Coulson et al.: Biological features of Achoerodus gouldii 
69 
number of fish >35 years old were caught by this fishery 
and the distribution of the lengths of the males did not 
appear to be truncated, mesh selectivity did not appar- 
ently exert a major influence on the upper end of the 
length distribution. 
Relative abundance analysis demonstrated that, in 
the three successive 12-month periods between August 
and July, 11 of the year classes between 1936 and 1992 
differed significantly from the average level of recruit- 
ment. The strengths of the 1972, 1980, 1983, and 1990 
year classes were 3.6, 2.6, 2.1, and 1.5 times greater, re- 
spectively, than the average level of recruitment (Fig. 9). 
Among the other year classes that differed significantly 
from the average level of annual recruitment, the 1981, 
1958, 1971, and 1944 year classes were strong, whereas 
the 1991, 1992, and 1985 year classes were weak (Fig. 9). 
The total mortality estimate, Z, of 0.086/year (0.08- 
0.092/year), derived from the catch curve analysis with 
the assumption of constant recruitment, was slightly 
less than the 0.093/year (0.08-0.092/year), derived for Z 
by using the relative abundance analysis and assuming 
variable recruitment. The estimate for M, derived by 
refitting Hoenig’s (1983) equation for fish and insert- 
ing the maximum age for A. gouldii , was 0.072/year 
(0.022-0.180/year). The approach of Hall et al. (2004), 
which was used to refine the above estimate of M , yield- 
ed a lower value, i.e., 0.054/year (0.021-0.090/year). 
Fishing mortality, F, was estimated to be 0.039/year 
(0.003-0.073/year). 
The spawning potential ratio at the current estimated 
level of F, on the basis of spawning stock biomass per 
recruit and calculated by using the von Bertalanffy 
growth curves for the females and males of A. gouldii 
shown in Figure 4D, was 0.88 (0.75-0.99) for females 
and 0.52 (0.27-0.96) for males (Fig. 10). 
Discussion 
Demonstration of protogynous hermaphroditism 
Our histological examination of the gonads of a large 
number of fish from throughout the year and covering a 
wide size range of A. gouldii fulfils one of the important 
requirements for demonstrating, with certainty, that a 
fish species is a functional hermaphrodite (Sadovy and 
Shapiro, 1987). Because all of the numerous A. gouldii 
captured with lengths of 100 to 655 mm were females 
and the prevalence of males subsequently rose with 
increasing body size to the extent where they consti- 
tuted 100% of the larger fish, all males of A. gouldii 
are apparently derived from females. This conclusion 
is supported by the fact that all of the 164 individu- 
als <15 years old, and whose gonads were examined 
histologically, were shown to be females. Moreover, the 
proportion of males subsequently increased progres- 
sively with age to the point where over 50% of individu- 
als >35—39 years old were males. Because the length 
at which the prevalence of males reached 50% of fish 
(821 mm) was substantially greater than the TL 50 at 
Table 2 
Estimates of the total lengths (and their lower and upper 
95% confidence limits [CLs], in mm) at which 50% <TL 50 ) 
and 95% (TL 95 ) of western blue groper ( Achoerodus goul- 
dii) change sex to male and the total lengths at which 
50% and 95% of A. gouldii change from green (females) to 
blue (males) in southwestern Australia. 
tl 50 
tl 95 
Sex change 
Estimate 
821 
930 
Lower CL 
800 
890 
Upper CL 
845 
1102 
Color change 
Estimate 
779 
961 
Lower CL 
753 
910 
Upper CL 
809 
1028 
Table 3 
Results of logistic regression analysis for determining 
whether color or total length (TL, in mm), or color and 
total length combined, were the best predictors of whether 
an individual of western blue groper ( Achoerodus goul- 
dii) was male. LL= log-likelihood; a, p v /f 2 are constants; 
AIC = Akaike’s information criterion. 
Parameter 
Color 
TL 
Color and TL 
LL 
-84.424 
-48.644 
-45.457 
a 
-4.631 
-17.275 
-15.505 
w 
0.021 
0.018 
ft 
4.979 
1.477 
AIC 
172.847 
101.289 
96.913 
maturity (653 mm) for females and greater than that 
at which all female fish were apparently destined to 
become mature, a large number of male A. gouldii must 
have been derived from mature females. Thus, the char- 
acteristics of A. gouldii fulfil the criteria of Sadovy de 
Mitcheson and Liu (2008) for designating this species as 
a protogynous hermaphrodite. Although, on their own, 
these features do not provide concrete evidence that 
A. gouldii is monandric, i.e., all males are derived from 
mature females (Sadovy de Mitcheson and Liu, 2008), 
they do raise the strong possibility that this is the case 
for this labrid. As with other protogynous labrids (e.g., 
Reinboth, 1970; Sadovy and Shapiro, 1987), the pres- 
ence of a central lumen in all mature testes of A. gouldii 
is considered to represent the retention of the ovarian 
lumen and to account for the peripheral location of the 
sperm ducts in those testes. 
Aging methods, length and age compositions, 
and growth patterns 
Although the von Bertalanffy growth curve provided 
a good fit to the lengths at age for all aged individuals 
