McKenna: Trophic relationships of Antarctic demersal fish of South Georgia Island 



651 



of salps is also a feature of the diet of N. squamifrons 

 found at Kerguelen I. (Duhamel and Hureau 1985). 

 Notothenia squamifrons diet displayed the least 

 amount of overlap of all the species examined. It 

 reached a maximum of 32% when compared with that 

 of N. gibberifrons (Table 6). 



Polychaetes were a major component (>23%) of the 

 diets of the three remaining species. Notothenia gib- 

 berifrons is an important commercial species. It was 

 the dominant species (numerically and by biomass) 

 during the 1988-89 AMLR survey of the South Georgia 

 region (McKenna 1989). It is a benthic species, which 

 uses its relatively small, sub terminal mouth to 'slurp 

 up' benthic epi- and infauna (Daniels 1982). It had the 

 most diverse diet of the species examined (Table 2); 

 however, over 50% of its diet consisted of polychaetes. 

 The remainder was comprised of invertebrate epi- 

 fauna and a small amount of krill (5%, Table 3). Its 

 diet overlapped most (56%) with that of M. microps 

 (Table 6). 



Mureanolepis microps is a scaleless, eel-like species 

 with a diverse diet (Table 2). It was the only 'benthic 

 invertebrate feeder' to eat fish (Table 3). Nine per- 

 cent of its diet consisted of salps and 2% of krill. The 

 other items in the diet were benthic organisms. Poly- 

 chaetes were the dominant prey item. Its diet over- 

 lapped most (56%) with that of the N. gibberifrons 

 (Table 6). 



Artedidraco mirus was the smallest species (~8cm 

 TL) consistently caught in the trawl. Its diet was com- 

 posed mostly of isopods and polychaetes (Table 3). This 

 species was represented by only four stomachs, which 

 may not accurately display the true diet of this species 

 at South Georgia. The greatest overlap of its diet (34%) 

 occurred with N. nudifrons (Table 6). 



Discussion 



These results support the conclusions of earlier 

 workers, in that the most abundant species (including 

 those of commercial value) of the South Georgia demer- 

 sal fish community can be classified into three groups 

 based on their summer diets. The largest group con- 

 tained those fish that depended heavily on krill and in- 

 cluded C. gunnari and N. rossii. Fish-eaters repre- 

 sented a second distinct group. Three of the four 

 members of this group (D. eleginoides, C. aceratus, and 

 Pseudochaenichthys georgianus) are commercially valu- 

 able. The third group contained a loose association of 

 species which tended to feed mostly on benthic inver- 

 tebrates. Notothenia gibberifrons and N. squamifrons 

 are the commercially important members of this group. 

 Notothenia gibberifrons is one of the most abundant 

 species in the region, while N. squamifrons is rare on 



the continental shelf around South Georgia (<500m) 

 (McKenna and Saila 1989, McKenna 1989). 



The fishes at South Georgia were using two different 

 food resources: pelagic organisms (mostly krill) and 

 benthic organisms. The large benthic biomass of the 

 Antarctic appears incapable of supporting a large or 

 diverse fish fauna because most of it is in the form of 

 non-food species (e.g., sponges, sea stars) (Belyaev and 

 Ushakov 1957, Andriashev 1965). This may explain 

 why N. gibberifrons, and a few less-common species, 

 were the only fish that depended heavily on benthic 

 organisms. To survive in this environment, these fish 

 have had to diversify their diets and draw upon more 

 prey species than the pelagic fish of the region. 



The majority of fish in the vicinity of South Georgia I. 

 relied on the pelagic food resource. The adaptations of 

 many of these basically benthic fishes to pelagic feeding 

 (Nybelin 1947, Permitin 1970, Eastman 1985, Kock 

 1985b) indicate the relative superiority, in quality 

 and/or availability, of food in the pelagic realm. Krill 

 was the only food resource consumed by all species of 

 Antarctic fish examined in this study. Five of the 

 fifteen species examined relied on krill for greater than 

 50% of their diet. 



There were direct and indirect trophic links between 

 krill and piscivores. All piscivores ate some krill, and 

 most of the identifiable fish that they ate were krill- 

 eaters. However, most of the fish consumed by pisci- 

 vores in this study were unidentifiable, and a stong link 

 between a piscivore (N. rossii) and the benthos (N. gib- 

 berifrons) has been demonstrated at Kerguelen I. 

 (Linkowski et al. 1983). Thus, the relative magnitudes 

 of links between piscivores and krill or the benthos re- 

 main unknown. 



Diet overlap and interspecific competition 



In paired comparisons, there was at least some overlap 

 in the diets, especially in those species relying heavily 

 on krill. However, in most cases the overlap was less 

 than 50%, indicating that resources were effectively 

 partitioned within the South Georgia community dur- 

 ing the austral summer. 



Despite the heavy dependence on krill and high 

 overlap (90%) of the diets of a few species, competi- 

 tion is probably not important in the Antarctic demer- 

 sal fish community during the summer (Targett 1981, 

 Daniels 1982). Competition occurs only when the 

 resource in common use is limiting (Larkin 1963). The 

 availability of krill to Antarctic fish varies (Permitin 

 and Tarverdieva 1979), but whether it is limiting is 

 unknown. In some years, krill is abundant enough to 

 come in contact with the bottom and is then available 

 to even strict benthic feeders like N. gibberifrons 

 (Targett 1981). 



