Journal of Heredity 2011:102(5):584–592
doi:10.1093/jhered/esr055
Advance Access publication June 24, 2011

Ó The American Genetic Association. 2011. All rights reserved.
For permissions, please email: journals.permissions@oup.com.

Mitochondrial Evidence for Panmixia
despite Perceived Barriers to Gene Flow
in a Widely Distributed Waterbird
REBEKAH A. OOMEN, MATTHEW W. REUDINK, JOSEPH J. NOCERA, CHRISTOPHER M. SOMERS, M.
CLAY GREEN, AND CHRISTOPHER J. KYLE

Address correspondence to R. A. Oomen at the address above, or e-mail: rebekahoomen@gmail.com.

Abstract
We examined the mitochondrial genetic structure of American white pelicans (Pelecanus erythrorhynchos) to: 1) verify or refute
whether American white pelicans are panmictic and 2) understand if any lack of genetic structure is the result of
contemporary processes or historical phenomena. Sequence analysis of mitochondrial DNA control region haplotypes of
367 individuals from 19 colonies located across their North American range revealed a lack of population genetic or
phylogeographic structure. This lack of structure was unexpected because: 1) Major geographic barriers such as the North
American Continental Divide are thought to limit dispersal; 2) Differences in migratory behavior are expected to promote
population differentiation; and 3) Many widespread North American migratory bird species show historic patterns of
differentiation resulting from having inhabited multiple glacial refugia. Further, high haplotype diversity and many rare
haplotypes are maintained across the species’ distribution, despite frequent local extinctions and recolonizations that are
expected to decrease diversity. Our findings suggest that American white pelicans have a high effective population size and
low natal philopatry. We suggest that the rangewide panmixia we observed in American white pelicans is due to high
historical and contemporary gene flow, enabled by high mobility and a lack of effective physical or behavioral barriers.
Key words: American white pelican, colonial waterbird, gene flow, mitochondrial DNA, panmixia, Pelecanus erythrorhynchos

Past events of global climate change, as observed during the
Pleistocene glaciations, produced barriers that drastically
altered the genetic structure of many taxa through
population bottlenecks, local adaptation, and genetic drift
(Avise 2009). Many avian species occupied 2 major glacial
refugia in North America during the Pleistocene: 1 east and
1 west of the North American Continental Divide (NACD),
where the Rocky Mountains and Great Plains acted as
historical barriers to gene flow. As a result, east–west genetic
differentiation is common among widespread migratory bird
species, including Swainson’s thrush (Catharus ustulatus;
Ruegg and Smith 2002), wood duck (Aix sponsa; Peters
et al. 2005), and numerous warblers (reviewed in Kelly and
Hutto 2005). In addition to historical barriers, genetic
structuring is influenced by behavioral and ecological lifehistory characteristics, such as migration and dispersal,
which affect gene flow. To understand genetic patterns
resolved by molecular markers, it is important to take both

584

contemporary and historical processes into account (Bossart
and Prowell 1998).
The evolution of ecologically important and complex lifehistory traits, such as migratory behavior, is better
understood with the use of genetic markers (Webster et al.
2002; Davis et al. 2006). In some migratory bird species,
some populations migrate whereas others do not, and in
many species, populations differ in the migratory routes used
(Davis et al. 2006). The effect of different migration patterns
on genetic variation is not well understood. Nonmigration in
some populations may promote differentiation through
limiting dispersal and gene flow and can be a cause of
speciation. For example, southern nonmigratory populations
of yellow warblers are a different subspecies (Dendroica
petechia bryanti) than northern migratory populations
(D. p. aestiva; Salgado-Ortiz et al. 2009). Further, population
differentiation resulting from different migratory routes has
been observed in some species (Sylvia atricapilla, Perez-Tris

Downloaded from http://jhered.oxfordjournals.org/ at Thompson Rivers University on November 21, 2011

From the Forensic Science Department, Trent University, Peterborough, Ontario, Canada (Oomen, Reudink, Nocera, and
Kyle); the Ontario Ministry of Natural Resources, DNA Building, Trent University, Peterborough, Ontario, Canada (Reudink
and Nocera); the Department of Biology, University of Regina, Regina, Saskatchewan, Canada (Somers); and the
Department of Biology, Texas State University, San Marcos, TX (Green).

Oomen et al.  Mitochondrial Evidence for Panmixia in a Widely Distributed Waterbird

differences. In fact, some pelagic seabirds with life-history
characteristics similar to pelicans exhibit no broad-scale
structure with microsatellites but show differences in
mitochondrial DNA within ocean basins (e.g., Morris-Pocock
et al. 2008). Mitochondrial DNA may reveal genetic
differentiation of the southern nonmigratory populations
stemming from a historic divergence in migratory behavior or
enhanced genetic drift in the isolated southern colonies. In
light of the recent range expansion and large distances
between new colonies (.500 km; Pekarik et al. 2009), we
expect to see differentiation in the newly established and
peripheral colonies as a result of founder effects and
enhanced genetic drift (Vucetich and Waite 2003). Mitochondrial markers may also reveal a deep phylogenetic break
across the NACD that stemmed from separate eastern and
western glacial refugia during the Pleistocene. Alternatively,
significant differences in haplotype composition between
populations, but very few nucleotide differences between
haplotypes, would suggest recent expansion from a single
glacial refugium, which is rare for widespread bird species
(Colbeck et al. 2008).
Using DNA samples obtained from colonies across the
breeding range of American white pelicans, we sequenced an
812–base pair (bp) fragment of the mitochondrial control
region to construct a map of the observed haplotypes and
their geographic distributions. Herein, we test for the
following: 1) large-scale genetic structure between the proposed eastern and western metapopulations and the southern
nonmigratory colonies; 2) fine-scale genetic structure between
pelican colonies; and 3) distinct ancestral lineages indicative of
more than 1 historic glacial refugium.

