Journal of Heredity Advance Access published February 28, 2013
Journal of Heredity
doi:10.1093/jhered/est002
© The American Genetic Association. 2013. All rights reserved.
For permissions, please email: journals.permissions@oup.com.
Dispersal at a Snail’s Pace: Historical
Processes Affect Contemporary
Genetic Structure in the Exploited
Wavy Top Snail (Megastraea undosa)
Alison J. HAupt, FiorenzA MicHeli, And stepHen r. pAluMbi
Address correspondence to Alison J. Haupt at the address above, or e-mail: ajhaupt@gmail.com.
Abstract
We used population genetics to assess historical and modern demography of the exploited wavy top snail, Megastraea undosa,
which has a 5–10 day pelagic larval duration. Foot tissue was sampled from an average of 51 individuals at 17 sites across
the range of M. undosa. Genetic structure at the mtDNA locus is strikingly high (ΦST of 0.19 across 1000 km), and a major
cline occurs in northern Baja California (ΦCT of 0.29 between northern and southern populations). Genetic data indicate
that the northern region is highly connected through larval dispersal, whereas the southern region exhibits low genetic structure. However, additional analyses based on patterns of haplotype diversity and relationships among haplotypes indicate that
M. undosa has likely recently expanded into the Southern California Bight or expanded from a small refugial population, and
analysis using isolation by distance to calculate dispersal distance indicates surprisingly short estimates of dispersal from 30
m to 3 km. This scenario of a northward expansion and limited larval dispersal is supported by coalescent-based simulations
of genetic data. The different patterns of genetic variation between northern and southern populations are likely artifacts of
evolutionary history rather than differences in larval dispersal and this may have applications to management of this species.
Speciically, these data can help to inform the scale at which this species should be managed, and given the potentially very
small dispersal distances, this species should be managed at local scales. Consideration of the evolutionary history of target
species allows for a more accurate interpretation of genetic data for management.
Key words: isheries, larval dispersal, management, phylogeography, population genetics, range expansion
Knowledge of population connectivity—the movement of
individuals between subpopulations—is critical to understanding population dynamics and the effective design of
management strategies (Kinlan and Gaines 2003; Palumbi
2003; Cowen and Sponaugle, 2009). A majority of marine
fishes and invertebrates are characterized by a bipartite life
cycle that includes a pelagic larval stage that may last several months, possibly allowing larvae to travel hundreds or
thousands of kilometers (Thorson 1950; Strathman 1987).
Assuming successful recruitment to adult populations, the
spatial scale of larval dispersal defines the scale of contemporary population connectivity (Kinlan and Gaines 2003;
Palumbi 2004).
Many of the problems encountered in managing natural
resources arise because of a mismatch between the scale of
management and the scale(s) of the ecological processes
being managed (Prince et al. 1998; Shanks et al. 2003;
Hilborn et al. 2005; Cumming et al. 2006; Prince 2010;
White and Costello 2011). For example, abalone in Australia
exhibit regional variation in growth, reproduction, and
the size at which sexual maturity is attained (Prince 2005).
If minimum size limits for the fishery are the same across
the entire region, then quickly growing abalone may enter
the fishery before first reproduction (Prince 2005, 2010).
Effective management strategies, including stock delineation
and marine reserve placement, require knowledge of the
spatial scale at which target species disperse (Sale et al. 2005;
Costello et al. 2010). By modeling the effects of networks of
marine protected areas, Costello et al. (2010) demonstrated
that spatial information, specifically knowledge of dispersal
scale, greatly improves management efficiency and increases
the economic value of a fishery. However, because of the
difficulty of tracking small larvae, larval dispersal scales are
often unknown and are rarely integrated into management.
1
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
From the Hopkins Marine Station, Stanford University, 100 Oceanview Blvd, Paciic Grove, CA 93950 (Haupt, Micheli, and
Palumbi); and the Northwest Fisheries Science Center, 2725 Montlake Blvd East, Seattle, WA 98112 (Haupt).
Journal of Heredity
2
equilibrium, and if populations are not yet at equilibrium,
contemporary genetic patterns may be more reflective of
evolutionary processes such as a range expansion rather
than contemporary dispersal patterns (Wright 1978). Work
by Slatkin (1993) on patterns of IBD indicates that the
amount of time for a population to reach equilibrium may
be a strong function of effective population size, which in
many marine populations may be much lower than local census size (Roman and Palumbi 2003). In marine populations
where effective management is contingent on assessment of
larval dispersal, it is critical to elucidate whether observed
genetic patterns reflect contemporary connectivity or historical population changes (Bird et al. 2007).
In this study, we seek to discover if and how past demographic processes influence contemporary genetic estimates
of larval dispersal and connectivity. We combine analyses
of empirical genetic data and coalescent-based simulated
data sets to explore the contribution of past demographic
processes and contemporary dispersal in influencing the
genetic structure of a harvested marine benthic invertebrate, the wavy turban snail Megastraea undosa, which ranges
from southern California, United States to southern Baja
California, Mexico. We use these data to investigate regional
structure and assess population connectivity across the range
of M. undosa. Our empirical results reveal two regimes of
genetic structure, which could be driven by either differences
in dispersal or historical processes. Coalescent-based simulations support the hypothesis that these patterns arise from a
historical range expansion and that dispersal may be limited
throughout the range despite differences in genetic structure.
Methods
Study Species and Management
Megastraea undosa is a large (5–10 cm), herbivorous gastropod
ranging from Point Conception, California, United States
(34.5 °N) to Abreojos, Baja California Sur, Mexico (26.7 °N;
M. undosa is present south of this area to Bahia Magdalena,
25.5 °N, at much lower densities) in rocky intertidal and subtidal
habitats (Taniguchi and Rogers-Bennett 2001). This snail is a
broadcast spawner that releases gametes into the water column
prior to fertilization. Megastraea undosa lives to be 12 years and
reproduction begins at age 4 and high levels of fecundity are
reached beginning at 8 years (Martone and Micheli 2012).
Reproduction can occur throughout the year but peaks in
October (Belmar-Pérez 1991). Larvae are competent to settle
after 5–10 days (Guzmán del Próo et al. 2003).
Megastraea undosa is harvested in California, United States,
and Baja California, Mexico, by divers. In California, catch
is largely unregulated: the only requirement is the need for
a valid fishing license and—other than no-take marine protected areas (MPAs) that afford protection to all species—
there is no spatial management for this species (Taniguchi
and Rogers-Bennett 2001). The only sampled site that falls
within a no-take MPA is Anacapa Island Landings Cove
(Site 1). The major concentration of the fishery is in Baja
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
Without the ability to physically track larvae, many researchers
have turned to population genetic techniques for estimation
of dispersal (Hellberg et al. 2002; Selkoe and Toonen 2006;
Hedgecock et al. 2007).
Ecological, physical, historical, and anthropogenic factors act in concert to drive contemporary patterns of dispersal for marine species (Grantham et al. 2003; Cowen
and Sponaugle 2009). When dispersal distances are nearly
equal to the geographic range of the species, this can lead to
panmixia, an openly mixing population. In contrast, when
dispersal occurs on smaller geographic scales, the potential for population subdivision exists. The combination of
a long pelagic larval phase and ocean currents may aid in
long distance dispersal and more open populations, reflected
by a lack of genetic structure throughout the range (Largier
2003; Siegel et al. 2003; Gawarkiewicz et al. 2007; Addison
et al. 2008). Conversely, larval behavior and oceanographic
features may act to limit dispersal (Cowen 2000; Kingsford
et al. 2002; Largier 2003). These contemporary processes
drive dispersal and successful delivery of larvae from one
population to another as well as influence genetic structure
of a species. Population genetic studies can give us insight
into whether marine populations are connected by larval
transport or if there are barriers to dispersal and therefore
successful recruitment is more local (Hedgecock et al. 2007;
Hellberg 2007; Weersing and Toonen 2009; Dawson et al.
2010; Pinsky et al. 2010). Particularly, studies that use patterns of genetic isolation by distance (IBD) to estimate dispersal kernels may provide further insight into connectivity
of separated populations (Kinlan and Gaines 2003; Palumbi
2003; Pinsky et al. 2010).
