low-head hydroelectric dam. Fish. Bull., U.S. 84:220-222. 



Boyd Kynard 

 John P. Warner 



Massachusetts Cooperative Fishery Research Unit 

 University of Massachusetts 

 Holdsworth Hall 

 Amherst, MA 01003 



HABITAT PARTITIONING BY SIZE IN 



WITCH FLOUNDER, GLYPTOCEPHALUS 



CYNOGLOSSUS: A REEVALUATION WITH 



ADDITIONAL DATA AND ADJUSTMENTS 



FOR GEAR SELECTIVITY 



In 1970, Powles and Kohler hypothesized separa- 

 tion of habitats of adult and juvenile witch flounder, 

 Glyptocephalus cynoglossus, by depth based on sur- 

 veys of Nova Scotia Banks and in the Gulf of St. 

 Lawrence. Juveniles were sampled with a small 

 mesh Icelandic shrimp trawl on the Nova Scotia 

 Banks. These data were supplemented by data ob- 

 tained from Squires' (1961) field records collected 

 during shrimp surveys in the Cabot Strait and Gulf 

 of St. Lawrence in the summers of 1957 and 1958 

 using a Norwegian deep-sea shrimp trawl. The 

 authors concluded that during the summer months 

 newly metamorphosed and small (<30 cm) witch 

 flounder were found in the 180-288 m depth range. 



Adult witch flounder (>30 cm) were sampled with 

 a No. 36 Yankee otter trawl on the Nova Scotia 

 Banks from May to October and from November to 

 April. Powles and Kohler (1970) concluded that adult 

 witch flounder were most abundant at a depth range 

 of 92-162 m. In winter months both adults and juven- 

 iles were found together in deeper water while in 

 the summer both groups were separated. 



Powles and Kohler (1970) suggested that this 

 deepwater distribution of juvenile witch flounder 

 could prevent direct competition with young of more 

 abundant species such as Atlantic cod, Gadus 

 morhua, and American plaice, Hippoglossoides 

 platessoides , and provide a natural conservation 

 against fishery exploitation. Their otter trawl 

 catches, over a depth range of 36-450 m, yielded few 

 juvenile witch flounder, although many small 

 American plaice were captured. Escapement of 

 juvenile witch flounder through the mesh in the 

 wings of the trawl was ruled out because many small 

 plaice were captured on the same grounds. The 



authors concluded that juvenile witch flounder were 

 absent unless American plaice and witch flounder 

 differed radically in behavior. Other studies of witch 

 flounder depth distribution on the continental slope 

 off Virginia (Markle 1975) and in the Gulf of St. 

 Lawrence, NAFO (Northwest Atlantic Fisheries 

 Organization) Divisions 4R and 4S (LaFleur and 

 Lussiaa-Berdou 1982) supported the habitat separa- 

 tion hypothesis. 



However, recent studies showed that a No. 36 

 Yankee shrimp trawl was more efficient in catch- 

 ing juveniles whereas a No. 41.5 Yankee otter trawl 

 was more efficient in catching adult witch flounder 

 (Walsh 1984). In that study juvenile American plaice 

 and witch flounder co-occurred in the shrimp trawl 

 catches; differential catches of witch flounder in the 

 otter trawl was due to the escapement of juveniles. 

 Apparent depth separation proposed by Powles and 

 Kohler (1970) may have been based on data biased 

 by gear selection. 



Accurate descriptions of life history patterns of 

 witch flounder are important for sound fisheries 

 management, especially with regard to competition 

 with other species and with regard to presumed 

 mechanisms which protect from overfishing. Powles 

 and Kohler (1970) derived their results from sum- 

 mer and winter surveys, and the conclusions were 

 tentative because of potential gear selectivity prob- 

 lem. Therefore, I reevaluated the depth separation 

 hypothesis with additional data taking gear selec- 

 tivity into consideration. 



Materials and Methods 



Data used in the analysis were obtained from 

 regular groundfish biomass surveys of the Gulf of 

 St. Lawrence, NAFO Divisions 4R and 4S, by re- 

 search vessels of the Northwest Atlantic Fisheries 

 Centre, St. John's, Newfoundland, during the period 

 1978-80. In addition, two juvenile flatfish surveys 

 were used: one in the northern Gulf of St. Lawrence, 

 NAFO Division 4R, 1980; and one in the areas of 

 Hermitage Bay and Fortune Bay, NAFO Division 

 3Ps, in 1981 (Fig. 1). 



Fishing Gears and Research Designs 



Groundfish surveys in September and October of 

 1978-80, NAFO Divisions 4R and 4S by the A. T. 

 Cameron (side trawler) were conducted with a stan- 

 dard No. 41.5 Yankee otter trawl with a stretched 

 mesh size of 127 mm in the wings and reducing to 

 111 mm in the cod end and a 30 mm mesh cod end 

 liner was used. A total of 188-30 min fishing sets 



FISHERY BULLETIN: VOL. 85, NO. 1, 1987. 



147 



