Clark and Tracey: Population changes of Hoplostethus atlanticus on the Challenger Plateau 



243 



suggest low values of F in recent years, which seems 

 unlikely given the substantial effort in the fishery 

 yet catches being less than the TAC. 



There is considerable uncertainty in all the data 

 sets. However, assuming a virgin biomass of 110,000 t 

 (as an approximation of the trawl survey and win- 

 ter CPUE values), the decline in mid-year biomass 

 of the population was rapid and the level in 1991 

 was about 20% of the virgin level (Table 8). 



Other species 



Biomass indices of the 12 main bycatch species caught 

 in the trawl surveys from 1987 to 1989 are presented 

 in Table 9. The coefficients of variation of these mean 



values range from 11% to 69% and differ between 

 years and species, which limits their comparability. 



However, there were no strong indications of in- 

 creasing abundance of any species relative to abun- 

 dance in 1987. There was little apparent change in 

 abundance of ribaldo (Mora moro), leafscaled gulper 

 shark (Centrophorus squamosus), widenosed chi- 

 maera (Rhinochimaera pacifica), spiky oreo (Neoc- 

 yttus rhomboidalis), Owston's spiny dogfish (Centro- 

 scymnus owstoni), or white rattail (Trachyrinchus 

 sp.). Declining abundance was suggested for big 

 scaled brown slickhead (Alepocephalus sp.), basket- 

 work eel (Diastobranchus capensis), Johnson's cod 

 (Halargyreus johnsoni), smallscaled brown slickhead 

 [Alepocephalus australis), shovelnosed spiny dogfish 

 (Deania calcea), and seal shark (Dalatias licha). The 

 biomass of species relative to orange roughy has 

 generally increased for all species except seal shark. 

 This change is strongest for ribaldo, Owston's dog- 

 fish, widenosed chimaera, leafscaled gulper shark, 

 and white rattail. 



Size structure 



Length-frequency distributions of orange roughy 

 from the entire survey area were similar in all years 

 with no marked differences from the overall 

 weighted length frequency (Fig. 4). There was a 

 strong unimodal distribution with the peak at 32- 

 33 cm standard length. Fish ranged in standard 

 length from 9 cm to 44 cm. 



Sex ratios varied between years, but were gener- 

 ally about 1:1. However, females always dominated 

 the size distribution above about 35 cm. The mean 



