Content uploaded by Carmen A Villegas-Sánchez
Author content
All content in this area was uploaded by Carmen A Villegas-Sánchez on Dec 06, 2014
Content may be subject to copyright.
REPORT
Small-scale genetic connectivity of bicolor damselfish
(Stegastes partitus) recruits in Mexican Caribbean reefs
C. A. Villegas-Sa
´nchez •R. Rivera-Madrid •
J. E. Arias-Gonza
´lez
Received: 27 June 2009 / Accepted: 22 May 2010 / Published online: 11 June 2010
ÓSpringer-Verlag 2010
Abstract The analysis of genetic similarities among
marine populations is a key method for use in connectivity
studies intended to provide information for management
strategies. The present study aimed at assessing the con-
nectivity levels of subpopulations of bicolor damselfish
(Stegastes partitus) recruits at a small scale (*200 km)
among four reefs in the Mexican Caribbean. Samples were
collected from 13 sites nested in two Marine Parks
(Cozumel and Xcalak), one Biosphere Reserve (Chin-
chorro Bank) and one unprotected area (Mahahual). A total
of 713 samples were genetically characterized by means of
seven microsatellite DNA markers and were analyzed on a
hierarchical basis. A strong genetic structure was detected
among sites with a weak but significant genetic structure
among reefs, the combination of which has not been
reported in previous studies. These results appear to be
related to a ‘‘sweepstake-chance effect’’ combined with
oceanographic factors. An isolation by distance test, in
addition to a hierarchical Bayesian method, revealed that
neither distance among sites and reefs nor any of 10
environmental factors tested could be used to explain the
genetic differences observed. The results suggest that
conservation strategies for S. partitus based on local scales
are likely to be effective.
Keywords Caribbean Stegastes partitus Genetic
Connectivity Microsatellites Habitat
Introduction
The term connectivity refers to the amount of biological
material (larvae, recruits, juveniles, or adults) exchanged
between populations of a particular species throughout its
distribution range (Palumbi 2003). This concept has been
proposed as a fundamental approach to understanding reef
degradation and recovery patterns, and hence designing
better management strategies for protected areas (Hughes
and Tanner 2000; Lesser 2004; Jones et al. 2009). Some
studies have shown that systems which are copiously
supplied by external larval sources are less reliant on local
management than systems where local retention prevails
(Roberts 1997; Cowen et al. 2000; Palumbi 2003). As is the
case for most marine species, coral reef fishes have a
pelagic larval phase; biological dispersal usually occurs
during this early stage of life and is regarded as the main
genetic interchange mechanism between distant popula-
tions (Leis and McCormick 2002). Little direct evidence is
available on the mean dispersal distance of reef fishes, with
most existing estimates based on natural and artificial
Communicated by Biology Editor Prof. Philip Munday
Electronic supplementary material The online version of this
article (doi:10.1007/s00338-010-0643-0) contains supplementary
material, which is available to authorized users.
C. A. Villegas-Sa
´nchez J. E. Arias-Gonza
´lez (&)
Laboratorio de Ecologı
´a de Ecosistemas de Arrecifes Coralinos,
Departamento de Recursos del Mar, Centro de Investigacio
´nyde
Estudios Avanzados del I.P.N.-Unidad Me
´rida, Ant. Carr. a
Progreso km 6, A.P. 73, Cordemex, 97310 Me
´rida, Yucata
´n,
Me
´xico
e-mail: earias@mda.cinvestav.mx; jeariasg@mac.com
URL: www.mda.cinvestav.mx
C. A. Villegas-Sa
´nchez
e-mail: cavs005@gmail.com
R. Rivera-Madrid
Unidad de Bioquı
´mica y Biologı
´a Molecular de Plantas, Centro
de Investigacio
´n Cientı
´fica de Yucata
´n, A.C., Calle 43 No. 130,
Col. Chuburna
´de Hidalgo, 97200 Me
´rida, Yucata
´n, Me
´xico
e-mail: renata@cicy.mx
123
Coral Reefs (2010) 29:1023–1033
DOI 10.1007/s00338-010-0643-0
marks presented on, or marked onto, the fishes0otoliths (ear
bones; Jones et al. 1999,2005; Swearer et al. 1999;
Almany et al. 2007). Direct studies on larvae are not easy
to accomplish due to the logistical difficulties related to the
extremely high fecundity and mortality rates at early life
stages (Thorrold et al. 2002). For this reason, dispersal
levels have more often been evaluated by means of alter-
native indirect methods, such as those based on oceano-
graphic circulation and larval behavior (Roberts 1997;
Cowen et al. 2000,2006; Paris and Cowen 2004), in
addition to molecular genetic techniques, which have been
widely used (Shulman and Bermingham 1995; Planes et al.
1998; Fauvelot and Planes 2002; Taylor and Hellberg
2003; Purcell et al. 2006; Haney et al. 2007). Relatively
new technologies, such as the application of genetic par-
entage analysis (Jones et al. 2005; Planes et al. 2009), have
also been used to evaluate connectivity; however, this
technique is limited to species where a large proportion of
the adult population can be captured (Jones et al. 2009).
Even though genetic evaluations contribute to the under-
standing of the dispersal levels of marine organisms, their
results must be considered carefully, since both temporal
and spatial variations have an influence on this type of
analysis (Hedgecock et al. 2007).
Genetic differences among reef fish populations can
occur at both large and small spatial scales. Shulman
(1998) reported that in ten studies published prior to 1998,
significant spatial differentiation was detected in five of
fourteen species (36%) at large scales ([1,000 km), seven
of twenty-six species (27%) at medium scales (200–
1,000 km), and six of ten species (60%) in at least one of
the local populations at small scale (\200 km). Since 1998,
a number of other studies have reported significant genetic
differences in reef fish populations at small spatial scales
(Planes et al. 1998; Fauvelot and Planes 2002; Taylor and
Hellberg 2003; Purcell et al. 2006; Hepburn et al. 2009).
Similar results have also been observed in other reef
organisms (Ruiz-Za
´rate and Arias-Gonza
´lez 2004), with
some studies reporting significant genetic variance at small
scales (\1 km; Sale et al. 1994; Pandolfi 2002). Although
genetic differentiation is thought to mostly occur at large
spatial scales in marine systems, biological dispersal may
be limited by habitat, local ocean conditions, and specific
biological characteristics; consequently, partially isolated
populations may occur quite commonly (Palumbi 1994).
Understanding and identifying the processes that lead to
genetic structure is crucial for biodiversity conservation
and management (Kittlein and Gaggiotti 2008). A recently
developed method based on Bayesian statistics that com-
bines genetic and nongenetic data allows potential inter-
actions between environmental factors and genetic
variation to be tested (Foll and Gaggiotti 2006). In pelagic
fishes, several studies have related genetic structure to
environmental factors recognized as restrictive to the flow
of organisms among subpopulations (Bekkevold et al.