Methods
Field Methods
We collected tissue, blood, and feather samples from 19
colonies across the breeding range of American white
pelicans (see Reudink et al. 2011 for additional details). The
location of study colonies is presented in Figure 1.
Laboratory Methods
We extracted DNA using a QIAGEN DNAeasy Tissue
Extraction kit (QIAGEN) according to the manufacturer’s
instructions. For feathers, the final elution was performed
using 100 ll of elution buffer heated to 65 °C. We amplified
an 812-bp section of the mitochondrial control region in
367 individuals with PCR primers Av438FDloopB
(5#TCACGTGAAATSAGCAACCC) and Av16137tPro
(5#ARAATRCCAGCTTTGGGAGTTGG) (Gibb et al.
2007). Amplification was done in 15-ll volumes with final
reaction concentrations of 1.5
 PCR buffer, 1.5 mM
MgCl2, 200 lM of each dNTP, 0.4 mg/ml bovine serum
albumin, 0.4 lM of each primer, 0.05 U/ll of Taq
polymerase, and 10 ng of template DNA. The PCR
conditions were as follows: 94 °C for 5 min followed by
30 cycles of 94 °C for 30 s, 57 °C for 60 s, and 72 °C for 60 s

585

Downloaded from http://jhered.oxfordjournals.org/ at Thompson Rivers University on November 21, 2011

et al. 2004) and not in others (Phylloscopus trochilus, Bensch
et al. 1999; Dendroica caerulescens, Davis et al. 2006). Many
widespread migratory songbirds in North America show
migratory and genetic differentiation between eastern and
western populations (reviewed in Kelly and Hutto 2005),
with few exceptions (Colbeck et al. 2008). Determining the
timing of sequence divergence of mitochondrial DNA
between populations with different migratory routes may
suggest the amount of population segregation required for
complex and ecologically important traits to diverge.
Studies of genetic variation commonly use different
markers with varying rates of evolution to help ensure that
inferences made from contemporary patterns of variation are
placed in the appropriate historical context (Lukoschek et al.
2008; Avise 2009). Nuclear (e.g., microsatellites) and
mitochondrial (mtDNA) markers often reveal contrasting
patterns of genetic variation, largely due to differing modes of
inheritance, effective population sizes (Birky et al. 1983), and
rates of evolution (e.g., Brunner et al. 1998; Hurles et al. 1998;
Lu et al. 2001; Yang and Kenagy 2009). Moreover, greater
genetic structure has been detected using mtDNA compared
with nuclear markers in some species exhibiting the uncommon phenomenon of rangewide panmixia (e.g., European
plaice [Pleuronectes platessa], Hoarau et al. 2004). Therefore,
employing mitochondrial markers when nuclear markers such
as microsatellites provide little or no resolution is a more
comprehensive approach to examining patterns of genetic
variation and the processes by which they are influenced.
The American white pelican (Pelecanus erythrorhynchos) is
a migratory waterbird with a breeding range consisting of
colonies scattered over much of the continental United
States and southern Canada: from the Pacific coast to the
Great Lakes (Evans and Knopf 1993). These colonies
experience drastic fluctuations in numbers that lead to local
extinctions and recolonizations (Anderson and King 2005).
The breeding range of American white pelicans is currently
expanding eastward (Pekarik et al. 2009) and has traditionally been divided into eastern and western metapopulations
(as in Hanski and Gilpin 1991) separated by the NACD
(Anderson and King 2005). Band-recovery data, which
suggest largely separate migratory pathways for breeding
individuals from colonies east and west of the NACD
(Vermeer 1977; Anderson JGT and Anderson KB 2005),
have been used to substantiate the designation of 2 discrete
metapopulations because they suggest low levels of gene
flow between them (Anderson and King 2005). Though this
small amount of gene flow appears sufficient to prevent
differentiation in nuclear microsatellite markers (Reudink
et al. 2011), it is unclear if it is enough to obscure historic
patterns of differentiation. In addition, a few nonmigratory
colonies exist in Mexico and coastal Texas (Chapman 1988).
The extent to which individuals from these colonies may be
differentiated genetically from individuals from migratory
colonies is not certain, although microsatellites suggest they
are not genetically different (Reudink et al. 2011).
The apparent lack of genetic differentiation in American
white pelicans has not been explored using mitochondrial
markers, which can reveal historically rooted genetic

Journal of Heredity 2011:102(5)

followed by a final extension of 72 °C for 10 min. We
quantified the amount of amplified DNA by electrophoresis
in a 1% agarose gel stained with ethidium bromide and
comparison with a low mass ladder (Invitrogen).
We diluted the amplified samples to 2.86 ng/ll (;2.5
ng/100 bp in 7 ll) with distilled deionized water and
cleaned them using an exonuclease I/shrimp alkaline
phosphatase (ExoSAP) protocol (New England Biolabs,
Beverly, MA) to remove excess reactants prior to
sequencing. PCR products were sequenced in both the
forward and reverse directions using BigDye Terminator
Cycle Sequencing reagents according to the manufacturer’s
instructions (Applied Biosystems, Foster City, CA) and were
left at 4 °C overnight. We tested the quality of the
PCR reaction relative to a positive control of a known
sequence. We purified the sequencing reaction products by
ethanol precipitation and resuspended the DNA in HiDi
formamide before visualizing the DNA fragments using an
ABI Prism 3730 genetic analyzer (Applied Biosystems).
To ensure that mitochondrial DNA sequences were
generated from this process, as opposed to nuclear
homologs, which are common in avian species (Sorenson
and Quinn 1998), we cloned PCR products from 2
individuals using Promega pGemTEasy Vector system and
Invitrogen Max efficiency DH5a competent cells. Thirteen
clones were sequenced from each individual to screen for
the presence of multiple alleles, which would suggest the
simultaneous amplification of nuclear homologs. Of the 8
unique clonal sequences that differed from the consensus
sequence, nucleotide differences were not observed more
than once per individual, indicating that these differences
are most likely a result of PCR error and not the
amplification of multiple alleles from nuclear pseudogenes.

