344 
Fishery Bulletin 99(2) 
Ensenada and Mazatlan (Mexico), Mayaguez and San 
Juan (Puerto Rico), and Cumana (Venezuela). Sampling 
was restricted to catches from east of 150°W from Janu- 
ary through December 1995. The sampling program was 
designed to select specimens greater than 50 cm in length, 
in order to efficiently sample sexually mature fish (Orange, 
1961), if there were any. 
The area sampling strata (Fig. 1) used by the IATTC 
for length data (Tomlinson et al., 1992) were also used for 
sampling skipjack tuna specimens for the present study. 
The sampling was designed to collect 25 recognizable ova- 
ries from a single purse-seine set from each area and 
month stratum for one year. A quota was stipulated of no 
more than three such samples, for a total of 75 ovaries, 
from each area and month stratum from which skipjack 
tuna were caught and unloaded. A total of 76 samples (in- 
cluding 1 1 replicates), each consisting of approximately 25 
female skipjack tuna, for a total of 1822 ovaries, were col- 
lected and processed. Skipjack tuna specimens were not 
available for sampling from various area and month stra- 
tum during this sampling period because of the distribu- 
tion of fishing effort. 
Each fish was measured to the nearest centimeter with 
calipers. The gonads were removed, sex was determined, 
and the ovaries placed in a plastic bag with identification 
labels and frozen. The capture locations, dates, set speci- 
fications, and sea surface temperatures (SSTs) were ob- 
tained from records of scientific observers or abstracts of 
the vessel logbooks made by IATTC employees, or from 
both. 
Analysis of ovaries 
Chi-square contingency tests were employed to compare 
distributions of oocyte diameters from t issue samples from 
the rostral, medial, and caudal regions of both ovaries 
from 30 fish. Because no significant differences were found 
(P>0.5), oocytes from an approximately 1-g tissue sample 
from the medial region of the right or left ovary were 
measured to the nearest 0.03 mm at 27x magnification 
with an ocular micrometer in a dissecting microscope. The 
mean diameter (random axis) was determined from 10 
oocytes per fish (with oocytes >0.21 mm) present in the 
most advanced mode of oocytes. Ovaries were slit longitu- 
dinally and the ovarian lumen examined microscopically 
for the presence of residual hydrated oocytes, indicative of 
recent spawning activity. 
Female skipjack tunas can be classified as immature, 
maturing, reproductively active, or resting. Skipjack have 
asynchronous oocyte development and are considered to 
he multiple or batch spawners (Schaefer, 1998; Schaefer, in 
press). Histological and morphological descriptions of oo- 
genesis in the ovary of skipjack tuna (Cayre and Farrugio, 
1986; Goldberg and Au, 1986; Stequert and Ramcharrun, 
1996) were used for classification of reproductive condi- 
tion of individuals in the present study. The characteris- 
tics used to classify the reproductive condition of the indi- 
vidual fish were 1) the mean diameter of the oocytes of the 
most advanced mode and 2) the presence or absence of re- 
sidual hydrated oocytes. A fish is classified as immature if 
Figure 1 
Areas of the eastern Pacific used for sampling skipjack 
tuna gonads. 
the mean diameter of the most advanced group of oocytes 
is less than or equal to 0.2 mm and if there are no residual 
hydrated oocytes present. 
Results 
Sex ratios 
There were 65 purse-seine sets on schools of skipjack tuna 
in which a minimum of 24 fish and a maximum of 87 fish, 
equal to or greater than 50 cm in length, were sampled 
(mean=52.5 cm, SE=1.44 cm). The percentage of males in 
the samples ranged from 26.5 to 71.3 (mean=51.5, SE=1.21). 
A total of 3284 fish, of which 1737 (52.9%) were males and 
1547 (47.1%) were females, were sampled. The overall sex 
ratio for the pooled data from the 65 samples was signifi- 
cantly different ( / ^'o.o.5.i = 10.99 ) from the expected 1:1 ratio. 
Chi-square tests of males and females grouped into 5-cm 
length classes for the pooled data (Table 1) indicated a sig- 
nificant deviation in the 50-54.9 cm class (56.0% males), 
and the 65—69.9 cm class (57.0% males). 
Replicate samples 
For the 11 replicate samples collected, chi-square tests 
indicated no significant differences in the proportions 
of reproductively active female skipjack tuna specimens 
among the paired replicates (P>0.05). Thus, the sample 
sizes of approximately 25 female skipjack tuna from a 
single purse-seine set appeared to be adequate for assess- 
ing the reproductive status of the fish from the sets sam- 
pled. The data from the replicate samples were not used in 
any of the subsequent analyses. 
