Fisheries Research 105 (2010) 172–177

Contents lists available at ScienceDirect

Fisheries Research
journal homepage: www.elsevier.com/locate/fishres

Population genetic structure of dolphinfish (Coryphaena hippurus) in the Gulf of

California, using microsatellite loci
Miguel A. Tripp-Valdez a , Francisco J. García de León a , Sofía Ortega-García b , Daniel Lluch-Cota a ,
Juana López-Martínez a , Pedro Cruz a,∗
a
Centro de Investigaciones Biológicas de Noroeste (CIBNOR), Mar Bermejo 195, Col. Playa Palo de Santa Rita, La Paz, B.C.S. 23096, Mexico
b
Centro Interdisciplinario de Ciencias Marinas, Instituto Politécnico Nacional (CICIMAR), N Av. Instituto Politécnico Nacional s/n, Col. Playa Palo de Santa Rita,
La Paz, B.C.S. 23096, Mexico

a r t i c l e i n f o a b s t r a c t

Article history: We assayed genetic variations at five microsatellite loci of dolphinfish captured at five sites in 2005
Received 15 January 2010 and eight sites in 2006 to detect genetic population structure of dolphinfish Coryphaena hippurus in the
Received in revised form 29 March 2010 Gulf of California and surrounding waters. Results show high genetic variation, similar to other pelagic
Accepted 30 March 2010
fishes with large populations. Pairwise FST values and hierarchical AMOVA detected subtle but significant
spatial and temporal heterogeneity, mainly in samples of 2005 and some of 2006. However, Bayesian
Keywords:
assignment analysis failed to detect genetic differentiation, which was also supported by the Mantel test
Dolphinfish
and gene flow estimates among the sampled sites. This suggests that, despite the slight heterogeneity
Microsatellite
Population structure
detected in this region, the dolphinfish forms a single panmictic population with high genetic variation
Gulf of California and gene flow.
© 2010 Elsevier B.V. All rights reserved.

1. Introduction differentiation between populations (Graves, 1998). However,

The dolphinfish Coryphaena hippurus, also known as dorado
or mahi–mahi, is a migratory pelagic fish inhabiting tropical and
subtropical regions of all oceans (Palko et al., 1982). It is found
from the northern Pacific coast of the Baja California Peninsula
to the Gulf of Tehuantepec and inside the Gulf of California. This
species grows quickly, reach sexual maturity at 50 cm fork length
and has the ability to spawn several times a year (Beardsley,
1967).
By Mexican law, this species is reserved exclusively for the recre-
ational fishing industry within 50 nautical miles of the coast. Given
high market demand, this species is also captured by artisanal fish-
ermen in many parts of Mexico, especially in the northwest. This
has led authorities to consider modifying fisheries laws so dolphin-
fish can be exploited by commercial fishing cooperatives. Whether
dolphinfish is a single population or segregates into local popula-
tions on the west coast of Mexico will be useful for management
plans of this resource. Studies of genetic diversity of marine popula-
tions are important tools for determining whether local stocks are
distinct or determining the degree of flow of individuals between
stocks (Ward, 2000).
Dolphinfish, like other pelagic fishes, have high mobility, large
population size and high reproductive capacity that limit genetic
∗ Corresponding author. Tel.: +52 612 123 8484x3345; fax: +52 612 125 3625.
E-mail address: pcruz@cibnor.mx (P. Cruz).
genetic heterogeneity has been found in some species after ana-
lyzing polymorphic microsatellite markers over an extensive area,
as in the case of striped marlin (McDowell and Graves, 2008) and
yellowfin tuna in the Pacific Ocean (Appleyard et al., 2001) and in
a more restricted area, as the case of bluefin tuna in the Mediter-
ranean (Carlsson et al., 2004) and Atlantic herring in Norway (Shaw
et al., 1999). This suggests that some biological or oceanographic
factors limit genetic flow.
In the Pacific area of Mexico, molecular markers were used for
observing population structure of dolphinfish. Rocha-Olivares et
al. (2006) found genetic heterogeneity in samples from Hawaii,
Sinaloa, and Baja California Sur using RFLPs of the mitochondrial
gene NADH1. Diaz-Jaimes et al. (2006) failed to detect any seasonal
and spatial genetic differences in samples of coastal Baja Califor-
nia Sur, Sinaloa, Sonora and Chiapas using a 750 bp fragment of
this gene. Consequently, it is not certain whether exploitation of
dolphinfish in one area affects the whole population. If discrete
populations are present, this knowledge is useful to establish man-
agement plans for each population.
We analysed five microsatellite loci of dolphinfish captured in
2005 and 2006 from the Gulf of California and the Pacific coast
of Mexico. Using microsatellite loci provided better resolution for
detecting population structure in moderately large geographical
areas, even in species with capacity for high dispersion (Waples,
1998). The main objective was to determine if dolphinfish in the
Pacific coastal waters of Mexico form relatively discrete popula-
tions or a single panmictic population.

0165-7836/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.fishres.2010.03.023
 M.A. Tripp-Valdez et al. / Fisheries Research 105 (2010) 172–177 173

2.2. DNA extraction and amplification

Genomic DNA was extracted using DNAeasy Tissue Kit (Qiagen).