2005; Jørgensen et al. 2005; Gaggiotti et al. 2009). In the
case of certain reef fishes, although habitat variation has
been identified as relevant for the survivorship of juveniles
(Beukers and Jones 1997; Nemeth 1998), and hence for the
effective dispersal of organisms, the effects of this envi-
ronmental factor on their genetic structure have not pre-
viously been explored.
In the present study, we estimate genetic connectivity of
the bicolor damselfish (Stegastes partitus) among four
Mexican Caribbean reefs. Previous studies of genetic
structure among S. partitus populations differ in their
conclusions. Lacson et al. (1989) found significant differ-
ences between sampled sites in the Florida Keys (USA) at a
small scale, as did Hepburn et al. (2009), who studied reefs
at small and medium scales. In contrast, Purcell et al.
(2009) only found significant genetic differences (based on
a stepping stone dispersal model) at a medium scale
(500 km), despite sampling individuals at small, medium,
and large scales in the Caribbean Basin. On the other hand,
some studies have reported genetic homogeneity among
populations of S. partitus at both small and medium scales
(Lacson and Morizot 1991; Lacson 1992; Ospina-Guerrero
et al. 2008).
Given the variability in the results of previous studies,
the present study was based on a hierarchical analysis with
13 sites nested in four reefs, in order to perform a robust
evaluation of the genetic connectivity of S. partitus sub-
populations at small spatial scales. We focused on recruits
because the genetic population structure at this stage differs
significantly from those of juveniles and adults, and it is
closer to the genetic structure of larvae, the key stage at
which dispersal processes take place (Jones and Barber
2005). Seven polymorphic microsatellite DNA markers
were analyzed, using an average of 55 individuals per site
to assess genetic structure. In addition, we collected data on
habitat structure at each site to test whether genetic vari-
ation was associated with environmental variation. We
hypothesized that small-scale genetic structure in S. par-
titus recruits would be correlated with variation in benthic
habitat composition.
Materials and methods
Study species and sites
The bicolor damselfish tends to be found closely associated
with the substrate of shallow coral reefs or in isolated patch
reefs in deeper waters (Cervigo
´n1993). Females lay suc-
cessive clutches of eggs in a nest kept by males (Robertson
et al. 1988). The larval pelagic stage of S. partitus lasts
1024 Coral Reefs (2010) 29:1023–1033
123
36.5 days on average, and both spawning and settlement
follow unimodal lunar cycles; although these events occur
throughout the year, there are two peaks, one in April and
another in November (Robertson et al. 1988). Larvae are
approximately 13 mm long when they settle to reef habi-
tats. For sampling, recruits were considered to be individ-
uals with a total length of less than 3 cm, following
Almada-Villela et al. (2003).
Samples of S. partitus recruits were collected from four
coral reefs, two of which are fringing reefs: Mahahual
(MA) and Xcalak (XC), while Chinchorro Bank (CH) and
Cozumel (CO) are oceanic reefs (Fig. 1). All the reefs are
part of the northern section of the Mesoamerican Barrier
Reef System (Almada-Villela et al. 2003). Two of these
reefs belong to Marine Parks (Cozumel and Xcalak), one
belongs to a Biosphere Reserve (Chinchorro Bank) and the
other (Mahahual) is an unprotected area. Chinchorro Bank
is located 27 km east of the southeastern coast of the
Yucatan Peninsula (Jorda
´n and Martin 1987), while
Cozumel National Park’s reefs are on the southwestern
coast of the island with the same name, which is located
109 km north of Chinchorro Bank and 18 km east of the
coast (Cetina et al. 2006). The two fringing reefs are
located along the southeastern coast of the Yucatan Pen-
insula, the southernmost region of the Mexican Caribbean.
The unprotected area (Mahahual reef) was included in this
study because of the notably high levels of tourism and
fishing exploitation it has faced in the last decade, in
addition to being reported as a reef with high diversity and
richness values (Arias-Gonza
´lez et al. 2008; Rodrı
´guez-
Zaragoza and Arias-Gonza
´lez 2008).
Sampling protocol
A hierarchical spatial sampling design was used with three
or four sites per reef in order to compare genetic structure
of S. partitus among sites and among reefs. Stegastes
partitus recruits were collected from northern (n), central
(c), and southern (s) sites of each reef; except in the case of
Chinchorro Bank, where samples were collected from n, s,
eastern (e) and western (w) sites (Fig. 1). A total of 713
individuals were obtained between June and September
2006 (Cozumel’s sites were sampled in September, Chin-
chorro’s in June/August, Mahahual’s in August, and
Xcalak’s in July/August). Between 50 and 64 samples were
analyzed for each site, except for the north and south of
Xcalak for which only 24 and 37 samples were analyzed,
respectively. Recruits were captured using clove oil solu-
tion (10% clove oil, 90% ethanol; Frisch et al. 2007) and
hand nets. Fish were preserved in 80% ethanol and trans-
ported to the laboratory where the ethanol was renewed
before storage at -20°C for posterior analysis.
Habitat surveys were conducted to examine the rela-
tionship between benthic habitat structure and any
observed genetic structure at the site level. Benthic habitat
composition was estimated from 10 video transects, 50 m
long by 0.4 m wide, conducted at each site. The percentage
composition of 10 habitat variables (seaweed, sand, scle-
ractinian corals, dead coral, sponges, hydrocorals, octoco-
rals, rocks, calcareous rubble, and reef matrix) was
estimated for each site using 40 still images extracted from
each video transect. Habitat type was recorded at 13 points
per image viewed on a high-resolution computer monitor
(method modified from Aronson et al. 1994). The point
count data (520 points per transect) were used to calculate
the percentage cover for each of the 10 habitat variables at
each site.