586

In addition, a single band of expected size was observed for
all amplified samples on the agarose gel, supporting the
conclusion that only 1 region was amplified. Sequence
identity was confirmed through a nucleotide–nucleotide
basic alignment search tool search of the National Center
for Biotechnology Information database (Altschul et al.
1990; http://blast.ncbi.nlm.nih.gov/Blast.cgi), which identified the mitochondrial control region of the Australian
pelican (Pelecanus conspicillatus; Gibb et al. 2007) as the most
similar sequence with 94% homology (no American white
pelicans were in the database previously). The sequence
homology between clones also confirms that only 1 control
region was amplified, which is important because the
presence of a duplicate control region was suspected in the
pelican family (Pelecanidae; Gibb et al. 2007).
Sequence Data Analysis
We edited and aligned sequences manually using MEGA
version 4.1 (Kumar et al. 2008). We confirmed forward and
reverse sequences by alignment with each other, and
haplotypes sampled only once were confirmed by reamplification and sequencing of these types. We estimated
genetic diversity using ARLEQUIN version 3.11
(Excoffier et al. 2005) to calculate haplotype diversity
(h, the probability that 2 haplotypes drawn randomly from
a population are different; Halldorsson et al. 2004) and
nucleotide diversity (p, the mean number of pairwise
differences per site between 2 sequences; Nei 1987).
Haplotype richness and the number of private haplotypes
in a sampling location were calculated manually, and
a rarefaction test was performed using ADZE version 1.0
(Szpiech et al. 2008) to assess differences in these diversity

Downloaded from http://jhered.oxfordjournals.org/ at Thompson Rivers University on November 21, 2011

Figure 1. Sampling locations across the American white pelican (Pelecanus erythrorhynchos) breeding range. Each circle represents 1
of the 19 colonies sampled and the bold black line represents the North American Continental Divide.

Oomen et al.  Mitochondrial Evidence for Panmixia in a Widely Distributed Waterbird

Results
Haplotype Diversity
In addition to the 3# end of the control region, the 812-bp
sequence amplified contained a short variable fragment

(116 bp) resembling part of the 3# end of the cytochrome
b gene and 57 bp of transfer RNA threonine. This gene
arrangement is consistent with that observed for the
Australian pelican (Pelecanus conspicillatus; Gibb et al. 2007).
The region contained 27 variable sites producing 32
distinct haplotypes in 367 individuals (Supplementary
Table S1). Sequences have been deposited in GenBank
(accession numbers: HQ315688–HQ315719). Haplotypes
1 and 2 were extremely common, accounting for 72% of
individuals, whereas rare haplotypes (,1% of haplotypes)
comprised a large portion of the remaining individuals
(7%; Table 1). Haplotype diversity (h) was high, whereas
nucleotide diversity (p) was very low (Table 2). Given the
high haplotype diversity observed, sampling error at some
colonies due to insufficient sample size is likely and was
taken into account in subsequent analyses. Allelic richness
and the number of private alleles were similar among
colonies after adjusting for sample size (Table 2). All but
one of Tajima’s D and of Fu’s Fs tests for neutrality
produced negative values (single exceptions being Pelican
Lake, Alberta, and Pipestone Rocks, Manitoba), although
only Pipestone Rocks had a significant D value whereas 11
colonies had significant Fs values (Table 2). However,
these tests cannot differentiate between a recent demographic expansion and weak selection acting on mildly
deleterious mutations located elsewhere in the mitochondrial genome (e.g., Fry 1999), to explain these deviations
from neutrality.
Genetic Structure
An AMOVA performed with sampling locations partitioned
according to eastern, western, and southern regions showed
that 100% of the total variation was explained by variation
within groups (uST 5 0.000, P 5 0.341; Table 3). Negative
pairwise FST values calculated between regions were
interpreted as zero and are due to an imprecision in the
algorithm used by the ARLEQUIN software when very low
deviations in frequencies across regions are observed
(Excoffier et al. 2005; Table 4).
Pairwise FST values between colonies were generally
extremely low (Supplementary Table S2). Small yet
significant FST values were obtained between Pipestone
Rocks and 5 other colonies, Lake Nipigon and 7 other
colonies, and Portage Lake and 3 other colonies. However,
after a Bonferroni correction for multiple comparisons
(which reduces a to 0.002), there were no significant FST
values.
When dividing the breeding range into 3 groups,
a SAMOVA conducted on all colonies found Pipestone
Rocks and Lake Nipigon, independently, to be maximally
differentiated from the remaining colonies, with 6.89% of
the variation explained by variation between groups (uST 5
0.059, P 5 0.006; Table 3). However, when only colonies
with sample sizes .15 were included, a SAMOVA found
Lake Nipigon, Portage Lake, and the remainder of the
sampled colonies to be the most different groups, with
variation between these groups explaining 4.33% of the

587

Downloaded from http://jhered.oxfordjournals.org/ at Thompson Rivers University on November 21, 2011

indices when adjusted to the lowest sample size (N 5 6)
and to N 5 15 (excluding those locations where N , 15).
The rarefaction test estimates the number of distinct and
private haplotypes expected in a random subsample of
specified size drawn from the population (Hurlbert 1971;
Petit et al. 1998). Both Tajima’s D and Fu’s Fs tests were
applied to test for in situ population growth and neutrality.
D is calculated by comparing the number of segregating
sites in relation to the average number of nucleotide
differences between DNA sequences (Tajima 1989), and
Fs is the probability of exhibiting an excess of rare alleles
compared with a neutral population (Fs should be
considered significant if P , 0.02; Fu 1997). All calculations
were permuted 10 000 times.
To identify genetic subdivisions between the eastern,
western, and southern regions, we performed an analysis of
molecular variance (AMOVA) using ARLEQUIN (Excoffier et al. 2005), which compares haplotype variation within
groups with variation between groups. Pairwise FST statistics
based on haplotype frequencies were also calculated using
ARLEQUIN (a 5 0.05) to estimate genetic distances
between regions and between colonies. Additionally, we
performed a spatial analysis of molecular variance (SAMOVA) using SAMOVA version 1.0 (Dupanloup et al. 2002),
which aims to define groups of samples that are maximally
differentiated from each other yet geographically homogenous. We performed a SAMOVA for a range of K (2–10)
and 100 simulated annealing processes on: 1) all sampling
locations and 2) only locations with a sample size .15 to
derive the most probable model of genetic clustering. We
compared AMOVA and SAMOVA results for the a priori
hypothesis of K 5 3 (east, west, and south) to see if they
produced similar clustering of genetic variation.
We used FaBox (Villesen 2007) to calculate haplotype frequencies and Network version 4.5 (http://www.
fluxus-engineering.com/sharenet.htm) to construct a phylogenetic network using the median-joining option (Bandelt
et al. 1999) with maximum parsimony construction (Polzin
and Daneschmand 2003) to illustrate the relationships
between haplotypes. The best model of nucleotide substitution was determined by evaluating Akaike’s Information
Criterion (Akaike 1974) in jModelTest v. 0.1.1 (Posada
2008). For Bayesian inference of phylogeny, we used the
Markov chain Monte Carlo analysis in MrBayes v. 3.1
(Ronquist and Huelsenbeck 2003) using 40 000 000 generations and a sample frequency of 5000. After discarding
25% as burn-in, 6000 trees were produced. We used
TreeView (Page 1996) to visualize the resulting phylogenetic
tree, which was rooted with a mitochondrial control region
sequence from the Australian pelican (GenBank accession
number: DQ780883; Gibb et al. 2007).

