Intraspecific evolutionary relationships among peregrine falcons in western North American high latitudes
Intraspecific evolutionary relationships among peregrine falcons in western North American high latitudes
Sandra L. Talbot 1 2 3
George K. Sage 1 2 3
Sarah A. Sonsthagen 1 2 3
Meg C. Gravley 1 2 3
Ted Swem 1 3
Jeffrey C. Williams 0 1 3
Jonathan L. Longmire 1 3
Skip Ambrose 1 3
Melanie J. Flamme 1 3
Stephen B. Lewis 1 3
Laura Phillips 1 3
Clifford Anderson 1 3
Clayton M. White 1 3 4
0 Alaska Maritime National Wildlife Refuge, U. S. Fish and Wildlife Service, Homer, Alaska, United States of America, 4 Los Alamos National Laboratory, Los Alamos, New Mexico, United States of America, 5 Yukon- Charley River National Preserve and Gates of the Arctic National Park and Preserve, National Park Service, Fairbanks, Alaska, United States of America, 6 Migratory Bird Management, U. S. Fish and Wildlife Service, Juneau, Alaska, United States of America , 7 Alaska Regional Office , National Park Service, Anchorage, Alaska, United States of America, 8 Falcon Research Group, Bow , Washington , United States of America
1 Funding: Funding for this research was provided by the Alaska Science Center, U. S. Geological Survey (SLT), U. S. Fish and Wildlife Service (JCW , USA
2 U.S. Geological Survey Alaska Science Center, Anchorage, Alaska, United States of America, 2 Fairbanks Fish and Wildlife Field Office, U. S. Fish and Wildlife Service , Fairbanks, Alaska , United States of America
3 Editor: Michael E. Douglas, University of Arkansas Fayetteville , UNITED STATES
4 Department of Plant and Wildlife Sciences and Monte L. Bean Life Science Museum, Brigham Young University , Provo, Utah , United States of America
Subspecies relationships within the peregrine falcon (Falco peregrinus) have been long debated because of the polytypic nature of melanin-based plumage characteristics used in subspecies designations and potential differentiation of local subpopulations due to philopatry. In North America, understanding the evolutionary relationships among subspecies may have been further complicated by the introduction of captive bred peregrines originating from non-native stock, as part of recovery efforts associated with mid 20th century population declines resulting from organochloride pollution. Alaska hosts all three nominal subspecies of North American peregrine falcons±F. p. tundrius, anatum, and pealei±for which distributions in Alaska are broadly associated with nesting locales within Arctic, boreal, and south coastal maritime habitats, respectively. Unlike elsewhere, populations of peregrine falcon in Alaska were not augmented by captive-bred birds during the late 20th century recovery efforts. Population genetic differentiation analyses of peregrine populations in Alaska, based on sequence data from the mitochondrial DNA control region and fragment data from microsatellite loci, failed to uncover genetic distinction between populations of peregrines occupying Arctic and boreal Alaskan locales. However, the maritime subspecies, pealei, was genetically differentiated from Arctic and boreal populations, and substructured into eastern and western populations. Levels of interpopulational gene flow between anatum and tundrius were generally higher than between pealei and either anatum or tundrius. Estimates based on both marker types revealed gene flow between augmented Canadian populations and unaugmented Alaskan populations. While we make no attempt at formal taxonomic revision, our data suggest that peregrine falcons occupying habitats in Alaska
and the North Pacific coast of North America belong to two distinct regional groupings±a
TS, SA, SBL), Brigham Young University (CMW),
National Park Service (MJF, LP), Falcon Research
Group (CA), Los Alamos National Laboratory (JLL),
and U. S. Bureau of Land Management (TS). The
funders had no role in the study design, data
collection and analysis, decision to publish, or
preparation of the manuscript.
Competing interests: The authors have declared
that no competing interests exist.
coastal grouping (pealei) and a boreal/Arctic grouping (currently anatum and tundrius)±
each comprised of discrete populations that are variously intra-regionally connected.