Molecular analyses
Genomic DNA was extracted from the caudal fin tissue of
all samples using the DNeasy Blood and Tissue Kit (Qia-
gen-69506), following the protocol for DNA purification
from animal tissues designed by the company. Nine spe-
cies-specific microsatellite loci were amplified (Table 1;
Williams et al. 2003; Thiessen and Heath 2007). The PCRs
were performed with 0.3–0.5 units of Amplitaq Gold
(Applied Biosystems), 200 lM of each dNTP, 2.0 mM of
MgCl
2
, 0.5 lM of each primer, approximately 100 ng of
DNA, 1.25 lL of PCR Gold Buffer 109without MgCl
2
Fig. 1 Geographic positions of coral reefs surveyed: Cozumel,
Chinchorro Bank, Mahahual, and Xcalak. Triangles indicate sampling
sites. Marine protected areas are delimited by solid lines, while the
unprotected area is delimited by a dashed line
Coral Reefs (2010) 29:1023–1033 1025
123
(Roche), and the ddH
2
O necessary for a final volume of
12.5 lL. The thermocycling program for the amplification
of the loci SpGGA7, SpTG16, SpTG10, SpTG53, and
SpTG8 (Thiessen and Heath 2007) consisted of an initial
Taq activation step at 95°C for 10 min, and a denaturation
step at 94°C for 2 min, followed by 29–30 cycles of
denaturation at 94°C for 15 s, annealing at specific locus
temperature (Table 1) for 15 s, and extension at 72°C for
30 s, ending with a final extension step at 72°C for 90 s
(except for the locus SpTG8 which had a final extension
step of 60 min). In the case of the loci, SpTG13, SpA-
AC41, SpGATA40, and SpAAT40 (Williams et al. 2003;
Thiessen and Heath 2007), the amplification program
consisted of an initial Taq activation step at 95°C for
10 min, and a denaturation step at 94°C for 1 min, fol-
lowed by 29 cycles of denaturation at 94°C for 30 s,
annealing at the specific locus temperature (Table 1) for
30 s, and extension at 72°C for 45 s, ending with a final
extension step at 72°C for 2 min (except for the loci
SpGATA40 and SpAAT40 which had a final extension step
of 60 min). The fragments obtained were analyzed on an
ABI Prism 377 DNA sequencer (Applied Biosystems), and
microsatellite allele sizes were determined with Genscan
3.1.2 analysis software (Applied Biosystems).
Statistical analyses
The loci were selected for the analyses according to their fit
to the Hardy–Weinberg equilibrium (HWE), which was
tested for each locus and subpopulation by means of
Wright’s fixation index (Fis) as described by Weir and
Cockerham (1984), using the software Genetix 4.05
(Belkhir et al. 1996–2004). The presence of null alleles,
mis-scoring due to stuttering, and allelic dropout due to
short allele dominance were detected using the program
Microchecker 2.2.3 (Van Oosterhout et al. 2004). Since the
loci SpTG8 and SpAAC41 exhibited significant deviations
from HWE, they were eliminated from the remaining
analyses. For each site and reef, allelic frequencies, number
of alleles per locus, observed heterozygosity, and gene
diversity (heterozygosity expected from Hardy–Weinberg
equilibrium assumptions) were calculated across different
loci using Genetix 4.05.
In order to eliminate the possibility of temporal varia-
tions influencing the spatial structure of the sites sampled
in two different months (CHn, CHe, CHw CHs, XCc and
XCs), genetic differences between paired months were
estimated by means of Wright’s index (Fst) as described by
Weir and Cockerham (1984), using the program Genetix
4.05. Fis and Fst indices were tested for significant dif-
ferences from zero by performing bootstrapping (10,000
permutations) over loci using Genetix 4.05; a modified
false discovery rate procedure (referred to as the B–Y
method; Benjamini and Yekutieli 2001) was used as a
correction for multiple comparisons.
As the presence of null alleles led to an overestimation
of Fst in cases of significant population differentiation,
when the presence of null alleles was inferred, null allele
frequencies for each locus and subpopulation were esti-
mated following the expectation maximization (EM)
Table 1 Characteristics of the
nine primers originally
proposed for the molecular
analysis of Stegastes partitus
recruits
T
a
is the annealing temperature
used in degrees Celsius
Locus name Primer sequences 50–30Repeat
motif
Size range
(bp)
No. of
alleles
T
a
(°C)
SpGGA7 CGATATGTTTAATGTGAGGAACG GGA
7
243–255 5 48
TTTCAGGAGGTAATAGTCCACCA
SpTG16 GTGAGACAGTGGGTCACCTG TG
16
149–205 23 52
GTTTTCCCCCTCCTCACACT
SpGT10 GGCCTGTTTAAAGGTCACCA GT
10
235–289 12 55
CACCAACGAGCTACGGTGTA
SpTG53 GATGGCCTCTGGTGTAATGC TG
53
150–254 29 48
CAACTGGGAAGGAGGTCAAG
SpTG8 ACGCCGAAATGCATCTTAAT TG
8
169–211 19 48
ACACATATTCCCTCGGTTTTT
SpTG13 CTTGTTCCTTGGCTTCTTGG TG
13
228–236 5 48
TGATAGTGGCAAGCAATGGA
SpAAC41 GCCCGTCACTGACAGTCTGTG AAC
13
171–259 28 56
CGAAGGCTGTGTTCAGTTATACA
SpGATA40 TTGCCTGCTGATAATTAACG GATA
6?29
116–295 39 48
ATGCCTCACAATGATGTATATTT
SpAAT40 CTGGGCTTCTCTTTTTTGTT AAT
22
164–228 24 48
AGTGGTCCCAGAGTTTTCTC
1026 Coral Reefs (2010) 29:1023–1033
123
algorithm of Dempster et al. (1977). With the allelic fre-
quencies corrected, pairwise Fst tests (following Weir
1996) were calculated using the excluding null alleles
(ENA) method described by Chapuis and Estoup (2007).
This correction method has been reported as being suitable
for the positive bias induced on Fst values by the presence
of null alleles in several studies on different organisms
(Bryja et al. 2007;The
´riault et al. 2007; Jordan and Snell
2008; Purcell et al. 2009). FreeNA software was used to
perform both the EM algorithm and the ENA method
(Chapuis and Estoup 2007). Genetic differences were also
evaluated with the unbiased genetic distance (Ds; Nei
1978) using the original allelic frequencies. The Ds index
was used because of its relatively low sensitivity to the
presence of null alleles, even with high frequencies (Cha-
puis and Estoup 2007). Both Ds values and significance
level of each test (with bootstrapping and 10,000 permu-
tations) were evaluated using the program Genetix 4.05,
while the B–Y method was used as a correction for mul-
tiple comparisons.
The relationship between genetic structure and habitat
characteristics estimated at each site was analyzed using
the hierarchical Bayesian method of Foll and Gaggiotti
(2006), which is implemented in the program Geste 2.0
(Foll and Gaggiotti 2006). This method estimates Fst val-
ues and relates them to nenvironmental factors by means
of 2
n
generalized linear models; the results provide pos-
terior probabilities for each model, assigning the highest
one to the model that best explains the data (Gaggiotti et al.
2009). Due to the relatively high number of factors with
respect to 13 subpopulations, the set of 10 factors was split
into two groups of three and one group of four factors; each
group was analyzed separately. The analyses using Geste
2.0 were based on a sample size of 30,000 and an addi-
tional burn in of 100,000 iterations with 10 pilot executions
of 5,000 iterations and a thinning interval of 50.
To assess the relative partitioning of genetic variation
within and between reefs and sites, an analysis of molec-
ular variance (AMOVA) was applied using the software
Arlequin 3.11 (Excoffier et al. 2005). For the hierarchical
AMOVA, samples were clustered in four levels: (1) among
reefs, (2) among sites within reefs, (3) among individuals
within sites, and (4) within individuals. The genetic
structure of the microsatellite data was also examined using
the factorial correspondence analysis implemented in
Genetix 4.05. The resulting eigenvalues were plotted with
the s.class function using the ade4 library (Thioulouse
et al. 1997) for Rsoftware (Ihaka and Gentleman 1996).
