Purves et al.: Distribution, demography, and discard mortality of crabs caught as bycatch in the South Atlantic 887 



handled on a pot vessel, as they would be on a trawl 

 vessel, mortalities are higher. 



Other factors, which could not be tested in the 

 re-immersion experiment, may also affect the rate 

 of crab survival. We re-immersed crabs in pots, 

 whereas normally they would be simply dropped 

 into the sea and would be subject to predation from 

 birds and fish before they reached the bottom. The 

 effect of discarding crabs away from their original 

 habitat is unknown, but our results demonstrate 

 a clear depth separation between the two species; 

 therefore one would expect at least an energetic 

 cost if crabs have to relocate. The crabs subjected 

 to re-immersion experiments were sampled imme- 

 diately before being discarded. They might suffer 

 further damage through the actual discard process; 

 for instance Stevens (1990) speculated that, while 

 traveling through offal chutes, they could become 

 entangled in machinery or suffer further damage 

 upon impact with the surface of the water Ide- 

 ally crabs should have been sampled after being 

 through the full discarding procedure, but this was 

 not practical. Finally, eggs often became dislodged 

 during handling and this loss possibly impacted 

 reproductive success. 



Zhou and Shirley (1995) presented results that indi- 

 cate that there are no long-term effects of handling on 

 crab survival, feeding rate, or crab condition; therefore 

 we might reasonably expect that the survival rates seen 

 in our experiments would also be the relevant long-term 

 survival rates. However, even with relatively low discard 

 mortality, the impacts of repeated catching and discarding 

 of individuals will have a cumulative effect on crab popula- 

 tions. Both retained and discarded bycatch should there- 

 fore continue to be reported and be incorporated into crab 

 population models. The presence of such a large discarded 

 bycatch might provide the opportunity for the retention 

 (and removal from the population) of parasitized crabs, as 

 suggested by Basson (1994). 



Crabs are an inconvenience in a fishery targeting tooth- 

 fish. In situ observations made during the AUDOS ex- 

 periments on the UK's January 2000 survey confirm that 

 toothfish seek to avoid direct contact with crabs (Yau et al., 

 2002), although crabs do form a component of their food 

 (Pilling et al., 2001). An inverse relationship was found 

 in the present study between toothfish numbers in pots 

 and crab numbers in pots, suggesting toothfish avoid pots 

 with large crab populations (Fig. 9). Therefore, conducting 

 the toothfish pot fishery in an area of low crab abundance 

 is sensible, and our data do suggest that, at intermedi- 

 ate depths, the crab catch should be low and composed 

 primarily of large P. formosa. Avoidance of areas of high 

 crab bycatch will also reduce the mortality associated with 

 discarding female and undersize male crabs. These discard 

 levels are very high (>93%) — considerably higher than 

 those in the Bering Sea (85%: Stevens, 1996). Such high 

 discard levels could be reduced further by developing new 

 pot designs to limit crab catches to larger, legal-size ani- 

 mals — for instance designs with excluder panels (Stevens, 

 1996) — or perhaps by reducing the minimum size limit. 



14 



12 ro 

 u 



10 o 



D Frequency of 

 number/pot 

 toothfish 



 Number/pot 

 P spinosissima 



0-1 



1-2 



2-3 



CPUE (number/pot) of toothfish 



Figure 9 



The frequency distribution of the number of toothfish caught per 

 pot compared with the CPUE's of Paralomis spinosissima and 

 P. formosa (numbers/pot) and of toothfish {Dissostichus eleginoides) 

 catches (numbers/pot), as found during the first cruise of the Argos 

 Georgia during March to May 2000. 



Acknowledgments 



We wish to acknowledge the assistance of Argos Ltd. and 

 the excellent cooperation of the captains of the two ves- 

 sels. Captain Joaquin Abraldes Gonzalez and Captain Jose 

 Andres Sampedro and their crew. Permission to publish 

 these data was kindly granted by Argos Ltd. The Govern- 

 ment of South Georgia and the South Sandwich Islands 

 funded the research by MRAG Ltd. on South Georgia 

 fisheries. 



Literature citations 



Agnew, D. J., T. M. Daw, M. G. Purves, and G. M. Pilling. 



2001. Fishingfor toothfish using pots: results of trials under- 

 taken around South Georgia, March-May 2000. CCAMLR 

 (Commission for the Conservation of Antartic Marine 

 Living Resources) Sci, 8:93-105. 

 Agnew, D. J., L. Heaps, C. Jones, A. Watson, K. Berkieta, and 

 J. Pearce. 



1999. Depth distribution and spawning pattern of Dissosti- 

 chus eleginoides at South Georgia. CCAMLR Sci. 6:19-36. 

 Arnauld, P. M. 



1976. Peches experimentales de Lithodes murrayi Hender- 

 son, 1888 (Crustacea, Anomura) aux iles Crozet (SW de 

 I'Oceanlndien). Thetys 3:167-172. 



Arnauld, R M., and T. Do-Chi. 



1977. Donnees biologiques et biometriques sur les lithodes 

 Lithodes murrayi (Crustacea: Decapoda: Anomura) des lies 

 Crozet (SW de I'Ocean Indien). Mar Biol. 39:147-159. 



Basson, M. 



1994. A preliminary investigation of the possible effects of 

 rhizocephalan parasitism on the management of the crab 

 fishery around South Georgia. CCAMLR Sci. 1:175-192. 

 Benedict, J. E. 



1894. Descriptions of new genera and species of crabs of 

 the family Lithodidae, with notes on the young oi Lithodes 