Clarification of the relationship between phenotypic and molecular characters is particularly
relevant to studies involving northern high-latitude avian species, as many vertebrate
populations there represent the leading edge of post-Pleistocene expansions from temperate glacial
]. Such populations are often little differentiated, since insufficient time has passed
for neutral genetic markers to coalesce [
], leading to discordance between morphological and
genetic variation at neutral genetic markers. Molecular genetic phylogenies have failed to
support many avian subspecies designations that are based on phenotypic characters [
including designations in species of raptor that occupy high latitude habitats [5,6], although the
criteria for designating subspecies based on genetic characteristics vary across investigators
]. Nevertheless, difficulties in understanding underlying genetic and environmental bases of
phenotypic characters have contributed to inconsistencies when making evolutionary
inferences based on either character class, and this has also contributed to controversy involving
subspecies designations that are based partially or largely on melanin-based plumage traits [
which can vary with individual body condition, age, and exposure to oxidative stress induced
by environmental factors [
Differences in levels of melanism, which can be due to variation at one or a number of
genes, can also emerge from both neutral and adaptive evolutionary processes, such as genetic
] for the former, and divergent and disruptive evolution, preferential selection for
mates displaying a specific color [
], and morph-specific variation in immunity as well as
pathogen pressure [
] for the latter. For example, a single non-synonymous change in the
melanocortin-1 receptor gene (MC1R), which codes for primary plumage color in avian
species, is perfectly associated with the presence of melanism in two Arctic species, the snow
goose (Chen caerulescens), and Arctic skua (Stercorarius parasiticus), in which polymorphisms
in color apparently influence mate choice [
]. Although white and melanistic snow geese
were once thought to comprise different species or subspecies [
], the MC1R
genotype-phenotype association in snow geese, as well as skuas, is specific to the MC1R locus and does not
reflect shared demographic history of populations containing different morphs [
a single nucleotide polymorphism (SNP) found in the MC1R gene is a determinant of leucistic
] and wild [
] gyrfalcons (Falco rusticolus), a raptor species that occupies Arctic
habitats, but for which variation in plumage color (from all-white to dark gray/brown) is not
currently associated with subspecies designations [16±18]. In gyrfalcons, however, there is no
evidence of assortative pairing based on plumage color, although there may be a relationship
between plumage color and timing of egg-laying and reproductive output [
]. In contrast,
even though subspecies designations in the red-tailed hawk (Buteo jamaicensis) in western
North America are based in part on the presence of certain melanin-based plumage
characteristics, there is no apparent correlation between specific variants of the MC1R gene and
variation in plumage color [
Similar to red-tailed hawks, designations among the named subspecies of peregrine falcons
(F. peregrinus) in North America are based in part on morphological characteristics, including
plumage traits associated with melanism [
]. Prior population genetics studies suggest that
these morphological characteristics do not fully predict the distribution of genetic variation
2 / 25
Fig 1. Map of peregrine falcon subspecies distributions in Alaska and Canada, and general sampling locales.
among representative populations, at least within the Canadian portion of the species' North
American range [
]. It is unclear whether this discordance ensues wholly from historical and
demographic processes associated with natural colonization, coupled with inadequate passage
of time for coalescence [
], or from recovery efforts associated with the late 20th century
augmentation or reestablishment of extirpated peregrine falcon populations using non-native
], or a combination of these and/or other factors.
