NOTE Abitia-Cardenas et al : Food habits and energy values of prey of Tetrapturus audax 
361 
Figure 1 
Map showing the location of the study area off the tip of Baja 
California. 
x t -X /SD, 
where: x ; = the absolute IRI value of each prey species; 
X = the mean value of the IRI; and 
SD = standard deviation. 
The caloric content of each prey, based on three 
samples obtained from stomach contents, was mea- 
sured with a Parr 1241 adiabatic calorimeter and 
expressed as calories per gram of dry weight, wet 
weight, and ash-free dry weight following Phillipson 
(1964). One-way analysis of variance was used to 
evaluate differences between ash-free dry weight 
caloric values of particular prey. Also a post-hoc test 
T-method (Sokal and Rohlf, 1981) was used to com- 
pare the means of dry-weight caloric values. 
The calories provided by each prey species were 
calculated by multiplying the values (calories/g wet 
weight) of each prey by the sum of their total contri- 
bution (weight) in the diet. To convert prey volumes 
to calories we assumed a density of 1.0 g/mL. 
Results 
Food habits 
Striped marlin (403) were sampled. The mean pos- 
torbital length was 177 ± 15 cm (standard deviation) 
and the mean weight was 58.4 ± 12.8 kg. Of those 
specimens sampled, 27 (6.7%) had empty stomachs 
and 26 (6.5%) had regurgitated their stomach con- 
tents. A total of 33 prey taxa were identified that 
comprised fish, cephalopods, and crustaceans. Only 
17 prey types could be identified to species (Table 2). 
The most important prey by volume were fish 
(86.2%), including S. japonicus (25.7%), California 
pilchard, Sardinops caeruleus, (18.8%), and E. teres 
(10.2%). Cephalopods made up 12.8% of the total 
volume, and jumbo flying squid, Dosidicus gigas, was 
particularly important (11.3%). Crustaceans, mainly 
red crab, Pleuroncodes planipes, represented only 1% 
of the total volume. 
A total of 2,679 organisms were enumerated, 68.6% 
of which were fish, 21.3% cephalopods, and 10.2% 
crustaceans. The dominant fish prey by number were 
S. caeruleus (18.9%), S. japonicus (14.3%), and Pa- 
cific hake, Merluccius productus, (9.6%). The cepha- 
lopod D. gigas represented 14.9%, and Argonauta 
spp. 3.0% of the total stomach contents by number. 
Pleuroncodes planipes was the most abundant crus- 
tacean, representing 7.2% of the total number of food 
items. 
In frequency of occurrence, fish were the most im- 
portant food in the diet of striped marlin (93.4%), 
particularly S. japonicus (45.4%), S. caeruleus 
(27 .7%), and E. teres (12.6%). Cephalopods occurred 
in 32.9 % of the samples; and D. gigas was the most 
common species (28.3%). Crustaceans, mainly P. 
planipes , occurred in 6.3% of samples. 
According to the IRI, fish were the most impor- 
tant prey (80.7%) of striped marlin, followed by 
cephalopods (18.5%), and crustaceans (0.8%). 
Scomber japonicus , S. caeruleus x and D. gigas were 
the most important fish prey (Fig. 2). 
Relative importance of several prey varied season- 
ally (Table 1). During 1988, fish were the most im- 
portant prey in spring and fall, cephalopods the most 
important prey in summer. In spring 1988, S. 
caeruleus was the most important fish in the diet, 
followed by S. japonicus and E. teres. In summer 
1988, the most important species was D. gigas, fol- 
lowed by the fish Selar crumenophtalmus , S. 
japonicus, and E. teres. In fall 1988, the highest IRI 
values were for S. japonicus, D. gigas, E. teres, and 
M. productus. 
During 1989, fish were the most important prey 
in all seasons, followed by cephalopods and crusta- 
ceans. In winter, the dominant species were S. 
japonicus, M. productus x and S. caeruleus. In spring, 
S. japonicus, D. gigas, S. caeruleus x and E. teres were 
the most important species. In summer, Caranx 
caballus was the most important prey. In fall, the 
highest IRI values were for S. caeruleus, S. japonicus, 
and Decapterus hypodus. The MANOVA showed no 
significant differences among seasons in the IRI val- 
ues of food groups consumed (F=1.96; df=4; P=0.11). 
However, when we considered taxa consumed (33 
recorded), we found significant differences (F- 17.6; 
df= 32; P<0.005), probably caused by the greater 