Journal of Heredity 2011:102(5)
Table 1

Frequencies of control region haplotypes among sampling locations

Sampled region

N

1

2

3

1

2

1

1 (4%)
2

1
1

2
1 3
1
2 1
1

1

2

2

1
1 1
1
1 2 1
1

1

2 (9%)
1 (5%)
1 (6%)

2
1
1

2 (8%)
1 (7%)
6 (17%)

1
3
1
1
1

2 (10%)
1
1

2 1
2 3
3 1
1
4 4 3 3 2 8
2 4 2 4 2
1

2

3
1

7
1

2
2

2
1

2

1
2
1

2
2

4 (12%)
2 (4%)
1 (4%)
20 (8%)
6 (6%)
1 (4%)

The first row contains the haplotype label. Percentages indicate the proportion of samples with a particular haplotype from a sampling location and are only
given for haplotypes with relatively high frequencies. The frequencies of rare haplotypes (those present in ,1% of individuals) are combined for each region.

variation observed (uST 5 0.039, P 5 0.011; Table 3).
Significant support was not obtained for any value of K, and
FCT increased as K approached 1 (data not shown).
Phylogenetic Analysis
The haplotype-spanning network revealed a star-like topology
with many rare haplotypes (diverged by only 1 nucleotide)
radiating out from 2 common haplotypes (1 and 2; Figure 2).
There were no observed differences in haplotypes originating
from each of the 3 regions and low incidences of homoplasy.
jModelTest selected TrNþG (Tamura and Nei 1993) as the
most appropriate model of nucleotide substitution. The
Bayesian phylogeny created using MrBayes was unable to
resolve haplotype groupings, with the majority of bootstrap
values below 50 (data not shown).

Discussion
Our analysis of mitochondrial control region haplotypes
from 367 individuals revealed a lack of genetic structure
across the entire breeding range of American white pelicans.
This lack of structure was unexpected because: 1) The
NACD, a presumed major geographic barrier to dispersal,
should limit gene flow; 2) Differences in migratory behavior
should promote population differentiation; and 3) Many
widespread North American migratory bird species show
historic patterns of differentiation resulting from inhabiting
multiple glacial refugia (Ruegg and Smith 2002; Kelly and
Hutto 2005; Peters et al. 2005). Many species with lifehistory characteristics similar to pelicans (e.g., coloniality,

588

high mobility, broad ranges, low annual fecundity, long life
span), such as pelagic seabirds, show fine-scale mtDNA
structure (Friesen et al. 2007). For example, marbled
murrelets (Brachyramphus marmoratus; Friesen et al. 2005),
red-legged kittiwakes (Rissa brevirostris; Patirana et al. 2002),
razorbills (Alca torda; Moum and Árnason 2001), and Cory’s
shearwater (Calonectris diomedea spp.; Gomez-Diaz and
Gonzalez-Solis 2007) show differences in mtDNA within
ocean basins. The lack of genetic structure detected using
microsatellite markers (Reudink et al. 2011) did not negate
our expectation of significant genetic structure in mtDNA
because a lack of concordance between mitochondrial and
nuclear DNA is common (e.g., Brunner et al. 1998; Hurles
et al. 1998; Lu et al. 2001; Morris-Pocock et al. 2008; Yang
and Kenagy 2009), particularly in species with high dispersal
capabilities and recent colonization histories (Hoarau et al.
2004). Additionally, mitochondrial evidence for rangewide
panmixia is rare (Coltman et al. 2007; Neethling et al. 2008).
The lack of mtDNA structure in this study corroborates
results on American white pelican population genetic
structure obtained using microsatellite markers (Reudink
et al. 2011), suggesting that gene flow in American white
pelicans precludes genetic differentiation of both genomes
at broad and fine scales. We suggest that the lack of genetic
differentiation between regions on either side of the NACD
refutes the current classification of American white pelicans
into eastern and western metapopulations (Anderson and
King 2005). Despite banding data indicating largely separate
eastern and western migratory pathways (Anderson JGT and
Anderson KB 2005), the lack of broad-scale structure
detected in this study suggests that gene flow is high enough

Downloaded from http://jhered.oxfordjournals.org/ at Thompson Rivers University on November 21, 2011

Akimiski Island, Nunavut
9
4
2
Lake of the Woods, Ontario
6
3
2
Lake Nipigon, Ontario
25 10 (40%) 10 (40%)
Pipestone Rocks, Manitoba
8
6
Last Mountain Lake, Saskatchewan
24 13 (54%) 6 (25%) 1
Reed Lake, Saskatchewan
23 10 (43%) 6 (26%)
Dore Lake, Saskatchewan
19 11 (58%) 3 (16%)
Portage Lake, Alberta
18 12 (67%) 2 (11%) 2
Pelican Lake, Alberta
11
5
5
1
Marsh Lake, MN
24 12 (50%) 4 (17%) 1
Chase Lake, ND
15
7 (47%) 4 (27%) 1
Lacreek National Wildlife Refuge, SD
35 16 (46%) 5 (14%) 3
Bitter Lake, SD
10
6
2
Medicine Lake, MT
21 10 (48%) 3 (14%) 1
Blackfoot Reservation, ID
10
5
2
Minidoka National Wildlife Refuge, ID
6
3
1
1
Anaho Island, NV
33 17 (52%) 5 (15%) 2
Clear Lake, CA
46 23 (50%) 8 (17%)
Padre Island, TX
24 13 (54%) 7 (29%) 1
Pooled region
East
248 125 (50%) 54 (22%) 10
West
95 48 (51%) 16 (17%) 3
South
24 13 (54%) 7 (29%) 1