Isolation by distance was assessed by means of the
Mantel procedure with 1,000 permutations in the software
IBD (Bohonak 2002), using genetic distances obtained with
the method described above (Ds) and geographical dis-
tances. Both allelic richness and private allelic richness
were calculated by means of the rarefaction method pro-
posed by Kalinowski (2004), which is used to reduce the
effect of sample size on these indices. Allelic richness was
used as a simple measure of genetic diversity, which is
related to the response potential of populations facing
selection pressure, while the private allelic richness was
used as a measure of genetic distinctiveness (Kalinowski
2004). In the present study, the samples were evaluated
with a hierarchical sampling design (selecting three sites
per reef and a minimum sample size of 36 genes from each
site) using the program HP-Rare 1.0 (Kalinowski 2005).
Paired tests based on sign analysis were implemented in
order to detect significant differences among multilocus
richness values at site and reef levels, as suggested by
Kalinowsky (2005).
Results
Genetic variation and Hardy–Weinberg equilibrium
(HWE)
Significant deviations from HWE, after correction with the
B–Y method, were observed in 62 of the 91 tests in the
seven loci, with an average Fis of 0.121 (Range =-0.075
to 0.397) at the site level (electronic supplementary mate-
rial; ESM). All loci showed significant deficits of hetero-
zygotes in at least 6 of the 13 sites (P\0.016; ESM). The
Fis index values at the reef level are not shown, since
combining samples divergent from HWE is likely to pro-
duce a Wahlund effect. The technical cause for the het-
erozygote deficits was most likely the presence of null
alleles, as indicated by the software Microchecker 2.2.3.
A total of 422 alleles were detected, with an average of
60 alleles per locus, a minimum value of 22 alleles (cor-
responding to locus SpGGA7), and a maximum of 127
alleles (corresponding to locus SpGATA40). The average
expected heterozygosity at site level was 0.874 with a
range of 0.599 to 0.970 (ESM), while at the reef level
average expected heterozygosity was 0.889 with a range of
0.655 to 0.974.
Genetic structure
The values of the Fst index used for testing temporal
variation were not significant (ranging from -0.002 to
0.012); therefore, samples from the same site were con-
sidered to belong to the same pool, regardless of the
sampling date. The analysis performed by the program
FreeNA showed that the null allele frequencies calculated
per locus across all subpopulations ranged from 0.0 to
0.347 at site level (ESM), while at reef level they ranged
from 0.027 to 0.220. No elevated null allele frequencies
Coral Reefs (2010) 29:1023–1033 1027
123
were reported for a specific locus or subpopulation. Global
Fst was 0.015 at site level (95% CI 0.003 to 0.034), and
pairwise Fst values ranged from 0.0 to 0.046 (Table 2).
After using the B–Y method to correct the Ds results, they
showed that significant genetic distances existed between
the 13 sites, with variations from 0.0 to 0.466 (Table 2). At
reef level, the global Fst was 0.006 (95% CI 0.001 to
0.012), and the results of the pairwise Fst index ranged
from 0.001 to 0.008 (Table 3). Ds values showed signifi-
cant differences among the four reefs, ranging from 0.016
to 0.089 (Table 3).
Both south Mahahual (MAs) and south Cozumel (COs)
presented important differences when compared to the
remaining sites, as shown by high pairwise Fst values
(Table 2). However, when compared to each other, a low
pairwise Fst value was obtained. These results were also
supported by the Ds index values, since comparisons with
MAs produced the highest values (ranging from 0.280 to
0.466), followed by those with COs (ranging from 0.121 to
0.281; Table 2).
The results of the Bayesian analysis performed by Geste
2.0 revealed that none of the benthic factors were associ-
ated with the observed genetic structure. In each of the
three analyses based on partial sets of factors, the highest
posterior probabilities, with values of 0.752, 0.804, and
0.812, were assigned to models made up of only a constant.
This means that in each analysis, the model excluding all
variables has at least a 75.2% probability of being the one
that best fits the genetic structure observed.
Significant genetic structure was identified among sites
within reefs, among individuals within sites, and within
individuals, but not among reefs (Table 4). The greatest
variation was detected within individuals (86.6%), fol-
lowed by among individuals within sites (11.8%), among
sites within reefs (1.4%), and among reefs (0.2%; Table 4).
In the factorial correspondence analysis, some sites and
reefs appeared separated (Fig. 2a, b). At site level, a clear
separation of MAs and COs was observed; similarly, the
whole group of Chinchorro Bank was separated from the
remaining sites (Fig. 2a). At reef level, four partially
overlapping groups appeared, with Chinchorro reef rela-
tively more separated from the others (Fig. 2b).
The isolation by distance test revealed a nonsignificant
correlation between Nei’s genetic distances and geographic
distances among sites and reefs (R
2
=-0.078, P=0.584
and R
2
=-0.593, P=0.880, respectively).
No significant differences were detected in private
allelic richness among sites or reefs (P[0.05). At site
level, allelic richness was only significantly higher in the
case of CHw with respect to CHs and XCs (P\0.05),
while at reef level no differences were detected.
Table 2 Multilocus estimates for pairwise Fst (above diagonal) and Ds (below diagonal) at 7 microsatellite loci in the 13 sites sampled
COn COc COs CHn CHe CHw CHs MAn MAc MAs XCn XCc XCs
COn 0.001 0.020 0.009 0.006 0.005 0.006 0.003 0.006 0.034 0.004 0.003 0.000
COc 0.011 0.019 0.011 0.009 0.007 0.014 0.005 0.011 0.031 0.005 0.002 0.005
COs 0.207 0.227 0.027 0.020 0.017 0.028 0.023 0.026 0.004 0.013 0.023 0.030
CHn 0.084 0.112 0.281 0.009 0.005 0.001 0.011 0.021 0.042 0.013 0.015 0.016
CHe 0.064 0.087 0.201 0.085 0.004 0.006 0.003 0.005 0.039 0.002 0.011 0.007
CHw 0.057 0.090 0.207 0.044 0.048 0.004 0.007 0.009 0.034 0.002 0.007 0.008
CHs 0.055 0.120 0.254 0.002 0.054 0.027 0.009 0.015 0.042 0.009 0.013 0.011
MAn 0.033 0.044 0.245 0.136 0.048 0.092 0.113 0.002 0.040 0.004 0.006 0.005
MAc 0.064 0.096 0.261 0.187 0.054 0.087 0.139 0.023 0.046 0.009 0.009 0.007
MAs 0.325 0.332 0.039 0.412 0.402 0.374 0.378 0.411 0.466 0.031 0.036 0.046
XCn 0.032 0.044 0.121 0.098 0.013 0.033 0.067 0.035 0.070 0.280 0.005 0.005
XCc 0.038 0.020 0.238 0.145 0.122 0.089 0.120 0.065 0.094 0.346 0.046 0.004
XCs 0.000 0.020 0.238 0.110 0.054 0.054 0.085 0.031 0.060 0.373 0.028 0.029
In the case of the Ds index, all values are presented in bold as they are significantly greater than zero after correction with the B–Y method.
