Lucena and O'Brien: Effects of gear selectivity and different calculation methods on growth parameters of Pomatomus sa/tatnx 
437 
Table 3 
Mean back-calculated total lengths (mm) of bluefish for the purse-seine and gill-net fishery. Lc = Mean observed lengths-at-age. 
Mean back-calculated lengths-at-age derived from the last annulus are shown in bold. 
Age-group 
Lc 
SD 
n (last annulus only) 
Method 
1 
2 
3 
4 
5 
6 
7 
Purse-seine fishery 
1 
313.5 
22 
27 
202.2 
2 
385.0 
26 
31 
189.0 
315.5 
3 
460.4 
23 
36 
227.9 
358.0 
437.3 
4 
518.1 
26 
35 
234.5 
373.1 
462.4 
504.3 
5 
552.5 
8 
72 
225.4 
366.9 
455.4 
504.5 
547.1 
6 
622.7 
3 
30 
219.1 
352.5 
447.2 
502.0 
552.7 
607.9 
7 
664.5 
1 
— 
247.9 
352.6 
512.1 
574.6 
615.9 
636.5 
647.4 
Mean (all annuli) 
109 
214.6 
350.6 
451.3 
506.1 
554.4 
615.4 
647.4 
SD 
36 
47 
45 
46 
64 
28 
— 
Gill-net fishery 
1 
354.3 
18 
27 
236.2 
2 
408.7 
59 
46 
230.4 
346.1 
3 
438.9 
75 
30 
239.8 
358.2 
430.3 
4 
467.7 
21 
41 
244.9 
366.2 
445.5 
480.8 
5 
498.0 
1 
— 
196.3 
350.3 
403.7 
426.5 
460.0 
Mean (all annuli) 
174 
236.7 
354.9 
433.3 
483.4 
460.0 
SD 
41 
39 
58 
41 
— 
Growth occurs from May to October and peaks in 
August. From these data, we believe that the growth 
increments are annuli. 
Back-calculated length-at-age and estimation 
of growth parameters 
Back-calculated lengths-at-age calculated by using the 
last annulus only ( Ila) were smaller than lengths-at- 
age calculated by using all annuli (lib) (Table 3). There 
were also significant differences in lengths-at-age for 
these two criteria between gear types (7-test, P<0.05 
for all age classes). Sexes were combined because there 
was no significant difference between them (7-test, 
P<0.05). Back-calculated lengths-at-age for gill-net 
samples showed a decreasing pattern for older fish (5 
years old), indicating the presence of Lee’s phenom- 
enon and the selective effects of this gear. 
Estimates of theoretical growth parameters indi- 
cated differences by gear and method of estimation 
(Table 4, Fig. 5). For criterion I, estimates with mean 
back-calculated lengths-at-age (criterion la) produced 
higher values of (and smaller values of k ) for 
purse-seine samples than estimates with individual 
back-calculated lengths-at-age (criterion lb). The op- 
posite trend was observed for the gill-net samples. 
On the other hand, for criterion II, especially for purse- 
seine samples, estimates of mean back-calculated lengths- 
at-age derived from the last annulus only (criterion Ila) 
produced smaller values of L ^ (and higher values of k) 
than estimates with mean back-calculated lengths-at-age 
for all annuli (criterion lib) (Table 4). Furthermore, for the 
purse seine, the growth parameters derived from methods 
3 (lb, Ila) and 4 (lb, lib) were distinct (Table 4). The at- 
