LANGTON: DIET OVERLAP BETWEEN SEVENTEEN NORTHWEST ATLANTIC FINFISH 



Meganyctiphanes norvegica, is responsible for 

 this since it alone accounts for >63% of each pred- 

 ator's diet. Similarly, the smaller peaks in the 

 figure are also the result of having euphausiids 

 as a common prey item. 



Silver Hake — Yellowtail Flounder 



Silver hake and yellowtail flounder have, for 

 the most part, low levels of dietoverlap (Fig. 7C). 

 The few intermediate levels that do occur are, in 

 all but one instance, related to yellowtail floun- 

 der that are 11-15 cm in length (size class 3) 

 which have preyed on Crangon septemspinosa, 

 Diehelopandalus leptocerus, or small unidenti- 

 fied fish. The one exception is for 6-10 cm (size 

 class 2) yellowtail flounder and silver hake. 

 These fish preyed primarily on Crangon septem- 

 spinosa, Neomysis americana, and amphipods 

 with a resulting diet overlap of 39%. Even with 

 these low overlap levels an overall pattern is 

 apparent; the greatest overlap occurs between 

 the smaller size classes of both species. 



Silver Hake — Fourspot Flounder 



The diets of silver hake and fourspot flounder 

 overlap at low to intermediate levels (Fig. 7D). 

 The highest value, 54%, occurs between 6-10 cm 

 (size class 2) silver hake and 16-20 cm (size class 

 4) fourspot flounder. The peak, as with most of 

 the other intermediate values, is the result of 

 predation on crustaceans such as Crangon sep- 

 temspinosa, Neomysis americana, and Diehelo- 

 pandalus leptocerus. 



DISCUSSION 



The diet overlap comparisons presented here 

 are one way to simplify fish food habits data and 

 to identify real or, at least, potential pathways of 

 energy exchange. As with any method of data re- 

 duction, however, certain compromises have to 

 be accepted which must be kept in mind when 

 discussing the results. The limitations of per- 

 centage similarity calculations have been de- 

 scribed by several authors (Day and Pearcy 

 1968; Moyle 1977; Keast 1977; Langtonand Bow- 

 man 1980; MacPherson 1981)and these limits in- 

 clude both biotic and abiotic factors. Such factors 

 as the taxonomic level of prey identification, the 

 actual quantity of prey consumed (especially 

 since percentage similarity is a relative measure 

 of dietary constituents), predator/prey distribu- 



tion and abundance, and temporal factors that 

 influence both predator and prey behavior all 

 have to be considered in evaluating the meaning 

 of diet overlap data. 



The present data consider the entire northwest 

 Atlantic as a single homogeneous ecological sys- 

 tem since all the available data were grouped by 

 species for the diet overlap calculations. This is a 

 first attempt to examine size-specific finfish 

 predation in the northwest Atlantic and, without 

 more extensive basic biological information on 

 finfish and invertebrate community structure, 

 there was no reason to subdivide the data set. The 

 research survey cruises on which the fish stom- 

 achs were collected were, however, planned for 

 discrete geographic regions (e.g., Gulf of Maine, 

 Georges Bank) and employed stratified random 

 sampling based primarily on depth dependent 

 strata (Grosslein 1969; Clark and Brown 1977). If 

 the research survey catch data were analyzed 

 statistically, using techniques such as cluster 

 analysis, to identify fish species associations or 

 assemblages, then there may be justification for 

 subdividing the data set. Such methods have 

 been utilized recently to identify northwest Pa- 

 cific finfish assemblages (Gabriel and Tyler 

 1980; Tyler et al. in press) and have been used, to 

 a limited degree, for northwest Atlantic fishes 

 (Tyler 1972, 1974; Knight and Tyler 1973). What- 

 ever techniques are employed the basic problem 

 is the same: defining what constitutes an eco- 

 logically homogeneous system. 



From the figures presented, it is clear that 

 completely dissimilar diets occur very rarely. 

 This raises the question of the significance of diet 

 overlap and whether such measures are indica- 

 tive of resource competition. The limits of diet 

 overlap calculations have been dealt with briefly 

 above and the significance of any given numeri- 

 cal value for diet overlap has also been men- 

 tioned. Diet overlap has some value as an indi- 

 cator of potential energy flow pathways but it is 

 not an absolute measure of trophic linkages. The 

 overlap values are obviously indicators of co- 

 existence rather than competition, especially 

 since overlap values have been observed to de- 

 crease rather than increase when resources are 

 limited (Zaret and Rand 1971; Keast 1978; Mac- 

 Pherson 1981). The ideas of competition versus 

 coexistence have been considered for gadoid 

 fishes by Jones (1978). Jones pointed out some of 

 the more subtle distinctions between the diets of 

 three gadoid species which, on cursory examina- 

 tion, appear to overlap. For example, although 



757 