Five microsatellite loci registered in the GenBank (Robert Chap-
man; South Carolina Department of Natural Resources, Charleston
SC., pers. comm.) were amplified (Chi002, Chi008, Chi008a,
Chi023 and Chi032; Table 1). Amplification of the loci DNA took
place in 12.5 ␮l reactions with the following concentrations: 1×
buffer, 3 mM MgCl2 , 0.2 mM dNTP, 0.003 mM of each primer and
0.02 U ␮l−1 Taq polymerase. We used the same PCR conditions as
in Yoshida et al. (2005) to reduce stutter bands. This consisted of
an initial denaturing period of 2 min at 94 ◦ C, five cycles at 94 ◦ C for
5 s, annealing and extension for 60 s at 59 ◦ C, 20 cycles at 94 ◦ C for
0 s, annealing and extension for 60 s at 59 ◦ C, and a final extension
for 60 min at 59 ◦ C. Amplified fragments were run in denaturizing
6% polyacrylamide electrophoresis at 2500 V, 100 mA and 90 W for
∼2.5 h. Alleles were observed using SYBR Gold nucleic acid gel stain
(Rodzen et al., 1998) and a scanner (FMBIO II Multi-viewer, Hitachi
Software Engineering). Alleles were sized using a sequencing lad-
der (Sequitherm Cycle Sequencing kit–Epicentre Tech) using the
PUC 19 plasmid and M13 primer.

2.3. Statistical analyses

Levels of genetic variability, expressed in number of alleles (NA ),

Fig. 1. Sampling sites of the dolphinfish Coryphaena hippurus. Sites with a triangle,
collected only in 2005; sites with a circle collected only in 2006; sites with a square,
collected in both years, n is sample size: a samples collected in 2005 and b samples
collected in 2006.
2. Materials and methods
2.1. Tissue collection
Muscle tissue (∼2 g) was collected from 731 dolphinfish over
2 years (2005 and 2006) in the Gulf of California and three out-
side sites (see Fig. 1). Three sites were sampled in 2005 and 2006
(Cabo San Lucas, Loreto and Guaymas), two sites were sampled
only in 2005 (Puerto Libertad, Topolobampo), and five sites were
sampled only in 2006 (Mazatlan, Nayarit, La Paz, Punta Lobos and
an open ocean sample incidentally collected from a tuna ves-
sel (10◦ 44 N, 131◦ 12 W). The samples were collected from June
through October (summer), with the exception of Nayarit, which
was collected in January. Individuals were captured by local fish-
ermen at all locations except the recreational fishing fleet at Cabo
San Lucas and La Paz and the incidental capture of the open ocean
sample. Tissue samples were preserved in 80% ethanol until DNA
extraction.
observed heterozygosity (HO ) and expected heterozygosity (HE )
were obtained for each locus from every sample using the soft-
ware GenAlEx 6.1 (Peakall and Smouse, 2006). Deviations from the
Hardy–Weinberg equilibrium were examined for each population
at each locus by calculating the fixation index FIS with software
Genepop 4.0 (Raymond and Rousset, 1995) with 10,000 dememo-
rizations and 1000 iterations. The null hypothesis of independence
between loci (no linkage disequilibrium) was tested using the
same software. Levels of genetic variation were compared among
sampled sites using the Kruskal–Wallis ANOVA, which is a nonpara-
metric alternative to one-way ANOVA, using software STATISTICA
v. 7 (StatSoft, 2004).
Homogeneity of microsatellite allele frequencies between all
pairs of samples in both years was assessed using Fisher’s exact
test with software Genepop 4.0 (Raymond and Rousset, 1995)
with 10,000 dememorizations, 500 batches and 5000 iterations
per batch. Sample pairs that did not differ in allele frequency were
pooled. Levels of genetic differentiation between all sampling sites
were analysed by calculating Wright’s pairwise FST statistics (Weir
and Cockerham, 1984). A global AMOVA test including all sam-
ples was conducted to obtain the global FST value. Hierarchical
genetic structuring of the samples was undertaken by assessing
the relative contribution among groups, within groups, and within
populations (AMOVA). For this, the sample sites that did not show

Table 1
Characteristics of the five microsatellite loci for the dolphinfish Coryphaena hippurus.

