290 



Fishery Bulletin 102(2) 



rate them into strategies designed to manage exploited 

 species (Avise, 1998). High dispersive capabilities of many 

 marine fish often correlate with low levels of population di- 

 vergence over vast areas (Ward et al., 1994; Graves, 1998) 

 and may be particularly true for species characterized by 

 high fecundity, large population size, and potentially long- 

 distance egg and larval dispersal. Although marine fish 

 predominantly have high dispersal rates and low levels of 

 population structuring, migratory species with continuous 

 distributions may develop and maintain stock structure 

 if they show fidelity to natal spawning sites or limited 

 egg and larval dispersal. Fidelity to natal grounds has 

 been shown in Greenland-Iceland cod (Frank, 1992) and 

 Georges Bank haddock (Polacheck et al., 1992). Genetic 

 divergence between areas originates when populations are 

 formed or through the restriction of gene flow. Cod in some 

 regions are known to migrate long distances, whereas in 

 other regions they are nearly stationary (Lear and Green, 

 1984). Tagging studies in the Gulf of Maine show little 

 exchange between the region east of Browns Bank and 

 Georges Bank, and the inner Gulf of Maine (Hunt et al., 

 1999); however exchange has been reported among Bay of 

 Fundy, southern Nova Scotia, Browns Bank, and Georges 

 Bank populations (Klein-MacPhee, 2002). Such exchange 

 among cod from different management areas may be im- 

 portant for stock assessments and management practices. 

 Determining underlying genetic structure of spawning 

 stocks is paramount to the conservation and management 

 of overexploited species. 



In the last 30 years the use of molecular-based stud- 

 ies in fisheries science has become common (Shaklee 

 and Bentzen, 1998). In cod, a number of studies have 

 used allozymes (Moller, 1968; Jamieson, 1975; Cross and 

 Payne, 1978; Dahle and Jorstad, 1993), but their use and 

 sensitivity are limited because of weak statistical power 

 resulting from low levels of polymorphism and because 

 of processes of balancing selection (Mork et al., 1985; 

 Pogson et al, 1995). Mitochondrial DNA (mtDNA) char- 

 acterization among Northwest Atlantic cod indicates that 

 there is limited, albeit significant, population structuring 

 throughout most the species' range (Smith et al., 1989; 

 Carr and Marshall, 1991; Pepin and Carr, 1993; Carr et 

 al., 1995; Arnason and Palsson, 1996). Genetic divergence 

 at the vesicle membrane protein locus, pantophysin (Panl), 

 originally called GM798 and identified as synaptophysin 

 {SypT) (Fevolden and Pogson, 1997), has been reported 

 among populations of cod from the Northwest Atlantic 

 (Pogson, 2001; Pogson et al., 2001), Norway and the Arctic 

 (Fevolden and Pogson, 1997), and Iceland (Jonsdottir et 

 al., 1999, 2002). High levels of variation have been re- 

 ported at nuclear RFLP loci (Pogson et al., 1995; Pogson 

 et al., 2001), and especially at microsatellite loci (Bentzen 

 et al., 1996; Ruzzante et al., 1996a, 1996b, 1997, 1998; 

 Beacham et al., 1999; Miller et al., 2000; Ruzzante et al., 

 2000, 2001). By using microsatellites, significant genetic 

 structuring has been detected among cod populations on 

 major continental shelves and on neighboring banks that 

 are separated by deep channels and have gyre-like circula- 

 tion patterns hypothesized to act as retention mechanisms 

 for eggs and larvae (Ruzzante et al., 1998). Although both 



Browns and Georges Bank maintain persistent gyre-like 

 circulation patterns that may act to retain eggs and lar- 

 vae, they are separated by the Fundian Channel (>260 m ) 

 which may pose a barrier to juvenile and adult migration 

 (Klein-MacPhee, 2002). Evaluation of Northwest Atlantic 

 haddock by using microsatellites showed similarly signifi- 

 cant stock structuring from Newfoundland to Nantucket 

 Shoals (Lage et al., 2001). Current assessment and man- 

 agement of cod in U.S. waters combine Georges Bank with 

 the regions to its south including Nantucket Shoals. This 

 study investigates genetic stock structure among cod from 

 this region and provides additional insight for scientists 

 and managers. 



Materials and methods 



Sampling 



Samples of adult cod were collected through the U.S. 

 National Marine Fisheries Service and the Canadian 

 Department of Fisheries and Oceans groundfish surveys 

 between 1994 and 2000. Adult cod were obtained from each 

 of the following spawning grounds (Fig. 1): Browns Bank 

 (July 1994, n = 30), Georges Bank (March 1994, n = 48; 

 March 1999. >*=96; In = 144), and Nantucket Shoals 

 (March 2000. n = 97). Blood or tissue (or both) was obtained 

 from individual fish and preserved in 95% ethanol for 

 subsequent DNA extraction. 



DNA extraction, amplification, and visualization 



DNA was extracted by using either a Qiamp DNA Mini 

 Kit ( Qiagen Inc., Valencia, CA) or by following a published 

 protocol designed for nucleated blood cells ( Ruzzante et al., 

 1998). Five microsatellite loci — Grnol, Gmol32 (Brooker 

 et al., 1994), Gmo8, Gmol9, Gmo34 (Miller et al, 2000). 

 and the pantophysin locus. Panl (Fevolden and Pogson, 

 1997; Pogson, 2001) — were used to evaluate genetic diver- 

 sity. Polymerase chain reactions (PCR) of all loci were per- 

 formed in an Eppendorf Mastercycler Gradient thermal 

 cycler. Final concentrations of reagents in a 25 uL PCR 

 cocktail were as follows: -10 ng of genomic DNA. lxPCR 

 buffer pH 9.5 110 mM KC1, 20 mM Tris-HCl pH 8.3, 10 mM 

 (NH 4 ) 2 SOJ, 1.5 mM MgCl 2 , 200 /iM each dNTP, 0.15 nM 

 forward primer. 0.15 ,«M reverse primer (unlabeled for the 

 Panl locus and 5-labeled with a TET, FAM, or HEX ABI 

 dye for all microsatellite loci), and 0.75 units of Taq DNA 

 polymerase. PCR conditions were as follows: initial 5 min 

 at 95°C. 30 cycles of denaturing at 95°C for 1 min, anneal- 

 ing at 50°C (Gmo8, Gmol9, and Gmo34>. 55°C (Pan I), and 

 57°C (Gmol and Gmol32) for 1 min 30 s, and extending 

 at 72°C for 1 min 30 s with a final extension of 72°C for 10 

 min. Gmol9 and G/?2o34, as well as Gmol and Gmol32, 

 were multiplexed in two 25 «L PCR reactions. Flourescent 

 microsatellite PCR products were visualized on an ABI377 

 automated DNA sequencer (Perkin-Elmer Corporation. 

 Foster City, CA) and were analyzed by using GeneScan 

 (vers. 2.1) and Genotyper (vers. 2.1) software programs 

 (Perkin-Elmer Corporation, Foster City, CA). Panl PCR 