In North America, peregrines breed from Mexico to the middle Arctic and from the Pacific
to the Atlantic coast. Morphological variation among peregrine falcons is sufficient that 19
subspecies have been described worldwide [
], three of which nest during summer months in
high latitude North America. Peale's falcon (F. p. pealei Ridgway 1873), the largest subspecies, is
distributed in maritime habitats coastally from Washington State through the Aleutian Islands,
Alaska, to the Commander Islands, Russia (Fig 1). Peregrines within the geographic distribution
of F. p. pealei are characterized by dark, heavily pigmented plumage, a diet comprised mostly of
alcids and procellarids, year-round residency, and relative stability of component populations
when the other two North American subspecies declined due to reproductive failure resulting
3 / 25
from organochloride contamination. Plumage coloration varies within the Peale's falcon;
peregrines occupying the Aleutian Islands are uniformly dark, whereas those from Haida Gwaii
(formerly Queen Charlotte Islands) have grayed tones and show a greater range of variation
The Nearctic subspecies (Fig 1), F. p. tundrius [
], is generally smaller and paler than the
other two North American subspecies, and they exhibit highly migratory behavior. The
breeding range of F. p. tundrius is in the Nearctic tundra region (herein, Arctic Alaska or Arctic
Canada), from about 77Ê N south to treeline, east to Greenland and west to Cape Prince of
Wales, Alaska, and individuals within the range of F. p. tundrius winter from Florida south to
Argentina (38Ê S) and Chile (41Ê S). The third subspecies, F. p. anatum (Bonaparte 1838), has
a continental distribution (Fig 1), and is considered intermediate between the other two
subspecies in size, plumage color, and migratory behavior. Based on analysis of morphological
characteristics of museum specimens, White [
] found that peregrine falcons in interior
boreal habitats in Alaska are on average larger than peregrines occupying locales within the
southwestern distribution of F. p. anatum, and darker than the pale, small, highly migratory
peregrines occupying the Arctic (that is, peregrines within the distribution of F. p. tundrius).
However, there is overlap in plumage characteristics and morphological measurements in
Alaskan F. p. tundrius and F. p. anatum [
Studies conducted in the early part of the 20th century suggest the peregrine falcon was
never abundant in North America [
]. Reproductive failure due to organochloride
pollution caused a severe decline in most temperate and high latitude North American peregrine
falcon populations, starting in the late 1940s and continuing into the early 1970s . This
well-documented and unprecedented collapse extirpated peregrine falcons in many regions of
North America [
], including major former strongholds such as the Appalachian region [
In some locales (e.g., the Yukon and Tanana rivers, Alaska [
]), localized human activity,
such as road construction and take for falconry [
], may have exacerbated declines
attributed to the pesticide 1,1,1-trichloro-2,2-bis (p-chlorophenyl) ethane (DDT). Insular,
non-migratory populations, including F. p. pealei along the Pacific Coast of Alaska and British
Columbia (with the exception of the population on Haida Gwaii) and along the Aleutian
Island chain, were considered least affected [
]. By 1975, only 324 nesting pairs of
peregrine falcons were confirmed on the entire North American continent  although numbers
were probably larger by perhaps 25% (estimate based on personal experience of authors and
others). Peregrines occupying boreal (F. p. anatum) and Arctic (F. p. tundrius) habitats were
listed as endangered in 1970 under the Endangered Species Conservation Act of 1969 (P.L.
91±135, 83 Stat. 275), a law that preceded the Endangered Species Act of 1973. Among the
three subspecies, F. p. anatum was the focus of the most intense recovery efforts.
Recovery of the peregrine falcon in North America proceeded after the ban of DDT
for agricultural use in Canada in 1970, and the United States in 1972, and the subsequent,
although gradual, decrease of the breakdown product 1,1-dichloro-2,2-bis (p-chlorophenyl)
ethylene, DDE, in North American food chains. Recovery efforts in Canada and much of the
contiguous United States that occurred between 1974 and 1999 included the introduction of
captive-bred peregrine falcons to bolster remnant or replace extirpated populations [
Brown et al.  report that only native birds were released during the Canadian reintroduction
efforts. However, this claim applies perhaps to only a portion of the Canadian reintroductions,
begun in the early 1980s, that used only native-captured F. p. anatum to augment eastern
Canadian populations [
]. Falcons released into portions of Canada and the continental
United States prior to the mid-1970s were captive stock derived from seven different
subspecies, including (in addition to the three North American subspecies) four subspecies found on
other continents (Europe: F. p. brookei, F. p. peregrinus; South America: F. p. cassini; and
4 / 25
Australia: F. p. macropus [
]). Further, peregrines released in the contiguous United
States were known to have dispersed into Canada, where they bred [
]. Although monitored
F. p. tundrius populations had declined by approximately 80%, this subspecies was downlisted
from endangered to threatened in 1984 (49 FR 10520, March 20, 1984) and delisted in October
1994 (59 FR 50796, October 5, 1994). Populations of F. p. anatum were delisted in the United
States on August 25, 1999 (64 FR 46541±46558).