Although genetic data can be used to infer potential dispersal distances of larvae, they also reflect evolutionary processes, which can make estimates difficult and possibly lead
to incorrect inferences of contemporary dispersal (Avise
et al. 1987; Hewitt 1996; Benzie 1999; Hewitt 2000; Jacobs
et al. 2004, Marko and Hart 2011). Population expansions,
contractions, and bottlenecks may leave indelible clues within
genetic patterns (Slatkin and Hudson 1991; Harpending et al.
1998). Environmental and anthropogenic drivers (e.g., climate
change, human fishing pressure, and habitat destruction)
can also change species distributions and population sizes
and consequently, alter genetic structure (Selkoe et al. 2010;
Puritz and Toonen 2011). Genetic data are often employed to
examine these evolutionary patterns in species demographics (Benzie 1999; Hewitt 2000; Jacobs et al. 2004; Hart and
Marko 2010; Marko and Hart 2011), which can cause a population to be out of equilibrium and affect estimates of gene
flow (Rousset 1997). Range expansions can lead to disparate
patterns of genetic structure in the historical range and in a
newly expanded range (Hellberg et al. 2001). In the historical
portion of the range, the population may be closer to equilibrium and have had time to diversify, whereas in the more
recently colonized area, the genetic diversity may be low due
to founder effects and a population further from equilibrium
(Ibrahim et al. 1996).
Using population genetics to estimate gene flow and
infer rates of dispersal assumes that the population is at
Haupt et al. • Historical Processes and Dispersal
California Sur, Mexico, from Punta Eugenia 27.8 °N to La
Bocana 26.8 °N, and fishermen in this region are organized
into 9 fishing cooperatives, where management of invertebrate species occurs cooperatively with government agencies
that regulate Mexican fisheries (McCay et al. forthcoming).
Unlike California—where M. undosa is managed as 1 region—
Baja California Sur, Mexico is managed at the local scale of
the fishing cooperative. These cooperatives are granted concessions, or exclusive access, for certain invertebrate fisheries,
including M. undosa, and have the authority to make decisions
about fishing season, quotas, and size limits as long as these
are within the limits set by the federal government (McCay
et al. forthcoming).
Total genomic DNA was extracted with NucleoSpin
DNA extraction kit (BD Biosciences, San Jose, CA) per
the manufacturer’s protocol. A 605-base-pair segment of
the mitochondrial cytochrome oxidase subunit-I (COI)
gene was amplified and sequenced using primers HCO2198 and LCO-1490 (Folmer et al. 1994) following standard PCR and sequencing protocol from Folmer et al. (1994)
using EconoTaq. PCR products were sequenced on an ABI
3100 sequencer (Applied Biosystems, Inc., Foster City, CA).
Individuals were sequenced in forward and reverse directions
and visually scored for accuracy and polymorphisms using
Sequencher version 4.8 (Gene Codes Corporation).
Population Structure
Foot tissue samples (~1 cm3) of 27–62 (average = 51; standard deviation [SD] = 9.6) individuals were collected nonlethally via SCUBA from 17 sites across the range of M. undosa
from the years 2005–2008 (Figure 1; Table 1). Tissue samples
were preserved in 95% ethanol. This study was specifically
designed to sample across the range of M. undosa rather than
to examine 1 particular area of interest.
Aligned sequences were exported from Sequencher into
Arlequin 3.5 (Excoffier et al. 2005) to calculate Φ statistics: a measure of the amount of variance among samples
that can be attributed to differences among populations.
Using Arlequin, we ran an analysis of molecular variance
(AMOVA) to calculate ΦST, ΦSC, and ΦCT values as well as
pairwise ΦST values between all sites. All AMOVAs were performed using pairwise distance with 10 000 permutations in
Figure 1. Distribution of COI mtDNA haplotypes along the range of Megastraea undosa. (a) Each pie chart represents
distribution of haplotypes at each site. Each color represents a unique haplotype and numbers above pie charts correspond to site
numbers in Table 1. (b) Red dots represent sampling locations.
3
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
Sample Collection and Sequencing
Journal of Heredity
Table 1 Sampling localities of Megastraea undosa and corresponding site numbers
Site name
Region
Latitude
Longitude
N
H
Hu
h
He
π
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Anacapa Landings Cove
Santa Cruz Island Yellowbanks
Catalina Island Salta Verde
Catalina Island Eagle Rock
Palos Verdes
Santa Barbara Island Graveyard
Ensenada
Isla San Martin
Punta Baja
Isla San Geronimo
Esmerelda
Punta Eugenia Sur
Punta Eugenia Norte
Bahia Tortugas Punta Quebrada
Bahia Tortugas Punta Clam Bay
Bahia Asuncion
Punta Abreojos
North
North
North
North
North
North
North
South
South
South
South
South
South
South
South
South
South
34.014
33.991
33.32
33.473
33.711
33.474
31.723
30.485
29.929
29.784
28.517
27.846
27.839
27.722
27.618
27.148
26.7
−119.418
−119.565
−118.453
−118.604
−118.318
−119.029
−116.718
−116.097
−115.753
−115.79
−114.072
−115.084
−114.931
−114.99
−114.842
−114.329
−113.644
27
42
60
54
52
50
40
48
55
48
57
55
48
62
61
66
44
4
4
7
6
5
5
7
8
14
16
18
19
22
22
17
13
13
0
0
1
1
0
1
2
1
2
2
3
10
10
10
11
2
4
0.4900
0.2230
0.3571
0.4416
0.3122
0.2261
0.4000
0.5824
0.8074
0.8777
0.8546
0.8387
0.8108
0.7546
0.7617
0.6770
0.6723
2
1
2
2
1
1
2
2
5
8
7
6
5
4
4
3
3
0.00173
0.00090
0.00121
0.00171
0.00105
0.00089
0.00146
0.00209
0.00300
0.00321
0.00303
0.00287
0.00304
0.00237
0.00273
0.00213
0.00195
Region refers to if the site was assigned to the northern or southern region; N is the number of samples per site; H is the number of haplotypes found; Hu
is the number of unique haplotypes; h is haplotype diversity; He = 1/(1 − h) is the number of effective haplotypes; and π is nucleotide diversity for the COI
mtDNA locus.
Arlequin. As post hoc test, we manually performed hierarchical groupings of sites to determine the geographic areas that
contained the largest ΦCT differences. The largest ΦCT was
found by dividing the populations into 2 groups rather than
3 or 4 groups. There is no clear genetic break in the data,
rather a cline in central Baja Mexico. Because there is no clear
genetic break between northern and southern populations, it
is difficult to draw a clear line between north and south for
further analyses. However, the common southern haplotype
disappears between Ensenada (Site 7) and Isla San Martin
(Site 8) and Ensenada and from here on, these groups of
sites will be referred to as north and south (Table 1). We also
looked within these groupings to see how genetic structure
differed throughout the range of M. undosa by performing
AMOVAs separately to obtain global ΦST values within the
northern region and within the southern region separately to
get an estimate of ΦSC for the 2 regions. These ΦST values
allowed for assessment of potential relative levels of genetic
connectivity within the northern and southern regions. We
were specifically interested to see if genetic structure was
evenly distributed along the range: if levels of ΦST for the
north were similar to levels of ΦST in the south. We were
interested to see if the different management schemes in the
2 different regions matched the patterns of larval dispersal
inferred through ΦST.
haplotypes, and effective number of haplotypes (Jost 2007;
Timmers et al. 2012) calculated in Arlequin, throughout the
range. We used mismatch distributions to look for a possible
pattern of expansion in the northern and southern populations. Population expansion and contraction influence patterns of genetic sequence divergence and frequency and
so mismatch distributions can be used to infer patterns of
past demography (Slatkin and Hudson 1991; Rogers and
Harpending 1992). Populations that have undergone recent
expansion will have a higher frequency of 0 mismatches.
This is because the population is likely to be dominated by a
few common haplotypes present in the founding individuals
(Harpending 1994). The statistic Harpending’s Raggedness
Index (HRI), which evaluates the raggedness of the mismatch distribution, was calculated using Arlequin. The resulting mismatch distributions for the northern and southern
regions were compared with a chi-square test to see if the
distributions were significantly different from each other.
