254 
Fishery Bulletin 109(3) 
Open waters 
Closed waters 
B 
1988 
2006 
2007 
20 
15 
10 
5 
0 
Summer Autumn Winter Spring 
Figure 3 
Back-transformed mean catch rates ±95% 
confidence limits for estuary cobbler (Cnido- 
glanis macrocephalus ) in Wilson Inlet in (A) 
waters open and closed to fishing, (B) 1988, 
2006, and 2007, and (C) each season. 
Gillmet selectivity and mortality 
Analyses of the mesh selectivity of the composite gill net 
for C. macrocephalus indicated that stretched meshes of 
35, 51, 63, 76, 89, and 102 mm predominantly caught fish 
with lengths of 160-379, 190-479, 200-549, 270-619, 
410-679, and 490-709 mm, respectively (Fig. 4A). The 
values for d x and 0 2 , derived from the model of Kirkwood 
and Walker (1986), were 6.1 and 6802, respectively. 
Trends in estimates of selectivity at age indicated that 
this species became fully susceptible to capture by the 
above suite of mesh sizes at ~2 years of age and essen- 
tially remained so in the immediately ensuing years 
(Fig. 4B). 
Fishing mortality ( F ) in Wilson Inlet during 1988-89 
was 0.57 per year (95% CL = 0.27-0.90 per year) and 
thus far less than the 1.47 per year (95% CL=1. 14-1.84 
per year) estimated for the population in this estuary 
during 2006-08, and there was no overlap of the 
95% CLs. Natural mortality (M) of C. macrocephalus 
was calculated to be 0.35 per year on the basis of a 
maximum age of 14 years (L. J. B. Laurenson, unpubl. 
data) and thus total mortality (Z) was estimated to be 
0.92 per year during 1988-89 and 1.82 per year during 
2006-08. 
Length and age at maturity 
The percent contributions of mature female C. 
macrocephalus to all females in the 400-449 and the 
450-499 mm TL length classes during 2005-08 were 
62% and 94%, respectively, and thus far greater than 
the corresponding values of 22% and 67% for 1987-89 
(Fig. 5). These differences help account for the L go and 
L 95 of females at maturity during 2005-08, i.e., 419 mm 
TL (95% CL=399-438 mm TL) and 490 mm TL (95% 
CL=448-538 mm TL), respectively, being significantly 
different (P<0.05) from those during 1987-89, i.e., 
449 mm TL (95% CL=434-463 mm TL) and 542 mm TL 
Table 1 
Mean squares and significance levels for three-way analysis of variance of catch rates of estuary cobbler ( Cnidoglanis macro- 
cephalus) derived from samples obtained with gill nets in regions open and closed to fishing in Wilson Inlet, Western Australia, 
in each season of 1988, 2006, and 2007. * P< 0.05. ** P<0.01. *** PcO.001. 
Main effects 
Region (R) 
Year (Y) 
Season (S) 
Residual 
Degrees of freedom 
1 
2 
3 
47 
Mean square 
9.986*** 
3.048** 
1.312* 
0.384 
Interactions 
RxY 
RxS 
YxS 
RxYxS 
Degrees of freedom 
2 
3 
6 
6 
Mean square 
0.720 
0.696 
0.344 
0.831 
