NOTE Link and Almeida: Opportunistic feeding of Myoxocephnkis octodecemspinosus on Georges Bank 



383 



100 



80 



> 60 



F, 40 



20 



1 





Rock crab 



1 



c^ 



 8B| 

 □ 7e! 

 D8E^ 

 H7D 



i 



Gammarid Other 



crustaceans 



Hermit 

 crab 



Sea mouse Polychaete 



Scallop 

 viscera 



Figure 2 



Diet composition {7i by volume t of longhorn sculpin at two dredged and two undredged loca- 

 tions in closed area II. Bars are 959i confidence intervals. Sample sizes are 15, 8, 11,6 fish 

 at stations 8B. 7E, 8E, and 7D respectively. 



a location, the amount of food consumed by sculpins in- 

 creased significantly, suggesting that sculpins continued 

 to feed on crabs and small crustaceans, but opportunis- 

 tically gorged on the remains of scallops that had been 

 shucked and thrown overboard from scallop fishing ves- 

 sels operating in the region. Although anecdotal, the stom- 

 achs of all sculpins at station 8B, the station most intense- 

 ly dredged, were distended beyond normal proportions for 

 20-25 cm fish, often extending below the pectoral fins. 



The mean stomach contents of sculpin were effectively 

 constant across the time series, averaging approximately 

 2 cm^ (Fig. 3). However, the maximal stomach contents, 

 the percent frequency of bivalve viscera, and the index of 

 gorge feeding showed distinct patterns across the time se- 

 ries. All three of these metrics were generally coincident. 

 Years with a higher gorge index corresponded to years with 

 a high occurrence of bivalve viscera in the diet of sculpins 

 (r=0.70. P<0.01). This finding implies that many, but not 

 all. of the gorging events were on scallop viscera and simi- 

 lar discards. The highest values of the gorge index were ob- 

 served in 1987 and 1998. 



Longhorn sculpin abundance exhibited an initial peak 

 and then a relatively steady period for the first 15 years 

 of the survey, followed by a period of lower abundance dur- 

 ing the mid-1980s and an increasing trend in the 1990s 

 (Fig. 4). In most years sculpin abundance ranged from 10 

 to 20 fish per tow. The years with the highest index of scul- 

 pin abundance were 1966 and 1998. In relation to the pre- 

 ceding years, the index of sculpin abundance notably in- 

 creased during 1966, 1987, and 1998. The latter two years 



correspond to years when the index of gorge feeding was 

 highest. The correlation between the gorge index and scul- 

 pin abundance was weak (r=0.21). but significant and pos- 

 itive (P<0.01). 



Discussion 



Much of the information describing the impact of fishing 

 gear on benthic communities and habitats has been decid- 

 edly negative. It is unclear whether populations of scaven- 

 gers such as crabs, flatfish, or other demersal fish benefit 

 from indirect effects of fishing (Kaiser and Spencer, 1994; 

 Greenstreet and Hall, 1996; Ramsay et al, 1996) if they 

 are not caught by the fishing gear. It is known that in the 

 short term, scavengers, especially those small enough to fit 

 through the mesh, are attracted to trawl and dredge tracts 

 (e.g Ramsay et al., 1998; Demestre et al.. 2000). Addi- 

 tionally, some studies have actually shown that the abun- 

 dance of scavenger populations increases in areas that 

 have been trawled (reviewed in Greenstreet and Rogers, 

 2000). Longhorn sculpin on Georges Bank have main- 

 tained their ubiquity and abundance, albeit at relatively 

 low levels compared with other species, during a period 

 of intense fishing pressure in this ecosystem (Fogarty and 

 Murawski. 1998). In recent years the sculpin population 

 on Georges Bank has begun to increase. 



Opportunistic feeding on scallop viscera appears to have 

 a positive influence on sculpin populations. The relation- 

 ship between gorge feeding and sculpin abundance is ad- 