Unique haplotypes were identified using DNAsp and
then imported into TCS version 1.21 (Clement et al. 2000)
to create haplotype networks. TCS calculates the number
of differences among haplotypes to create a parsimonious
connection and then joins haplotypes into networks (Clement
et al. 2000). The proportion of individuals from the northern
or southern group was determined for each haplotype.
Historical Demography
Estimating Dispersal: IBD
To explore the historical demography of this species, we
employed 3 techniques: first, we looked at diversity measures;
second, at mismatch distributions of haplotypes; and finally,
we examined the relationships among haplotypes for signals of
past demographic processes in contemporary genetic patterns.
We looked for patterns of genetic diversity by comparing nucleotide diversity (π) (Nei and Jin 1989), number of
We looked for a pattern of IBD by plotting pairwise genetic
distance (pairwise FST/(1 − FST) and pairwise Euclidian geographic distance. For the IBD curve, we used only comparisons from the southern portion of the range. Conventional
FST as calculated by Arlequin 3.5 is based solely on frequencies of haplotypes unlike ΦST, which considers relatedness (number of mutations) between haplotypes as well
4
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
Site no
Haupt et al. • Historical Processes and Dispersal
σ=
1
,
4Dem
where σ is the spread, or width, of the dispersal kernel,
m is the slope of IBD plot, and De is effective density.
The significance of the relationship was determined by
performing a Mantel test in Arlequin. Rousset (1997) uses a
one-dimensional formula, where the length of the habitat is
greater than the width; the coastline along the eastern Pacific
basin generally fits this assumption. However, the habitat
is not continuous and this patchy nature could influence
estimates of dispersal (Pinsky et al. 2012).
To obtain estimates of effective density (De), we integrated
census density estimates with data on reproductive maturity
of M. undosa. We used data from a previous study that surveyed densities of M. undosa in 2006 at 15 sites from Bahia
Tortugas to Punta Abreojos, Baja California Sur, Mexico
(Martone et al. unpublished data). Surveyed M. undosa were
broken into 10-mm size bins and these data were integrated
with known percentage of reproductively mature females at
each size bin (Martone and Micheli 2012). Densities and sizes
were similar among sites in Baja; so for ease of calculation, we
chose to use size frequency data from one site (Punta Clam
Bay, Bahia Tortugas 27.6 °N) to estimate density of reproductive individuals. With this method, we were able to translate census density of M. undosa to an estimate of density of
reproductive individuals. This was used as an estimate of De
over the entire range but could have been an over estimation
of De because we cannot estimate the number of individuals who are actually contributing to the next generation and
this number is likely smaller than the density of reproductive
individuals. To evaluate how sensitive our measure of dispersal distance is to De, we recalculated σ given values of De up
to 4 orders of magnitude lower.
Coalescent Simulations
We used the coalescent-based program SIMCOAL 2.0
(Laval and Excoffier 2004) to determine expected patterns
of genetic diversity with different dispersal and evolutionary
histories. SIMCOAL 2.0 uses Kingman’s backward-based
coalescent approach to construct gene genealogies based
on demographic patterns such as migration and population
growth and simulates DNA sequences forward from the
most recent common ancestor (Kingman 1982a, 1982b). The
purpose of these simulations was to explore what patterns of
genetic diversity might be produced under different scenarios
of dispersal and evolutionary history of expansion. We
examined combinations of 2 demographic history scenarios—
northward range expansion or no range expansion—and 2
dispersal scenarios—low (10 km) and high (100 km) dispersal
distances. These simulated data were then analyzed using
Arlequin to assess genetic diversity in each simulation output
and compared results with our empirical data.
We simulated 30 populations along a linear coastline each
separated by 30 km with effective population sizes of 10 000
individuals. Our empirical sampling design included 17 sites
separated by an average of 65 km (range: 8–240 km), and our
estimate of effective population size using our empirical estimate of De is of the same order of magnitude of the 10 000
individuals used for the simulation. To mimic our sampling
design as closely as possible, the first 12 sites were part of
the northern population and the remaining 18 were part of
the southern population. We ran 1000 replicate runs of each
scenario. Fifty individuals were sampled at each of these sites.
Migration matrices were calculated using a Gaussian dispersal kernel in R (R Development Core Team 2008) centered
around 10 and 100 km for the low and high dispersal simulations, respectively. We simulated 605 base pairs of DNA
sequence at a single locus and mutation rate for all simulations scenarios was set at 10–7 per base pair (Hugall et al.
2002; Kelly et al. 2010). Simulations with a northward range
expansion were simulated with a historical range contraction
1000 generations ago when the first 12 populations (simulated northern range) were reduced to 0 and the founder
population for the expansion was 100 individuals. Megastraea
undosa reaches reproductive maturity at 4 years, reaches high
levels of fecundity at 8 years, and lives for approximately
12 years (Martone and Micheli 2012). The generation time
of M. undosa is likely about 8 years and so an expansion 1000
generations about would correspond to roughly 8000 years
ago and fall sometime after the last glacial maximum when
conditions in the Southern California Bight were similar to
contemporary conditions—rather than colder conditions
during Pleistocene glacial cycles—and may have allowed for
the species to expand northward (Kennett and Ingram 1995;
Kennett and Venz 1995; Graham et al. 2003). Output files
were run in Arlequin as batch files to assess differences in
5
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
as differences in frequency (Excoffier et al. 2005). When
pairwise conventional F statistics were calculated, we saw a
strange pattern for pairwise comparisons of northern sites
with other northern sites where these values were much
higher than expected (Supplementary Figures 1 and 2). For
example, the FST between Sites 1 and 4 (both dominated by
the same haplotype) is 0.670, whereas the FST between Sites
1 and 15 (which have entirely different suites of haplotypes)
is 0.353. By contrast, the pairwise ΦST values are consistent
with the observed pattern of differences in haplotype frequencies among populations. Pairwise FST and ΦST values are
generally highly correlated, but this relationship breaks down
in these data for pairwise comparisons of northern sites with
other northern sites (Supplementary Figure 2). It is possible
that the low diversity in the northern portion of the range
combined with these sites being dominated by 1 main haplotype creates FST values that are higher than expected for
comparisons of northern populations. These FST values are
correct estimates of differentiation based solely on frequency
but may be potentially misleading. Because of this, northern
sites were excluded from the IBD plot. Pairwise geographic
distance was calculated by importing latitude and longitude
coordinates into a geographic distance matrix generator. This
pattern should represent a balance between drift and migration and allow us to estimate dispersal distance using the following equation (Rousset 1997):
Journal of Heredity
genetic diversity between northern and southern sites and
were then compared with the diversity measurements of the
actual data.
Results
Population Structure
Historical Demography
We found that nucleotide diversity in the north (π = 0.00125,
SD = 0.001019) was less than half of that in the south
Table 2 AMOVA results for Megastraea undosa at the COI mtDNA locus
Global
Among populations
Within populations
North only
Among populations
Within populations
South only
Among populations
Within populations
California vs. Mexico
Among groups
Among populations within groups
Within populations
North (7: Ensenada) vs. South (8: Isla San Martin)
Among groups
Among populations within groups
Within populations
North (8: Isla San Martin) vs. South (9: Punta Baja)
Among groups
Among populations within groups
Within populations
North (9: Punta Baja) vs. South (10: Isla San Geronimo)
Among groups
Among populations within groups
Within populations
North (10: Isla San Geronimo vs. South (11: Esmerelda)
Among groups
Among populations within groups
Within populations
Variance
% Total
0.14755
0.64203
18.69
81.31
−0.00158
0.37703
−0.42
100.42
0.03651
0.80038
4.36
95.64
ΦST = 0.04363a
<0.0001
0.21420
0.04674
0.64203
23.72
5.18
71.1
ΦSC = 0.06787
ΦST = 0.28898
ΦCT = 0.23721
<0.0001
<0.0001
0.00020
0.25036
0.02247
0.64203
27.37
2.46
70.18
ΦSC = 0.03350
ΦST = 0.29521
ΦCT = 0.27078
<0.0001
<0.0001
0.0001
0.26177
0.01082
0.64203
28.62
1.18
70.2
ΦSC = 0.01570
ΦST = 0.29674
ΦCT = 0.28553
<0.0001
0.0003
0.0003
0.23750
0.02115
0.64203
26.37
2.35
71.28
ΦSC = 0.03217
ΦST = 0.28790
ΦCT = 0.26423
<0.0001
<0.0001
<0.0001
0.21470
0.03446
0.64203
24.09
3.87
72.04
ΦSC = 0.05094
ΦST = 0.27958
ΦCT = 0.24091
<0.0001
<0.0001
0.0002
Φ Statistics
P
ΦST = 0.18687
<0.0001
ΦST = −0.00419a
0.6835
Global ΦST refers to the entire range of the species. Northern ΦST refers to north of and including Ensenada, Baja California, Mexico, to the northern edge
of the range. Southern ΦST refers to sites south of and including Isla San Martin, Baja California, Mexico, to the southern edge of the range. Hierarchical
AMOVA results are also reported.
a This is an approximation of Φ .
