Newman et al : Variability in the population structure of Lutjsnus adetu and L qumquelineatus 315 



IT 



•Jjf ' Myrmidon 



-$ Kelson • » 



John Brewer ■'•> 

 8 \\ Lodestone ' 



Davies ^ 



"v 



JT-v-^ 



: :u 



Figure 1 



The locations of the reefs sampled for Lutjanus adetu and L. gum- 

 quelineatus from October 1991 to December 1993 in the central region of 

 the Great Barrier Reef, Australia. 



confounding differences among reefs with differences 

 between sexes). Both reefs and sexes were treated 

 as fixed and orthogonal factors in the analysis. Mul- 

 tiple comparisons were performed with Tukey's hon- 

 estly significant difference (HSD) test (a=0.05). 



The Kruskal-Wallis one-way analysis of variance 

 by ranks was used to test for differences in the age 

 structures of each species among reefs (Siegel and 

 Castellan, 1988). Multiple pair-wise comparisons 

 were performed, by using Kolmogorov-Smirnov (K- 

 S) tests, to determine differences in the age struc- 

 ture of each species between individual reefs (Zar, 

 1984). 



Growth rates for each species at each reef were 

 also examined. The von Bertalanffy growth function 

 (VBGF) was fitted to lengths at age for each species 

 from each reef by using nonlinear least-squares 

 estimation procedures. 9 The VBGF is defined as 

 follows: 



L, = L„ ( 1 - e 



-Kil-t.. 



where L t 



k 



length at time t\ 

 asymptotic length; 

 Brody growth coefficient; 



9 Prager, M. H., S. B. Saila, and C. W. Recksiek. 1989. 

 FISHPARM: a microcomputer program for parameter estima- 

 tion of nonlinear models in fishery science, 2nd ed. Old Do- 

 minion University Oceanography Technical Report 87-10. 



= age of the fish; and 

 = theoretical origin of the growth curve. 



Von Bertalanffy growth curves were compared by 

 using the likelihood-ratio test (Kimura, 1980), con- 

 sidered the most robust measure (Cerrato, 1990). A 

 modified analysis of the residual sum of squares 

 (ARSS) was also employed to compare VBGF's among 

 reefs (Zar, 1984; Chen et al., 1992). Further, one-way 

 analysis of variance was used to determine whether 

 there were significant differences in the mean length 

 (FL) of the early age classes of each species among 

 reefs (a=0.05). 



Analysis of covariance was used to determine 

 whether there were significant differences in the 

 clean weight at length ( FL ) relationships among reefs 

 and between sexes for each species. Length and 

 weight data were transformed to a natural logarithm 

 function (log or) to satisfy assumptions of normality 

 and homogeneity as detected by Cochran's test 

 (Winer, 1971). Both reefs and sexes were treated 

 as fixed and orthogonal factors in the analysis. 

 Again, multiple comparisons were performed by 

 using Tukey's honestly significant difference (HSD) 

 test. 



Estimates of the annual instantaneous rate of to- 

 tal mortality (Z) for each species among individual 

 reefs were obtained by using the age-based catch- 

 curve method of Beverton and Holt ( 1957) and Ricker 

 (1975). The natural logarithm of the number offish 