In Alaska, which like Canada hosts all three nominal North American subspecies, recovery
of peregrines occupying boreal and Arctic habitats commenced in the late 1970s, following the
prohibition of DDT for agricultural purposes in 1972 [
]. Unlike elsewhere in North
America, however, reintroductions were not conducted in Alaska; populations there were
allowed to recover through natural recruitment and recolonization [
]. As a result,
genetic assessment of relationships among Alaskan populations comprising the three named
subspecies may be less confounded by the introduction of genetic signatures of peregrine
falcons from other regions, as in Canada [
] and Scandinavia [
]. Nevertheless, it is not known
whether peregrines released in the contiguous United States and Canada during the late 20th
century recovery effort dispersed into Alaska. As well, previous inferences associated with
levels and distribution of genetic diversity within North American F. p. pealei [
] did not include
information from falcons occupying the Aleutian Island chain, which represents the largest
and most geographically isolated component of this subspecies' distribution.
To provide a phylogeographic framework within which to assess the relationships among
the named subspecies, we used genetic markers from the mitochondrial and nuclear genomes
to investigate taxonomic relationships among the three North American subspecies occupying
non-augmented northwestern-most locales. We focus on two central questions: 1) is the
current subspecies taxonomy, which is based largely on plumage characteristics, predictive of the
distribution of neutral gene variation within Alaskan peregrine falcons; and 2) what are the
relationships among populations within named subspecies in western high latitude North
America? We use sequence data from the control region of the mitochondrial DNA and
fragment data from 11 polymorphic microsatellite loci to investigate relationships among
maritime populations occupying the Aleutian Island chain, including the Commander Islands,
Russia, and coastal North Pacific locales (F. p. pealei), as well as boreal (F. p. anatum) and
Arctic habitats (F. p. tundrius) of Alaska. We also include comparative data from a population of
falcons nesting on the San Juan Archipelago in western Washington State, a locale thought to
comprise a contact zone between southeastern populations of F. p. pealei, and southern
populations of F. p. anatum . To infer levels of gene flow from potentially augmented
populations, and, given the mid-to-late 20th century decline in two of the three North American
subspecies, to test for genetic signatures of recent demographic change against a background
of more historical population change, we include data from falcons of western Canadian Arctic
and boreal populations.
Materials and methods
Samples included in this study were obtained from: A) specimens in museum collections (the
Monte L. Bean Museum, Provo, Utah (n = 16) and the Museum of the Zoological Institute in
St. Petersburg, Russia (n = 7); B) a single blood sample from the Bird Treatment and Learning
Center, collected by veterinarians or other clinical staff with permission of State and Federal
wildlife authorities; and C) blood, eggs, epithelial swabs, or feather samples collected as part of
past and ongoing field research [27,32,36±40], all collected with appropriate Federal and State
or Provincial permits. The latter include appropriate permits from the U. S. Fish and Wildlife
5 / 25
Service (all collections in the United States) and the Canadian Wildlife Service (all collections
in Canada), site-specific permits for collections on lands administered by the National Park
Service-and the U. S. Fish and Wildlife Service's National Wildlife Refuge-system, the
Washington State Department of Fish and Game (the San Juan Island samples), and the Alaska
Department of Fish and Game (for scientific collection in Alaska). All field sampling
procedures (blood withdrawal) were reviewed prior to sample collection. Field activities subsequent
to 1985, with the exception of those by U. S. Fish and Wildlife Service personnel, were
conducted following appropriate Institutional Animal Care and Use Committee (IACUC), or
equivalent, approval. U. S. Fish and Wildlife Service did not implement requirements for
IACUC permitting until 2008; all field research conducted by U. S. Fish and Wildlife personnel
(TS, SA, SBL) subsequent to 2008 were conducted under an appropriate IACUC. In all cases,
blood withdrawal procedures followed recommendations by the Ornithological Council for
handling wild birds and sampling tissues [
]. Year and location of sample collection are
]. Samples are held at the U. S. Geological Survey Alaska Science Center but are not
available for loan.
