Sarre and Potter: Variation in age compositions and growth rates of Acanthopagivs butcheri 



797 



tainly have flushed out to sea any eggs or lai-vae of this 

 species that were produced during that period. However, it 

 should also be recognized that freshwater discharge was so 

 strong during that period that even many large A. butcheri 

 were flushed out of the estuary, with the result that some 

 of these fish were subsequently caught by anglers along 

 the nearby coast (Spurr*"). Furthermore, when freshwater 

 discharge is very high, the salinities fall to such low levels 

 that they are unlikely to be conducive to spawning by A. 

 butcheri (Haddy and Pankhurst, 2000). Thus, the absence 

 of the 1984 year class in samples may be due to the loss of 

 eggs, larvae, or maturing and mature fish to the ocean in 



1984, or to the inhibitory effect of low salinities on spawn- 

 ing, or to a combination of the latter. 



The fact that the 1985 year class oi A. butcheri was also 

 not caught may reflect a low return of large A. butcheri to 

 the estuary by the commencement of the spawning period in 



1985. The absence of the 1986, 1987, and 1989 year classes 

 and the paucity of the 1988 year class can probably be 

 attributed to the fact that, although freshwater discharge 

 was not as strong as in 1984, it was still sufficient to breach 

 the bar at the estuary mouth during the spawning period in 

 each of those years (SpuiT*^). It would thus also have been 

 likely to result in a loss to the ocean of eggs and larvae pro- 

 duced during the spawning periods in those years or to the 

 emigration of maturing or mature fish, or in both of these 

 effects (Hodgkin and Clark'^). The view that heavy freshwa- 

 ter flushing or very low salinities, or both, were the main 

 contributors to the lack of spawning success of A butcheri 

 between 1984 and 1989 is consistent with the observation 

 that, when freshwater discharge was not sufficiently strong 

 to breach the estuary mouth, as was the case between 1990 

 and 1995 (Spun-''), there was at least a reasonable recruit- 

 ment of each of the 1990 to 1995 year classes. 



Comparisons between von Bertalanffy growth 

 parameters for females and males 



The likelihood ratio test showed that the growth coeffi- 

 cients {/}) for female and male fish were significantly dif- 

 ferent in only one of the four estuaries, i.e. the Nornalup 

 Walpole Estuary, and even then the probability level was 

 close to 0.05, which is consistent with the fact that the 

 values for the 95'7f confidence intervals for this parameter 

 for the two sexes overlap in each estuary other than the 

 Nornalup Walpole Estuary. The lack of a marked dis- 

 tinction between the growth rates of female and male 

 fish is hardly surprising because A. butcheri undergoes a 

 substantial amount of growth before the gonads start to 

 become mature for the first time (Sarre and Potter, 1999). 

 However, the maximum length and asymptotic length (L^J 

 were always greater for female than male fish in each of 

 the four estuaries, thereby paralleling the situation with 

 many other fish species in both southwestern Australia 

 (Laurenson et al.. 1994; Wise et al., 1994; Hyndes et al. . 

 1996; 1998) and elsewhere (e.g. Kenchington and Augus- 

 tine, 1987; McPherson, 1992; Crabtree et al., 1995). 



Spurr, P. 199.5. Local resident and former commercial fisher- 

 man. Personal commun. Bremer Bay, Western Australia. 



The relatively low values determined for age at length 

 zero for the two sexes in the four estuaries and for males 

 in Lake Clifton, i.e. -0.13 to -0.61 years, reflects in part 

 the good fit of the growth cui-ves to the points for the age 

 at length of the small fish. These low values for t„ contrast 

 with the -5.21 years for females and -3.70 years for males 

 that were calculated by Morison et al. (1998) for A. butch- 

 eri in the Gippsland Lakes. Furthermore, the fork lengths 

 at age zero for female and male fish in the Gippsland 

 Lakes were ca. 110 and 100 mm, respectively. Thus, the 

 von Bertalanffy growth equations recorded for A. butcheri 

 in the Gippsland Lakes do not provide a good description 

 of the pattern of growth throughout the full size range of 

 fish. 



Variations in von Bertalanffy growth parameters 

 among populations 



Although the patterns of growth of female and male A. 

 butcheri in the Swan River, Wellstead, and Nornalup Wal- 

 pole estuaries followed the same overall trends, with length 

 increasing rapidly with time initially and then forming 

 asymptotes, the values for k and L ., for each sex varied 

 significantly amongst the populations in those estuaries. 

 The initial rate of increase in length in these three estuar- 

 ies was greatest in the Swan River Estuary and least in 

 the Nornalup Walpole Estuary. Although the value for k 

 for male A. butcheri in Lake Clifton differed significantly 

 from that of this sex in the Swan River Estuary, the same 

 was not true for L, . However, the value for L^ for male A. 

 butcheri in Lake Clifton was still similar to that of this sex 

 in the Swan River Estuary. The above comparisons dem- 

 onstrate that the growth rate in Lake Clifton was similar 

 to that in the Swan River Estuary, which is located only 

 ca. 90 km farther to the north (see also Fig. 1). The von 

 Bertalanffy growth parameters demonstrated that male A. 

 butcheri grew more rapidly and attained greater asymp- 

 totic lengths in the Swan River Estuary and Lake Clifton 

 than in either the Wellstead or Nornalup Walpole estuar- 

 ies. The presence of faster growth rates in the Swan River 

 Estuary and Lake Clifton, which are located at latitudes 

 of ca. 32° on the lower west coast of Australia, than in the 

 Nornalup Walpole and Wellstead estuaries, which are sit- 

 uated much farther south at a latitude of ca. 34° on the 

 south coast of Western Australia, may reflect the greater 

 temperatures found in more northern regions. 



The pattern of growth of A. butcheri in the Moore River 

 Estuary differed from those of this species in each of the 

 other three estuaries and Lake Clifton, in that the increase 

 in length with time was initially slower and the gi-owth 

 curve did not exhibit a marked asymptote. This pattern 

 suggests that some factor or factors were less than opti- 

 mal for growth during the first few years of life, but that 

 conditions for gi'owth improved later in life. The slow ini- 

 tial rates of increase in length of A. butcheri in the Moore 

 River Estuary during early life may be related to the excep- 

 tionally high densities of this species in nearshore, shallow 

 waters, the region which constitutes the typical habitat of 

 the juveniles of this species (Sarre, 1999). The far greater 

 density of this species in these waters, than in correspond- 



