32 
Fishery Bulletin 1 13(1) 
Primary Reader 
Figure 2 
Age bias plot for 198 gray triggerfish (Batistes capriscus ) sampled from the 
southeastern United States during 1990-2012 and aged by 2 primary readers. 
The first reader’s mean age estimates are plotted against the second reader’s 
age estimates. Error bars indicate standard deviations. 
An ANCOVA of length at age for individual ages 
revealed no significant differences in size at age be- 
tween the 2 sectors (P=0.33) for 7 of the 9 ages for 
which comparisons could be made (Table 3). When the 
ANCOVA was performed for state, 9 of 11 tests were 
nonsignificant. On the basis of these results, we pooled 
data across sectors and states. The resulting von Ber- 
talanffy growth equation was 
L t = 457( l-e _0 - 33(t + L58) ) (4) 
for all sectors, states, and sexes combined (n- 6267) 
(Table 4; Fig. 4). 
Weight-length relationships 
Statistical analyses revealed an additive error term 
(variance not increasing with size) in the residuals of 
the W-TL relationship, indicating that a direct nonlin- 
ear fit was appropriate. This relationship is described 
by the following regression: 
W = 3.1 x 10- 5 TL 2 - 88 (n=7618). (5) 
Residuals of the W-FL relationship exhibited multipli- 
cative error, indicating that a linearized ln-transform 
fit of the data was appropriate. This relationship is de- 
scribed by the following regression: 
ln( Wj = 2.98 ln(FL) - 17.5, (ra=20,431; r 2 =0.86) (6) 
where r 2 is the coefficient of determination. This equa- 
tion was transformed back to the form 
W = axFL h (7) 
after adjustment of the intercept for log-transformation 
bias with the addition of one-half of the mean square 
error (MSE) (Beauchamp and Olson, 1973), resulting in 
this relationship: 
W = 2.55 x 10- 5 FL 2 - 98 (rc=20,431; MSE=0.035). (8) 
The relationship between FL and TL is described by 
the following equation: 
FL = 30.33 + 0.79 x TL (72=8065; /- 2 =0.84). (9) 
Natural mortality 
Hewitt and Hoenig’s (2005) method, which uses maxi- 
mum age or life span (15 years in this study), esti- 
mated that M was 0.28. The method of Charnov et 
al. (2013), which produces age-specific estimates of M 
with the use of von Bertalanffy growth parameters, 
resulted in estimates of 0.65 for age-1 fish, 0.38 for 
age-5 fish, 0.34 for age-10 fish, and 0.33 for age-15 
fish (Table 2). 