4 5 7 8 9 10 11 14 19 24 27 Rare (,1%)

Oomen et al.  Mitochondrial Evidence for Panmixia in a Widely Distributed Waterbird
Table 2 Haplotype diversity (±standard deviation [SD]), nucleotide diversity (±SD), allelic richness (Ar), and private allele (Ap)
estimates for the raw haplotype frequencies (actual N) and frequencies adjusted to N 5 6 and N 5 15 by rarefaction analysis
Actual
N
Sampled region

N

h

p

Ar

Ap Ar
2
0
1
1
0
2
0
0
0
1
0
2
0
1
0
0
2
0
0

Ap

Adjusted
N 5 15
Ar

Ap

Tajima’s Fu’s
D
Fs

3.917 1.715
1.04
3.000 0.000
0.05
3.093 0.397 4.650 0.871 0.59
2.500 1.290
21.53
2.955 0.431 4.208 0.296 0.77
3.627 0.852 6.150 1.379 1.21
3.195 0.339 5.329 0.583 1.10
2.804 0.276 4.627 0.481 1.35
2.541 0.000
0.20
3.596 0.714 6.227 1.150 1.14
3.502 0.613 6.000 0.329 0.77
3.872 0.767 7.032 1.326 1.35
3.067 0.211
0.66
3.864 0.708 7.199 0.981 1.37
3.667 0.406
0.94
4.000 0.013
1.23
3.539 1.007 6.168 1.506 0.76
3.523 0.548 6.008 0.727 1.33
2.905 0.400 4.500 0.351 0.95

1.69
0.43
2.21
0.20
1.51
23.93
22.60
22.44
0.02
23.45
22.61
28.61
1.18
25.47
22.10
21.81
24.57
24.90
2.02

22.00
21.64
0.95

229.62
213.11
2.02

Significance (indicated in bold) assigned to P , 0.05 for Tajima’s D and P , 0.02 for Fu’s Fs tests for neutrality.

across the NACD to prevent population differentiation. In
contrast, many other widespread North American bird
species exhibit structuring on either side of the divide
because the Rocky Mountains and Great Plains act as
a barrier to gene flow (Ruegg and Smith 2002; Lovette et al.
2004; Kelly and Hutto 2005; Peters et al. 2005). Alternatively, if pelicans have an extremely high effective
population size, gene flow across the NACD may be low
and insufficient time may have passed for east–west
differences to evolve. However, given the additional

evidence from microsatellite markers (Reudink et al. 2011)
and banding data (Anderson JGT and Anderson KB 2005)
suggesting some degree of movement between eastern and
western regions, it is reasonable to conclude that gene flow
may exceed the 1-to-10-migrant-per-generation rule leading
to genetic homogeneity (Mills and Allendorf 1996).
Contrary to our expectations, neither differences in
haplotype diversity nor significant genetic differentiation
was observed in recently established or peripheral colonies.
In addition, high haplotype diversity and many rare

Table 3 Comparison of explained variance calculated using AMOVA for pooled regions (east, west, south) and the optimal group
selection using SAMOVA (all sampling locations and only those with N  15)
Source of variation

% of variation

P value

AMOVA (all colonies), groups: 1) east, 2) west, and 3) south
Among groups
0.61
0.81
Among groups within colonies
0.59
0.24
Within colonies
100.02
0.34
usr
0.00
SAMOVA (all colonies), groups: 1) Pipestone Rocks, 2) Lake Nipigon, and 3) remaining colonies
Among groups
6.89
0.006
Among groups within colonies
0.97
0.32
Within colonies
94.07
0.30
usr
0.06
SAMOVA (only colonies where N  15), groups: 1) Lake Nipigon, 2) Portage Lake, and 3) remaining colonies
Among groups
4.33
0.01
Among groups within colonies
0.44
0.28
Within colonies
96.11
0.16
usr
0.04

589

Downloaded from http://jhered.oxfordjournals.org/ at Thompson Rivers University on November 21, 2011

Akimiski Island, Nunavut
9 0.806 ± 0.120 0.00172 ± 0.00131 5
Lake of the Woods, Ontario
6 0.806 ± 0.120 0.00107 ± 0.00100 3
Lake Nipigon, Ontario
25 0.697 ± 0.060 0.00112 ± 0.00090 6
Pipestone Rocks, Manitoba
8 0.464 ± 0.200 0.00124 ± 0.00105 3
Last Mountain Lake, Saskatchewan
24 0.652 ± 0.081 0.00097 ± 0.00081 5
Reed Lake, Saskatchewan
23 0.759 ± 0.071 0.00144 ± 0.00108 8
Dore Lake, Saskatchewan
19 0.655 ± 0.112 0.00114 ± 0.00092 6
Portage Lake, Alberta
18 0.556 ± 0.130 0.00079 ± 0.00072 5
Pelican Lake, Alberta
11 0.636 ± 0.089 0.00090 ± 0.00081 3
Marsh Lake, MN
24 0.732 ± 0.086 0.00158 ± 0.00114 8
Chase Lake, ND
15 0.743 ± 0.094 0.00132 ± 0.00103 6
Lacreek National Wildlife Refuge, SD
35 0.771 ± 0.067 0.00143 ± 0.00105 12
Bitter Lake, SD
10 0.644 ± 0.152 0.00107 ± 0.00093 4
Medicine Lake, MT
21 0.767 ± 0.090 0.00151 ± 0.00111 9
Blackfoot Reservation, ID
10 0.756 ± 0.130 0.00132 ± 0.00107 5
Minidoka National Wildlife Refuge, ID
6 0.800 ± 0.172 0.00124 ± 0.00110 4
Anaho Island, NV
33 0.716 ± 0.077 0.00137 ± 0.00102 9
Clear Lake, CA
46 0.720 ± 0.063 0.00140 ± 0.00103 10
Padre Island, TX
24 0.641 ± 0.081 0.00121 ± 0.00095 6
Pooled region
East
248 0.696 ± 0.027 0.00127 ± 0.00094
West
95 0.715 ± 0.046 0.00136 ± 0.00100
South
24 0.641 ± 0.081 0.00121 ± 0.00095

Adjusted
N56

Journal of Heredity 2011:102(5)
Table 4
regions

Pairwise differentiation between pooled sampling
East

East
West
South

0.51
0.94

West

South

0.001

0.015
0.008

0.63

FST values are shown above the diagonal and P values are shown below the
diagonal.

