376 



Fishery Bulletin 98(2) 



Sevigny and de Lafontaine, 1992; Gagne, 1995; Seeb, 

 1998). In the Northwest Atlantic, differences in the 

 electrophoretic mobiUty patterns of the dimeric pro- 

 tein malate dehydrogenase (MDH; EC 1.1.1.37) from 

 Hver tissue appear to distinguish S. fasciatus from 

 S. mentella (Payne and Ni, 1982; McGlade et al., 

 1983; Rubec et al., 1991). The presence of two alleles 

 (MDWAl and MDH*A2} segregating at this locus 

 results in three genotypes that can easily be iden- 

 tified by electrophoresis. A low-mobility banding 

 pattern, corresponding to the genotype MDH'-'A2A2 

 predominates in S. fasciatus whereas a high-mobility 

 pattern, corresponding to the genotype MDH'AIAI, 

 is more characteristic of S. mentella. However, there 

 are two restrictions to using this approach. First and 

 foremost, the heterozygous individuals sharing alleles 

 of both S. fasciatus and S. mentella cannot be assigned 

 unambiguously to one species. Second, electropho- 

 retic mobility of MDH does not differ between S. 

 mentella and S. norvegicus (McGlade et al., 1983). 

 Despite these limitations and until a more reliable 

 taxonomic tool becomes available, electrophoretic 

 mobility patterns of liver MDH remain the best 

 approach for identification of redfish species in the 

 Gulf of St. Lawrence, especially at larval stages. 

 However, one should remember that the taxonomic 

 status of redfish is still under debate and although, 

 for convenience, we use and refer to the species 

 names S. fasciatus and S. mentella throughout the 

 text, these may be putative species. 



Redfish of the North Atlantic are ovoviviparous; 

 their eggs are fertilized by sperm stored in the 

 oviducts. Although the reproductive biology of the 

 redfish of the Gulf of St. Lawrence is not well under- 

 stood, mating (transfer of spermatozoa from male to 

 female) probably takes place during late fall or early 

 winter. Fertilization and embryogenesis take place 

 in winter, and larvae hatch internally and ai-e extru- 

 ded during late spring and early summer (St-Pierre 

 and de Lafontaine, 1995, and references therein). 

 In consequence, the location of larval extrusion may 

 differ significantly from the location where copula- 

 tion has taken place. In order to avoid confusion, the 

 term "larval extrusion," rather than "spawning," is 

 used to distinguish between the hatching-extrusion 

 phase and the mating phase of the redfish reproduc- 

 tive cycle. The importance of gene flow within and 

 between species is established during the mating and 

 fertilization phase of the reproductive cycle, where- 

 as the location of larval hatching and extrusion will 

 determine the geographic distribution of the species 

 at the larval stage. 



It is commonly assumed that the life cycle of the 

 two species of redfish is completed inside the Gulf 

 and that the redfish lai-val population represents a 



mixture of S. fasciatus and iS. me77^e//a. Although the 

 presence of newly hatched redfish larvae in plank- 

 ton samples from the Gulf did confirm that redfish 

 extrude larvae in this area (de Lafontaine, 1990, and 

 references therein; de Lafontaine et al., 1991), the 

 species composition of these larvae has never been 

 elucidated and their distribution in the Gulf has 

 never been described. The objectives of the present 

 study were 1) to describe the species composition of 

 larval redfish in order to verify that the two species 

 use the Gulf of St. Lawrence as an extrusion site, 2) 

 to determine the spatial co-occurrence of S. fasciatus 

 and S. mentella within the Gulf, and 3) to describe 

 the larval size distribution of the two species as an 

 indication of the variability in extrusion times. 



Materials and methods 



Sample collection 



Sampling was conducted at 32 stations along three 

 transects on the western and eastern side of Anti- 

 costi Island in the Gulf of St. Lawrence between 22 

 June and 4 July 1991 and at 74 stations, located 

 mainly to the east of Anticosti Island, between 10 

 June and 20 June 1992 (Figs. 1 and 2). Numerous 

 sampling stations were selected in order to obtain 

 the largest spatial coverage for larval redfish distri- 

 bution in the Gulf of St. Lawrence. Plankton sam- 

 ples were collected with a modified opening-closing 

 l-m- rigid "Tucker" trawl equipped with two, 333-m 

 mesh nets and two G.O. model 2030 flowmeters. 

 The gear was hauled in double-oblique tows from 

 the ship's side at a cruising speed of 2.5-3.0 knots. 

 Sampling was restricted to the 0-50 m upper layer 

 where redfish larvae are concentrated (Kenchington, 

 1991; Runge and de Lafontaine, 1996). Maximum 

 sampling depth was determined by a Vemco acous- 

 tic transducer attached to the gear frame and by cal- 

 culation from the amount of wire length deployed 

 and the wire angle. Tow duration varied between 6 

 and 10 min. Once back on board, nets were rapidly 

 rinsed and samples were concentrated in codends 

 and transferred to the laboratory on the vessel for 

 larval sorting. 



Upon collection, a visual examination of the sam- 

 ples was made. When the estimated number of larvae 

 in one sample was greater than fifty, a subsample 

 was taken for further sorting. All redfish larvae from 

 the subsample were sorted individually and placed 

 gently, while alive, in a drop of water on a plexi- 

 glass plate (95 mm diameter and 0.61 mm thickness) 

 next to an inscribed number that served to identify 

 individuals for the videotape recordings (see below) 



