Shulzitski et al.: Population connectivity among Dry Tortugas, Florida, and Caribbean populations of Lutjanus analis 
505 
Table 2 
Genetic diversity at eight microsatellite loci in five sampled populations of mutton snapper (Lutjanus analis). n = number of geno- 
typed individuals; A = number of alleles; A = allelic richness; a = number of private alleles; H E = expected heterozygosity; H 0 = 
observed heterozygosity; test for Hardy-Weinberg equilibrium: * P<0.05, ** P<0.01, *** P<0.001. 
Locus 
Population 
La25 
La39 
Lal8a 
La27a 
La34a La45a La49a 
LaC-16 
Overall 
Belize (BZ) 
n = 49 
A 
21 
11 
31 
25 
14 
13 
14 
10 
139 
A 
18.9 
9.6 
27.0 
22.3 
13.4 
11.5 
13.0 
9.4 
125.1 
A 
0 
0 
3 
2 
0 
0 
2 
0 
7 
H e 
0.943 
0.806 
0.968 
0.945 
0.910 
0.856 
0.911 
0.864 
Ho 
0.939 
0.750 
0.957 
0.872* 
0.875 
0.898 
0.896 
0.896* 
Honduras (HN) 
n = 53 
A 
27 
13 
28 
26 
17 
14 
12 
10 
A 
21.5 
11.3 
24.5 
22.3 
15.6 
12.3 
11.5 
9.5 
128.5 
a 
5 
1 
0 
2 
0 
0 
0 
0 
8 
H e 
0.942 
0.813 
0.962 
0.939 
0.926 
0.887 
0.905 
0.847 
Ho 
0.868 
0.585*** 
0.962 
0.825 
0.942 
0.865 
0.827 
0.830 
Puerto Rico (PR) 
n = 46 
A 
25 
12 
27 
22 
16 
12 
15 
10 
A 
21.0 
10.6 
23.7 
21.0 
14.9 
11.2 
13.9 
9.3 
a 
3 
0 
0 
1 
0 
1 
2 
0 
7 
H e 
0.934 
0.771 
0.956 
0.937 
0.911 
0.873 
0.908 
0.846 
Ho 
0.957 
0.622** 
0.957 
0.906 
0.976 
0.884 
0.762* 
0.841 
Dry Tortugas (DT) 
n = 39 
A 
24 
11 
31 
19 
16 
13 
12 
9 
135 
A 
21.8 
10.3 
28.1 
19.0 
15.4 
12.0 
11.3 
8.7 
126.6 
a 
0 
1 
0 
1 
0 
0 
0 
0 
2 
H e 
0.942 
0.794 
0.964 
0.920 
0.915 
0.876 
0.894 
0.859 
H 0 
0.895 
0.500*** 
0.947 
0.862 
0.921 
0.821 
0.789 
0.846 
Jupiter (JP) 
n = 55 
A 
25 
11 
32 
29 
16 
12 
14 
10 
149 
A 
21.2 
9.7 
26.8 
23.9 
14.8 
11.4 
12.8 
9.3 
129.9 
a 
2 
0 
2 
2 
0 
0 
1 
0 
7 
H e 
0.949 
0.802 
0.966 
0.945 
0.917 
0.891 
0.916 
0.858 
H 0 
0.964 
0.537*** 
0.982 
0.841 
0.836 
0.873 
0.891 
0.759 
Only 46 individuals (18.8%) were correctly assigned 
to their source populations with the frequency-based 
assignment test (Table 3). Individuals were assigned 
to the population in which their multilocus genotypes 
exhibited the highest likelihood of occurrence. However, 
for most individuals, likelihoods of occurrence were 
similar across populations and success for assignments 
to origin was similar across populations. Thus, we were 
unsuccessful at detecting the origin of an individual 
based on its multilocus genotype and population allele 
frequencies — not surprising given the similarity in al- 
lele frequencies across population samples. 
With the Bayesian methods applied in the program 
STRUCTURE we identified a single genetic cluster with 
a posterior probability of 1.0. Posterior probabilities 
were essentially zero for the presence of two, three, 
four, or five clusters. In the absence of prior informa- 
tion regarding the geographic origin of individuals, we 
did not detect any genetic structure corresponding to 
geographic location. 
Results from the Geneland analysis broadly agreed 
with those from STRUCTURE, with subtle differences. 
Ten independent runs incorporating no uncertainty, or 
with uncertainty of 120 km 2 in geographic position of 