Locus name Size (pb) Primers Sequence Accession key

Chi002 104 F:GAAAAACTCACACGG TCACTTG (CT)8 (CA)2 CG(CA)3 AY135025

R:GGCTTGCCAACCTGA GATTA

Chi008 134 F:ATTGATGAGGGTTCAGACGG (TC)6 (TGTC)3 TGTA(TGTC)3 (TC)3 AY189832

R:GGCAGCAGTTCAGGAGGTTA

Chi008A 106 F:GGGCTCATGACACAAATT CC (TG)12 AY135026

R:CCAAACATGTGAGTGCTGCT

Chi023 139 F:GATGGGAGACTCCAA CCTGA (TG)2 AG(TG)3 TCAG(TG)3 CAGTAG(TG)3 TCAG(TG)3 AY135027
R:CCCATCTTGTGGAGGTTGAT

Chi037 82 F:GATATCAGGCCTCCT GCT TG (TG)8 AY135028

R:GGGATTGGT TCCCTCACTCT

Source: Robert Chapman (South Carolina Department of Natural Resources, Charleston SC, pers. comm.).
174 M.A. Tripp-Valdez et al. / Fisheries Research 105 (2010) 172–177
genetic differentiation with pairwise FST analysis were pooled and
compared with the rest of the samples. All the analyses were per-
formed using software Arlequin 3.0 (Excoffier et al., 2005) using
10,000 permutation in each case. In all cases with multiple tests,
significance levels were adjusted using the sequential Bonferroni
technique (Rice, 1989).
We also used a Bayesian approach for assignment of individuals
to samples for each year with software Structure 2.3.1 (Pritchard
et al., 2000) to complement the results obtained with F statistics.
For this, we selected the admixture model and the correlated allele
frequencies between populations. The Markov Chain Monte Carlo
consisted of 105 steps with a burn-in period of 25,000 steps. We
explored a range of K from 1 to 8 with 5 runs of each K value.
The Mantel test was used to test correlation between genetic
distances, expressed as FST /(1 − FST ), and geographic distance,
expressed as logKm with the program ISOLDE (part of Genepop;
Raymond and Rousset, 1995).
We used the software Migrate 2.1.3 (Beerli, 2008) to infer the
population size parameter  (4NE ), where NE is the effective pop-
ulation size and  is the mutation rate per site) and migration
rate (M, m/), where m is the immigration rate per generation.
For this analysis, we used the maximum likelihood approach with
the infinite-alleles model. Values of FST were used as the starting
parameters for the estimation of  and M. The ML was run for
ten short and three long chains with 10,000 and 100,000 recorded
genealogies, respectively, with a burn-in of 10,000 genealogies for
each chain. We ran the software two times for verifying consistency
of results.
3. Results
3.1. Genetic diversity
The mean number of alleles (NA ) was between 11.4 and 17.6
at the five loci. The observed heterozygosity (HO ) ranged from
0.80 to 0.93 and expected heterozygosity (HE ) ranged from 0.8
to 0.86 (Table 2). After the sequential Bonferroni correction, no
loci were in linkage disequilibrium in either year (P > 0.05), which
indicates that the assortment of alleles is independent at the five
loci. Of 65 Hardy–Weinberg equilibrium tests, using FIS (data not
shown) for every sampled site from both years with five loci, only
10 tests showed disequilibrium, seven of them with heterozy-
gote excess and three with heterozygote deficits (Table 2). The
Kruskal–Wallis test showed that there was no statistical difference
in values of NA (P = 0.48), HO (P = 0.10) and HE (P = 0.91) between
locations.
3.2. Population structure
After sequential Bonferroni correction, the exact test for genetic
differentiation in allele frequencies revealed only two exact tests
with insignificant values (P > 0.025), These were Loreto 2005 ver-
sus Loreto 2006, which were pooled for subsequent analyses in a
single “Loreto” sample, and Cabo San Lucas 2006 versus Guaymas
2006, which was treated separately because of the different place of
origin. Pairwise multi-locus FST analysis showed small but highly
significant values in 41 out of 66 comparisons (FST from −0.017
to 0.031; P < 0.002; Table 3). From table, significant values for FST
occurred mainly between samples collected in 2005, between them
and samples the collected in 2006. The samples from 2006 collected
at the more geographically distant locations (Nayarit 2006, Punta
Lobos 2006 and Ocean 2006) had significant differences in values
for FST and with some other 2006 localities inside the Gulf. The other
samples collected in 2006 (Cabo San Lucas 2006, Guaymas 2006, La
Paz 2006 and Mazatlan 2006) had smaller or insignificant values
for FST (P > 0.002).
The global AMOVA test that included all samples resulted in a
small but significant value for FST (FST = 0.01; P = 0.0). For the hierar-
chical AMOVA test, we pooled the samples that showed no genetic
differentiation from values for FST (Loreto, Cabo San Lucas 2006,
Guaymas 2006, La Paz 2006 and Mazatlan 2006) into a single group
and compared this with the other samples. Results indicated sig-
nificant differences among tested groups (FCT = 0.01; P = 0.00) and
within populations (FST = 0.01; P = 0.00), but no significant differ-
ences among populations within groups (FSC = −0.004; P = 0.99).
These results denote a subtle heterogeneity over time within the
Gulf and subtle spatial heterogeneity with samples in surrounding
waters.
The Bayesian assignment test made with the Structure software
showed that, after five iterations, the higher value of K were for K = 1
(log likelihood ln P(K) for K = 1 was −16,022, from which the ln P(K)
values began to fall and variance increased). These results suggest
that dolphinfish has no genetically differentiated groups within the
area that we sampled. According to the Mantel test, there was no
significant correlation between genetic and geographic distance in
the sampled locations (R2 = 0.005; P = 0.32).
3.3. Effective population size and migrate rates
Maximun likelihood estimates of the values of  made with
the Migrate software ranged from 1.05 (Ocean site) to 3.97
(Loreto); these values were translated to effective population sizes
(NE = /4) from 2625 to 9925, assuming a microsatellite mutation
rate of 10−4 per locus per generation occurs (Table 4).
In general, migration rates estimated from Migrate software
were small, with values from 0.06 to 4.04. Gene flow in the sam-
ples is asymmetrical, where the higher values of M are found
among samples inside the Gulf of California. Samples from out-
side the Gulf (Punta Lobos and Ocean) appear to migrant into the
Gulf.
4. Discussion
4.1. Genetic variability
This is the first study of genetic variation using microsatel-
lite markers of the dolphinfish. Analyses of five microsatellite loci
revealed high levels of genetic variation (expressed in number of
alleles and heterozygosity) during the 2-year study. These results
are comparable in variability to the report by DeWoody and Avise
(2000) for other marine fishes (NA : 19.9, HE : 0.77) and pelagic fishes
of commercial importance, such as the swordfish Xiphias gladus
(NA : 14.3–17.5, HE : 0.771–0.784 in Muths et al., 2009), bigeye tuna
Thunnus obesus (NA : 8.7–20.4, HE : 0.897–0.909 in Gonzalez et al.,
2008), and sardine Sardina pilcharus (NA : 27.4–31, HE : 0.944–0.953
in Gonzalez and Zardoya, 2007), which have large populations and
high gene flow. The Hardy–Weinberg disequilibrium is common
in many marine fishes, but deviations to equilibrium generally
prevail over heterozygotic deficits (Appleyard et al., 2001; Diaz-
Jaimes and Uribe-Alcocer, 2006; Ruzzante et al., 1996), which
are a consequence of factors involving the fish’s reproductive
systems, presence of null alleles, or a Wahlund effect (reduc-
tion of heterozygosity in a population caused by subpopulation
structure). In the dolphinfish, we found more deviations toward
heterozygous excess, possibly as a consequence of the mixture and
reproduction of individuals of different cohorts that could have
different allelic composition. However, we could not find a clear
tendency of Hardy–Weinberg disequilibrium in a specific loci or
location, so there is a chance that those deviations of the equilib-
rium are a consequence of stochastic events in allelic frequency
distributions.
 M.A. Tripp-Valdez et al. / Fisheries Research 105 (2010) 172–177 175