Abbreviations for sample locations are as follows: north Cozumel, COn; central Cozumel, COc; south Cozumel, COs; north Chinchorro, CHn;
east Chinchorro, CHe; west Chinchorro, CHw; south Chinchorro, CHs; north Mahahual, MAn; central Mahahual, MAc; south Mahahual, MAs;
north Xcalak, XCn; central Xcalak, XCc; south Xcalak, XCs
Table 3 Multilocus estimates for pairwise Fst (above diagonal) and
Ds (below diagonal) at 7 microsatellite loci in the four reefs sampled
Cozumel Chinchorro Mahahual Xcalak
Cozumel 0.007 0.001 0.004
Chinchorro 0.072 0.008 0.007
Mahahual 0.016 0.089 0.005
Xcalak 0.027 0.063 0.047
In the case of the Ds index, all values are presented in bold as they are
significantly greater than zero after correction with the B–Y method
1028 Coral Reefs (2010) 29:1023–1033
123
Discussion
Significant genetic differences were found in subpopula-
tions of S. partitus recruits in the study area, even though
the maximum spatial scale used in the present study
(*200 km) is relatively small. This is the third analysis
that shows genetic structure at a small spatial scale in the
case of this species. Lacson et al. (1989) reported large
values of Nei’s (1972) genetic distances between two sites
on a scale of 0.6 km (D=0.020) in the Florida Keys, and
in a study carried out in Honduras, Belize, and Mexico
reefs, Hepburn et al. (2009) described genetic differences
(maximum Fst =0.004) among sites of the same reef
separated by distances less than 50 km. In a later study,
Lacson and Morizot (1991) revisited their data from 1989,
and like Hepburn et al. (2009), found temporal instability
in their results and consequently considered such differ-
ences to be an effect of stochastic events rather than
deterministic ones. Nevertheless, no other genetic study of
S. partitus has reported strong structure among sites com-
bined with weak but significant structure among reefs.
There are three factors that could help explain the small-
scale genetic variation exhibited in the current study and
the conflicting differences between studies: (1) the absence
of a relationship between genetic and geographic distances,
which may indicate that the observed genetic structure is
temporally unstable (e.g., ‘‘sweepstake-chance effect’’); (2)
the unusual isolation of MAs and COs from the rest of the
sites, together with the apparent connectivity between
them, even when they constitute one of the more separated
pairs of sites; and (3) the unlikely influence of null alleles
on the genetic structure.
The ‘‘sweepstake-chance effect’’ (Hedgecock 1994)is
suggested as the primary explanation for the pattern of
genetic structure we observed. This effect has previously
been proposed as an explanation for genetic differences in
some fish species and invertebrates (Selkoe et al. 2006;
Arnaud-Haond et al. 2008; Hepburn et al. 2009). Hedge-
cock (1994) suggested that in some organisms, such as
fishes, with very high fecundity and mortality at larval
stage for a given year, most of the young recruits may
come from very few parents due to ‘‘a sweepstake-chance
matching of reproductive activity with oceanographic
conditions conducive to spawning, fertilization, larvae
development, and recruitment’’. This would result both in a
high variance in the progeny number and in a very small
effective population size. We propose that larvae from few
Table 4 Analysis of molecular variance (AMOVA) among the four
groups of Stegastes partitus from the Mexican Caribbean
Source of variation Variance
components
Percentage
variation
Pvalues
Among reefs 0.007 0.224 0.182
Among sites within reefs 0.043 1.392 0.000
Among individuals within sites 0.370 11.765 0.000
Within individuals 2.726 86.617 0.000
Pvalues in bold mark significant differences
Axis1
Axis2
Axis1
Axis2
(b)
(a)
Fig. 2 Scatter plots of factorial correspondence analysis with seven
microsatellite loci; each line represents the distance of one individual
to the gravity center of the population (site or reef), and ellipses
around gravity centers are confidence intervals of 95%. The labels in
the margins of the graphs represent the position of each gravity center
on axis one (horizontal) and axis two (vertical). aSite classes with
18.68% of variation on axis 1 and 15.29% on axis 2. bReef classes
with 50.13% of variation on axis 1 and 26.93% on axis 2.
Abbreviations are as described in Table 2
Coral Reefs (2010) 29:1023–1033 1029
123
parents (different in each generation and not always from
the same site) may have been settling at a particular local
site due to random oceanographic factors and larval
behavior. If the sampling area is small enough, the recruits
might appear inbred and isolated; however, sampling in
larger areas could make this pattern less evident. As a
result, different patterns appear at site and at reef scales.
The differences found with respect to Hepburn et al. (2009;
subtle small-scale genetic structure) and Purcell et al.
(2009; subtle genetic structure over 100’s of km, but no
pattern at a small-scale) would depend on the precise nat-
ure of the ‘‘small’’ and ‘‘large’’ scales sampled.
Currents along the Mexican Caribbean coast have a
predominant northeastward direction; however, important
weeklong reversal periods (southward flows) have been
observed around Chinchorro Bank. In addition, some
models suggest that the dynamics of the entire area might
be strongly influenced by the occurrence of eddies
throughout the region (Cetina et al. 2006). This complex
current system, probably in combination with several other
variables that affect the dispersal of larvae, such as active
behavior (Cowen and Castro 1994; Paris and Cowen 2004),
variation in mortality rates (Cowen et al. 2000), and other
local processes (Shulman and Bermingham 1995), may be
contributing to the genetic structure observed.
The resulting Fst values shown here present substan-
tially higher levels of genetic structure (global Fst 0.015)
than all previous studies performed with the same species
and microsatellites (Ospina-Guerrero et al. 2008; Hepburn
et al. 2009; Purcell et al. 2009). In this sense, Cowen et al.
(2006) showed recruitment in the Mexican Caribbean to be
strongly limited; they related their results to the life his-
tories and larval capabilities of the fishes, as well as to the
oceanographic features of the region. Furthermore, in a
study developed with satellite ocean color imagery, Soto
et al. (2009) showed Chinchorro and Cozumel reefs to be
among the most isolated from land-based water sources as
well as from plumes flowing through other reefs in the
Mesoamerican Barrier Reef System. Consequently, the
high Fst values reported in this study are believed to be
attributed, in part, to the oceanographic regime of the
Mexican Caribbean.
