Fishery Bulletin 116(2) 
1 78 
800- 
600- 
400- 
200 - 
Kengfang (NE) 
— - Females, vonB 
•- Males, vonB 
- Females, inverse vonB 
- Males, inverse vonB 
800- 
600 - 
400- 
200 - 
Tsukuan (SW) 
0 
8 
0 
8 
Age (years) Age (years) 
Figure 4 
Sex-specific growth trajectories for populations of a cutlassfish, Trichiurus japonicus, near the ports of 
Kengfang and Tsukuan off the northeast and southwest coasts of Taiwan, derived from both the von 
Bertalanffy growth model (vonB, thick black lines) and the inverse model (inverse vonB, thin gray lines). 
For Kengfang data, one female sample with a large pre-anal length (752 mm) was excluded prior to fit¬ 
ting regression to enhance quality of fitting given the very high leverage and some uncertainty in the 
age estimate for this fish. 
agreement with the general patterns of larger length 
at age for older fish in Kengfang than Tsukuan (Fig. 3, 
B and C); however, for early ages, the larger K for T. 
japonicus in Tsukuan compensates for their lower L„, 
in agreement with length at age of immature fish being 
larger in Tsukuan than in Kengfang (Fig. 3A). 
Age and length at 50% maturity 
The estimates of A 50 and L 50 indicate consistent pat¬ 
terns between the areas and sexes: fish taken in Keng¬ 
fang (NE) were older and larger at 50% maturity than 
those collected in Tsukuan (SW), and females reached 
50% maturity at older ages and larger lengths than 
males (Fig. 5). Nonetheless, randomization tests re¬ 
vealed between-area differences in A 50 and L 50 were 
significant for males (P=0.005 and P=0.Q14 for A 50 and 
L 50 , respectively) but not for females (P=0.116 and 
P=0.066 for Ar l0 and L 50 , respectively; results of ran¬ 
domization tests of between-sex differences in A 50 and 
are shown in Suppl. Table 5) (online only). When data 
for the 2 areas were combined, ranges of A 50 and L 50 
were 1.2-1.7 years and 153-184 mm, respectively. 
Probabilistic maturation reaction norms 
The PMRNs could be estimated for T. japonicus at 
ages of 1-2 years for both sexes and areas (Fig. 6), 
encompassing the age range during which 50% matu¬ 
rity was reached (Fig. 5). The 95% CIs of the sex- and 
area-specific midpoints of PMRNs overlapped (Fig. 6). 
Randomization tests showed no significant differences 
in these midpoints of PMRNs between sexes or areas 
Figure 5 
Estimates of length and age at 50% maturity (L 50 and 
A 50 ) for populations of a cutlassfish, Trichiurus japoni¬ 
cus, sampled at the ports of Kengfang (K) and Tsukuan 
(T) in Taiwan in 2013-2015. Open squares indicate es¬ 
timates for-females (f), and black circles indicate values 
for males (m). Error bars indicate bootstrap-derived 
95% confidence intervals. 
(between-areas: P= 0.24 and P= 0.25 for males at ages 1 
and 2, respectively; P=0.Q9 and P=0.44 for females at 
ages 1 and 2, respectively; for between-sexes random¬ 
ization tests, see Suppl. Table 5) (online only). 
Comparison of the midpoints of PMRNs and the 