SC
6
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
We detected highly significant genetic structure throughout
the range (ΦST = 0.187, P < 0.0001; Table 2). Populations in
the north were dominated by 1 haplotype that declined but
was still present in the southern portion of the range at a
lower frequency (Figure 1). In contrast, the southern populations are dominated by a different haplotype that is nearly
exclusive to this region: only 1 individual at Catalina Island
(Site 6) possessed this haplotype (Figure 1).
Hierarchical AMOVAs among groups of sites revealed
that the largest difference was between Isla San Martin
(Site 8) and Punta Baja (Site 9), (ΦCT = 0.28620, P < 0.0001).
Table 2 shows the results of different hierarchical analyses
with different north–south groupings of samples. There is
no clear genetic break but rather a gradient present in north
central Baja. Because the dominant southern haplotype disappears between Ensenada (Site 7) and Isla San Martin (Site 8;
Figure 1), for all other analyses, sites from Ensenada to the
northern range limit are considered northern and sites from
Isla San Martin to the southern limit are southern (Table 1).
We found different patterns of genetic connectivity
among populations in the northern and southern regions
(Table 2). Among the northern sites, no genetic structure
was detected (ΦST = −0.004, P = 0.683), whereas in the
southern sites there was significant structure (ΦST = 0.044,
P < 0.001). Among the northern populations, there are no
significant pairwise ΦST population comparisons, but among
the southern populations, there are multiple significant
comparisons (Table 3). The southern region encompasses a
slightly larger area (northern = 359 km of coastline; southern = 483 km of coastline), and so it is possible that the
differences we see in structure are due to the smaller northern area. However, we can reject this hypothesis because
if we isolate a smaller area in the center of the southern
region, significant genetic structure is present (Ensenada–
Esmerelda, 294 km; ΦST = 0.058, P < 0.001), and if we
consider the island coastlines of the Channel Islands, the
northern area may be larger.
0.61261
<0.00001 0.24324
0.00901 <0.00001 0.42342
<0.00001 0.31532 0.86486
<0.00001 <0.00001 0.57658
0.00901 0.0991 0.2973
<0.00001 <0.00001 0.45045
<0.00001 0.00901 0.18018
0.03604
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
0.61261
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
0.04505
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
0.02257
0.08558
0.05479
0.02085
0.07656
0.0916
0.04227
−0.01152
−0.01378
−0.01537
−0.01174
−0.01611
−0.00771
0.01515
−0.01669
−0.00102
0.02097
−0.00508
0.63964
0.1982
0.93694
0.97297
0.76577
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
0.00377 −0.01076 0.00402
−0.00802 0.00917 −0.01611
−0.0067 −0.01086
0.57658
−0.00002
0.78378 0.36937
0.41441 0.06306 0.55856
0.91892 0.7027
0.82883
0.02703 0.09009 <0.00001
<0.00001 <0.00001 <0.00001
<0.00001 <0.00001 <0.00001
<0.00001 <0.00001 <.00001
<0.00001 <0.00001 <0.00001
<0.00001 <0.00001 <0.00001
<0.00001 <0.00001 <0.00001
<0.00001 <0.00001 <0.00001
<0.00001 <0.00001 <0.00001
<0.00001 <0.00001 <0.00001
0.00773
0.2973
0.30631
0.55856
0.27928
0.12613
0.63964
0.17117
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
<0.00001
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Signiicant comparisons (α = 0.05) are boldfaced and these comparisons that are signiicant after Bonferroni corrections (corrected α = 0.0003 for 153 comparisons) are underlined. P values are listed below the
diagonal.
0.30006 0.2529
0.2919
0.30057 0.31098 0.30933
0.41242 0.36097 0.39906 0.40926 0.4213
0.43348
0.38345 0.33172 0.36579 0.37966 0.38255 0.38666
0.30757 0.25933 0.29181 0.30694 0.30442 0.29896
0.39937 0.35006 0.38284 0.39806 0.40202 0.40817
0.4324
0.37865 0.4175
0.42537 0.43964 0.45412
0.34676 0.2983
0.33471 0.34652 0.35252 0.3531
0.20734 0.15438 0.19025 0.19797 0.2015
0.20206
0.06646 0.03283 0.05802 0.05656 0.05986 0.06357
0.0418
0.00237 0.03529 0.0272
0.03498 0.04782
−0.00058 −0.00847 −0.00217 0.00151 −0.00267 0.00691
0.00647 −0.00385 0.00315 −0.00574 0.00382
0.17117
0.00595 −0.00543 0.00655 0.02246
0.6036
0.20721
0.0121 −0.01071 −0.00589
0.21622 0.62162 0.12613
0.01111 0.03408
0.71171 0.18919 0.99099 0.14414
−0.00549
0.28829 0.00901 0.76577 0.03604 0.71171
0.23751
0.34062
0.3111
0.24021
0.32796
0.35973
0.27936
0.14491
0.02993
0.00551
0.11481
0.20625
0.17657
0.11826
0.19588
0.22005
0.1517
0.03396
0.15942
0.26462
0.23318
0.16494
0.25373
0.28119
0.20303
0.07
−0.00503
12
11
10
9
8
7
6
5
4
3
2
1
Table 3 Above the diagonal pairwise ΦST comparisons are listed for Megastraea undosa at the COI mtDNA locus
(π = 0.002755, SD = 0.001793; t-test: P < 0.0001), with 13
and 82 haplotypes, respectively. Mismatch distributions of
both northern and southern regions indicate a pattern of
expansion measured by HRI (north: HRI = 0.3996, P = 0.41
and south: 0.0376, P = 0.41; Harpending, 1994; Figure 2).
However, the large proportion of 0 mismatches in the northern region indicates that the northern expansion may be
more recent and pronounced. These two distributions were
significantly different (χ2 = 41851.638, degrees of freedom
[df] = 6, P < 0.0001).
The haplotype network shows that the dominant southern haplotype is the root of the diversity for this species
(Figure 3). All of the diversity in this system may radiate
off of this 1 southern haplotype, whereas the most common northern haplotype is just one of many radiating off
of the dominant southern haplotype. This star phylogeny—
with one very common haplotype surrounded by many less
common variants—also indicates an expansion (Slatkin and
Hudson 1991; Avise 2000).
Estimating Dispersal
We found a significant relationship between geographic and
genetic distance (Mantel test: correlation coefficient = 0.5708,
P = 0.005; Figure 4) using only pairwise comparisons from
the southern portion of the range. Our estimate of effective
density, which incorporated data on reproductive maturity of
M. undosa (Martone and Micheli 2012), was 0.76 individuals/
m2. Using this De estimate and the slope of the IBD relationship (0.0005), the estimated dispersal kernel is 25 m. With a
De of 1, 2, 3, and 4 orders of magnitude lower the estimated
dispersal kernel is 79, 250, 793, and 2508 m, respectively.
Coalescent Simulations
Genetic diversity estimates of northern versus southern populations were consistent with our empirical results only for
the scenario that included a northward range expansion with
limited dispersal (10 km; Table 4). Scenarios that included
long distance dispersal and/or lack of a range expansion
resulted in no difference in genetic diversity (measured by π)
between northern and southern regions.
