Perkinson et al.: Evaluation of the stock structure of Rachycentron canadum in the southeastern United States 
221 
Carolina, North Carolina, and Virginia as water tempera¬ 
tures approach 20°C (Richards, 1967; Smith, 1995; Lefebvre 
and Denson, 2012). The results of work that involved the col¬ 
lection of eggs and larvae, as well as ovarian histology, indi¬ 
cate that spawning occurs in these inshore locations during 
the spring and summer (Smith, 1995; Franks and Brown Pe- 
terson, 2002; Lefebvre and Denson, 2012), although spawn¬ 
ing may also occur with fish aggregating on the continental 
shelf (Hassler and Rainville, 1975). When estuarine and 
nearshore waters drop below 20°C in the fall, cobia move out 
of these areas, although overwintering locations are not well 
known. Because cobia are a popular target for recreational 
anglers throughout their range, state and federal regulations 
have been established to promote sustainable fishing. 
Formal management measures for cobia in the United 
States began with the implementation of the coastal migra¬ 
tory pelagic resources fishery management plan (FMP) in 
1983 (GMFMC and SAFMC, 1983), which established a 
single stock of cobia extending from Texas through the bor¬ 
der of North Carolina and Virginia (later extended through 
New York) and established a size limit of838 mm fork length 
(FL). Management authority was shared by the Gulf of 
Mexico and South Atlantic Fishery Management Councils. 
In 2012, Amendment 18 to the FMP (GMFMC and SAFMC, 
2011) established GOM and Atlantic Ocean migratory 
groups of cobia. Although early genetic analysis revealed 
no differences between GOM and western North Atlantic 
Ocean cobia (Hrincevich, 1993), differences in life-history 
characteristics, such as maximum age and growth rate, 
required the change. The stock boundary was established 
in Monroe County, Florida, at the current demarcation of 
jurisdiction between the management councils. The Monroe 
County stock boundary was chosen on the basis of the doc¬ 
umented seasonal migration of GOM cobia from overwin¬ 
tering grounds in the Florida Keys to the northern GOM 
and the presumption that Atlantic Ocean migratory group 
cobia overwinter in the Florida Keys as well (Williams, 
2001). Initial genetic and conventional tagging data ana¬ 
lyzed in preparation for Southeast Data, Assessment, and 
Review 28 (SEDAR, 2013a, 2013b) refuted the Florida Keys 
as an overwintering location for western North Atlantic 
Ocean cobia and, as a result, Amendment 20B to the FMP 
(GMFMC and SAFMC, 2014) established a new boundary 
between GOM and western North Atlantic Ocean stocks at 
the border of Georgia and Florida. Data that led to the new 
stock delineation are presented herein. 
Identification of fish stocks is necessary to properly allocate 
catch among multiple user groups and effectively manage the 
species under the requirements of the Magnuson-Stevens 
Fishery Conservation and Management Act (MSFCMA, 
2007). Successful stock delineation is also critical to the stock 
assessment process, as most population models assume that 
the stock will have homogeneous life-history characteris¬ 
tics and a closed life cycle in which recruitment occurs from 
within that stock (Cadrin et al., 2005). Because these char¬ 
acteristics differ between GOM and western North Atlantic 
Ocean cobia, an appropriate delineation of the stock bound¬ 
ary is essential to accurately assigning life-history param¬ 
eters such as growth rate, fecundity, and age structure for 
each stock. Stock identification methods include analyzing 
a variety of characteristics, such as meristics, reproduction, 
morphometries, otolith composition and shape, parasite tags, 
and fatty acid profiles (Hilbom and Walters, 1992; Izzo et al., 
2017); however, mark-recapture and genetic analysis are 2 of 
the most commonly used methods of identifying stocks. 
Mark-recapture methods that involve external tags have 
been used for over a century to provide information on fish¬ 
eries (Ricker, 1948). Tagging studies have been used to deter¬ 
mine migratory patterns and stock structure of species such 
as the Atlantic salmon (Salmo salar) (Hansen and Jacobsen, 
2003), billfish (Istiophoridae) (Orbesen et al., 2008), 
Queensland school mackerel ( Scomberomorus queenslandi- 
cus), Australian spotted mackerel (S. munroi) (Begg et al., 
1997), and pollock ( Pollachius virens ) (Neilson et al., 2006). 
In contrast to the use of tag-recapture methods in studies of 
stock structure, the use of genetic analysis in stock identifica¬ 
tion is a relatively recent and rapidly evolving field. Over the 
last 2 decades, genetic methods have been more frequently 
employed to distinguish population structure of fish; genetic 
analyses have included the use of random amplified poly¬ 
morphic DNA in species such as the striped bass ( Morone 
saxatilis) (Bielawski and Pumo, 1997), Pacific cod (Gactus 
microcephalies) (Saitoh, 1998), and Antarctic toothfish (Dis- 
sostichus mawsoni ) (Parker et al., 2002). More recently, 
microsatellite markers have been used to differentiate stock 
structure in Atlantic cod (Gadus morhua ) (Knutsen et al., 
2011) , steelhead ( Oncorhynchus mykiss) (Campbell et al., 
2012) , and eulachon ( Thaleichthys pacificus) (Flannery 
et al., 2013). Our study combined the more traditional 
tag-recapture analysis with modem genetic methods. 
Preliminary genetic and tag-capture analyses of cobia 
stock structure in the southeastern United States con¬ 
ducted in preparation for the 2012 benchmark stock assess¬ 
ment (Perkinson and Denson 4 ; Darden 5 ) cast doubt on the 
accepted stock boundary in the Florida Keys. Recently, 
Dippold et al. (2017) examined cobia migratory patterns 
by using tag-recapture data; however, the study focused 
primarily on 5 zones within the GOM. All regions north of 
the Florida Keys were combined into a single zone, making 
a thorough evaluation of the stock boundary delineation 
between the GOM and Atlantic Ocean stocks difficult. To 
identify the most biologically appropriate delineation, we 
conducted a meta-analysis of all available tag-recapture 
data, adding additional zones to the east coast of Florida 
to provide greater resolution of movement. Additionally, 
we analyzed microsatellite genetic data from cobia to com¬ 
pare the genotypes of fish collected in locations through¬ 
out the GOM and western North Atlantic Ocean. Our 
combination of data sources provided a complementary 
4 Perkinson, M., and M. Denson. 2012. Evaluation of cobia 
movements and distribution using tagging data from the 
Gulf of Mexico and South Atlantic coast of the United States. 
SEDAR28-DW05,17 p. Southeast Data Assessment and Review 
(SEDAR), North Charleston, SC. [Available from website.] 
5 Darden, T. 2012. Cobia preliminary data analyses—US Atlan¬ 
tic and GOM genetic population structure. SEDAR28-DW01, 
2 p. Southeast Data Assessment and Review (SEDAR), North 
Charleston, SC. [Available from website.] 
