both species. There is, therefore, 
a scarcity of information regarding whether competi¬ 
tive trophic interactions play a role in the association 
between yellowfin and skipjack tunas throughout the 
entire ETPO. Apart from documenting diet overlap and 
potential competition, diet studies of predatory fishes 
also provide valuable information on ingested prey and 
their spatial and temporal variation in abundance and 
biomass. Understanding interactions in the food web is 
a prerequisite for gaining insight into the role of pred¬ 
ators, commercial fisheries, and environmental effects 
on ecosystem structure and dynamics. Prey species are 
often the central key link in such interactions (Galvan- 
Magana, 1999; Olson and Watters, 2003; Griffiths et ah, 
2013). 
The goals of our study were 1) to analyze the diets 
of associated yellowfin and skipjack tunas caught in 
the ETPO, and 2) to examine diet variability in terms 
of the area, season, capture time of day, and tuna size 
class and sex in order to assess the hypothesis that 
competitive trophic interactions form the basis for the 
association between these 2 tuna species. Our data pro¬ 
vide important information on the feeding strategies 
of yellowfin and skipjack tunas on mid-trophic-level 
communities at various spatiotemporal scales in the 
pelagic ETPO. 
Materials and methods 
The study area was located in the ETPO between 35° 
and 5°N, and from 140°W to the coastline (Fig. 1). This 
area is characterized by a well-developed, relatively 
shallow thermocline, generally less than 100 m deep 
and is influenced by 6 major surface currents and 4 
subsurface currents (Kessler, 2006). Zone 1 is influ¬ 
enced partially by the California Current. Zone 3 is 
influenced by the North Equatorial Current, and both 
zones 2 and 3 are influenced by the North Equatorial 
Counter Current, and the north and south subsurface 
Counter Current (Lavin et al., 1997). 
Samples of yellowfin and skipjack tunas were col¬ 
lected simultaneously from 25 purse-seine sets during 
15 trips by observers from the Inter-American Tropical 
Tuna Commission, between January through Novem¬ 
ber 2005. In the ETPO, the tuna fleet uses 3 types of 
capture methods: sets associated with floating objects. 
sets associated with dolphins, and sets associated with 
free-swimming schools of tuna that are not associated 
with floating objects or with larger marine species. The 
tunas were identified, the capture location, date, time 
of day (morning, afternoon, or evening), fork length 
(FL), and sex of each specimen were recorded, and 
the stomach contents were collected and immediately 
frozen. 
The stomach analyses were conducted at the Cen¬ 
tro Interdisciplinario de Ciencias Marinas in La Paz, 
Mexico. In the laboratory, we thawed the stomach con¬ 
tents and categorized the digestive state of the prey 
species according to the digestive levels described by 
Galvan-Magana (1988) who assigned 4 digestive levels: 
l=food includes recently consumed items; 2=food items 
with little to no skin remaining; 3=presence of fish 
skeletons; and 4=presence of hard structures like fish 
otoliths, crustacean remains, and cephalopod beaks. 
For prey items (fish, crustaceans, and cephalopods) 
at digestive level 1, we used identification keys by Allen 
and Robertson (1994), Fischer et al. (1995), and Thom¬ 
son et al. (2000); whereas for prey items at digestive 
levels 2 and 3, we used the taxonomic key by Clothier 
(1950), which is based on vertebral characteristics (e.g., 
number, position, and form of the vertebrae). Finally, 
prey items at digestion level 4 were identified by us¬ 
ing keys by Fitch and Brownell (1968) for fish species, 
Brusca (1980) for crustaceans, and Wolff (1984) and 
Clarke (1986) for cephalopods. 
Once the prey items were identified, the data were 
stratified by area (based on Galvan-Magana, 1999), 
month, time of capture, sex, and size class. Size classes 
were divided at 85 cm FL for yellowfin tuna, with small 
defined as 1-85 cm FL and large defined as >85 cm FL, 
(Schaefer, 1998); and at 50 cm FL for skipjack tuna, 
with small defined as 1-50 cm FL and large defined 
as >50 cm FL (lATTC^). Capture time intervals were 
morning (0900-1059), afternoon (1100-1359), and eve¬ 
ning (1400-1600). 
We used the program Estimates (Colwell, 2006) to 
construct cumulative prey curves to determine wheth¬ 
er the number of stomachs collected were sufficient to 
represent the diet of both yellowfin and skipjack tu¬ 
nas (Ferry and Cailliet, 1996). The cumulative curves 
were generated by randomizing the species richness 
and abundance data (100 times) for the total number 
of stomachs to obtain cumulative Shannon-Wiener di¬ 
versity values. Then, we calculated the coefficient of 
variation (CV) of the diversity values as an indicator 
of the degree of variability of the diet. CVs <0.05 were 
considered adequate for the representation of the tro¬ 
phic spectrum of yellowfin and skipjack tunas (Steel 
and Torrie, 1992). Finally, cumulative diversity and CV 
values were plotted in relation to the number of stom¬ 
achs analyzed. 
® lATTC (Inter-American Tropical Tuna Commission). 
2002. Annual report of the Inter-American Tropical Tuna 
Commission 2000, 171 p. lATTC, La Jolla, CA. [Available 
from website.] 
