Comparisons between generalized 

 growth curves for two estuarine 

 populations of the eel tailed catfish 

 Cnidoglanis macrocephalus 



Laurie J. B. Laurenson 



School of Biological and Environmental Sciences. Murdoch University 

 Murdoch, Western Australia, 6 1 50, Australia 



Ian C. Potter* 



School of Biological and Environmental Sciences, Murdoch University, 

 Murdoch. Western Australia. 6 1 50. Australia 



Norm G. Hall 



Western Australian Marine Research Laboratories 

 Perth. Western Australia 6020, Australia 



The eel tailed catfishes (Plotosidae) 

 are distributed throughout the 

 Indo-west Pacific region and com- 

 prise approximately 30 species. 

 Just over half of these species are 

 found in Australian waters (Hoese 

 and Hanley, 1989). The estuarine 

 catfish or cobbler, Cnidoglanis 

 macrocephalus Glinther, is one of 

 three plotosid species that are 

 found in the marine and estuarine 

 waters of the southwestern region 

 of Australia (Kowarsky, 1976; 

 Hutchins and Swainston, 1986). 

 Cnidoglanis macrocephalus can 

 complete its life cycle in estuaries 

 as well as in coastal marine waters 

 (Laurenson et al., 1993a), suggest- 

 ing that the populations of this spe- 

 cies in each of the different estuar- 

 ies represent separate demes, a 

 view supported by the results of 

 electrophoretic studies (Ayvazian et 

 al., 1994). 



Cnidoglanis macrocephalus is 

 the most valuable of several teleo- 

 sts fished commercially in Western 

 Australian estuaries (Lenanton 

 and Potter, 1987). While the perma- 

 nently open Swan and Peel-Harvey 

 estuaries on the southwestern coast 



of Western Australia were previ- 

 ously the main contributors to the 

 fishery for this species (Laurenson 

 et al., 1992), this role has now been 

 assumed by Wilson Inlet on the 

 southern coast of the state (Lauren- 

 son, 1992; Laurenson et al., 1993b). 

 In contrast to the Swan and Peel- 

 Harvey estuaries, Wilson Inlet is 

 seasonally closed and, because of its 

 more southerly location, does not 

 reach as high a temperature in the 

 summer (c.f. Loneragan et al., 1989; 

 Potter etal., 1993). 



Fish are commonly aged by 

 counting the number of annuli on 

 hard structures, such as scales, 

 otoliths, vertebrae, or spines (e.g. 

 Beamish and McFarlane, 1983; 

 Casselman, 1987). However, prior 

 to carrying out such counts, it is 

 important to validate that each of 

 the sequential growth zones is 

 formed annually (e.g. Beamish and 

 McFarlane, 1983; Beckman et al., 

 1989; Collins et al., 1989; Hyndes 

 et al., 1992). Although Nel et al. 

 ( 1985) showed that the translucent 

 zones in the asterisci of C. macro- 

 cephalus from the Swan Estuary 

 tended to be formed annually, their 



results were based on pooled data 

 for all fish and, thus, did not verify 

 that this applied equally to each of 

 the sequential translucent zones. 

 Moreover, since the data for males 

 and females were pooled, it was not 

 possible to determine whether the 

 growth rates of the two sexes in this 

 system were the same. 



A variety of different forms of 

 growth equations can be calculated 

 from 1) the lengths at given ages 

 and 2) back calculations of body 

 length at each annulus, using the 

 relationship between body length 

 and otolith radius. Both calcula- 

 tions use a predetermined "birth 

 date" for the species. The effective- 

 ness of using length-at-age data 

 relies on obtaining representative 

 samples of all age classes. Back 

 calculations are particularly useful 

 when certain age classes have not 

 been sampled effectively but may 

 produce biased estimates of the 

 lengths of younger fish, i.e. Lee's 

 phenomenon (Ricker, 1975). Fur- 

 thermore, the lack of independence 

 of the multiple measures for 

 lengths at annulus formation ob- 

 tained for a single fish by this 

 method may introduce a statistical 

 bias. 



The aims of our study were 1) to 

 validate that each of the sequential 

 translucent growth zones on 

 otoliths of C. macrocephalus in Wil- 

 son Inlet and the Swan Estuary 

 correspond to an annulus and 2) to 

 construct growth curves for each 

 sex in both populations, using both 

 lengths offish at age of capture and 

 back-calculated lengths. These 

 curves were then used to compare 

 a) growth between sexes within 

 each estuary, b) growth between 

 estuaries, and c) growth calculated 

 using lengths at age and back-cal- 

 culated lengths. 



* Send reprint requests and correspondence 

 to the second author. 



Manuscript accepted 11 April 1994. 

 Fishery Bulletin 92:880-889. 



880 



