FISHERY BULLETIN: VOL. 81. NO. -4 



calculated for each sex reflected the length- at- age 

 data with similar values for both sexes through age 4 



120-, 



100 



L.,1113(l-e- 0l30(, - 0216) )°* 

 Lt = 97 6 (l-e-° 162(, - 0742) )A" 

 Lt=90 2(l-e-° 153(t -° 026) )£ 



10 15 20 25 



Age (years) 



30 



35 



FIGURE 7.— von Bertalanffy growth formulae for male, female, and 

 combined tilefish. 



and faster growth rates for males of age 5 and older 

 (Table 1). 



Length and Weight Relationships 



The length:weight regressions for both whole and 

 eviscerated weights were not significantly different 

 between males and females. Log-tansformed 

 regressions were most appropriate. The data were 

 examined with ANCOVA between 50 and 95 cm FL 

 where there were about equal numbers of obser- 

 vations for each sex. Differences between slopes for 

 males and females were more nearly significant for 

 the whole weight regressions than the eviscerated 

 weight regressions (ANCOVA, P = 0.08 and 0.26, 

 respectively). Final regressions were therefore com- 

 puted for all tilefish combined and converted to Flick- 

 er's (1973) functional regression equations. These 

 were 



In (wt) = -5.32 + 3.26 In (FL) 

 and 



Table 1.— Mean back- calculated fork length (cm) at age 1 , empirical length at age, annual increment, and population 

 growth rate (exponential) for female, male, and all tilefish combined in 1978. 



because of apparent otolith growth patterns; back-calculations were not possible for females above age 30 or males above age 25. 

 2 Back-calculated lengths of < 1 8 cm FLmay be inaccurate because no otoliths from fish <1 8cm FL were used to fit the otolith size fork length regres- 

 sion line 



758 