Table 2
Summary statistics for five microsatellite loci in 13 samples of the dolphinfish Coryphaena hippurus.

Sample Locus

chi002 chi008 chi008a chi023 chi037 Mean population

Cabo San Lucas 2005 N 57 58 59 58 58 58

NA 13 26 15 25 9 17.6
HO 0.82 0.90 0.93 0.93 0.86 0.89
HE 0.78 0.89 0.82 0.93 0.82 0.85

Loreto 2005 N 55 54 55 54 55 55
NA 13 19 12 31 16 18
HO 0.98 0.94 0.89 0.91 0.93 0.93
HE 0.80 0.85 0.83 0.92 0.83 0.85

Guaymas 2005 N 62 63 61 61 63 62
NA 11 19 19 25 12 17.2
HO 0.92 0.92 0.70 0.98 0.92 0.89
HE 0.85 0.88 0.86 0.93 0.82 0.86

Puerto Libertad 2005 N 32 31 31 31 32 31.4

NA 14 13 16 12 11 13.2
HO 0.78 0.65 0.87 0.87 0.91 0.81
HE 0.86 0.82 0.91 0.89 0.80 0.86

Topolobampo 2005 N 73 70 72 69 72 71.2

NA 13 21 14 20 12 16
HO 0.77 0.90 0.86 0.96 0.83 0.86
HE 0.76 0.85 0.86 0.89 0.77 0.82

Cabo San Lucas 2006 N 56 68 63 47 58 58.4

NA 6 14 12 15 11 11.6
HO 0.68 0.85 0.90 0.81 0.78 0.80
HE 0.74 0.88 0.83 0.90 0.75 0.82

Loreto 2006 N 33 32 33 34 32 32.80

NA 7 18 10 16 7 11.60
HO 0.82 0.88 0.91 0.91 0.75 0.85
HE 0.74 0.88 0.85 0.89 0.72 0.82

Guaymas 2006 N 56 52 48 51 48 51
NA 7 15 13 16 10 12.2
HO 0.73 0.87 0.98 0.86 0.85 0.86
HE 0.75 0.86 0.85 0.87 0.78 0.82

Mazatlan 2006 N 71 71 68 64 74 69.6

NA 11 22 13 14 14 14.8
HO 1.00 0.94 0.87 0.97 0.84 0.92
HE 0.79 0.88 0.86 0.89 0.79 0.84

Nayarit 2006 N 78 77 70 73 77 75
NA 9 16 12 16 16 13.8
HO 0.95 0.79 0.76 0.78 0.83 0.82
HE 0.79 0.85 0.79 0.87 0.76 0.81

La Paz 2006 N 41 34 37 41 40 38.6

NA 8 16 10 14 9 11.4
HO 0.85 0.74 0.84 0.93 0.80 0.83
HE 0.71 0.85 0.78 0.90 0.74 0.80

Punta Lobos 2006 N 38 38 35 34 37 36.4

NA 8 17 12 16 8 12.2
HO 0.82 0.87 0.83 0.94 1.00 0.89
HE 0.73 0.84 0.84 0.89 0.72 0.80

Ocean 2006 N 45 48 50 48 48 47.8

NA 9 17 10 15 7 11.6
HO 0.89 0.90 0.92 1.00 0.92 0.92
HE 0.80 0.88 0.82 0.90 0.73 0.82

Mean of loci N 58.08 58.00 56.83 55.42 57.83 57.23

NA 10.17 18.33 13.25 18.25 11.33 14.27
HO 0.84 0.85 0.86 0.91 0.87 0.87
HE 0.78 0.86 0.84 0.90 0.77 0.83

Sample size (N); number of alleles (NA ); observed heterozygosity (HO ); expected heterozygosity (HE ). Data in bold are Hardy–Weinberg disequilibrium for FIS (P < 0.01).

