Southern flounder iParalichthys 
lethostigma ) stock structure inferred 
from otolith shape analysis 
1 Department of Biology and Marine Biology 
University of North Carolina Wilmington 
601 South College Road 
Wilmington, North Carolina 28403 
Present address for contact author: Coastal Carolina University 
Department of Biology 
P.O. Box 261954 
Conway, South Carolina 29528 
Abstract — Examination of otolith 
morphometric variation has been 
shown to provide improved descrip- 
tions of stock structure for several 
marine fish species. We examined 
spatial variation in otolith shape 
of southern flounder (Paralichthys 
lethostigma) to understand popula- 
tion structure at the following geo- 
graphic levels: ocean basin (Atlantic 
and Gulf of Mexico); regional coastal 
waters (Texas, Florida) and (Georgia, 
South Carolina); and local coastal 
waters (North Carolina). To reduce 
variability, we considered only age- 
1 female fish. From digitized otolith 
images, we extracted descriptions for 
common shape indices and elliptical 
Fourier coefficients and found strong 
evidence for differences at the ocean 
basin scale, but only weak evidence 
for structure at either within-basin 
(i.e., among states) or within-state 
(local) spatial scales. Our finding of 
inferred stock structure differences 
between the ocean basins aligns with 
the geographic break in the distri- 
bution of this species — the absence 
of this species from the southern 
portion of Florida — as well as with 
recent genetic findings. Currently, 
state-level management of southern 
flounder in both areas does not ac- 
count for any basin-wide population 
mixing and, therefore, by default, as- 
sumes a separate unit stock for each 
state, although our findings indicate 
that mixing could be extensive. Ad- 
ditional sources of information (e.g., 
genetics, life history traits) collected 
at appropriate spatial scales should 
be examined to confirm suspected 
levels of mixing and to determine 
suitable management strategies for 
the conservation of southern floun- 
der stocks throughout their ranges. 
Manuscript submitted 7 August 2013. 
Manuscript accepted 9 September 2014. 
Fish. Bull. 112:326-338 (2014). 
doi:10.7755/FB.112.4.9 
The views and opinions expressed or 
implied in this article are those of the 
author (or authors) and do not necessarily 
reflect the position of the National 
Marine Fisheries Service, NOAA. 
Stephen R. Midway (contact author ) 1 
Steven X. Cadrin 2 
Frederick S. Scharf 1 
Email address for contact author: srm30@psu.edu 
2 School for Marine Science and Technology 
University of Massachusetts Dartmouth 
200 Mill Road, Suite 325 
Fairhaven, Massachusetts 02719 
Although many marine populations 
once were considered to be panmic- 
tic on the basis of large geographi- 
cal ranges and larval dispersal over 
long distances, results from im- 
proved stock identification methods 
(e.g., mtDNA, parasite community, 
and shape analyses) are calling into 
question some initial assumptions 
of population homogeneity (Cadrin 
et al., 2005). For example, although 
only very small amounts of gene flow 
may be required to homogenize con- 
siderable genetic variation without 
selection (Palumbi, 2003), evidence 
is accumulating from multiple spe- 
cies to support the existence of fine- 
scale geographic structure in several 
adaptive traits (Conover et al., 2006). 
Regardless of their genetic simi- 
larities or differences, fish stocks 
possessing variable traits that can 
affect their responses to harvest still 
must be delineated clearly to achieve 
management objectives related to 
yield maximization and biomass con- 
servation (Ricker, 1958; Begg et al., 
1999; Conover et al., 2006). Spatial 
structure of fish stocks and the al- 
location of fishing effort, therefore, 
should be considerations in the man- 
agement of any species (Stephenson, 
1999; Ying et al., 2011) because ig- 
noring population structure can lead 
to negative outcomes, such as loss of 
genetic diversity and reduction in the 
yield-generating potential of a stock 
(Pawson and Jennings, 1996; Bailey, 
1997; Booke, 1999). Within fisheries 
science, stock definitions vary but 
are focused largely on consistency 
of unique traits — the characteristics 
that distinguish a stock should re- 
main constant through time and be 
unique to that stock (Ihssen et al., 
1981; Booke, 1999) for both conser- 
vation and harvest purposes (Cadrin, 
2000 ). 
Methods of phenotypic stock iden- 
tification have expanded greatly 
from abundance and meristic ap- 
proaches to now include the use of 
both natural and artificial tags, ex- 
amination of life history traits, pop- 
ulation genetics, and morphometric 
outlines (reviewed in Cadrin et al., 
2005). Recently, the study of closed- 
form structures, such as otoliths and 
scales, has increased with the advent 
of computers that are able to rapidly 
analyze large amounts of data. In 
addition, otoliths are collected rou- 
