cm, the reverse was found to be true. Our data from the northwest 

 Atlantic are very similar to Rae's (1%9). Annelids accounted for 

 almost three-quarters of the diet overall, and on a regional basis 

 they constituted up to 80-90% of the prey (Tables 1, 5). Other 

 groups such as echinoderms, arthropods, and molluscs made up the 

 remainder of the diet. Although there are regional differences in 

 diet, presumably due to differing environmental conditions and 

 species availability, the food habits of the witch flounder are 

 reasonably constant throughout its geographic range. 



American plaice.— Information on the food of Hippoglossoides 

 plalessoides was first published by Huntsman (1918). He stated that 

 once the young plaice take up a bottom mode of living, they eat 

 small crustaceans such as amphipods and cumaceans as well as 

 small worms. As they grow. Huntsman observed a size-related shift 

 in the fishes' diet with the larger plaice preying very heavily on 

 echinoderms, in particular, sand dollars, sea urchins, and serpent 

 stars. In addition, he found a variety of other crustaceans, worms, 

 and clams in the stomachs but very few fish. Powles (1965) 

 examined the stomachs of American plaice taken from the 

 Magdalen Shallow area in the southwestern part of the Gulf of St. 

 Lawrence. Plaice >30 cm TL (total length) were found to feed 

 more frequently on mysids, amphipods, small echinoderms, and 

 annelids than other types of prey, while larger plaice ( > 30 cm TL) 

 fed on echinoderms and bivalves. Pitt (1973), in an extensive study 

 of Grand Bank American plaice food (International Commission 

 for the Northwest Atlantic Fisheries, Divisions 3L and 3N), found 

 that, overall, echinoderms were the principal prey, especially brittle 

 stars, Ophiura sarsi and O. robusia; sand dollars, Echinarachnius 

 panna; and sea urchins, Sirongylocenironis droebachiensis. Other 

 prey included representatives of most crustacean forms, 

 polychaetes, molluscs, and fish. Pitt, like Powles and Huntsman, 

 found a size-related shift in predation. Our data on American plaice 

 food (Table 1) from farther south in the northwest Atlantic are 

 similar to the above reports. Echinoderms such as sand dollars, 

 brittle stars, and sea urchins were the major prey. Our data is not 

 broken down into size classes but instead is a subsample of the 

 entire plaice survey catch. The average fish size was, however, 

 around 30 cm in all the geographic areas where fish were collected 

 (Table 6), and according to Pitt's data, at this size the most 

 noticeable shift in diet begins to show up. For Grand Banks plaice 

 > 30 cm in length, fish became an increasingly important prey. We 

 did not find fish to be important at all in the American plaice we ex- 

 amined, and this may be due to the relatively small average size for 

 the entire population within our survey area (Fig. 1). In all other 

 respects, the diet of American plaice off the U.S. coast is very 

 similar to its Canadian counterpart, preying heavily on echinoderms 

 and to a lesser degree on arthropods and molluscs. 



Yellowtail flounder.— Although yellowtail fiounder are of com- 

 mercial importance, their feeding habits have been little studied. 

 Sumner et al. (1913 a,b) in a survey of the fauna of Woods Hole, 

 Mass., described the stomach contents of yellowtail flounder as 

 containing enormous numbers of Crustacea, especially amphipods, 

 shrimp, mysids, and small crabs. The two crustacean genera they 

 listed are Caprella and Squilta. In addition to the Crustacea, they 

 also found annelids, bivalve and univalve molluscs, and small 

 fishes. Bigelow and Schroeder (1953) also described the diet of 

 yellowtail fiounder as consisting mainly of the above-mentioned 

 items. More recently, there have been several papers which quan- 

 titatively describe the food of yellowtail flounder for Southern New 

 England, Georges Bank, and the Canadian Grand Banks (Efanov 

 and Vinogradov 1973; Pitt 1976; Bowman and Langton 1978). In 



these three areas, as in the five geographic areas we investigated 

 (Table 7), arthropods and annelids were the major prey. From our 

 data and the published reports on the food of the northwest Atlan- 

 tic yellowtail flounder stocks, one can conclude that this species is 

 exclusively a benthic predator with similar food habits throughout 

 its geographic range. 



Winter nounder. — Winter flounder is an omnivorous fish whose 

 feeding behavior and food have been studied extensively. Much of 

 the data have been summarized by Klein-MacPhee (1978) in a 

 synopsis of the biological data on this species. As far as the food of 

 winter flounder is concerned, the most extensive studies are those of 

 Linton (1921), Pearcy (1962), Richards (1963), Mulkana (1966), 

 MacPhee (1969), Kennedy and Steele (1971), Tyler (1972), Wells et 

 al. (1973), and Frame (1974). Both MacPhee (1969) and Frame 

 (1974) have shown that the diet of winter flounder reflects the en- 

 vironmental conditions under which the fish lives, and this may ex- 

 plain some of the variation in the published data on its diet. Never- 

 theless, based on the prey lists from the above-mentioned papers, it 

 is clear that this flounder is a bottom-feeding fish, usually relying 

 heavily on polychaete worms, bivalve molluscs, and crustaceans as 

 food. It is also interesting to note that because of its habit of feeding 

 in the interiidal zone, algae have been found in the stomach 

 contents, and it has been postulated that the fish may be using 

 plants as a food source (MacPhee I%9; Kennedy and Steele 1971; 

 Wells et al. 1973). Our data are for offshore fish and, consequently, 

 plants contributed an insignificant quantity to the diet, but many of 

 the other taxa listed in the papers mentioned above, especially an- 

 nelids and molluscs, were found to be important prey groups. In 

 addition to annelids and molluscs, however, the Cnidaria also con- 

 tributed substantially to the diet (Table 1). The occurrence of 

 coelenterates in the diet is widespread, as they occurred in the diet in 

 significant quantities (up to 53% of the diet in the Middle Atlantic) 

 in four out of five of the geographic areas surveyed (Table 8) and 

 therefore must be considered a normal prey item of offshore winter 

 flounder. 



Species Comparisons « 



In comparing the prey of the flatfish populations described in 

 this paper, and considering only phyla that account for > 10% of 

 the diet, there are five arthropod, four annelid, three molluscan, 

 two fish, one echinoderm, and one coelenterate predators. 



Arthropods were a major dietary component of five of the flat- 

 fish species examined (Table 1). Arthropods accounted for 86.8% 

 of the diet of windowpane, 42.2% of the Gulf Stream flounder's 

 diet, 40.7% of the fourspot flounder's diet, 37.2% of the yeUowtail 

 flounder's diet, and 16.0% of the American plaice's diet (Table 1). 

 Two of these arthropod predators preyed on pandalid shrimp: the 

 windowpane (22.4%) and fourspot flounder (15.8%). The win- 

 dowpane, however, relied heavily on mysids (41.8%) but also 

 preyed on the Crangonidae (18.4%), while the fourspot flounder 

 consumed a variety of arthropods, including brachyurans and other 

 caridians. Only two species of fish relied on the same arthropod 

 group as prey. Both the yellowtail and Gulf Stream flounders 

 preyed on amphipods, 31.4 and 24.2%, respectively. 



Annelids were the major prey of witch flounder (72.8%) and also 

 accounted for approximately half the diet of Gulf Stream (51.2%) 

 and yellowtail flounders (42.0%) and one-quarter of the diet of 

 winter flounder (26.6%) (Table 1). For all these predators, the an- 

 nelid prey was represented by a number of different families, and, 

 in all but one instance, none of the families individually added up to 

 > 10% of the diet. Only the Nephthyidae accounted for a high 



14 