4.2. Population structure
Small but highly significant values of pairwise FST revealed
that dolphinfish have subtle genetic differentiation between sam-
pled sites, mainly in the samples collected in 2005 but in some
of the samples of 2006. These results and those from AMOVA
showed a slight genetic heterogeneity both in temporal and spatial
scales. However, Bayesian assignment analysis made with Struc-
ture software and an isolation-by-distance test, which correlates
geographic distance with genetic distance, failed to detect any dif-
ferent genetic group in the entire data set. This means that even
when there was some genetic heterogeneity among the samples,
176 M.A. Tripp-Valdez et al. / Fisheries Research 105 (2010) 172–177

Table 3
Pairwise values of FST (below the diagonal) and their respective P values (above the diagonal).

CSL05 LOT GUA05 PLI05 TOP05 CSL06 GUA06 MAZ06 NAY06 LAP06 PLO06 OCE06

CSL05 – 0.000 0.000 0.000 0.000 0.038 0.004 0.000 0.000 0.000 0.000 0.000
LOT 0.019 – 0.000 0.000 0.000 1.000 0.270 0.725 0.011 0.105 0.000 0.001
GUA05 0.027 0.023 – 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
PLI05 0.020 0.020 0.012 – 0.000 0.373 0.005 0.000 0.000 0.000 0.000 0.000
TOP05 0.024 0.008 0.027 0.015 – 0.974 0.002 0.000 0.000 0.002 0.000 0.000
CSL06 0.004 −0.015 0.010 0.000 −0.005 – 0.997 0.988 0.012 0.999 0.735 0.761
GUA06 0.007 0.000 0.021 0.010 0.007 −0.010 – 0.119 0.002 0.090 0.003 0.006
MAZ06 0.017 −0.002 0.016 0.014 0.009 −0.007 0.001 – 0.000 0.164 0.000 0.004
NAY06 0.021 0.004 0.025 0.019 0.017 0.005 0.007 0.009 – 0.019 0.000 0.000
LAP06 0.016 0.002 0.022 0.018 0.009 −0.017 0.003 0.001 0.006 – 0.061 0.001
PLO06 0.018 0.013 0.031 0.021 0.015 −0.004 0.009 0.011 0.020 0.004 – 0.000
OCE06 0.022 0.006 0.021 0.018 0.010 −0.003 0.006 0.005 0.015 0.009 0.010 –

Bold number of values of FST are significant (P < 0.002; after sequential Bonferroni test); CSL (Cabo San Lucas); LOT (Loreto); GUA (Guaymas); PLI (Puerto Libertad); TOP
(Topolobampo); MAZ (Mazatlan); NAY (Nayarit); LAP (La Paz); PLO (Punta Lobos); OCE (Ocean) sites.

