248 
Fishery Bulletin 109(3) 
The results of several studies indicate that the 
growth of certain species can alter in response to 
changes in their density. For example, the growth 
of porbeagle (Lamna nasus) increased following a 
marked decline in the abundance of this species due 
to heavy exploitation (Cassoff et al., 2007) and, on 
the basis of data for eight populations, the growth of 
immature walleye ( Sander vitreus) was estimated to 
have increased 1.3 times when abundance was low 
rather than high (Venturelli et al., 2010). Furthermore, 
the growth of Atlantic herring ( Clupea harengus) 
on the Georges Bank in the northwestern Atlantic 
decreased when the abundance of that stock rose after 
the collapse of that fishery (Melvin and Stephenson, 
2007). Although these authors concluded that such 
changes in growth were related to changes in density, 
such density-dependent phenotypic expressions are 
likely to occur only when there is strong competition 
for a limited supply of food or other essential resources 
(Brander, 2007). 
Maturation reaction norms, which describe the 
probability that a fish will mature as a function of 
its length and age, have been used to assess whether 
changes in the length and age at maturation of some 
species are likely to be due to genotypic or phenotypic 
influences, or both (Stearns and Koella, 1986; Heino 
et al., 2002a, 2002b; Dieckmann and Heino, 2007). For 
example, this approach has shown that the maturation 
reaction norm for European plaice ( Pleuronectes 
platessa) shifted toward younger ages and smaller 
lengths with intensive exploitation, indicating that 
short-term phenotypic responses were overlaid by a 
longer-term genetic response (Grift et al., 2003). 
In none of the previous studies had the question of 
whether the biological characteristics of a stock of a 
species have changed in response to fishing pressure 
been investigated by using fishery-independent data 
for a heavily fished species whose life cycle is confined 
to an estuary. The estuary cobbler ( Cnidoglanis 
macrocephalus), which is endemic to southern 
Australia and represented by discrete populations in 
estuaries (Laurenson et al., 1993a; Ayvazian et al., 
1994), is the greatest contributor to the overall value 
of the commercial estuarine fishery on the south coast 
of Western Australia (Smith and Brown 1 ) and is a 
major component of the ichthyofaunal communities 
of seasonally open estuaries in this region (Chuwen 
et al., 2009a). The vast majority of the catch of this 
benthic plotosid is taken in Wilson Inlet, and long- 
term, experienced commercial fishermen have become 
increasingly concerned that the abundance of large 
C. macrocephalus has declined during recent years 
(McIntosh 2 ; Miller 3 ). 
1 Smith, K., and J. Brown. 2008. South coast estuarine 
managed fishery status report. In State of the fisheries 
report 2007/08 (W. J. Fletcher and K. Santoro, eds.), p. 
216—223. Dept. Fisheries, Perth, Western Australia. 
2 McIntosh, O. 2009. Personal commun. Commercial fish- 
erman, P.O. Box 565, Denmark, Western Australia, 6333. 
The collection of sound biological data for C. 
macrocephalus in Wilson Inlet in 1987-89 (Laurenson 
et al., 1993a, 1993b, 1994) provided an excellent 
opportunity to replicate that fishery-independent 
sampling regime in 2005-08, and thus elucidate 
whether the biological characteristics of this plotosid 
have changed in a manner that would be consistent 
with continued heavy exploitation. Because individuals 
of the population of C. macrocephalus complete their 
entire life cycle within Wilson Inlet, we had the 
particular advantage of sampling the full distribution 
of that population. Initial comparisons between the 
fishery-independent data for 1987-89 and 2005-08 
confirmed the opinion of commercial fishermen that 
the prevalence of larger (and thus probably older) 
C. macrocephalus and the abundance (catch rate) of 
this species declined between the two periods. These 
comparisons also revealed that the decline in abundance 
of C. macrocephalus in Wilson Inlet was matched 
by, and consistent with, a corresponding increase in 
fishing mortality. In view of these findings, focus was 
subsequently placed on elucidating 1) whether the 
increased and heavy exploitation of C. macrocephalus 
in Wilson Inlet was accompanied by a reduction in the 
length and age at maturity; and 2) whether growth 
changed between the two periods. Reaction norms 
relating the probability of maturation to length and 
age were examined to elucidate whether these norms 
changed in a way that would be consistent with genetic 
rather than phenotypic changes. 
Materials and methods 
Sampling regime 
Cnidoglanis macrocephalus was sampled in each season 
between winter 2005 and autumn 2008 at the same 
six sites in the Wilson Inlet basin that were sampled 
between winter 1987 and autumn 1989 (Laurenson et 
al., 1993a, 1993b, 1994). These sites included two that 
were in the small region closed to commercial fishing 
and four that were open to such fishing (Fig. 1). The 
data derived from these samples were used to compare 
the catch rates and mortality during the two periods. 
Additional samples were collected from other sites (Fig. 
1) to augment the numbers used for determining the 
length and age at maturity and growth. Note that, to 
ensure comparability between periods, the raw fishery- 
independent data obtained during 1987-89 were used to 
estimate the catch rates, mortality, length and age at 
maturity, and growth of C. macrocephalus in that period 
in precisely the same manner as that employed for the 
corresponding data collected during 2005-08. 
At each site during 2005-08, nearshore, shallow wa- 
ters of Wilson Inlet were sampled with a seine net, and 
3 Miller, W. 2009. Personal commun. Commercial fisher- 
man, Crusoe Beach Road, Denmark, Western Australia, 
6333. 