Figure 2. Maximum parsimony network of 32 mitochondrial control region haplotypes in American white pelicans (Pelecanus
erythrorhynchos). The size of the circle is proportional to the frequency of the haplotype in the population; shading indicates the
region in which the haplotype occurs (east, dark gray; west, light gray; south, white).

590

Downloaded from http://jhered.oxfordjournals.org/ at Thompson Rivers University on November 21, 2011

haplotypes are maintained across the species distribution,
suggesting a high effective population size. A high effective
population size may be maintained despite local extinctions
and recolonizations if these processes redistribute individuals among breeding sites and thereby negate founder
effects and genetic drift. Further, the lack of genetic
structure across the species’ range counters the existence of
behavioral barriers to gene flow, such as natal philopatry and
sex-biased dispersal, which are common in species with
similar life-history characteristics, such as pelagic seabirds
(Friesen et al. 2007). Pelicans may exhibit lower levels of
philopatry than other colonial waterbirds due to fluctuating
colony numbers driven by variable water conditions, which
may force pelicans to disperse and breed elsewhere in any
given year (Reudink et al. 2011). These dispersal events may
happen in large groups over long distances, frequently
enough to prevent the loss of genetic variation at newly
established and peripheral colonies, and may facilitate

a gradual range shift in response to environmental changes
at breeding sites.
Nonmigration can promote genetic differentiation
(Buerkle 1999) and speciation (Salgado-Ortiz et al. 2009)
by limiting dispersal and gene flow, but we observed no
genetic differences between the isolated nonmigratory
pelican colony in coastal Texas and the northern migratory
colonies. Perez-Tris et al. (2004) also found little differentiation between migratory and nonmigratory blackcap (Sylvia
atricapilla) populations, which they attributed to either the
remnant of an ancestral nonmigratory population from
which all others diverged or secondary colonization of
sedentary populations since the initial demographic expansion that followed the last glacial cycle. However, in our
study, haplotype diversity in Texas did not differ from other
colonies, as one would expect in an ancestral population.
Moreover, breeding colonies in Mexico and coastal Texas
have likely been there for centuries (Oberholser and Kincaid
1974; Chapman 1988), and microsatellite markers intimate
high contemporary gene flow (Reudink et al. 2011). We
suggest that gene flow was likely maintained between the
migratory and nonmigratory colonies historically as well as
contemporarily, resulting in a lack of differentiation in
mtDNA. One possible explanation for high gene flow in the
nonmigratory colony is that some migratory individuals
remain in the south to breed with local adult nonmigrants.
However, our data do not allow us to rule out a secondary
colonization event.

Oomen et al.  Mitochondrial Evidence for Panmixia in a Widely Distributed Waterbird

Supplementary Material
Supplementary Tables S1 and S2 can be found at http://
www.jhered.oxfordjournals.org/.

Funding
Ontario Ministry of Natural Resources’ Climate Change
Forum to J.J.N (CC-08-09-022); Explorer’s Club to M.W.R.;
Canadian Wildlife Service to C.M.S.; Northern Scientific
Training Program to R.A.O.; Canada Research Chairs
Program to C.M.S. (950-208342); Natural Sciences and
Engineering Research Council (NSERC) of Canada Undergraduate Student Research Award to R.A.O. and NSERC
Discovery Grants to C.J.K. and C.M.S.

Acknowledgments
We thank M. Sovada, K. Tribby, S. Comeau-Kingfisher, A. McGregor,
B. Madden, D. Mauser, D. Withers, M. Wackenhut, J. DiMatteo, J. Laux,
D. Anderson, M. Obbard, S. Lockhart, N. Gorman, and numerous field
assistants for assisting with the collection of pelican tissues. S. Castillo,
J. Zigouris, R. Reudink, and the rest of the Kyle lab provided helpful
discussion, comments, and data interpretation. We also thank 3 anonymous
reviewers for their helpful comments and suggestions.