the variability is too high to find any prevalent group of dolpinfish
in this region. This supports the results obtained by Diaz-Jaimes et
al. (2006) with mtDNA. They did not detect any population struc-
ture of samples from the Pacific and the Gulf of California from 2002
to 2004. Both results suggest a single population in the study area.
In studies of microsatellites in cichlid fish (Duftner et al., 2006),
birds (Friesen et al., 2006) and mammals (Cegelski et al., 2003),
data were analysed simultaneously with the Structureı̌s Bayesian
approach and pairwise FST calculations; both methods revealed a
well-defined genetic population structure, contrary to what we
found in dolphinfish. The values of FST in our study (<0.027) are
low, compared to cichlid fish (0.04–0.2), birds (0.07–0.2) and mam-
mals (0.07–0.09). This implies that the genetic structure detected
by FST for dolphinfish is too weak for a multi-locus analysis (Struc-
ture software) to detect, providing it is present. Latch et al. (2006)
observed that Structure software infers the number of clusters even
when these were not well-differentiated (FST = 0.02–0.03), but with
results suggesting that FST must be at least 0.05 to reach an accuracy
>97%. Similar results were obtained for bigeye tuna (Gonzalez et al.,
2008), in which certain values of pairwise FST were very low, but
highly significant (FST between 0.001 and 0.027), but the Structure
software failed to detect any genetically differentiated group, sup-
porting the conclusion that there is no genetic structure differences
in tuna in this area.
In this context, it is important to distinguish between low but
biological meaningful genetic differences and a statistical effect
(Waples, 1998). In our study, significant values of FST found in
both years are too low to suggest genetically different popula-
tions. Besides, dolphinfish have biological features that support
the hypothesis of a single population in the northern part of Mex-
ico, including: (1) high migratory activity (Kingsford and Defries,
1999); (2) feed on many prey and change their diet according to
the epipelagic environment (Aguilar-Palomino et al., 1998); (3)
high reproductive capacity and capacity to spawn many times
within the year (Beardsley, 1967) and (4) along the Pacific coast
of Mexico, the highest abundance of dolphinfish larvae is present
within the 28 ◦ C isotherm in summer and the 24 ◦ C isotherm in
spring (Sanchéz-Reyes, 2008). That means that reproduction of dol-
phinfish is associated with sea surface temperature rather than a
specific geographic site. This view of the distribution of dolphinfish
is important, given the interannual variability of temperature in the
Gulf of California caused by ENSO events (Lluch-Cota, 2004).
4.3. NE estimate and migration rates
Effective population size of dolphinfish, estimated by Migrate
software based in  values ranged from 2625 to 9925. Even if there
is no census information on the real population size of dolphinfish
in this region to compare with estimates of NE , the values derived
from Migrate software are very similar to population size of big-
eye tuna (4175–15,175) obtained by Gonzalez et al. (2008) using
Migrate software with the same mutation rate. These authors men-
tion that for population size derived from annual catch data NE is
several orders of magnitude lower than the real population size,
which appears to be common in marine fish (Frankham et al., 2005;
Hauser et al., 2002). This means that the real population size of
dolphinfish could be several orders of magnitude larger than the
estimated NE .
Migration rates (M) estimated with Migrate software were small
comparing with other large pelagic fishes, such as striped marlin
Kajikia audax in the Pacific (immigration rates between 25.1 and
71.2; McDowell and Graves, 2008) and bigeye tuna T. obesus (immi-
gration rates between 10 and 60; Gonzalez et al., 2008). However,
these rates demonstrate that there is gene flow between samples
that prevents isolation of groups. Estimates of migration rates also
demonstrate that Cabo San Lucas in 2005 and 2006 seem to be an

Table 4
Maximun likehood estimates of migration rates (M) and effective population size (NE ) of the dolphinfish Coryphaena hippurus.

Location CSL05a LOTa GUA05a PLI05a TOP05a CSL06a GUA06a MAZ06a NAY06a LAP06a PLO06a OCE06a NE

