644 



Fishery Bulletin 89(4). 1991 



Figure 1 



Location of stations sampled around 

 South Georgia I., Antarctica, by the 

 1987-88 AMLR demersal fish survey. 



To effectively manage fish stocks, 

 it is necessary to understand the im- 

 pact of fishing on the fish community. 

 This includes understanding inter- 

 specific relationships among fish and 

 with other organisms in their envi- 

 ronment. Knowledge of trophic rela- 

 tionships is necessary to describe 

 these ecological links (Edwards and 

 Bowman 1979, Grosslein et al. 1980, 

 Langton 1983, Sissenwine 1984). 



A number of other studies have 

 examined the diets of Antarctic fish (Holloway 1969, 

 Permitin and Tarverdieva 1979, Yukhov 1971, Rakusa- 

 Suszczewski and Piasek 1973, McCleave et al. 1977, 

 Moreno and Osorio 1977). Everson (1984b) reviews the 

 feeding ecology of Antarctic fishes. However, few of 

 these studies have quantitatively measured the inter- 

 specific overlap of diets. The most extensive work is 

 that of Targett (1981) who gave a quantitative descrip- 

 tion of the community structure and trophic relation- 

 ships of demersal fish around three Antarctic islands. 

 However, his collections were made with a small net 

 (3 m) (not easily compared with those used commercial- 

 ly, (17-18 m)) and only two stations were sampled in 

 the South Georgia I. region late in the autumn of 1975. 



Targett's work emphasized the degree of resource 

 partitioning by demersal fish. In general, he found a 

 low degree of overlap in food utilization. In the case 

 of commercial species, however (Champsocephalus 

 gunnari, South Georgia icefish Pseudochaenichthys 

 georgianus, and painted notie Nototheniops larseni), 

 there was a relatively high degree of overlap. This 

 group consists mostly of predators on krill. At one 

 South Georgia station, all fishes preyed on krill. 



The present study reports on a quantitative exam- 

 ination of the trophic relationships within the demer- 

 sal fish community on the continental shelf (~5 km off- 

 shore) around South Georgia I. 



Methods 



The specimens for this study were collected from the 

 research vessel RV Profesor Siedlecki using commer- 

 cial fishing gear, and represent the fishable community 

 that was present during December 1987-January 1988. 

 Fish were collected by 30-minute tows of a P32/36 ot- 

 ter trawl (mouth opening of 17.5 m,' 43-52 mm mesh 



liner) as part of the 1987-88 Antarctic Marine Living 

 Resources (AMLR) survey of the fish stocks around 

 South Georgia I. Successful trawls were made at 108 

 stations within about 5 km of the island (Fig. 1). These 

 stations were randomly located over the continental 

 shelf within three depth strata (50- 150 m, 150-250 m, 

 250-500 m) (Gabriel 1987, McKenna and Saila 1989). 

 The collection methods are described in detail by 

 McKenna and Saila (1989). 



This sampling provided a substantial size range of 

 individuals from the common species and at least a few 

 representatives of most of the rare species (Table 1; 

 McKenna 1990, app. C). A total of 321 stomachs were 

 collected from 15 species of fish. Three of the fifteen 

 species from which stomachs were collected belonged 

 to the icefish family (Channichthyidae; Champsocepha- 

 lus gunnari, black-finned icefish Chaenocephalus acera- 

 tus, Pseudochaenichthys georgianus). Eight species 

 were Antarctic cods (Nototheniidae; Notothenia rossii, 

 gray rockcod Notothenia squamifrons, striped rockcod 

 Pagothenia hansoni, Nototheniops larseni, yellowfin 

 notie Nototheniops nudifrons, humped rockcod Noto- 

 thenia gibberifrons, Patagonian rockcod Patagonothen 

 brevicauda guntheri, and Patagonian toothfish Disso- 

 stichus eleginoides). Two species of Bathydraconidae 

 (South Georgia icedragon Parachaenichthys georgi- 

 anus, and bronze icedragon Psilodraco breviceps) were 

 represented. The remaining two were members of 

 some of the less-common families of the region (Arte- 

 didraconidae: fancy plunderfish Artedidraco mirus, 

 Mureanolepididae: smalleye morey cod Muraenolepis 

 microps. 



Length (TL and SL) and weight measurements were 

 taken for fish from each trawl. A stomach index value 

 was also recorded for each fish. This index ranged from 

 zero (empty stomachs) to five (full stomach). Individuals 

 with a stomach index (Permitin and Tarverdieva 1979) 



