Fujinami et at: Age determination and growth of Prionace glauca in the western North Pacific Ocean 
1 13 
Table 2 
Comparison of goodness of fit for statistical models developed for 3 dif¬ 
ferent periodicities of growth-band formation in blue sharks (Prionace 
glauca) of 2 size classes defined by precaudal length (PCL). The model 
with the lowest Akaike’s information criterion (AIC) is preferred because 
it is estimated to be closest to the unknown reality that generated the 
data. The AAIC statistic is the relative difference between the best model 
(which has an AAIC of zero) and each of the other models in the set. 
Size class 
Periodicity of 
band formation 
AIC 
AAIC 
Individuals <200 cm PCL 
Annual cycle 
1009.8 
0.0 
Biannual cycle 
1232.5 
222.7 
No cycle 
1283.7 
273.9 
Individuals >200 cm PCL 
Annual cycle 
169.0 
0.0 
Biannual cycle 
240.1 
71.1 
No cycle 
236.1 
67.1 
Table 3 
Results from likelihood ratio tests (Kimura, 1980) for the von Bertalanffy growth model used to examine differences in 
growth parameters between male and female blue sharks (Prionace glauca) in the western North Pacific Ocean. The param¬ 
eters of the von Bertalanffy growth model are the theoretical asymptotic length (L„), measured in centimeters in precaudal 
length; the annual growth coefficient (k); and the theoretical age, measured in years, at zero length U 0 ). H 0 =null hypothesis 
(all growth parameters differ by sex). 
Test 
Hypothesis 
L„ 
Male 
k 
to 
Female 
k 
to 
X 2 
P-value 
H 0 vs. Hj 
L„,(male) = L„ (female) 
267.4 
0.134 
-1.16 
267.4 
0.133 
-1.10 
16.36 
<0.001 
H 0 vs. H 2 
k (male) = k (female) 
268.6 
0.133 
-1.14 
266.5 
0.133 
-1.12 
14.94 
<0.001 
H 0 vs. H 3 
t 0 (male) = t 0 (female) 
273.4 
0.130 
-1.14 
263.0 
0.136 
-1.14 
12.37 
<0.001 
H 0 vs. H 4 
All parameters same between sexes 
267.1 
0.134 
-1.13 
267.1 
0.134 
-1.13 
21.66 
<0.001 
vex structures (Fig. 4A) and translucent bands (Fig. 
4B) tended to increase from the boreal autumn to win¬ 
ter, peaking in January, and to be lowest in summer. 
In contrast, concave structures (Fig. 4A) and opaque 
bands (Fig. 4B) were prevalent in summer and least 
common in winter. The results of the statistical anal¬ 
ysis of Okamura and Semba (2009) indicate that the 
annual cycle of band formation was most plausible be¬ 
cause the model with that periodicity had the smallest 
Akaike’s information criterion (Table 2). 
Growth parameters 
There was a significant difference between the sexes in 
estimates for parameters of the von Bertalanffy growth 
function (Table 3). Females had a higher growth coef¬ 
ficient (k-0. 146/year) than males (^=0.117/year), but 
their theoretical asymptotic length (L„= 257.2 cm PCL) 
was shorter than that for males (L„ =284.9 cm PCL) 
(Table 4, Fig. 5). Growth rates of both sexes were simi¬ 
lar until an age of approximately 7 years, after which 
the female growth rate gradually decreased relative to 
that for males of the same age. 
The maximum observed age was 17.3 years for 
males and 15.8 years for females. Theoretical longevity 
estimates, calculated by using methods of Taylor (1958) 
and Fabens (1965), were 24.3 and 29.6 years for males 
and 19.5 and 23.7 years for females. These values were 
greater than the maximum observed age for either sex. 
Age at maturity and maternity 
Maturity data were available for 414 males (33.4-252.0 
cm PCL) and 365 females (33.4-238.0 cm PCL). The 
youngest mature individuals were 4.0 years old, and, 
for both sexes, the maximum age recorded for an im¬ 
mature individual was 7.0 years. The age at which 50% 
of males were mature was 5.9 years (95% Cl: 5.3-6.4 
years), and that of females was 5.3 years (95% Cl: 4.7- 
5.7 years) (Fig. 6, A and B). Maternity condition data 
were available for 354 females (33.4-238.0 cm PCL); 
the age at which 50% of females were in maternal con- 
