244 



Fishery Bulletin 92(2). 1994 



size of females was significantly greater than that 

 of males «-test, P<0.05). Size data were also exam- 

 ined by the following subareas: Central Flat, Pin- 

 nacles, and Westpac Bank. Length frequencies by 

 sex were generally similar to those for the total 

 survey area and are not presented here. However, 

 the distributions were approximately normal in 

 shape and could be described by their means and 

 standard errors to provide a simpler comparison 

 between areas and between years (Fig. 5). There 

 were no apparent differences between years or be- 

 tween areas «-test, P<0.05), and no consistent trend 

 in size over time. 



Reproduction 



All trawl surveys occurred during the months of 

 June— July. There was considerable variation evident 

 in the overall proportions of fish of different repro- 

 ductive stage between years and areas (Table 10). 

 To a large extent this reflected the timing of the 

 survey (e.g. 1987 was earlier than the others) and 

 showed a high proportion of maturing fish. However, 

 in all years the majority of fish sampled were ma- 

 ture and were involved in that year's spawning 

 (stages 3-6). 



The Central Flat and Pinnacle regions were typi- 

 cally dominated by mature fish in or near spawn- 

 ing condition. In 1987 on the Westpac Bank there 

 was a high proportion of fish that were in very early 

 stages of maturation and hence unlikely to spawn 

 in that year. From 1988 to 1990 the proportion of 

 actively spawning fish increased. However, levels of 

 nonspawning fish have consistently been higher 

 here than in the Central Flat and Pinnacle areas. 



Timing of spawning 



The progression of gonad stages appeared consistent 

 between years for which data spanning several 

 weeks in July exist (Fig. 6). A pattern of maturing 

 fish declining to low levels was observed early in 