Discussion
We found high levels of global genetic structure throughout the contemporary biogeographic range of M. undosa.
However, when the range is parsed into northern and southern regions, there are conflicting patterns of genetic structure: no detectable structure in mtDNA ΦST was found in
the north and significant structure in the south. These differences translate into different levels of inferred genetic connectivity via larval dispersal: high connectivity in the north
and low in the south. This result might indicate that contemporary management regulations set at a fine spatial scale in
the south but not in the north appropriately reflect different
patterns of population connectivity within these 2 regions.
However, an examination of evolutionary processes, such
7
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
13
14
15
16
17
Haupt et al. • Historical Processes and Dispersal
Journal of Heredity
Figure 3. COI mtDNA haplotype network for Megastraea undosa. Each circle represents a unique haplotype. Size of circle is
proportional to number of individuals that possess the haplotype. White represents the proportion of individuals from the north
and black from the south. Red circles represent necessary missing—or unsampled—haplotypes to connect existing haplotypes.
8
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
Figure 2. Number of COI mtDNA sequence mismatches for Megastraea undosa is shown on the x axis and frequency on the
y axis. The northern population is represented in white and the southern in black.
Haupt et al. • Historical Processes and Dispersal
Table 4 Ratio of northern or southern diversity for simulated
scenarios and the empirical data for Megastraea undosa COI mtDNA
History
Dispersal
Expansion
Expansion
No Expansion
No Expansion
Actual data
100
10
100
10
Northern/southern
diversity
0.9253
0.4438
0.9247
1.0068
0.4250
The only scenario that matches the empirical ratio is with a northward range
expansion and limited dispersal.
as range expansions and bottlenecks, and potential demographic history indicates that this snail may have undergone
a northward range expansion and/or a northern population
expansion after the end of the last glacial maximum into the
Southern California Bight—the northern region—and this
process yielded lower genetic diversity in the north. A selective sweep could have also caused lower genetic diversity;
although, we are unable to rule out this possibility, it is
unlikely that only the north would have been affected by a
selective sweep as there is no obvious selective pressure that
is unique to the Southern California Bight.
Larval transport of M. undosa is likely to be influenced
by oceanography on a number of spatial scales throughout
the range. At a regional scale, the range is dominated by the
offshore-equatorward flowing California Current, but at
local scales, each portion of the range is subject to a different set of oceanographic conditions including coastal eddies,
upwelling centers, oceanographic fronts, and poleward
flowing water (Simpson and Lynn 1990; Hickey 1998;
Zaytsev et al. 2003; Woodson et al. 2012). Oceanographic
patterns in the Southern California Bight have been well
studied, and recent research using circulation simulations
to estimate potential and realized dispersal for fishes and an
invertebrate species found that larval connectivity was heterogeneous in the Southern California Bight (Watson et al.
2010). In the fall months when M. undosa spawns, higher flow
was seen from southern sites to northern sites, which likely
impacts connectivity between the Southern California Bight
and the southern portion of the range. Oceanographic patterns associated with the southern portion of the range in
Baja California are not as well understood or studied, but it
is likely that the major headland of Punta Eugenia influences
these patterns (Bernardi and Talley 2000; Zaytsev et al. 2003;
Haupt et al. in review).
Although differences in oceanography likely influence
local dispersal patterns of M. undosa, they are unlikely to fully
explain the observed differences in genetic differentiation
and diversity throughout the range. Coalescent-based computer simulations indicate that this pattern could have been
created by a range expansion and maintained by limited dispersal. Taken together, these results show that evolutionary
processes influence patterns of contemporary genetic structure and need to be taken into account when making recommendations for management and conservation.
Contemporary Patterns of Genetic Connectivity
Using the pattern of IBD, we estimated short dispersal distances ranging from 30 m to ~3 km, depending on estimates
9
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
Figure 4. IBD plot for Megastraea undosa at the COI mtDNA locus. Pairwise Euclidian geographic distance graphed against
pairwise genetic difference (FST/(1 − FST); Mantel test: correlation coeficient = 0.5708, P = 0.005).
Journal of Heredity
Historical Demography
Physiological tolerance to physical factors such as temperature can play an important role in structuring species ranges
in response to climatic changes and may be driving range
contractions and expansions (Roy et al. 1995). Marine flora
and fauna have adapted to climatic changes with range shifts
in both historical and contemporary times (Valentine and
Jablonski 1993; Barry et al. 1995; Roy et al. 1995; Zacherl
et al. 2003; Dawson et al. 2010). During the last glaciation in
the late Pleistocene, the water temperatures in the Southern
California Bight may have been much cooler, 7–8 °C, whereas
current temperatures range from 12 to 20 °C (Kennett and
Ingram 1995; Kennett and Venz 1995; Graham et al. 2003).
A study by Muhs et al. (2006) examined marine terrace fauna
of the eastern Pacific from fossil assemblages dating 120 000
and 80 000 years BP. They discovered that the species assemblages at 120 000 years BP were similar to contemporary
10
assemblages, indicating that paleotemperatures may have
been very similar to current temperatures. In contrast, fauna
from 80 000 years BP consisted of many species not currently found in this region; these species are currently found
in northern, colder areas, indicating that during this era, the
Southern California Bight was likely much colder (Muhs et al.
2006). For many southern species, including M. undosa, the
Southern California Bight may have been too cool during the
most recent glaciation to serve as usable habitat until after
recession of glaciers in the northern Pacific and subsequent
warming of these waters.
Repeated extirpations and recolonizations resulting in
founder effects could have kept genetic diversity in the system low. Megastraea undosa is found in the Southern California
Bight in paleontological studies from 120 000 years ago but
not found again until more contemporary studies of Native
American middens from 6 000 to 10 000 years ago in the
Channel Islands (Forgeng 1992; Erlandson et al. 1998; Roy K,
personal communication). During the Holocene, M. undosa
may have been able to expand into southern California from
Mexico; given its potential short dispersal distance combined
with the predominant southward currents, this expansion
might have occurred very slowly. In modeling kelp forest
populations throughout the late quaternary, Graham et al.
(2010) indicated large shifts in kelp abundance and productivity of kelp forests throughout time. This would likely
translate into major changes in kelp forest communities.
Although M. undosa is not dependent on the presence of kelp,
it is a kelp forest–associated species and is currently found
more often in kelp-forested areas than in deforested areas
and would likely be affected by large-scale changes in kelp
forests (Graham, 2004). Previous work by Selkoe et al. (2010)
found that kelp is an important predictor of genetic differentiation in the Southern California Bight and Baja California.
Kelp may also play an important role for M. undosa, though
kelp densities do not seem to correlate to changes in genetic
structure and although this species is a kelp forest–associated
species and its range extends past southern limit for the range
of Macrocystis pyrifera.
Effects of climatic changes due to glaciation are commonly found in genetic structure of species in both marine
and terrestrial environments (Hewitt 2000). As northern
areas cooled during glacial periods, many species contracted
into southern warmer areas and recolonized after climate
warming (Hewitt 2004). Recent work by Marko et al. (2010)
documented that although persistence through the last glaciation of marine species in the northeastern Pacific was
common, 5 of 14 species did show evidence of a postglaciation range expansion. Although coastal marine species may
have persisted in refugia in the northeastern Pacific (7 of 14
species), some species’ ranges were reduced to southern latitudes. Species, even ones closely related, may respond differently to climatic changes. Hickerson and Cunningham (2005)
found that with sister species of intertidal fishes, one species
persisted despite glaciation and the other experienced a range
contraction. These events are often observed in the northeastern Pacific where glaciers were present in coastal systems
(Rocha-Olivares and Vetter 1999; Marko 2004; Wilson 2006;
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
of effective density. Even with a dispersal distance of 1–3
km, obtained using the lowest estimates of effective density, we would still conclude that this snail has limited dispersal. Previous studies that examined larval dispersal of
marine gastropods have found both high and low levels
of genetic differentiation and inferred connectivity, but
gastropods also encompass a large range of pelagic larval
durations (PLD; Kelly and Palumbi, 2010). Recent work on
blacklip and black abalone, which have a comparable PLD
to M. undosa, show that these abalone species may have very
limited dispersal distance, on the order of 7–10 km for
blacklip abalone (Gruenthal and Burton 2008; Miller et al.
