Midway et al : Stock structure of Paralichthys lethostigma inferred from otolith shape analysis 
333 
0.80 
0.78 
0.76 -| 
0.74 
0.72 
0.70 
0.68 
0.66 -L 
South Atlantic 
Gufl of Mexico 
0.25- C 
0.20 
0.15 - 
0.10 - 
0.05 
0.8 
0.7 -| 
0.6 
0.5 H 
South Atlantic 
1.6 - 
1.5 - 
1.4 - 
1.3 - 
1.2 - 
1.1 
1.0 - 
South Atlantic Gulf of Mexico 
D 
Gufl of Mexico 
South Atlantic 
Gulf of Mexico 
Figure 3 
Boxplots of 4 shape indices for otoliths from female southern flounder ( Paralichthys lethostig- 
ma) collected in 1996 and in 2009-12 in the U.S. South Atlantic and the Gulf of Mexico as part 
of state or federal agency sampling programs and from seafood dealers. Boxplots for the fol- 
lowing shape indices are grouped by basin: (A) rectangularity, (B) form function, (C) ellipticity, 
and (D) aspect ratio. Significant difference (on the basis of Welch’s t-test with a=0.05) between 
groups is indicated by different shadings of the boxes. Boxes represent the interquartile range; 
whiskers indicate the first and third quartiles +1.5 x interquartile range; and dots represent 
points beyond those defined by the whiskers. 
gnon and Morat, 2010). Although the relative influence 
of these controls is the subject of ongoing research, ear- 
ly findings indicate that genetic influences determine 
the overall shape of an otolith and that environmental 
effects contribute finer morphological details (Hussy, 
2008; Vignon and Morat, 2010; Vignon, 2012). 
Almost undoubtedly, southern flounder ranging from 
Texas to North Carolina experience a wide range of en- 
vironmental conditions. Geographic gradients in envi- 
ronmental conditions may differentiate otolith shape 
sufficiently enough to enable detection of large-scale 
population structure, but there also may be consider- 
able local environmental variation (e.g., salinity, tem- 
perature, and food) that effectively masks larger geo- 
graphic patterns. Therefore, the ability to discriminate 
among even broadly spaced locales (e.g., North Caro- 
lina versus Florida) can be compromised. 
Broad and fine-scale processes that affect coastal en- 
vironments, which contribute to fish growth, and poten- 
tially otolith shape also can vary independently in time. 
One way to limit the influence of variable environmen- 
tal effects is by controlling for year class, although sig- 
nificant year-class effects have been detected in some 
otolith shape studies (Castonguay et al., 1991) and not 
in others (Begg and Brown, 2000; Galley et al., 2006). 
Furthermore, in a study of orange roughy ( Hoploste - 
thus atlanticus), Gauldie and Crampton (2002) ex- 
plored the idea of balancing selection — an alternating, 
generational morphology — operating to determine fish 
otolith morphology (which they observed on a 2-year 
cycle in their study). This balancing selection results in 
a persistent otolith polymorphism in populations that 
consist of multiple age classes. Polymorphism related 
to year-class effects could have contributed to within- 
group variation in our study because we included fish 
from multiple year classes, and polymorphism could 
have made it more difficult to detect broader regional 
differences in otolith shape. 
Our study was improved by removal of possible sex 
effects, in addition to our collections being limited to 
relatively young (age-1) and mostly immature (Midway 
and Scharf, 2012) individuals. Because variable envi- 