The Fis values showed a generalized heterozygote
deficiency in the majority of the loci and, although these
values were higher than those presented in previous studies
reporting genetic differences among reef fish populations
(Bell et al. 1982; Lacson et al. 1989; Planes 1993; Planes
et al. 1993,1998; Fauvelot and Planes 2002; Purcell et al.
2006), a recent study with the same species reported sim-
ilar HW deviations (Purcell et al. 2009) and proposed null
alleles to be the main cause of them. Taking into account
the Microchecker 2.2.3 results (which indicated null alleles
to be the principal cause of HW deviations) together with
the less probable factor of inbreeding (bearing in mind the
enormous population sizes of this species) and the con-
siderably high rate of changes in the flanking regions of the
microsatellites (Grimaldi and Crouau-Roy 1997), it is
proposed that null alleles are the principal cause of the
observed heterozygote deficiencies in the present study.
The influence of null alleles on the genetic structure was
analyzed and found unlikely to have changed the findings
significantly. The strength of the Fst and Ds indices, in
addition to the results of other statistical approaches, also
suggests something other than null alleles to be the cause of
the genetic differences observed. Furthermore, Dakin and
Avise (2004) reported that bias introduced by null alleles is
negligible when their frequencies are lower than 0.2.
Consequently, the HWE deviations reported here do not
seem to affect the Fst index values, since 108 of 117 null
allele frequencies were below 0.2, while five of the
remaining nine corresponded to the discarded loci (SpTG8
and SpAAC41).
We explored the possible relationship between genetic
differences and benthic habitat composition given that (1)
no significant relationship was found between geographic
and genetic distances (which could otherwise have
explained the genetic structure observed) and (2) the results
of Nemeth (1998) showed the substratum architecture to
have a significant influence on the survivorship of S. par-
titus juveniles. Considering the latter point, it is possible
that if the survivorship of recruits in inadequate habitats is
low, it can occur to such a degree that a random loss of
genetic variability is generated, producing genetic varia-
tion. This variation can be detected using highly poly-
morphic neutral markers such as microsatellites (Turner
et al. 2002; Rousset 2004). The results presented here
showed no clear relationship between benthic factors and
genetic differences; hence the effects of habitat composi-
tion on the survivorship of recruits were not significant at
genetic level. This can be explained by taking into account
the suggestion by Cowen and Sponaugle (2009) that, in
terms of population connectivity, in order to maintain
genetic homogeneity among local populations, a low
exchange of organisms is required, which may be several
orders of magnitude lower than the required to alter
demographic rates.
As no significant differences were detected among val-
ues of private allelic richness at site or reef level, this index
provides no proof of genetic isolation in the study area.
Furthermore, although the allelic richness showed two
significant differences in the case of sites (CHw vs CHs
and XCs), they were not consistent throughout the analy-
ses; thus they cannot be considered as evidence of a pattern
of genetic diversity. It is important to note that in the
particular case of XCn, due to the small sample size (less
than 30) combined with the high variability of the genetic
1030 Coral Reefs (2010) 29:1023–1033
123
markers used, the possibility of committing Type II error
cannot be underestimated; however, considering that this
uncertainty is present in only one of the 13 sites, the
general pattern of genetic structure observed can be con-
sidered valid.
Taking into account the genetic structure reported here,
it is proposed that conservation strategies for S. partitus,
and other species with similar characteristics, could be
successful at local scales. However, further studies should
be carried out with different biological models and on
broader temporal scales in order to explore the effect of
potential variations in current patterns.
Acknowledgements This project was supported by Mexican Sec-
retarı
´a de Educacio
´nPu
´blica (SEP) and Consejo Nacional de Ciencia
y Tecnologı
´a (CONACyT) funds. The first author also thanks CO-
NACyT for the PhD scholarship and the World Wildlife Fund (WWF,
Russell E. Train Education for Nature Program) for the grant. We
thank the Mexican Protected Marine Areas (Chinchorro Bank Bio-
sphere Reserve, Cozumel and Xcalak Marine Parks) for their assis-
tance in the field work and members of the LEEAC laboratory for
their help in sample collection. We thank Jose
´H. Lara Arenas, Paul
Barber, Elizabeth Jones, and Margarita Aguilar Espinosa for advice,
discussion, and technical help. Thanks also to the three anonymous
reviewers for improving the manuscript.
References
Almada-Villela PC, Sale PF, Gold-Bouchot G, Kjerfve B (2003)
Manual of methods for the MBRS synoptic monitoring program.
Selected methods for monitoring physical and biological
parameters for use in the Mesoamerican region. Mesoamerican
Barrier Reef Systems project (MBRS), Belize City
Almany GR, Berumen ML, Thorrold SR, Planes S, Jones GP (2007)
Local replenishment of coral reef fish populations in a marine
reserve. Science 316:742–744
Arias-Gonza
´lez JE, Legendre P, Rodrı
´guez-Zaragoza FA (2008)
Scaling up beta diversity on Caribbean coral reefs. J Exp Mar
Biol Ecol 366:28–36
Arnaud-Haond S, Vonau V, Rouxel C, Bonhomme F, Prou J, Goyard
E, Boudry P (2008) Genetic structure at different spatial scales in
the pearl oyster (Pinctada margaritifera cumingii) in French
Polynesian lagoons: beware of sampling strategy and genetic
patchiness. Mar Biol 155:147–157
Aronson RB, Edmunds PJ, Precht WF, Swanson DW, Levitan DR
(1994) Large-scale, long-term monitoring of Caribbean coral
reefs: simple, quick, inexpensive techniques. Atoll Res Bull
421:1–19
Bekkevold D, Andre
´C, Dahlgren TG, Clausen LAW, Torstensen E,
Mosegaard H, Carvalho GR, Christensen TB, Norlinder E,
Ruzzante DE (2005) Environmental correlates of population
differentiation in Atlantic herring. Evolution 59:2656–2668
Belkhir K, Borsa P, Chikhi L, Raufaste N, Bonhomme F (1996–2004)
Genetix 4.05, logiciel sous Windows TM pour la ge
´ne
´tique des
populations. Laboratoire Ge
´nome, Populations, Interactions,
CNRS UMR 5171, Universite
´de Montpellier II, Montpellier
France. http://www.genetix.univ-montp2.fr/genetix/genetix.htm
Bell LJ, Moyer JT, Numachi K (1982) Morphological and genetic
variation in Japanese populations of the anemonefish Amphip-
rion clarkii. Mar Biol 72:99–108
Benjamini Y, Yekutieli D (2001) The control of the false discovery
rate in multiple testing under dependency. The Annals of
Statistics 29:1165–1188
Beukers JS, Jones GP (1997) Habitat complexity modifies the impact
of piscivores on a coral reef fish population. Oecologia 114:50–59
Bohonak AJ (2002) IBD (isolation by distance): a program for
analyses of isolation by distance. J Hered 93:153–154.
http://www.bio.sdsu.edu/pub/andy/IBD.html
Bryja J, Charbonnel N, Berthier K, Galan M, Cosson JF (2007)
Density-related changes in selection pattern for major histocom-
patibility complex genes in fluctuating populations of voles. Mol
Ecol 16:5084–5097
Cervigo
´n F (1993) Los peces marinos de Venezuela. Vol 2.
