Kingsford and Hughes: Growth, mortality, and size of Acanthochromis polyacanthus 



569 



Age frequencies, therefore, were pooled 

 at the shelf level (within distance strata; 

 Fig 7). 



Acanthochromis polyacanthus mortality 

 rates did not differ significantly between 

 the inner-, mid- and outer-shelf strata 

 (test for slopes df, s 63) , F=0.367, P=0.70) 

 (Fig. 6). Although mortality estimates 

 were progressively lower with increased 

 distance from the coast, this trend was 

 not significant (inner shelf: -0.51, mid- 

 shelf: -0.48. outer shelf: -0.43; Fig. 6, Ta- 

 ble 4). Associated survival rate estimates 

 (S) varied between reefs by -9% per an- 

 num at inner- and mid-shelf strata and by 

 -6% per annum on the outer shelf (Table 

 3). The mean difference in survival rates 

 for A. polyacanthus between the inner and 

 mid-shelf was ~29c and between the mid- 

 and outer shelf was -3% (Table 4). 



Discussion 



110 

 105-1 

 100 

 95  

 90- 

 85 

 80 -I 

 75 



The Slashers 



Myrmidon 



Barnett Patches 



Havannah 



Pandora 



0.5 



0.75 



1 1.25 



K 



1.5 



1.75 



The demographic parameters of L x and 

 patterns of growth for populations of A. 

 polyacanthus varied across the shelf on the 

 central GBR. Although there was varia- 

 tion in body size and growth among reefs 

 within a distance stratum, it was minor 

 compared to overall cross-shelf patterns. In this study, 

 mortality estimates and maximum age were similar 

 for populations of fish across the shelf. Thus, in order 

 to explain the cross-shelf trend in body size, fish must 

 have grown faster with increasing distance from shore 

 (Fig. 7, Table 1). 



Despite the relative paucity of age-based studies on 

 reef fishes (Choat and Robertson, 2002), variable rates 

 of growth have been previously demonstrated for fish at 

 local scales (hundreds of metres to kilometers: Fowler 

 and Doherty, 1992), medium scales (kilometers to tens 

 of kilometers: Choat and Axe, 1996; Hart and Russ, 

 1996; Newman et al., 1996; Meekan et al., 2001; Gust 

 et al., 2002), and large scales (thousands of kilometers: 

 Choat and Robertson, 2002). Gust et al. (2002) found 

 that growth patterns of scarids varied between the reef 

 crests of mid- and outer-shelf sampling locations on the 

 northern GBR. In contrast to the results from the cur- 

 rent study, however, outer-shelf populations of scarids 

 had smaller asymptotic sizes and slower growth rates 

 than mid-shelf populations. The factors influencing pat- 

 terns of growth, therefore, vary by group. 



Differences in the shape of growth curves between 

 geographic regions or areas may be determined by both 

 genetic and environmental influences (Sebens, 1987). 

 Populations of reef fish are generally considered to be 

 genetically open systems (Sale, 1991) and it is consid- 

 ered unlikely that adaptation of such populations to 

 local conditions through genetic selection can occur 

 (Warner, 1991). Acanthochromis polyacanthus, how- 



Figure 6 



95*^ confidence ellipses for the von Bertalanffy growth parameters 

 K (growth coefficient) and L, (mean asymptotic lengthl for Acantho- 

 chromis polyacanthus from all reefs sampled. 



ever, possesses a unique life history trait among reef 

 fishes in that it lacks a dispersive larval phase. The 

 major implication of this characteristic is the potential 

 for genetic isolation of populations of these fish. Even 

 reefs that are in relatively close proximity to one an- 

 other (100's of m) may become "genetic islands" isolated 

 by any barrier that proves impassable to adults (e.g., 

 deep water). Without gene flow, reproductively isolated 