2008). By combining oceanographic connectivity matrices
with genetic connectivity data, White et al. (2010) determined that the whelk, Kelletia kelletii, which has a PLD of
40–60 days (Zacherl D, unpublished data), an order of magnitude longer than M. undosa, likely disperses on a scale of 30
km. Oceanographic modeling studies of two commercially
fished molluscs—a scallop and a gastropod—in the Gulf
of California, Mexico, predict that dispersal potential may
be much larger than this (100’s of km) and that larvae may
be exported out of marine reserves to fished areas, though
the authors point out that these models may overestimate
potential dispersal because larval behavior could limit distance traveled by larvae (Cudney-Bueno et al. 2009).
By considering the full range of M. undosa, rather than
focusing solely on northern populations out of context, we
see a clear pattern of IBD, which may indicate limited dispersal throughout the range. Separately, however, the north
and the south present different patterns: the north has no
structure and the south has significant structure. These differing patterns could lead us to infer different levels of connectivity via larval dispersal in these areas. Coalescent-based
simulations indicate that to maintain the pattern of reduced
genetic diversity in the north for 1000 generations after the
northward expansion that dispersal in this snail must be limited. Potentially, the northward expansion and the historical
processes drive the patterns in the north, rather than a difference in realized larval dispersal.
Haupt et al. • Historical Processes and Dispersal
Marko et al. 2010). With M. undosa, we see a possible signature of glacial events even though its range does not overlap
with glaciated areas.
Applications to Management
Supplementary Material
Supplementary material can be found at http://www.jhered.
oxfordjournals.org/.
Funding
National Science Foundation Graduate Research Fellowship
awarded to A.J.H.; NSF Biocomplexity in the Environment
Grant OCE-0410439. Partnership for the Interdisciplinary
Study of Coastal Oceans sponsored by the David and Lucile
Packard Foundation and the Gordon and Betty Moore
Foundation.
Acknowledgments
First and foremost the authors thank the fishing cooperatives of FEDECOOP for their logistical and field support in
Baja California. We thank R. Martone, R. Beas, L. Gonzales,
S. Guzmán del Próo, E. Serviere, J. Belmar, J. Carillo,
F. Lopez, J. Ramirez, C. White, S. Koch, S. Hamilton,
J. Wible, E. Hoagland, D. Zacherl, J. Lorda, and C. O’Connell
for assistance in the field. We thank A.O. Shelton, M. Pinksy,
and S.T. Nixon for computing help. Helpful discussion and
feedback were provided by M. Siegle, M. Pespeni, C. Wood,
D. Barshis, R. Martone, K. Ruegg, C. Tepolt, G. Somero,
M. Pinsky, J. Ladner, R. Kelly, K. Kroeker, and two anonymous reviewers.
References
Addison JA, Ort BS, Mesa KA, Pogson GH. 2008. Range-wide genetic
homogeneity in the California sea mussel (Mytilus californianus): a comparison
of allozymes, nuclear DNA markers, and mitochondrial DNA sequences.
Mol Ecol. 17:4222–4232.
Avise JC. 2000. Phylogeography: the history and formation of species.
Cambridge (MA): Harvard University Press.
Avise JC, Arnold J, Ball RM, Bermingham E, Lamb T, Neigel JE, Reeb CA,
Saunders NC. 1987. Intraspeciic phylogeography: the mitochondrial DNA
bridge between population genetics and systematics. Ann Rev Ecol and Syst.
18:489–522.
Barry JP, Baxter CH, Sagarin RD, Gilman SE. 1995. Climate-related, longterm faunal changes in a California rocky intertidal community. Science.
267:672–675.
Belmar-Pérez J. 1991. Madurez gonádica y ciclo reproductor del caracol
panocha (Astraea undosa Wood, 1828; Gasterópodo: Turbinidae) en Bahía
Tortugas, Baja California Sur. Anales del ICMyL UNAM. 18:169–187.
11
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
The fishery for M. undosa is the third most economically
important invertebrate fishery in Baja California, Mexico,
after lobster and abalone, and management occurs at the
local scale of fishing cooperatives (Singh-Cabanillas 1996;
Gluyas-Millán et al. 2002; McCay et al. forthcoming).
Given the limited dispersal of M. undosa, this small scale
of management is likely appropriate. Work by Martone
and Micheli (2012) suggests that ecological processes such
as growth rates and reproduction also vary across a small
spatial scale further supporting local rather than regional
management. In southern California, management for this
species is nearly nonexistent and occurs at the regional scale
(Taniguchi and Rogers-Bennett 2001). In this study, the lack
of genetic structure in California could support management at a large regional scale. However, when placed in the
context of historical demography and evolutionary process,
our results indicate that management likely should occur at
local scales in California as well. For this reason, managers
should be cautious when using data that are not placed in
context of an entire range or trying to generalize across
species (Bird et al. 2007). Mismatch distributions, genetic
diversity patterns, and simulations all support the hypothesis that a recent northward expansion into the Southern
California Bight drives the lack of genetic structure in the
north. Further, our simulations demonstrate that the most
likely dispersal and demographic scenario, of those considered, is a northward range expansion and limited dispersal. Making any assessment about larval dispersal with little
to no genetic structure is very difficult (Waples 1998). For
M. undosa, with the knowledge we gain by looking at the
entirety of the range we may be able to recommend that,
though we detect no structure in the northern region of the
range, management should occur at local scales in both the
southern and northern regions.
Newer tools and more quickly evolving markers may aid
in avoiding these issues of confusing contemporary and evolutionary patterns to assess connectivity via larval dispersal.
Assignment methods, particularly parentage analysis (Jones
and Ardren 2003), have shown great promise especially for
looking at larval spillover from within marine reserves to outside fished areas (Jones et al. 2005; Planes et al. 2009). The
advantage of these methods is that they give actual measurements of contemporary dispersal rather than a long-term
average. However, the limitations are that this is a one-time
snapshot and without repetition among years a generalization for dispersal patterns cannot be made. Also these types
of studies often only capture evidence of limited dispersal
because the sample area is small relative to the species range
and thus is not collecting potential long distance dispersers,
which may be of ecological and conservation significance.
Our results highlight some potential issues and solutions
for the use of genetic tools to infer contemporary patterns
of larval dispersal. Current genetic structure may reflect evolutionary history but consideration of the entire range of a
species as well as exploring historical processes may allow for
more accurate inferences of dispersal. In the case study of
M. undosa, we find that inclusion of the entire range as well
as consideration of historical demography indicates potentially limited dispersal in the northern region despite lack of
genetic structure.
Journal of Heredity
Benzie JH. 1999. Genetic structure of coral reef organisms: ghosts of dispersal past. Int Comp Biol. 39:131–145.
Harpending HC. 1994. Signature of ancient population growth in a low-resolution mitochondrial DNA mismatch distribution. Hum Biol. 66:591–600.
Bernardi G, Talley D. 2000. Genetic evidence for limited dispersal in
the coastal California killiish, Fundulus parvipinnis. J Exp Mar Biol Ecol.
255:187–199.
Harpending HC, Batzer MA, Gurven M, Jorde LB, Rogers AR, Sherry ST.
1998. Genetic traces of ancient demography. Proc Natl Acad Sci USA.
95:1961–1967.
Bird CE, Holland BS, Bowen BW, Toonen RJ. 2007. Contrasting phylogeography in three endemic Hawaiian limpets (Cellana spp.) with similar life
histories. Mol Ecol. 16:3173–3186.
Hart MW, Marko PB. 2010. It’s about time: divergence, demography, and the
evolution of developmental modes in marine invertebrates. Integr Comp
Biol. 50:643–661.
Clement M, Posada D, Crandall KA. 2000. TCS: a computer program to
estimate gene genealogies. Mol Ecol. 9:1657–1659.
Hedgecock D, Barber P, Edmands S. 2007. Genetic approaches to measuring
connectivity. Oceanography. 20:70–79.
Costello C, Rassweiler A, Siegel D, De Leo G, Micheli F, Rosenberg A. 2010.
The value of spatial information in MPA network design. Proc Natl Acad
Sci USA. 107:18294–18299.