Fundacio
´n cientı
´fica Los Roques, Caracas
Cetina P, Candela J, Sheinbaum J, Ochoa J, Badan A (2006)
Circulation along the Mexican Caribbean coast. J Geophys Res
111:C08021. doi:10.1029/2005JC003056
Chapuis MP, Estoup A (2007) Microsatellite null alleles and
estimation of population differentiation. Mol Biol Evol
24:621–631. http://www.ensam.inra.fr/URLB/
Cowen RK, Castro LR (1994) Relation of coral reef fish larval
distributions to island scale circulation around Barbados, West
Indies. Bull Mar Sci 54:228–244
Cowen RK, Sponaugle S (2009) Laval dispersal and marine
population connectivity. Annu Rev Mar Sci 1:446–466
Cowen RK, Paris CB, Srinivasan A (2006) Scaling of connectivity in
marine populations. Science 311:522–527
Cowen RK, Lwiza KMM, Sponaugle S, Paris CB, Olson DB (2000)
Connectivity of marine populations: open or closed? Science
287:857–859
Dakin EE, Avise JC (2004) Microsatellite null alleles in parentage
analysis. Heredity 93:504–509
Dempster AP, Laird NM, Rubin DB (1977) Maximum likelihood
from incomplete data via the EM algorithm. J Royal Stat Soc.
Series B (Methodological) 39:1–38
Excoffier L, Laval G, Schneider S (2005) Arlequin (version 3.0): An
integrated software package for population genetics data anal-
ysis. Evol Bioinform Online 1:47–50. http://anthro.unige.ch/
software/arlequin/
Fauvelot C, Planes S (2002) Understanding origins of present-day
genetic structure in marine fish: biologically or historically
driven patterns? Mar Biol 141:773–788
Foll M, Gaggiotti O (2006) Identifying the environmental factors that
determine the genetic structure of populations. Genetics
174:875–891. http://www-leca.ujf-grenoble.fr/logiciels.htm
Frisch AJ, Ulstrup KE, Hobbs JA (2007) The effects of clove oil on
coral: an experimental evaluation using Pocillopora damicornis
(Linnaeus). J Exp Mar Biol Ecol 345:101–109
Gaggiotti OE, Bekkevold D, Jørgensen HBH, Foll M, Carvalho GR,
Andre C, Ruzzante DE (2009) Disentangling the effects of
evolutionary, demographic, and environmental factors influenc-
ing genetic structure of natural populations: Atlantic herring as a
case study. Evolution 63:2939–2951
Grimaldi MC, Crouau-Roy B (1997) Microsatellite allelic homoplasy
due to variable flanking sequences. J Mol Evol 44:336–340
Haney RA, Silliman BR, Rand DM (2007) A multi-locus assessment
of connectivity and historical demography in the bluehead
wrasse (Thalassoma bifasciatum). Heredity 98:294–302
Hedgecock D (1994) Does variance in reproductive success limit
effective population size of marine organisms? In: Beaumont A
(ed) Genetics and evolution of aquatic organisms. Chapman and
Hall, London, pp 122–134
Hedgecock D, Barber PH, Edmands S (2007) Genetic approaches to
measuring connectivity. Oceanography 20:70–79
Hepburn RI, Sale PF, Dixon B, Heath DD (2009) Genetic structure of
juvenile cohorts of bicolor damselfish (Stegastes partitus) along
Coral Reefs (2010) 29:1023–1033 1031
123
the Mesoamerican barrier reef: chaos through time. Coral Reefs
28:277–288
Hughes TP, Tanner JE (2000) Recruitment failure, life histories, and
long-term decline of Caribbean corals. Ecology 81:2250–2263
Ihaka R, Gentleman R (1996) R: a language for data analysis and
graphics. J Comput Graph Stat 5:299–314. http://www.r-
project.org
Jones GP, Planes S, Thorrold SR (2005) Coral reef fish larvae settle
close to home. Curr Biol 15:1314–1318
Jones GP, Milicich MJ, Emslie MJ, Lunow C (1999) Self-recruitment
in a coral reef fish population. Nature 402:802–804
Jones GP, Almany GR, Russ GR, Sale PF, Steneck RS, van Oppen
MJH, Willis BL (2009) Larval retention and connectivity among
populations of corals and reef fishes: history, advances and
challenges. Coral Reefs 28:307–325
Jones ME, Barber PH (2005) Characterization of microsatellite loci
for the detection of temporal genetic shifts within a single cohort
of the brown demoiselle, Neopomacentrus filamentosus. Mol
Ecol Notes 5:834–836
Jorda
´n E, Martin E (1987) Chinchorro: morphology and composition
of a Caribbean atoll. Atoll Res Bull 310:1–27
Jordan MA, Snell HL (2008) Historical fragmentation of islands and
genetic drift in populations of Gala
´pagos lava lizards (Microl-
ophus albemarlensis complex). Mol Ecol 17:1224–1237
Jørgensen HBH, Hansen MM, Bekkevold D, Ruzzante DE, Loes-
chcke V (2005) Marine landscapes and population genetic
structure of herring (Clupea harengus L.) in the Baltic Sea. Mol
Ecol 14:3219–3234
Kalinowski ST (2004) Counting alleles with rarefaction: private
alleles and hierarchical sampling designs. Conserv Genet 5:
539–543
Kalinowski ST (2005) HP-Rare 1.0: a computer program for
performing rarefaction on measures of allelic richness. Mol
Ecol Notes 5:187–189. http://www.montana.edu/kalinowski/
Software/HPRare.htm
Kittlein MJ, Gaggiotti OE (2008) Interactions between environmental
factors can hide isolation by distance patterns: a case study of
Ctenomys rionegrensis in Uruguay. Proc R Soc B 275:
2633–2638
Lacson JM (1992) Minimal genetic variation among samples of six
species of coral reef fishes collected at La Parguera, Puerto Rico,
and Discovery Bay, Jamaica. Mar Biol 112:327–331
Lacson JM, Morizot DC (1991) Temporal genetic variation in
subpopulations of bicolor damselfish (Stegastes partitus) inhab-
iting coral reefs in the Florida Keys. Mar Biol 110:353–357
Lacson JM, Riccardi VM, Calhoun SW, Morizot DC (1989) Genetic
differentiation of bicolor damselfish (Eupomacentrus partitus)
populations in the Florida Keys. Mar Biol 103:445–451
Leis JM, McCormick MI (2002) The biology, behavior, and ecology
of the pelagic, larval stage of coral reef fishes. In: Sale PF (ed)
Coral reef fishes. Dynamics and diversity in a complex
ecosystem. Academic Press, San Diego, pp 171–199
Lesser MP (2004) Experimental biology of coral reef ecosystems.
