406 



Fishery Bulletin 89(3). 1991 



growing fish (e.g., due to size-selective fishing mortal- 

 ity). The same trend has been observed for yellowtail 

 rockfish and Pacific ocean perch 5. alutus (Leaman and 

 Beamish 1984). 



Our fitted von Bertalanffy growth curves (Table 2, 

 Fig. 4) differed considerably from those reported by 

 Lenarz (1980) and Phillips (1964). Our L m values 

 (Table 2) were lower than the estimates obtained from 

 whole otolith ages (Lenarz 1980) or the combined-sex 

 estimate obtained from scales (Phillips 1964). The 

 decreases in L^ as maximum age increased are con- 

 sistent with Hirschhorn's (1974) observation that a 

 change in the range of ages used to fit the von Ber- 

 talanffy growth curve can result in 

 changes to the parameter estimates. 

 Wilson (1985) reported that use of the 

 broken and burnt technique on S. diplo- 

 proa and S. pinniger significantly altered 

 von Bertalanffy parameter estimates. 

 Our estimate of k for males was lower 

 than that reported by Lenarz whereas the 

 value for females was higher. The com- 

 bined-sex estimate of k reported by 

 Phillips (1964) was greater than either 

 estimate obtained in this study. These dif- 

 ferences in growth rate may be due to the 

 increase in maximum age as well .as dif- 

 ferences in depths and areas sampled. 



35 

 30 



25 

 20 

 15 



142 188 128 

 166 



68 26 25 



24 16 , , ; 1 I ; ' 



L-fH+ttHtl 



Female 



11 6 2 , , 3 3 



3 1 2 

 ' I * 



oL 



-l , I , l , l_ 



2 4 6 8 10 12 14 16 18 20 22 



1 35 



en 



5 25 



_J 



O 20 



O 





Male 



1 2 6 4 



_i i i i i_ 



2 4 6 8 10 12 14 16 18 20 22 



30 



25 



20 



15 



Females 



10 12 



16 18 20 22 



Age (yrs) 



Figure 4 



Mean length-at-age and fitted von Bertalanffy growth curves 

 for male and female shortbelly rockfish (upper two panels) 

 (range of values = thin line; 1 SE range = thick line, asterisks 

 = length at which only a single fish was sampled; numbers 

 above each line = sample size). The lower panel facilitates 

 comparison of the sex-specific growth curves. 