References
Akaike H. 1974. A new look at statistical model identification. IEEE Trans
Automat Control. 19:716–723.
Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local
alignment search tool. J Mol Biol. 3:403–410.
Anderson DW, King T. 2005. Introduction: biology and conservation of
the American White Pelican. Waterbirds. 28:1–8.
Anderson JGT, Anderson KB. 2005. An analysis of band returns of the
American white pelican, 1922 to 1981. Waterbirds. 28:55–60.
Avise JC. 2009. Phylogeography: retrospect and prospect. J Biogeogr.
36:3–15.
Bandelt H-J, Forster P, Rohl A. 1999. Median-joining networks for inferring
intraspecific phylogenies. Mol Biol Evol. 16:37–48.
Bensch S, Andersson T, Åkesson S. 1999. Morphological and molecular
variation across a migratory divide in willow warblers, Phylloscopus trochilus.
Evolution. 53:1925–1935.
Birky CW Jr, Maruyama T, Fuerst P. 1983. An approach to population and
evolutionary genetic theory for genes in mitochondria and chloroplasts, and
some results. Genetics. 103:513–527.
Bossart JL, Prowell DP. 1998. Genetic estimates of population structure
and gene flow: limitations, lessons, and new directions. Trends Ecol Evol.
13:202–206.
Brunner PC, Douglas MR, Bernatchez L. 1998. Microsatellite and
mitochondrial DNA assessment of population structure and stocking
effects in Arctic charr Salvelinus alpinus (Teleostei: Salmonidae) from central
Alpine lakes. Mol Ecol. 7:209–223.
Buerkle CA. 1999. The historical pattern of gene flow among migratory and
nonmigratory populations of prairie warblers (Aves: Parulinae). Evolution.
53:1915–1924.
Chapman BR. 1988. History of the white pelican colonies in south Texas
and northern Tamaulipas. Colon Waterbird. 11:275–283.
Colbeck GJ, Gibbs HL, Marra PP, Hobson K, Webster MS. 2008.
Phylogeography of a widespread North American migratory songbird
(Setophaga ruticilla). J Hered. 99:453–463.
Coltman DW, Stenson G, Hammill MO, Haug T, Davis CS, Fulton TL.
2007. Panmictic population structure in the hooded seal (Cystophora cristata).
Mol Ecol. 16:1639–1648.
Davis LA, Roalson EH, Cornell KL, McClanahan KD, Webster MS. 2006.
Genetic divergence and migration patterns in a North American passerine
bird: implications for evolution and conservation. Mol Ecol. 15:2141–2152.
Dupanloup I, Schneider S, Excoffier L. 2002. A simulated annealing approach
to define the genetic structure of populations. Mol Ecol. 11:2571–2581.
Evans RM, Knopf FL. 1993. American White Pelican (Pelecanus
erythrorhynchos). In: Poole A, Gill F, editors. The birds of North America.
Philadelphia (PA): The Academy of Natural Sciences No. 57.
Excoffier L, Laval G, Schneider S. 2005. Arlequin (version 3.0): an
integrated software package for population genetics data analysis. Evol
Bioinf. 1:47–50.
Friesen VL, Birt TP, Piatt JF, Golightly RT, Newman SH, Hébert PN,
Gissing G. 2005. Population genetic structure in marbled murrelets
(Brachyramphus marmoratus), and the delineation of ‘distinct population
segments’ for conservation. Conserv Genet. 6:607–613.
Friesen VL, Burg TM, McCoy KD. 2007. Mechanisms of population
differentiation in seabirds. Mol Ecol. 16:1765–1785.
Fry AJ. 1999. Mildly deleterious mutations in avian mitochondrial DNA:
evidence from neutrality tests. Evolution. 53:1617–1620.
Fu Y-X. 1997. Statistical tests of neutrality of mutations against population
growth, hitchhiking and background selection. Genetics. 147:915–925.

591

Downloaded from http://jhered.oxfordjournals.org/ at Thompson Rivers University on November 21, 2011

Unlike nearly all other widespread North American
migratory bird species studied (e.g., Swainson’s thrush,
Ruegg and Smith 2002; wood duck, Peters et al. 2005;
American redstart [Setophaga ruticilla], Colbeck et al. 2008;
and all warblers with continental distributions, Kelly and
Hutto 2005), we found no evidence to suggest contemporary populations of pelicans may have originated from more
than 1 glacial refugium during the Pleistocene. The starshaped haplotype network and high haplotype diversity
coupled with low nucleotide diversity at all colonies suggest
a recent expansion from a single population following
a bottleneck (Grant and Bowen 1998). However, because of
the high mobility and dispersal capabilities of American
white pelicans (Evans and Knopf 1993; Pekarik et al. 2009),
we cannot discount the possibility that they may have
occupied multiple refugia during the Pleistocene and
experienced continuous gene flow between them or that
all but 1 lineage went extinct.
The lack of phylogeographic structure in the American
white pelican underscores the idea that species that share
similar life-history characteristics and occupy similar
contemporary distributions may have very different
evolutionary histories (Zink 1996; Colbeck et al. 2008).
Pelicans have maintained high genetic diversity and exhibit
no genetic differentiation despite putative barriers to
dispersal (such as the NACD) that promote differentiation
in other species. How genetic homogeneity is maintained
in pelicans warrants further study into the interplay
between demographic (e.g., frequent local extinctions and
recolonizations) and behavioral processes (e.g., natal
philopatry). Understanding these interactions may allow
us to predict how highly mobile species such as American
white pelicans will fare in the face of large-scale
environmental change.

Journal of Heredity 2011:102(5)
Gibb GC, Kardailsky O, Kimball RT, Braun EL, Penny D. 2007.
Mitochondrial genomes and avian phylogeny: complex characters and
resolvability without explosive radiations. Mol Biol Evol. 24:269–280.

Patirana A, Hatch SA, Friesen VL. 2002. Population differentiation in the
red-legged kittiwake (Rissa tridactyla) as reveled by mitochondrial DNA.
Conserv Genet. 3:335–340.

Gomez-Diaz E, Gonzalez-Solis J. 2007. Geographic assignment of seabirds
to their origin: combining morphologic, genetic, and biogeochemical
analyses. Ecol Appl. 17:1484–1498.

Pekarik C, Hodder C, Weseloh DVC, Matkovich C, Shutt L, Erdman T,
Matteson S. 2009. First nesting of American white pelican on Lake
Superior, Ontario, Canada. Ont Birds. 27:42–49.

Grant WAS, Bowen BW. 1998. Shallow population histories in deep
evolutionary lineages of marine fishes: insights from sardines and anchovies
and lessons for conservation. J Hered. 89:415–426.

Perez-Tris J, Bensch S, Carbonell R, Helbig AJ, Telleria JL. 2004. Historical
diversification of migration patterns in a passerine bird. Evolution.
58:1819–1832.

Halldorsson BV, Bafna V, Edwards N, Lippert R, Yooseph S, Istrail S. 2004.
A survey of computational methods for determining haplotypes. In: Istrail S,
Waterman MS, Clark AG, editors. Computational methods for SNPs and
haplotype inference. Berlin (Germany): Springer-Verlag. p. 26–47.