Hellberg ME. 2007. Footprints on water: the genetic wake of dispersal
among reefs. Coral Reefs. 26:463–473.
Cowen RK. 2000. Connectivity of marine populations: open or closed?
Science. 287:857–859.
Cudney-Bueno R, Bourillón L, Sáenz-Arroyo A, Torre-Cosío J, Turk-Boyer
P, Shaw WW. 2009. Governance and effects of marine reserves in the Gulf
of California, Mexico. Ocean Coast Manage. 52:207–218.
Hellberg ME, Burton RS, Neigel JE, Palumbi SR. 2002. Genetic assessment
of connectivity among marine populations. Bull Mar Sci. 70:18.
Hewitt G. 1996. Some genetic consequences of ice ages, and their role in
divergence and speciation. Biol J Linn Soc. 58:247–276.
Hewitt G. 2000. The genetic legacy of the Quaternary ice ages. Nature.
405:907–913.
Cumming GS, Cumming DHM, Redman CL. 2006. Scale mismatches in
social-ecological systems: causes, consequences, and solutions. Ecol Soc. 11:14.
Hewitt G. 2004. Genetic consquences of of climatic oscillations in the
Quaternary. Phil Trans R Soc Lond B. 359:183–195.
Dawson MN, Grosberg RK, Stuart YE, Sanford E. 2010. Population genetic
analysis of a recent range expansion: mechanisms regulating the poleward range limit in the volcano barnacle Tetraclita rubescens. Mol Ecol.
19:1585–1605.
Hickerson MJ, Cunningham CW. 2005. Contrasting quaternary histories in an
ecologically divergent sister pair of low-dispersing intertidal ish (Xiphister)
revealed by multilocus DNA analysis. Evolution. 59:344–360.
Erlandson JR, Tveskov MA, Byram RS. 1998. The development of maritime
adaptations on the southern coast of North America. Arctic Anthropology.
35:6–22.
Excofier L, Laval G, Schneider S. 2005. Arlequin (version 3.0): an integrated
software package for population genetics data analysis. Evol Bioinform
Online. 1:47–50.
Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R. 1994. DNA primers for
ampliication of mitochondrial cytochrome c oxidase subunit I from diverse
metazoan invertebrates. Mol Marine Biol Biotechnol. 3:294–299.
Forgeng EE. 1992. Archeology, ecology, and site-formation processes at the
Daisy Cave midden, San Miguel Island, California [master’s thesis]. [Eugene
(OR)]: Department of Anthropology, University of Oregon.
Gawarkiewicz GG, Monismith S, Largier J. 2007. Observing larval transport processes affecting population connectivity: progress and challenges.
Oceanography. 20:40–53.
Gluyas-Millán M, Quinonez-Velazquez C, Talavera-Maya J. 2002. Effect of
El Nino 1997–1998 on the snail Astraea undosa (Wood) population along the
Baja California western coast. J Shellish Res. 21:831–834.
Hickey BM. 1998. Coastal oceanography of western North America from
the tip of Baja California to Vancouver Island. In: Robinson AR, Brink KH,
editors. The sea. New York: John Wiley and Sons. p. 345–393.
Hilborn R, Orensanz JM, Parma AM. 2005. Institutions, incentives and the
future of isheries. Philos Trans R Soc Lond, B, Biol Sci. 360:47–57.
Hugall A, Moritz C, Moussalli A, Stanisic J. 2002. Reconciling paleodistribution models and comparative phylogeography in the Wet Tropics rainforest land snail Gnarosophia bellendenkerensis (Brazier 1875). Proc Natl Acad Sci
USA. 99:6112–6117.
Ibrahim KM, Nichols RA, Hewitt GM. 1996. Spatial patterns of genetic
variation generated by different forms of dispersal during range expansion.
J Hered. 77:282–291.
Jacobs DK, Haney TA, Louie KD. 2004. Genes, diversity, and geologic process on the Paciic Coast. Annu Rev of Earth Pl Sci. 32:601–652.
Jones AG, Ardren WR. 2003. Methods of parentage analysis in natural populations. Mol Ecol. 12:2511–2523.
Jones GP, Planes S, Thorrold SR. 2005. Coral reef ish larvae settle close to
home. Curr Biol. 15:1314–1318.
Graham M. 2003. Ice ages and ecological transitions on temperate coasts.
Trends Ecol Evol. 18:33–40.
Jost L. 2007. Partitioning diversity into independent alpha and beta components. Ecology. 88:2427–2439.
Graham MH. 2004. Effects of local deforestation on the diversity and
structure of southern California giant kelp forest food webs. Ecosystems.
7:341–357.
Kelly RP, Oliver TA, Sivasundar A, Palumbi SR. 2010. A method for detecting population genetic structure in diverse, high gene-low species. J Hered.
101:423–436.
Graham MH, Kinlan BP, Grosberg RK. 2010. Post-glacial redistribution and
shifts in productivity of giant kelp forests. Proc Biol Sci. 277:399–406.
Kelly RP, Palumbi SR. 2010. Genetic structure among 50 species of the
northeastern Paciic rocky intertidal community. PLoS ONE. 5:e8594.
Grantham BA, Eckert GL, Shanks AL. 2003. Dispersal potential of marine
invertebrates in diverse habitats. Ecol Appl. 13:108–116.
Kennett JP, Ingram BL. 1995. A 20,000-year record of ocean circulation and
climate change from the Santa Barbara basin. Nature. 377:510–514.
Gruenthal KM, Burton RS. 2008. Genetic structure of natural populations
of the California black abalone (Haliotis cracherodii Leach, 1814), a candidate
for endangered species status. J Exp Mar Biol Ecol. 355:47–58.
Kingman J. 1982a. On the genealogy of large populations. J Appl Probab.
19:27–43.
Guzmán Del Próo S. 2003. Larval and early juvenile development of the
wavy turban snail, Megastraea undosa (Wood, 1828) (Gastropoda: Turbinidae).
Veliger. 46:4.
Kingsford MJ, Leis JM, Shanks A, Lindeman KC, Morgan SG, Pineda J.
2002. Sensory environments, larval abilities and local self-recruitment. B
Mar Sci. 70:309–340.
12
Kingman J. 1982b. The coalescent. Stoch Proc Appl. 13:235–248.
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
Cowen RK, Sponaugle S. 2009. Larval dispersal and marine population connectivity. Ann Rev Mar Sci. 1:443–466.
Hellberg ME, Balch DP, Roy K. 2001. Climate-driven range expansion and
morphological evolution in a marine gastropod. Science. 292:1707–1710.
Haupt et al. • Historical Processes and Dispersal
Kinlan BP, Gaines SD. 2003. Propagule dispersal in marine and terrestrial
environments: a community perspective. Ecology. 84:2007–2020.
Rogers AR, Harpending H. 1992. Population growth makes waves in the
distribution of pairwise genetic differences. Mol Biol Evol. 9:552–569.
Largier JL. 2003. Considerations in estimating larval dispersal distances from
oceanographic data. Ecol Appl. 13:71–89.
Roman J, Palumbi SR. 2003. Whales before whaling in the North Atlantic.
Science. 301:508–510.
Laval G, Excofier L. 2004. SIMCOAL 2.0: a program to simulate genomic
diversity over large recombining regions in a subdivided population with a
complex history. Bioinformatics. 20:2485–2487.
Rousset F. 1997. Genetic differentiation and estimation of gene low from
F-statistics under isolation by distance. Genetics. 145:1219–1228.
Marko PB. 2004. “What’s larvae got to do with it?” Disparate patterns of postglacial population structure in two benthic marine gastropods with identical
dispersal potential. Mol Ecol. 13:597–611.
Marko PB, Hart MW. 2011. The complex analytical landscape of gene low
inference. Trends Ecol Evol (Amst). 26:448–456.
Marko PB, Hoffman JM, Emme SA, Mcgovern TM, Keever CC, Cox LN.
2010. The “Expansion-Contraction” model of Pleistocene biogeography:
rocky shores suffer a sea change? Mol Ecol. 19:146–169.
