Anderson et al.: Evolutionary associations between Cynoscion arenarius and C nothus 
15 
98° W 
96° W 
94° W 
30° N 
28° N 
26° N 
Map of the Texas gulf coast, showing the study area (hatched area), as 
well as major bays and inlets of Texas provided as points of reference. 
Trawls were collected in the offshore region of the gulf coast adjacent to 
Galveston Bay, TX, and sand seatrout (C. arenarius) and silver seatrout 
(C. nothus) were retained from each trawl. 
tiple individuals (Palmeirim, 1998). 
Morphometric and meristic measure- 
ments were taken on the left side of 
each fish. In the event that fins or 
scales were damaged, the right side 
was used, and no measurements 
were taken if both sides were dam- 
aged. Body weight, standard length 
(SL), pectoral-fin length, pelvic-fin 
length, anal-fin base length, and 
eye diameter were measured in 
each specimen. Weight was taken 
in grams (g) and length measure- 
ments were taken in millimeters 
(mm). Unless otherwise indicated, 
statistical analyses were performed 
with SAS software (vers. 8.02, SAS 
Inst., Inc., Cary, NO. Differences 
in weight and length between the 
species were assessed with a Sat- 
terthwaite t-test for unequal vari- 
ances (after failure of an equality- 
of-variance test). Length and weight 
were log-transformed to normalize 
extreme observations, and the inter- 
specific difference between length- 
to-weight ratios was tested with a 
pooled t-test of transformed values. 
Meristic counts were made with a 
dissecting microscope and included 
anal-fin soft rays and lateral-line 
scale counts. For anal-fin soft-ray counts, the last 
branched soft ray was counted as one ray (McEachran 
and Fechhelm, 2005). Four commonly used diagnostics 
were evaluated for identification to species. The ratios 
of pectoral-fin length to pelvic-fin length (Chao, 2002) 
and also the ratio of anal-fin base-length to eye-diam- 
eter (DeVries and Chittenden, 1982) were calculated, 
and differences between species were assessed with a 
pooled t-test on untransformed data. Anal-fin soft rays 
(Ginsburg, 1929) and lateral-line scales (Hoese and 
Moore, 1998) were counted, and differences between 
these meristics were assessed with a chi-square test 
of homogeneity. Following morphological analyses, 
dorsal-fin soft tissue was excised from each specimen 
and placed in 70% denatured ethanol. 
Total genomic DNA was extracted from each fin-clip 
with a Puregene® miniprep kit (Gentra Systems, Min- 
neapolis, MN) according to the manufacturer’s instruc- 
tions. The mtDNA methods used here were similar to 
those of Cordes et al. (2001) and Cordes and Graves 
(2003). A portion of the 12S/16S ribosomal gene lo- 
cus of the mtDNA was amplified by polymerase chain 
reaction (PCR) with the primers 12SAL and 16SAH 
(Cordes et al., 2001). Amplification products were run 
through a 2% agarose gel next to a size standard span- 
ning the ranges of 100-1500 base pairs (bp) to verify 
expected fragment length. Each amplicon was then di- 
gested with the restriction enzyme Rsal (New England 
Biolabs, Inc., Ipswich, MA) according to the standard 
protocol of the supplier, and restriction fragments were 
separated on a 2% agarose gel at 100 volts for 1 hour. 
A size standard was loaded onto each gel in order to 
approximate the size RFLPs. Gels were stained with 
ethidium bromide and RFLP bands were made visible 
(i.e., fluoresced) under an ultraviolet lamp. 
Before initiation of this study, few microsatellite 
markers had been effectively characterized in the lit- 
erature for sand or silver seatrout. However, a number 
of markers had been identified in a genomic library 
from the closely related sciaenid red drum ( Sciaetiops 
ocellatus) (Turner et ah, 1998; Saillant et ah, 2004), 
and some of these markers have subsequently been 
used successfully in members of the genus Cynoscion 
(Gold et al., 2003; Ward et ah, 2007). Here, sixteen 
previously described microsatellite markers were cho- 
sen for examination: SOC12, SOC50, SOC77, SOC85, 
SOC125, and SOC243 (Turner et ah, 1998), CNE133 
(Gold et al., 2003), SOC410, SOC412, SOC415, SOC416, 
SOC419 , SOC423, SOC424, SOC428, and SOC432 (Sail- 
lant et al., 2004). Eight individuals of each species were 
genotyped with the suggested primers in each case. 
Each reverse oligonucleotide was previously labeled 
with a WellRED dye (Proligo USA LLC, Boulder, CO), 
and amplified products were produced at each locus 
by means of PCR. Products from individual reactions 
were diluted 1:20 with water and separated with a 
Beckman-Coulter CEQ IM 8000 automated capillary sys- 
tem (Beckman Coulter, Inc., Fullerton, CA), according 
