42 



Fishery Bulletin 97(1), 1999 





175 195 215 235 255 

 Total length (mm) 



liMililiiu  



275 295 315 



Figure 2 



Length-frequency distribution of E. cimhriuf collected during the GM 93-20 survey between 2 

 August and 13 August 1993 plotted m groups of 5 le.g. 95=91-95), (n=727). 



from 95 to 328 mm TL (Fig. 2). Shrinkage was not 

 significant (P=0.2465), therefore, all total lengths 

 reported in this study are from previously frozen fish. 



None of the treatment effects (LD, HS, EDTA) 

 employed on both otoliths from fish collected at sta- 

 tion 40 was found to be significantly different 

 (P=0.9488). The EDTA method was selected because 

 it produced the highest r^ value (/•-=0.9961) in a re- 

 gression analysis. There was no significant differ- 

 ence between left and right otoliths (P=0.8934). 



Nine age classes were identified (Table 1). An 

 AN OVA was run on the assigned ages of the 281 rock- 

 lings. The three independent trials were not signifi- 

 cantly different from one another (P=0.8028). There 

 was a high correlation (P=0.984) between total length 

 and age (Fig. 3). Because of the difficulty in inter- 

 preting annular growth or disagreement about age 

 over the three independent readings, 28 out of 281 

 otoliths were discarded. 



Linear growth was significantly different in males 

 and females (P=0.0001) (Table 2). Nine male and 

 eight female age classes were observed. To account 

 for the sex differences in growth, the length frequency 

 has been subdivided to show the distribution of male, 

 female, and undetermined sexes (Fig. 4). Undetermined 

 sexes were either immature or badly preserved. 



There was a significant difference between the 

 weights of males and females (P^O.OOOl). The aver- 

 age weight of all sexed specimens regardless of age 



showed that males weighed more than females (mean 

 26.5 ±19.3 and 24.5 ±24.2 g, respectively). Females 

 were heavier at age than males for the first three 

 year classes (Fig. 5). A scattergram of age versus total 

 length (mm) and weight (g) is presented in Figure 6. 



Regression analysis of TL vs. OL is defined hy y = 

 0.014.V + 0.838 where 0.838 is the intercept used to 

 standardize the Fraser-Lee back-calculation equa- 

 tion. Back-calculated lengths at age (Table 3) and 



