316 



Fishery Bulletin 94(2). 1996 



in each age class (N t ) was plotted against their cor- 

 responding age (t), and Z was estimated from the 

 descending slope, b. Estimates of the survival rate 

 of both species on different reefs were then calcu- 

 lated from the Z that was derived from catch curves, 

 because Z = -log, S (Ricker, 1975). 



Results 



There were significant differences in mean length, 

 age, and weight of both species among reefs, and 

 these differences were independent of the sex of the 

 fish (Tables 1 and 2). Tukey (HSD) comparisons 

 showed that, in general, mean length and weight 

 were larger and mean age greater for both species at 

 KL, although the multiple comparisons among reefs 

 were not all definitive (Tables 1 and 2). Significant 

 differences in mean length, age, and clean weight of 

 both species between sexes were the same as those 



described in Newman et al., 7 with males larger than 

 females in species (Tables 1 and 2). 



Age structures 



Age structures of L. adetii differed significantly 

 among reefs (Kruskal-Wallis statistic: H=13.85, 

 P<0.01), with the greatest difference being that be- 

 tween RI and JB (Table 3). The relative abundance 

 of older fish (year classes 10-22) at RI was greater 

 than that at JB (Fig. 2). This pattern was seen also 

 at LO and KL, although the overall age structures 

 were not significantly different (Table 3). The pat- 

 tern of peaks in abundance of year classes was not 

 consistent across all reefs. The small sample size 

 taken from KL may have biased results; however, 

 the sample covered a range of 10 age classes and was 

 similar to the age structure at RI. The age structure at 

 KL was significantly different from that of both JB and 

 LO, and the age structure at JB was significantly dif- 



