236 



Fishery Bulletin 97(2), 1999 



160 



140 



120 



100 



BO 



60 



40 



3 20 



I 



120 

 100 

 80 

 60 

 40 

 20 



5 10 15 20 25 30 35 40 45 50 55 60 65 



Age (years) 



Figure 8 



Hapuku length-at-age data (R, ,,), with fitted von Bertalanffy growth 

 cur\'es for (A) all samples and both sexes combined (R, and R, ,1. and 

 (B) all samples combined and males and females separately (R2 2'- See 

 Table 2 for cur\'e equations. 



length-frequency data represent only small, imma- 

 ture hapuku. Over the age range 0-12 years, the 

 MULTIFAN growth curve was very similar to growth 

 curves based on otolith ages (Fig. 10). 



Tag recapture 



One tagged hapuku was recaptured, released with a 

 new tag, and recaptured a second time (Johnston. 

 19921. It grew from 73 cm to 83 cm during its first 

 800 days at liberty and from 83 cm to 96 cm during 

 its second 829 days at liberty. This fish was entered 

 into the growth data set three times: once for each of 

 the two periods at liberty and once for the combined 

 period at liberty. All other hapuku used in the growth 

 analysis were recaptured only once. 



One hundred and forty four recaptured hapuku had 

 sufficient information (lengths at tagging and recap- 

 ture, days at liberty) to be used in the growth analy- 

 sis. Most of these hapuku were between 50 and 90 

 cm long at tagging (Fig. 11) and would have been 

 immature. The GROTAG reference lengths a and /3 

 were set at 55 cm and 85 cm respectively. Periods at 

 liberty ranged from 9 to 3708 days (0-10.2 years), 

 but most fish were at liberty for less than four years 

 (Fig. 12). 



Initial GROTAG model fits identified six fish as 

 outliers (fish having absolute standardized residu- 

 als greater than 3.0). Five of these fish had negative 

 growth increments of 3-27 cm, and one had a posi- 

 tive increment of 24 cm after 347 days at liberty. 

 These outliers probably resulted from measurement 



