Ferreira and Russ: Age-validation and growth rate of Plectropomus leopardus 



53 



so 



B 

 <u 



-J 

 _<: 

 C 



P leopardus of 19 years in New Caledonia. 

 These higher estimates of longevity suggest 

 that coral trout at Lizard Island could also 

 attain older ages. In this case, the absence 

 of fishes older than 14 years of age in the 

 sample collected at Lizard Island could be 

 related to local levels of fishing pressure. 



In the present work, the results of tet- 

 racycline labelling indicated that in the 

 otoliths of P. leopardus the opaque zone 

 (annulus) was formed during the winter 

 and spring months whereas the translu- 

 cent zone was formed during summer and 

 autumn. Though the physiological basis for 

 the formation of optically distinct zones in 

 calcified structures has not been directly 

 established, their presence has been com- 

 monly associated with varying growth 

 rates, influenced by temperature, photo- 

 period, feeding rate, or reproductive cycle 

 (see Casselman, 1983, and Longhurst and 

 Pauly, 1987, for review). On a daily basis, 

 it has been demonstrated that the trans- 

 lucent zone, or accretion zone, is formed 

 during the phase of more active otolith 

 growth, and the opaque or discontinuous 

 zone is formed during growth stagnation (Mugiya et 

 al., 1981; Watabe et al., 1982). Mosegaard et al. 

 (1988) examined the effect of temperature, fish size, 

 and somatic growth rate on otolith growth rate and 

 suggested that metabolic activity, not necessarily so- 

 matic growth rate, governs otolith growth. Thus, if 

 the formation of the opaque zone in the coral trout 

 otoliths is associated with a period of reduced meta- 

 bolic activity, an external determining factor could 

 be temperature, as the lowest values for water tem- 

 perature around Lizard Island are observed during 

 winter and early spring. 4 Annulus formation oc- 

 curred in otoliths of juveniles and adults of coral 

 trout during the same period, suggesting that repro- 

 duction is not a determining factor. 



The growth of the otolith was continuous with age 

 but apparently related to somatic growth. A simi- 

 lar pattern has been observed for other species of 

 fish (Beckman et al., 1991). Otolith weight was the 

 best predictor of age in the linear model, explain- 

 ing as much variation in age as fork length in the 

 von Bertalanffy model. 



The main criteria for choosing a growth curve are 

 quality of fit and convenience, differing according to 

 whether the need is for a mathematical description 

 of a detailed physiological growth process or for fish- 

 ery managem ent (Moreau, 1987). The results ob- 



4 Lizard Island Research Station. 1992. LIRS, PMB 37, Cairns, 

 Queensland 4870, Australia. Unpubl. data. 



70 



60 



50 



- 40 



30 - 



20 - 



10 



+  



 I 



!!!! 



  



! i I i J ' « 



Age-at 

 Island 



2 2 4 6 8 10 12 14 16 



Age (years) 



Figure 7 



•length data for coral trout, P. leopardus, from Lizard 

 captured by each sampling gear used in this study. 



tained here indicated clearly that the von 

 Bertalanffy model adequately described the growth 

 of the coral trout. Schnute's model was useful because 

 of its flexibility and the stability of its parameters. 



As most fishing gears are selective towards a cer- 

 tain size (Ricker, 1969), and smaller sizes are not 

 usually available, it is common that growth curves 

 are fitted to truncated data representing only part 

 of the population. For the coral trout, because of 

 gear selectivity and legal size restrictions (legal 

 minimum=35 cm TL), only fish of 2+ years were 

 captured by line- and spear-fishing. However, the 

 first three years of life represent the period of fast- 

 est growth, after which the growth pattern changes 

 considerably. As a result, much slower growth rates 

 were obtained when the growth curve was fitted 

 only to the age classes recruited to the fishery. The 

 effects of different age ranges on estimated von 

 Bertalanffy growth parameters have been recog- 

 nized for many years (Knight, 1968; Hirschhorn, 

 1974) and greatly compromise comparisons of 

 growth rates between populations (Mulligan and 

 Leaman, 1992). 



Furthermore, one effect of size-dependent mortal- 

 ity is the selective removal of fast-growing individu- 

 als (Ricker, 1969; Miranda et al., 1987). Thus, it is 

 likely that the average size of the youngest age 

 groups recruited to the fishery will be biased to- 

 wards the largest, fast-growing individuals. This 



