FISHERY BULLETIN: VOL. 79, NO. 4 



Etive as in Johns Bay, use of an unweighted per- 

 centage reoccurrence of principal prey to evaluate 

 dietary overlap gives an exaggerated picture of 

 partitioning of prey types. 



Trophic partitioning by prey size was apparent 

 for the three flounder species examined from 

 Johns Bay (Figure 2). Keast and Webb (1966) have 

 stressed the importance of mouth morphology and 

 body form in channeling predators towards dis- 

 tinct prey. The small-mouthed winter flounder 

 selected small crustaceans, mainly amphipods, 

 and the larger mouthed windowpane concentrated 

 on larger prey, primarily mysids. The yellowtail 

 flounder had a mouth size intermediate between 

 the other two flounder species and it fed on prey 

 from both size ranges. Ross (1977) noticed a simi- 

 lar segregation of prey sizes by searobins as spa- 

 tial overlap increased. Resource partitioning by 

 prey size was at a minimum between the Atlantic 

 cod and longhorn sculpin (Figure 3). Both of these 

 species had a similar mouth shape and ingested 

 prey of the same size range. The similarity of prey 

 size utilization is reflected in the high food overlap 

 value (0.74) for these two species. Hespenheide 

 (1975) observed a strong correlation between prey 

 size overlap and prey type overlap for cohabiting 

 birds. 



An analysis of the benthic infauna was made to 

 determine potentially available food and selectiv- 

 ity of prey by the demersal fishes (Tables 12, 13). 

 Availability depends not only on prey abundance, 

 but also on the interactions of other factors, in- 

 cluding prey size, microdistribution, capture suc- 

 cess, and speed of movement (Griffiths 1975). Al- 

 though polychaetes and moUusks dominated in 

 the bottom sediments, crustaceans were the pre- 

 ferred food of the demersal fishes. Generally, pred- 

 ators consumed prey that were active either at the 

 sediment surface or in the upper few centimeters 

 of the bottom sediments. Some abundant food 

 items, such as Nucula proxima, Prionospio 

 steenstrupi, and Exogone hebes were not impor- 

 tant dietary constituents. The small size of P. 

 steenstrupi and E. hebes probably limits their 

 selection by predators. Predation on A'^. proxima 

 may be low because the feeding structures of some 

 predators prevent extensive burrowing in the sed- 

 iment or because of this bivalve's low caloric value. 

 Optimal feeding strategy predicts that animals 

 should feed on prey that give the maximum energy 

 yield per unit time and this will govern the degree 

 of palatability of a prey item (Schoener 1971; 

 Emlen 1973). 



Recent work by Virnstein (1977) in Chesapeake 

 Bay concluded that infaunal densities in soft- 

 bottom communities are predator controlled. The 

 cropping pressure of the demersal predators 

 checks the population growth of many prolific 

 benthic invertebrates. The benthos in Johns Bay 

 is subject to varying amounts of predation pres- 

 sure throughout the year. During the winter, the 

 fish community in Johns Bay was very depauper- 

 ate and it is likely that many fishes moved into 

 warmer water offshore (Edwards 1964). Many of 

 these demersal fishes show a decrease in feeding 

 rate as temperature drops (Tyler^) and the winter 

 flounder ceased feeding during the cold months. 

 Because of the reduced abundance and lowered 

 metabolism of the fishes, predation on the benthos 

 was probably at a minimum during the winter. 

 During the warmer months there was an influx of 

 fish species into the bay and an increase in fish 

 diversity and abundance. Environmental condi- 

 tions are favorable at this time and the food supply 

 may be abundant enough to support the expanded 

 fish community without competitive interactions. 



The demersal fishes in Johns Bay occupy the 

 same habitat and there is considerable spatial 

 overlap in their foraging zones. Active predators 

 (e.g., Atlantic cod, red hake) forage over a wider 

 area than sedentary predators (e.g., longhorn 

 sculpin, ocean pout). These wide-ranging species 

 may feed in the foraging zones of several seden- 

 tary individuals. My data suggest that the benthic 

 fishes partition food resources by selecting prey 

 from different depth strata (microhabitats) in the 

 environment. Predators may choose either in- 

 faunal, epifaunal, or nek tonic organisms and the 

 proportion of these prey types in the diet is a re- 

 flection of preferred foraging strata (Table 14). At 

 one extreme are predators that feed largely on 

 nektonic prey (e.g., windowpane), while other 

 fishes are strongly dependent on bottom-dwelling 

 organisms (e.g., yellowtail flounder). 



The trophic similarity of the demersal fishes in 

 this coastal community suggests that in a food- 

 limited environment many of these predators 

 would experience intense competition. However, 

 establishing food limitation is a difficult task be- 

 cause information is lacking both on benthic pro- 

 duction rates and the food rations required by the 



*ryler, A. V 1971. Monthly changes in stomach contents of 

 demersal fishes in Passamaquoddy Bay, N.B. Fish. Res. Board 

 Can.,Tech. Rep. 288, 114p. 



786 