CSL05 1.76 1.91 1.24 2.60 1.85 0.82 1.07 0.18 0.70 0.67 0.40 0.70 4400
LOT 3.27 3.97 1.67 1.25 2.00 2.45 0.89 1.58 1.20 1.20 1.30 2.05 9925
GUA05 2.01 2.69 1.76 1.86 1.42 0.75 0.35 0.91 0.53 0.13 0.89 1.43 4400
PLI05 0.66 0.49 1.80 1.57 2.38 1.63 0.32 1.42 1.50 0.11 0.26 0.26 3925
TOP05 3.39 1.68 1.24 1.86 1.16 0.60 0.39 1.90 0.95 0.50 0.39 1.62 2900
CSL06 0.31 0.21 0.38 0.18 0.33 1.26 1.67 1.67 2.42 1.01 0.83 1.21 3150
GUA06 0.14 1.33 0.89 0.56 0.47 1.36 1.21 1.43 2.18 3.35 1.94 0.80 3025
MAZ06 0.92 2.30 0.68 0.56 0.43 3.56 1.70 1.79 1.80 1.56 0.66 1.05 4475
NAY06 0.15 0.84 0.66 0.51 0.69 1.02 1.31 2.89 1.30 1.46 1.38 0.71 3250
LAP06 0.36 0.69 1.86 0.94 0.76 1.47 4.04 2.86 2.70 1.18 0.83 1.16 2950
PLO06 0.13 0.63 1.09 0.36 0.83 2.92 1.72 0.63 1.39 1.17 1.26 2.20 3150
OCE06 0.06 1.30 0.78 0.71 0.41 2.35 1.27 2.24 1.40 2.18 2.53 1.05 2625
a
Recipient population; values in bold correspond to  estimates of each sample; NE estimated with the formula /4.
 M.A. Tripp-Valdez et al. / Fisheries Research 105 (2010) 172–177
important center of gene flow. Zúñiga-Flores et al. (2008) observed
that, although dolphinfish are found throughout the year near Cabo
San Lucas, it is more abundant in summer when sea surface tem-
perature inside the Gulf of California increase and forms favorable
conditions for aggregation. Since the collection of tissue samples
in our study occurred between June and October, migration rates
may be a consequence of the movement of dolphinfish to warmer
waters.
Although the results for NE and migration rates for dolphinfish
are comparable with those reported for other species of pelagic
fishes, such as tuna, sardines and striped marlin (Carlsson et al.,
2004; Gonzalez and Zardoya, 2007; Gonzalez et al., 2008; McDowell
and Graves, 2008), these data need to be used with caution, espe-
cially if the objective is for managing the resource. Although the
coalescent method that uses Migrate software provides more accu-
rate and reliable estimates of genetic differentiation than methods
based on Wright’s FST (Beerli and Felsenstein, 2001; Gonzalez et al.,
2008), the estimates obtained with this software could have large
variations by a few orders of magnitude, depending on the muta-
tion rate that is used (Carlsson et al., 2004). Given these concerns,
the results for NE and gene flow are very important for management
of dolphinfish, there is still no reliable information on how individ-
uals are moving inside the Gulf of California (Zuñiga-Flores, Centro
Interdisciplinario de Ciencias Marinas-IPN, La Paz, B.C.S., Mexico,
pers. comm.).
Even when we found temporal and spatial genetic heterogeneity
among samples, no obvious grouping of dolphinfish was identified.
This means that under the genetic concept of stock, which estab-
lishes that a stock is a reproductively isolated unit and is genetically
different from other units (Begg and Waldman, 1999; Ward, 2000),
dolphinfish in the Gulf of California and surrounding waters form a
single panmictic stock with high genetic variability, comparable to
other commercially exploited pelagic fish. This information should
be considered before a management plan is established because any
local fishery could have potential effects on the entire population
and all areas where dolphinfish are captured.
Acknowledgments
We thank Robert Chapman of the South Carolina Department
of Natural Resources for data on microsatellite sequences; Eloisa
Herrera–Valdivia, Rufino Morales and Marcela Zuñiga for assistance
with tissue sampling. Jair Rangel for technical support in laboratory.
Funding was provided by the Comisión Nacional de Acuacul-
tura y Pesca (CONAPESCA grant 956-1), Centro de Investigaciones
Biológicas del Noroeste and the State Governments of Sonora and
Sinaloa.
References
Aguilar-Palomino, B., Galvan-Magaña, F., Abitia-Cárdenas, L., Muhlia-Melo, A.,
Rodriguez-Romero, J., 1998. Feeding aspects of the dolphin Coryphaena hippu-
rus Linnaeus, 1758 in Cabo San Lucas, Baja California Sur, Mexico. Cien. Mar. 24,
253–265.
Appleyard, S.A., Grewe, P.M., Innes, B.H., Ward, R.D., 2001. Population structure of
yellowfin tuna (Thunnus albacares) in the western Pacific Ocean, inferred from
microsatellite loci. Mar. Biol. 139, 383–393.
Beardsley Jr., G.L., 1967. Age, growth, and reproduction of the dolphin, Coryphaena
hippurus, in the Straits of Florida. Copeia 2, 441–451.
Beerli, P., 2008. Migrate-N Version 2.4., http://popgen.scs.fsu.edu/Migrate-n.html.
Beerli, P., Felsenstein, J., 2001. Maximum likelihood estimation of a migration
matrix and effective population sizes in n subpopulations by using a coalescent
approach. Proc. Natl. Acad. Sci. U.S.A. 98, 4563–4568.
Begg, G.A., Waldman, J.R., 1999. An holistic approach to fish stock identification. Fish.
Res. 43, 35–44.
Carlsson, J., McDowell, J.R., Diaz-Jaimes, P., Carlsson, J.E.L., Boles, S.B., Gold, J.R.,
Graves, J.E., 2004. Microsatellite and mitochondrial DNA analyses of Atlantic
bluefin tuna (Thunnus thynnus thynnus) population structure in the Mediter-
ranean Sea. Mol. Ecol. 13, 3345–3356.
177
Cegelski, C.C., Waits, L.P., Anderson, N.J., 2003. Assessing population structure
and gene flow in Montana wolverines (Gulo gulo) using assignment-based
approaches. Mol. Ecol. 12, 2907–2918.
DeWoody, J.A., Avise, J.C., 2000. Microsatellite variation in marine, freshwater and
anadromous fishes compared with other animals. J. Fish. Biol. 56, 461–473.
Diaz-Jaimes, P., Uribe-Alcocer, M., 2006. Spatial differentiation in the eastern Pacific
