FEEDING SELECTIVITY OF DOVER SOLE, MICROSTOMUS PACIFICUS, 



OFF OREGON 



Wendy L. Gabriel^ and William G. Pearcy'^ 



ABSTRACT 



Factors influencing the selection of food by Dover sole were investigated by analyzing stomach contents 

 offish and serially sectioned box core samples for benthic invertebrates from two areas of high Dover 

 sole abundance on the central Oregon continental shelf. At both locations (119 and 426 m depth), 

 polychaetes and ophiuroids were more important than molluscs and crustaceans as food in terms of 

 frequency of occurrence, weights, and numbers. Polychaetes and ophiuroids were generally positively 

 selected at both locations; i.e., they were more common in fish stomachs than in box core samples. 

 Molluscs were generally negatively selected at both locations. Crustaceans were positively selected at 

 426 m and consumed nonselectively at 119 m. The box core samples may, however, underestimate 

 crustaceans and hence give artificially high values of electivity. 



Significant changes in frequency of occurrence of principal prey taxa with fish size were observed for 

 27 principal prey taxa at 119 m and 7 prey taxa at 426 m. These changes indicate that composition of 

 fish diet varies with fish size. 



At the 119 m station, the larger the fish size at which a significant difference in prey frequency 

 occurred, the larger the increase in electivity across the size interval. This implies increased selectivity 

 by large fish. Body size of a prey taxon was positively correlated with fish length at which significant 

 difference in prey frequency occurred: larger fish consumed larger prey. Mean depth of a prey taxon 

 within the sediment was also positively correlated with the length offish at which a significant increase 

 in prey frequency occurred: larger fish consumed prey found deeper in sediment. 



Few size-related changes in diet were found at the 426 m location. Environmental abundance of a 

 preferred taxon, polychaetes, was lower at 426 m than at 119 m. Dover sole may therefore change 

 feeding strategy from that of a specialized predator, whose feeding habits vary with its body size where 

 polychaetes are abundant, to that of a generalist consuming more types and sizes where few 

 polychaetes are available. Vertical distribution of prey within the sediment at 426 m was shallower 

 than at 119 m; thus the advantage afforded large fish in removing deeply buried prey may be 

 eliminated. Implications of results are discussed in terms of optimal foraging strategy. 



The Dover sole, Microstomas pacificus , is a promi- 

 nent member of the deepwater continental shelf 

 community off Oregon (Pearcy 1978) and makes 

 the largest contribution to total biomass of 

 flatfishes landed commercially off the coast of 

 Oregon (Demory et al.^). Yet the published litera- 

 ture on the trophic role of this species in deepwater 

 continental shelf assemblages is sparse. Hager- 

 man (1952) listed "small bivalves ... scaphopods 

 .. .sipunculids, polychaetes [Nereis sp.), nema- 

 todes, echinoids (sea urchins), ophiuroids (brittle 

 stars),. . .gastropods (Thais sp.),...at times... 



'School of Oceanography, Oregon State University, Corvallis, 

 Oreg.; present address: Fisheries Program, Department of For- 

 estry and Wildlife Management, University of Massachusetts, 

 Amherst, MA 0100.3. 



^School of Oceanography, Oregon State University, Corvallis, 

 OR 97331. 



■^Demory, R. L., M. J. Hosie, N. TenEyck, and B. O. Fors- 

 berg. 1976. Groundfish surveys on the continental shelf off 

 Oregon, 1971-74. Oreg. Dep. Fish Wildl. Inf. Rep. 76-9, 7 p. 



shrimp and other crustacean forms" as principal 

 prey animals off California. Pearcy and Hancock 

 (1978) included a list of 35 common polychaete 

 species or taxa, 7 crustacean species or taxa, 9 

 mollusc species or taxa, and 2 echinoderm taxa 

 consumed by Dover sole collected on the central 

 Oregon continental shelf. 



Although selectivity has long been considered 

 an important aspect in resource partitioning with- 

 in and among species, few studies have included a 

 survey of available food items on which to base and 

 compare feeding habit descriptions of benthic 

 fishes. Early work by Steven ( 1930) described prey 

 available and consumed by demersal fishes in the 

 English Channel. Later, Jones (1952) related the 

 Cumberland coast bottom fauna and food of 

 flatfishes. More recently Arntz (1978) described 

 the benthic food web of the western Baltic, includ- 

 ing food selection by the two most common demer- 

 sal fish species found there (the cod, Gadus 



Manuscript accepted June 1981. 



FISHERY BULLETIN: VOL. 79, NO. 4, 1981. 



749 