Hoarau G, Piquet AM-T, van der Veer HW, Rijnsdorp AD, Stam WT,
Olsen JL. 2004. Population structure of plaice (Pleuronectes platessa L.) in
northern Europe: a comparison of resolving power between microsatellites
and mitochondrial DNA data. J Sea Res. 51:183–190.
Hurlbert SH. 1971. The nonconcept of species diversity: a critique and
alternative parameters. Ecology. 4:577–586.
Hurles ME, Irven C, Nicholson J, Taylor PG, Santos FR, Loughlin J,
Jobling MA, Sykes BC. 1998. European Y-chromosomal lineages in
Polynesians: a contrast to the population structure revealed by mtDNA. Am
J Hum Genet. 63:1793–1806.
Kelly JF, Hutto RL. 2005. An east-west comparison of migration in North
American wood warblers. Condor. 107:197–211.
Kumar S, Dudley J, Nei M, Tamura K. 2008. MEGA: a biologist-centric
software for evolutionary analysis of DNA and protein sequences. Brief
Bioinform. 9:299–306.
Lovette IJ, Clegg SM, Smith TB. 2004. Limited utility of mtDNA markers
for determining connectivity among breeding and overwintering locations
in three neotropical migrant birds. Conserv Biol. 18:156–166.
Lu G, Basley DJ, Bernatchez L. 2001. Contrasting patterns of mitochondrial
DNA and microsatellite introgressive hybridization between lineages of lake
whitefish (Coregonus clupeaformis); relevance for speciation. Mol Ecol. 10:965–985.
Lukoschek V, Waycott M, Keogh JS. 2008. Relative information content of
polymorphic microsatellites and mitochondrial DNA for inferring dispersal
and population genetic structure in the olive sea snake, Aipysurus laevis. Mol
Ecol. 17:3062–3077.
Mills LS, Allendorf FW. 1996. The one-migrant-per-generation rule in
conservation and management. Conserv Biol. 10:1509–1518.
Morris-Pocock JA, Taylor SA, Birt TP, Damus M, Piatt JF, Warheit KI,
Friesen VL. 2008. Population genetic structure in Atlantic and Pacific
Ocean common murres (Uria aalge): natural replicate tests of postPleistocene evolution. Mol Ecol. 17:4859–4873.

Petit R, Mousadik AE, Pons O. 1998. Identifying populations for
conservation on the basis of genetic markers. Conserv Biol. 12:844–855.
Polzin T, Daneschmand SV. 2003. On Steiner trees and minimum spanning
trees in hypergraphs. Oper Res Lett. 31:12–20.
Posada D. 2008. jModelTest: phylogenetic model averaging. Mol Biol Evol.
25:1253–1256.
Reudink MW, Kyle CJ, Nocera JJ, Oomen RA, Green MC, Somers CM.
2011. Panmixia on a continental scale in a widely distributed colonial
waterbird. Biol J Linn Soc. 102:583–592.
Ronquist F, Huelsenbeck JP. 2003. MrBayes 3: Bayesian phylogenetic
inference under mixed models. Bioinformatics. 19:1572–1574.
Ruegg KC, Smith TB. 2002. Not as the crow flies: a historical explanation
for circuitous migration in Swainson’s thrush (Catharus ustulatus). Proc R Soc
Lond B Biol Sci. 269:1375–1381.
Salgado-Ortiz J, Marra PP, Robertson RJ. 2009. Breeding seasonality of the
mangrove warbler (Dendroica petechia bryanti) from southern Mexico. Ornitol
Neotrop. 20:255–263.
Sorenson MD, Quinn TW. 1998. Numts: a challenge for avian systematics
and population biology. Auk. 115:214–221.
Szpiech ZA, Jakobsson M, Rosenberg NA. 2008. ADZE: a rarefaction
approach for counting alleles private to combinations of populations.
Bioinformatics. 24:2498–2504.
Tajima F. 1989. The effect of change in population size on DNA
polymorphism. Genetics. 123:597–601.
Tamura K, Nei M. 1993. Estimation of the number of nucleotide
substitutions in the control region of mitochondrial DNA in humans and
chimpanzees. Mol Biol Evol. 10:512–526.
Vermeer K. 1977. Comparison of white pelican recoveries from colonies
east and west of the Canadian Rocky Mountains. Murrelet. 58:79–82.
Villesen P. 2007. FaBox: an online toolbox for FASTA sequences. Mol Ecol
Notes. 7:965–968.
Vucetich JA, Waite TA. 2003. Spatial patterns of demography and genetic
processes across the species’ range: null hypotheses for landscape
conservation genetics. Conserv Genet. 4:639–645.

Moum T, Árnason E. 2001. Genetic diversity and population history of two
related seabird species based on mitochondrial DNA control region
sequences. Mol Ecol. 10:2463–2478.

Webster MS, Marra PP, Haig SM, Bensch S, Holmes RT. 2002. Links
between worlds: unraveling migratory connectivity. Trends Ecol Evol.
17:76–83.

Neethling M, Matthee CA, Bowie RCK, von der Heyden S. 2008. Evidence
for panmixia despite barriers to gene flow in the southern African endemic,
Caffrogobius caffer (Teleostei: Gobiidae). BMC Evol Biol. 8:325.

Yang D-S, Kenagy GJ. 2009. Nuclear and mitochondrial DNA reveal
contrasting evolutionary processes in populations of deer mice (Peromyscus
maniculatus). Mol Ecol. 18:5115–5125.

Nei M. 1987. Molecular evolutionary genetics. New York: Columbia
University Press.

Zink RM. 1996. Comparative phylogeography in North American birds.
Evolution. 50:308–317.

Oberholser HC, Kincaid EB Jr. 1974. The bird life of Texas. Vol. 1. Austin
(TX): University of Texas Press.

Received July 21, 2010; Revised April 21, 2011;
Accepted May 17, 2011

Page RDM. 1996. TREEVIEW: an application to display phylogenetic trees
on personal computers. Comput Appl Biosci. 12:357–358.

Corresponding Editor: Robert C Fleischer

592

Downloaded from http://jhered.oxfordjournals.org/ at Thompson Rivers University on November 21, 2011

Hanski I, Gilpin M. 1991. Metapopulation dynamics: brief history and
conceptual domain. Biol J Linn Soc. 42:3–16.

Peters JL, Gretes W, Omland KE. 2005. Late Pleistocene divergence
between eastern and western populations of wood ducks (Aix sponsa)
inferred by the ‘isolation with migration’ coalescent method. Mol Ecol.
14:3407–3418.