McCay BJ, Micheli F, Ponce-Diáz G, Murray G, Shester G, Ramirez-Sanchez
S, Weisman W. Forthcoming. Community-based concessions on the Paciic
coast of Mexico. Marine Policy.
Miller KJ, Maynard BT, Mundy CN. 2008. Genetic diversity and gene low in
collapsed and healthy abalone isheries. Mol Ecol. 18:200–211.
Muhs DR, Simmons KR, Kennedy GL, Ludwig KR, Groves LT. 2006. A
cool Paciic Ocean at the close of the last interglacial complex. Quat Sci
Rev. 25:235–262.
Nei M, Jin L. 1989. Variances of the average numbers of nucleotide substitutions within and between populations. Mol Biol Evol. 6:290–300.
Palumbi SR. 2003. Population genetics, demographic connectivity, and the
design of marine reserves. Ecol Appl. 13:146–158.
Palumbi SR. 2004. Marine reserves and ocean neighborhoods: the spatial
scale of marine populations and their management. Annu Rev Envir Res.
29:31–68.
Pinsky ML, Montes HR Jr, Palumbi SR. 2010. Using isolation by distance
and effective density to estimate dispersal scales in anemoneish. Evolution.
64:2688–2700.
Sale PF, Cowen RK, Danilowicz BS, Jones GP, Kritzer JP, Lindeman KC,
Planes S, Polunin NV, Russ GR, Sadovy YJ, et al. 2005. Critical science gaps
impede use of no-take ishery reserves. Trends Ecol Evol (Amst). 20:74–80.
Selkoe KA, Toonen RJ. 2006. Microsatellites for ecologists: a practical guide
to using and evaluating microsatellite markers. Ecol Lett. 9:615–629.
Selkoe KA, Watson JR, White C, Horin TB, Iacchei M, Mitarai S, Siegel DA,
Gaines SD, Toonen RJ. 2010. Taking the chaos out of genetic patchiness:
seascape genetics reveals ecological and oceanographic drivers of genetic
patterns in three temperate reef species. Mol Ecol. 19:3708–3726.
Shanks AL, Grantham BA, Carr MH. 2003. Propagule dispersal distance and
the size and spacing of marine reserves. Ecol Appl. 13:159–169.
Siegel D, Kinlan B, Gaylord B, Gaines S. 2003. Lagrangian descriptions of
marine larval dispersion. Mar Ecol-Prog Ser. 260:83–96.
Simpson JJ, Lynn RJ. 1990. A mesoscale eddy dipole in the offshore
California current. J Geophy Res. 95:13,009–13,022.
Singh-Cabanillas J. 1996. Pesqueria de caracol panocha. In Casas-Valdez M,
Ponce-Diaz G, editors. La Paz (Mexico): Estudio del Potencial Pesquero y
Acuícola de Baja California. p. 43–58.
Slatkin M. 1993. Isolation by distance in equilibrium and non-equilibrium
populations. Evolution. 47:264–279.
Slatkin M, Hudson RR. 1991. Pairwise comparisons of mitochondrial
DNA sequences in stable and exponentially growing populations. Genetics.
129:555–562.
Strathman MF. 1987. Reproduction and development of marine invertebrates of the northern Paciic Coast. Seattle (WA): University of Washington
Press.
Pinsky ML, Palumbi SR, Andréfouët S, Purkis SJ. 2012. Open and closed
seascapes: where does habitat patchiness create populations with high fractions of self-recruitment? Ecol Appl. 22:1257–1267.
Taniguchi, IK., Rogers-Bennett L. 2001. Wavy turban snail. In California’s
living marine resources: a status report. In: Leet WS, Dewess CM, Larson
Klingbeil RE, editors. CDFG, Univ. of Calif. Agr. Nat. Res. Publ. SG01-11,
pp. 140–141.
Planes S, Jones GP, Thorrold SR. 2009. Larval dispersal connects ish populations in a network of marine protected areas. Proc Natl Acad Sci USA.
106:5693–5697.
Thorson G. 1950. Reproductive and larval ecology of marine bottom invertebrates. Biol. Rev. 25:1–45.
Prince J. 2005. Combating the tyranny of scale for haliotids: micro-management for microstocks. B Mar Sci. 76:557–578.
Prince J. 2010. Rescaling isheries assessment and management: a generic
approach access rights, change agents, and toolboxes. B Mar Sci. 86:197–219.
Prince J, Walters C, Ruiz-Avila B, Sluczanowski P. 1998. Territorial user’s
rights and the Australian abalone (i.) ishery. In: Jamieson GS, Campbell A,
editors. Proceedings North Paciic Symposium on Invertebrate Stock
Assessment and Management. Ottawa (Canada): National Research Council
of Canada. p. 367–375.
Puritz JB, Toonen RJ. 2011. Coastal pollution limits pelagic larval dispersal.
Nature Commun. 2:226.
RDC Team (R Development Core Team). 2008. R: A Language and
Environment for Statistical Computing. Vienna (Austria): R Foundation for
Statistical Computing.
Rocha-Olivares A, Vetter R. 1999. Effects of oceanographic circulation on
the gene low, genetic structure, and phylogeography of the rosethorn rockish (Sebastes helvomaculatus). Can J Fish Aquat Sci. 56:803–813.
Timmers MA, Bird CE, Skillings DJ, Smouse PE, Toonen RJ. 2012. There’s
no place like home: crown-of-thorns outbreaks in the central paciic are
regionally derived and independent events. PLoS ONE. 7:e31159.
Valentine J, Jablonski D. 1993. Species diversity in ecological communities: historical and geographical perspectives. Chicago (IL): University of
Chicago Press.
Waples R. 1998. Separating the wheat from the chaff: patterns of genetic
differentiation in high gene low species. J Hered. 89:438.
Watson JR, Mitari S, Siegel DA, Caselle JE, Dong C, McWilliams JC. 2010.
Realized and potential larval connectivity in the Southern California Bight.
Mar Ecol Prog Ser. 401:31–48.
Weersing K, Toonen R. 2009. Population genetics, larval dispersal, and connectivity in marine systems. Mar Ecol-Prog Ser. 393:1–12.
White C, Costello C. 2011. Matching spatial property rights isheries with
scales of ish dispersal. Ecol Appl. 21:350–362.
White C, Selkoe KA, Watson J, Siegel DA, Zacherl DC, Toonen RJ. 2010.
Ocean currents help explain population genetic structure. P Roy Soc Lond
B. 277:1685–1694.
13
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
Martone RG, Micheli F. 2012. Geographic variation in demography of a
temperate reef snail: importance of multiple life-history traits. Mar EcolProg Ser. 457:85–99.
Roy K, Jablonski D, Valentine JW. 1995. Latitudinal range shifts of
Pleistocene marine mollusks thermally anomalous assemblages revisited:
patterns in the extraprovincial latitudinal range shifts of Pleistocene marine
mollusks. Geology. 23:1071–1074.
Journal of Heredity
Wilson AB. 2006. Genetic signature of recent glaciation on populations
of a near-shore marine ish species (Syngnathus leptorhynchus). Mol Ecol.
15:1857–1871.
Zacherl D, Gaines SD, Lonhart SI. 2003. The limits to biogeographical distributions: insights from the northward range extension of the marine snail,
Kelletia kelletii. J Biogeogr. 30:913–924.
Woodson CB, Tyburczy J, Barth JA, McManus MA, Caselle J, Carr MH,
Malone D, Raimondi PT, Washburn L, Menge B, Palumbi SR. 2012. Coastal
fronts set recruitment and connectivity patterns across multiple taxa. Limnol
Oceanogr. 57:582–596.
Zaytsev O, Cervantes-Duarte R, Montante O, Gallegos-Garcia A. 2003.
Coastal upwelling activity on the Paciic shelf of the Baja California
Peninsula. J Oceanogr. 59:489–502.
Wright S. 1978. Evolution and the genetics of populations. A treatise in
four volumes. Volume 4: variability within and among natural populations.
Chicago (IL): University of Chicago Press.
Received November 8, 2011; First decision January 24, 2012;
Accepted January 21, 2013
Corresponding Editor: Brian W Bowen
Downloaded from http://jhered.oxfordjournals.org/ at NW Fisheries Science Center on February 28, 2013
14