yellowfin tuna revealed by microsatellite variation. Fish. Sci. 72, 590–596.
Diaz-Jaimes, P., Uribe-Alcocer, M., Ortega-Garcia, S., Durand, J.D., 2006. Spatial and
temporal mitochondrial DNA genetic homogeneity of dolphinfish populations
(Coryphaena hippurus) in the eastern central Pacific. Fish. Res. 80, 333–338.
Duftner, N., Sefc, K.M., Koblmuller, S., Nevado, B., Verheyen, E., Phiri, H., Sturmbauer,
C., 2006. Distinct population structure in a phenotypically homogeneous rock-
dwelling cichlid fish from lake Tanganyika. Mol. Ecol. 15, 2381–2395.
Excoffier, L., Laval, G., Schneider, S., 2005. ARLEQUIN. Ver 3.0. An integrated software
package for population genetic data analysis. Evol. Bioinform. Online 1, 47–50.
Frankham, R., Ballou, J.D., Briscoe, D., 2005. Introduction to Conservation Genetics.
Cambridge University Press, Cambridge.
Friesen, V.L., Gonzalez, J.A., Cruz-Delgado, F., 2006. Population genetic structure and
conservation of the Galapagos petrel (Pterodroma phaeopygia). Conserv. Genet.
7, 105–115.
Gonzalez, E., Zardoya, R., 2007. Relative role of life-history traits and historical factors
in shaping genetic population structure of sardines (Sardina pilchardus). BMC
Evol. Biol. 7, 197.
Gonzalez, E.G., Beerli, P., Zardoya, R., 2008. Genetic structuring and migration pat-
terns of Atlantic bigeye tuna, Thunnus obesus (Lowe, 1839). BMC Evol. Biol. 8,
252.
Graves, J.E., 1998. Molecular insights into the population structures of cosmopolitan
marine fishes. J. Hered. 89, 427–437.
Hauser, L., Adcock, G.J., Smith, P.J., Bernal Ramãrez, J.H., Carvalho, G.R., 2002. Loss
of microsatellite diversity and low effective population size in an overexploited
population of New Zealand snapper (Pagrus auratus). Proc. Natl. Acad. Sci. U.S.A.
99, 11742–11747.
Kingsford, M.J., Defries, A., 1999. The ecology and fishery for Coryphaena spp. in the
waters around Australia and New Zealand. Sci. Mar. 63, 267–275.
Latch, E.K., Dharmarajan, G., Glaubitz, J.C., Rhodes, J.R.O.E., 2006. Relative perfor-
mance of Bayesian clustering software for inferring population substructre and
individual assignment at low levels of population differentiation. Conserv. Gen.
7, 295–302.
Lluch-Cota, S.E., 2004. Marine Ecosystems of the North Pacific. Pices Spec. Pub., Gulf
of California, pp. 1–7, #1.
McDowell, J.R., Graves, J.E., 2008. Population structure of striped marlin (Kajikia
audax) in the Pacific Ocean based on analysis of microsatellite and mitochondrial
DNA. Can. J. Aquat. Fish. Sci. 65, 1307–1320.
Muths, D., Grewe, P., Jean, C., Bourjea, J., 2009. Genetic population structure of the
swordfish (Xiphias gladius) in the southwest Indian Ocean: sex-biased differ-
entiation, congruency between markers and its incidence in a way of stock
assessment. Fish. Res. 97, 263–269.
Palko, J.B., Beardsley, G.L., Richards, W.J., 1982. Synopsis of the biological data on
dolphin-fishes, Coryphaena hippurus and Coryphaena equiselis L. NOAA Technical
Report NMFS Circular 443, FAO Fish Synopsis No. 130, 28.
Peakall, R., Smouse, P.E., 2006. Genalex 6: genetic analysis in excel, population
genetic software for teaching and research. Mol. Ecol. Notes 6, 288–295.
Pritchard, J.K., Stephens, M., Donnelly, P., 2000. Inference of population structure
using multilocus genotype data. Genetics 155, 945–959.
Raymond, M., Rousset, F., 1995. GENEPOP 4.0 population genetic software for exact
test and ecumenism. J. Hered. 86, 248–249.
Rice, W., 1989. Analyzing tables of statistical tests. Evolution 43, 223–225.
Rocha-Olivares, A., Bobadilla-Jimenez, M., Ortega-Garcia, S., Saavedra-Sotelo, N.,
Sandoval-Castillo, J.R., 2006. Mitochondrial variability of dolphinfish Coryphaena
hippurus populations in the Pacific Ocean. Cien. Mar. 32, 569–578.
Rodzen, J.A., Agresti, J., Tranah, G., May, B., 1998. Agarose overlay allow simplified
staining of polyacrilamide gels. Biotechnology 24, 584.
Ruzzante, D.E., Taggart, C.T., Cook, D., 1996. Spatial and temporal variation in the
genetic composition of a larval cod (Gadus morhua) aggregation: cohort contri-
bution and genetic stability. Can. J. Fish. Aquat. Sci. 53, 2695–2705.
Sanchéz-Reyes, N.A., 2008. Distribución de larvas de dorado Coryphaena hippurus
(Linnaeus, 1758) y Coryphaena equiselis (Linnaeus, 1758) en el Pacífico oriental
mexicano. Master’s Thesis. Centro Interdisciplinario de Ciencias Marinas-IPN, La
Paz, B.C.S., Mexico.
Shaw, P.W., Turan, C., Wright, J.M., O’Connell, M., Carvalho, G.R., 1999. Microsatellite
DNA analysis of population structure in Atlantic herring (Cuplea harengus), with
direct comparison to allozyme and mtDNA RFLP analyses. Heredity 83, 490–499.
StatSoft, 2004. STATISTICA, Version 7. StatSoft, www.statsoft.com.
Waples, R.S., 1998. Separating the wheat from the chaff: Patterns of genetic differ-
entiation in high gene flow species. J. Hered. 89, 438–450.
Ward, R.D., 2000. Genetics in fisheries management. Hydrobiologia 420, 191–201.
Weir, B.S., Cockerham, C.C., 1984. Estimating F-statistics for the analysis of popula-
tion structure. Evolution 38, 1358–1370.
Yoshida, K., Nakagawa, N., Wada, S., 2005. Multiplex PCR system applied for
analysing microsatellite loci of Schlegel’s black rockfish, Sebastes schlegeli. Mol.
Ecol. Notes 5, 416–418.
Zúñiga-Flores, M.S., Ortega-García, S., Klett-Traulsen, A., 2008. Interannual and sea-
sonal variation of dolphinfish (Coryphaena hippurus) catch rates in the southern
Gulf of California, Mexico. Fish. Res. 94, 13–17.