J Exp Mar Biol Ecol 300:217–252
Nei M (1972) Genetic distance between populations. Am Nat
106:283–292
Nei M (1978) Estimation of average heterozygosity and genetic
distance from a small number of individuals. Genetics 89:
583–590
Nemeth RS (1998) The effect of natural variation in substrate
architecture on the survival of juvenile bicolor damselfish.
Environ Biol Fish 53:129–141
Ospina-Guerrero SP, Landinez-Garcı
´a RM, Rodrı
´guez-Castro DJ,
Arango R, Ma
´rquez E (2008) Genetic connectivity of Stegastes
partitus in the South Caribbean evidenced by microsatellite
analysis. Cienc Mar 34:155–163
Palumbi SR (1994) Genetic divergence, reproductive isolation, and
marine speciation. Annu Rev Ecol Syst 25:547–572
Palumbi SR (2003) Population genetics, demographic connectivity,
and the design of marine reserves. Ecol Appl 13:S146–S158
Pandolfi JM (2002) Coral community dynamics at multiple scales.
Coral Reefs 21:13–23
Paris CB, Cowen RK (2004) Direct evidence of a biophysical
retention mechanism for coral reef fish larvae. Limnol Oceanogr
49:1964–1979
Planes S (1993) Genetic differentiation in relation to restricted larval
dispersal of the convict surgeonfish Acanthurus triostegus in
French Polynesia. Mar Ecol Prog Ser 98:237–246
Planes S, Bonhomme F, Galzin R (1993) Genetic structure of
Dascyllus aruanus populations in French Polynesia. Mar Biol
117:665–674
Planes S, Parroni M, Chauvet C (1998) Evidence of limited gene flow
in three species of coral reef fishes in the lagoon of New
Caledonia. Mar Biol 130:361–368
Planes S, Jones GP, Thorrold SR (2009) Larval dispersal connects fish
populations in a network of marine protected areas. Proc Natl
Acad Sci USA 106:5693–5697
Purcell JFH, Cowen RK, Hughes CR, Williams DA (2006) Weak
genetic structure indicates strong dispersal limits: a tale of two
coral reef fish. Proc R Soc B 273:1483–1490
Purcell JFH, Cowen RK, Hughes CR, Williams DA (2009) Population
structure in a common Caribbean coral-reef fish: implications for
larval dispersal and early life-history traits. J Fish Biol 74:403–
417
Roberts CM (1997) Connectivity and management of Caribbean coral
reefs. Science 278:1454–1457
Robertson DR, Green DG, Victor BC (1988) Temporal coupling of
production and recruitment of larvae of a Caribbean reef fish.
Ecology 69:370–381
Rodrı
´guez-Zaragoza FA, Arias-Gonza
´lez JE (2008) Additive diver-
sity partitioning of reef fishes across multiple spatial scales.
Caribb J Sci 44:90–101
Rousset F (2004) Genetic structure and selection in subdivided
populations. Princeton University, Princeton
Ruiz-Za
´rate MA, Arias-Gonza
´lez JE (2004) Spatial study of juvenile
corals in the Northern region of the Mesoamerican Barrier Reef
System (MBRS). Coral Reefs 23:584–594
Sale PF (2004) Connectivity, recruitment variation, and the structure
of reef fish communities. Integr Comp Biol 44:390–399
Sale PF, Guy JA, Steel WJ (1994) Ecological structure of assem-
blages of coral reef fishes on isolated patch reefs. Oecologia
98:83–99
Selkoe KA, Gaines SD, Caselle JE, Warner RR (2006) Current shifts
and kin aggregation explain genetic patchiness in fish recruits.
Ecology 87:3082–3094
Shulman MJ (1998) What can population genetics tell us about
dispersal and biogeographic history of coral-reef fishes? Aust J
Ecol 23:216–225
Shulman MJ, Bermingham E (1995) Early life histories, ocean
currents, and the population genetics of Caribbean reef fishes.
Evolution 49:897–910
Soto I, Andre
´foue
¨t S, Hu C, Muller-Karger FE, Wall CC, Sheng J,
Hatcher BG (2009) Physical connectivity in the Mesoamerican
Barrier Reef System inferred from 9 years of ocean color
observations. Coral Reefs 28:415–425
Swearer SE, Caselle JE, Lea DW, Warner RR (1999) Larval retention
and recruitment in an island population of a coral-reef fish.
Nature 402:799–802
Taylor MS, Hellberg ME (2003) Genetic evidence for local retention
of pelagic larvae in a Caribbean reef fish. Science 299:107–109
The
´riault V, Bernatchez L, Dodson JJ (2007) Mating system and
individual reproductive success of sympatric anadromus and
1032 Coral Reefs (2010) 29:1023–1033
123
resident brook charr, Salvelinus fontinalis, under natural condi-
tions. Behav Ecol Sociobiol 62:51–65
Thiessen RJ, Heath DD (2007) Characterization of one trinucleotide
and six dinucleotide microsatellite markers in bicolor damselfish,
Stegastes partitus, a common coral reef fish. Conserv Genet
8:983–985
Thioulouse J, Chessel D, Dole
´dec S, Olivier JM (1997) ADE-4: a
multivariate analysis and graphical display software. Statistics
and Computing 7:75–83
Thorrold SR, Jones GP, Hellberg ME, Burton RS, Swearer SE, Neigel
JE, Morgan SG, Warner RR (2002) Quantifying larval retention
and connectivity in marine populations with artificial and natural
markers. Bull Mar Sci 70:291–308
Turner TF, Wares JP, Gold JR (2002) Genetic effective size is three
orders of magnitude smaller than adult census size in an
abundant, estuarine-dependent marine fish (Sciaenops ocellatus).
Genetics 162:1329–1339
Van Oosterhout C, Hutchinson WF, Willis DPM, Shipley P (2004)
Micro-checker: software for identifying and correcting genotyp-
ing errors in microsatellite data. Mol Ecol Notes 4:535-538.
http://www.microchecker.hull.ac.uk
Weir BS (1996) Genetic data analysis II: methods for discrete
population genetic data. Sinauer Associates, Sunderland
Weir BS, Cockerham CC (1984) Estimating F-statistics for the
analysis of population structure. Evolution 38:1358–1370
Williams DA, Purcell J, Hughes CR, Cowen RK (2003) Polymorphic
microsatellite loci for population studies of the bicolor damsel-
fish, Stegastes partitus (Pomacentridae). Mol Ecol Notes 3:
547–549
Coral Reefs (2010) 29:1023–1033 1033
123
