HACUNDA: TROPHIC RELATIONSmPS AMONG DEMERSAL FISHES 



Table ll. — Mouth dimensions offish species. 



length. The flounders had different ratios (win- 

 dowpane 0.43, yellowtail flounder 0.76, winter 

 flounder 0.87). The longhorn sculpin (1.12) and the 

 Atlantic cod (1.11) had similar mouth shapes. 



Benthos Analysis 



A summary of the species composition by num- 

 bers and weights for the benthic samples is given 

 in Tables 12 and 13. The sediment at stations A-1, 

 A-2, B-1, and C-1 was silty sand, and remaining 

 stations were sand. A total of 55 species were iden- 

 tified. The polychaetes were the dominant group 

 and constituted 51.4% (by number) of the or- 

 ganisms present. Crustaceans (34.1%) and mol- 

 lusks (12.8%) were next in abundance. The re- 

 maining groups (sipunculids(?), nematodes, 

 echinoderms) accounted for only 1.7% of the total 

 number of individuals. The most abundant species 

 were the polychaetes Prionospio steenstrupi, 

 Exogone hebes, Tharyx acutus, Lumbrineris 

 fragilis; the crustaceans Unciola sp. and Am- 

 pelisca agassizi; and the mollusk Nucula proxima. 

 In terms of biomass ( percentage wet weight) mol- 

 lusks (41.2%) were the most important group fol- 

 lowed by polychaetes (41.0%) and crustaceans 

 (8.8%). The biomass was dominated by the mol- 

 lusk A^. proxima (34.6%) and to a lesser extent by 

 the polychaetes L. fragilis, Sternaspis scutata, and 

 P. steenstrupi, and the crustaceans A. agassizi and 

 Unciola sp. 



DISCUSSION 



Recent studies of temperate, coastal marine fish 

 communities have suggested that there is consid- 

 erable division of food resources among predators 

 (Tyler 1972; Kislalioglu and Gibson 1977). Tyler 

 (1972) examined the food utilization among de- 

 mersal fishes in Passamaquoddy Bay, New 

 Brunswick, and found relatively little overlap 

 among diets based on the percentage reoccurrence 

 of principal prey among predators (16% summer 



community; 24% winter community). In Johns 

 Bay the percentage reoccurrence of principal prey 

 among demersal fishes was 20% which is within 

 the range ( 10-24% ) that Tyler calculated for other 

 marine communities. However, assessing dietary 

 overlap by means of the method proposed by Tyler 

 may be misleading because all principal prey are 

 weighted equally in the calculation (see Methods). 

 For example, although the percentage reoccur- 

 rence of principal prey of the fishes in Johns Bay 

 suggests considerable resource division, a closer 

 examination of the data reveals that seven of the 

 eight predators rely primarily on two prey types, 

 Unciola and Leptocheirus (Table 10). If the princi- 

 pal prey items in each predator's diet are weighted 

 in terms of relative importance a more accurate 

 evaluation of dietary overlap may be determined 

 from the partition plot. For the demersal fishes in 

 Johns Bay the percentage reoccurrence of ranked 

 principal prey is 57%, which indicates that pred- 

 ators rely on many of the same major food sources. 

 This conclusion is supported by the food overlap 

 values that were obtained using Pianka's (1973) 

 formula (Table 9). 



There is insufficient information provided in 

 Tyler's (1972) paper to evaluate the relative im- 

 portance of his principal prey; however, a study by 

 Kislalioglu and Gibson (1977) provided another 

 source of data. These authors calculated a 14.7% 

 reoccurrence of principal prey for shallow-water 

 fishes from three habitats in Loch Etive, western 

 Scotland. Food resource partitioning, however, is 

 not as dramatic as this value would indicate be- 

 cause of the inclusion of five pelagic fishes in the 

 calculation. Moreover, almost all the demersal 

 species (13 out of 15) in Loch Etive were primarily 

 dependent on amphipods as their most important 

 food source (based on a points method of stomach 

 content analysis), and among these fishes 25.7% 

 had significant dietary overlap in terms of the 

 proportion of different amphipod species utilized. 

 If the percentage reoccurrence of principal prey is 

 recalculated using weighted principal prey, the 

 result is 56% ^ Trophic similarity is especially evi- 

 dent for the fishes from the open sand-shell mud 

 habitat (which corresponds to the habitat 

 examined in Johns Bay) where there was signifi- 

 cant overlap in amphipod species consumed be- 

 tween four of the five species examined. In Loch 



^Resource division is not strictly comparable to Johns Bay 

 because of the differing degrees of principal prey subdivision 

 which may affect the result of the calculation. 



783 



