18 
Fishery Bulletin 107(1) 
there was the likelihood that duplicate fragments were 
present in each haplotype at the 190 band. Additionally, 
multiple uncharacterized bands appeared that were <100 
18 -■ 
16 - 
14 - 
12 - 
10 - 
8 - 
6 - 
4 - 
2- 
0 - 
r r i i i r i i i 
0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 
30 
25 
20 
15 
B 
Anal-fin length/eye diameter 
n 
u 
1 
• 
r-7j 
1 1 [vCiPiA'j 
rm 
0.9 
o C 
1 1.1 1.2 
Pectoral-/pelvic-fin ratio 
1.3 
60 
50 
40 
30- 
20 - 
10 
0 
9 10 11 
Anal-fin rays 
12 
57 58 59 60 61 
Lateral-line scales 
62 63 
Figure 3 
Histograms representing the distribution of meristic and 
morphometric characteristics of silver seatrout (Cynoscion 
nothus) (white bars) and sand seatrout (C. arenarius) (black 
bars) collected offshore from Galveston Bay, TX, in July 2007. 
The four main diagnostics examined included (A) anal-fin/ 
eye- diameter ratio, (B) pectoral-fin/pelvic-fin ratio, (C) total 
anal-fin rays, and (D) total lateral-line scales. The abscissa 
represents each data class (ratio [A and B] or count [C and D] 
data), whereas the ordinate represents the total membership 
of each species sample in that class. 
bp in length. These uncharacterized bands likely account 
for the remainder of digested DNA (-320 bp expected in 
C. arenarius, and -420 bp expected in C. nothus ) that 
was not accounted for by primary bands, although 
this assumption was not explicitly tested. Correct 
assignment of RFLP haplotypes to each species 
was confirmed by comparison with anal-fin ray 
counts and anal-fin to eye-diameter ratios. 
Microsatellite DNA 
The nine microsatellites used for species compari- 
sons had a range of three to 51 alleles overall. One 
locus ( SOC415 ) had a dramatic allele range differ- 
ence between the species (Table 1), which resulted 
in almost complete disjunction between allele dis- 
tributions, possibly the result of a large insertion 
in sand seatrout, or deletion in silver seatrout. A 
deletion is most likely the reason because the allele 
range in sand seatrout is similar to that of the 
species in which the markers were initially cloned, 
S. ocellatus. Otherwise, the detected allele ranges 
of the remaining eight loci were similar between 
Cynoscion species and overlapped the range of S. 
ocellatus. 
The average allele diversity was 16.44 al- 
leles per locus in sand seatrout and 9.89 alleles 
per locus in silver seatrout, and diversity was 
(qualitatively) higher in sand seatrout at eight of 
nine loci (Table 3). Similarly, gene diversity was 
higher in sand seatrout at seven of nine loci and 
ranged from 0.11 to 0.97, compared to a range of 
0.02 to 0.90 in silver seatrout. There was no in- 
dication of genotypic disequilibrium at any locus 
in any population because all P-values fell at or 
above the adjusted critical level of a - 0.0014. 
It should be noted, however, that a single com- 
bination of loci ( SOC243 and SOC415 ) did show 
evidence of significant linkage before adjustment 
for multiple tests (P~0.0014). The inbreeding co- 
efficient (F is ) was not significantly different from 
0 at any locus in either population. As a result, 
all nine loci were included in downstream as- 
signment assays. 
Significant genetic divergence between the spe- 
cies was found at each of the nine loci, resulting 
in a range for 6 of 0.025-0.464, with a mean value 
of 0=0.117 (Table 3). Six loci accounted for ap- 
proximately 86% of the discriminatory power of 
the panel of microsatellites in Whichloci analyses 
(Table 4). Of these, only the four highest ranked 
loci were needed to correctly assign individuals 
to populations under an assignment stringency 
LOD-score of two. All six were needed when the 
assignment stringency was increased to a LOD- 
score of three. Unsuccessful assignment was due 
to stringency restrictions rather than to actual 
mis-assignment of individuals. In every case, all 
individuals were assigned correctly at a relaxed 
stringency. 