We collected whole blood, buccal swabs, and/or feathers from adult nesting female and
male peregrine falcons, or young in nests, from sites within the distribution of each
subspecies (Fig 1). The Arctic subspecies, F. p. tundrius, is represented in Alaska by samples taken
from the Sagavanirktok River (n = 1) and the Colville River (n = 46), pooled and designated
as COL, and in western Canada by samples collected from Hope Bay, Northwest Territories
(HB, n = 18, [
]). F. p. anatum in Alaska is represented by samples collected from sites
along the Porcupine (POR, n = 9), Tanana (TAN, n = 9) and Yukon (YUK, n = 17) rivers,
and in western Canada by samples collected from Alberta and Saskatchewan [
from the Lake Athabasca area (LKAT, n = 6), and the Mackenzie Valley (MCV, n = 8). The
eastern portion within the distribution of F. p. pealei is represented by samples collected
from 1) the North Pacific coast of North America, including the Alexander Archipelago in
southeast Alaska, and Haida Gwaii Archipelago of British Columbia (NPAC, n = 11), and
south central coastal Alaska, including locales near Yakutat, Kenai Fjords, and Kodiak
Island (SCCOA, n = 13). The western distribution of F. p. pealei is represented by samples
(n = 44) collected from across four island groups along the Aleutian island chain; these
include the Rat (Amchitka and Buldir islands; RAT, n = 17), Andreanof (Kasatochi and
Amatignak islands; ANDR, n = 17), Near (Attu Island; NEAR, n = 5) and Commander
(COMM, n = 7) island groups. Peregrines sampled from the San Juan Islands (SJI, n = 15),
thought to represent a contact zone between continental F. p. anatum and the North Pacific
segment of F. p. pealei , are treated as a separate population not designated to a
subspecies. Sample locales and acronyms are given in Fig 1, specific data associated with each
sample are provided in .
Blood samples were obtained from either nestlings or falcons trapped using standard
techniques (see [
]). Blood was taken from the brachial vein using non-heparinized
needles and stored in 1mL of blood preservation buffer (0.1M Tris-HCl [pH 8.0], 0.1M EDTA,
0.1M NaCl, and 0.5% SDS; [
]). Buccal swabs were collected using techniques similar to
Handel et al. [
], except after collection, swabs were stored in 500 μL of preservation
buffer. Feathers and eggshell membranes were kept in coin envelopes stored in silica gel
until DNA extraction. Samples were kept at room temperature until DNA was extracted,
and thereafter stored at -80ÊC (for blood and swabs) or in silica gel in dry cabinets (feathers
and eggshell membranes).
6 / 25
See S1 Methods for details of the laboratory analyses. Following extractions and quantification
of DNA, fragment data were collected at 11 polymorphic loci ([7,45±47]) and nucleotide
sequence data from a 559 base pair segment of domain 1 of the mtDNA control region, from
peregrine falcons collected from 14 locales across Arctic, boreal, coastal Alaskan forests and
the Aleutian Islands, the San Juan Islands in Washington State, and western Canada, following
procedures outlined elsewhere [
Phylogenetic analyses. An unrooted phylogenetic network was constructed for mtDNA
control region haplotypes using NETWORK 4.610 (Fluxus Technology, Clare, United
Kingdom) and the reduced median method [
Genetic variability and tests of neutrality. Details of analyses assessing levels of genetic
diversity and neutrality at microsatellite loci and the mtDNA control region are provided in S1
Methods. Briefly, we quantified genetic variation and tested for neutrality in microsatellite loci
and mtDNA control region sequences using a variety of computer programs routinely used to
analyze genetic data [49±51]. We note that tests of selective neutrality of mtDNA control
region sequence data (Fu's Fs [
] and Tajima's D [
]) can also provide evidence of
population expansion (see S1 Methods).
Population structure and regional differentiation. Details of analyses to assess levels of
population and regional genetic structure are provided in S1 Methods. Significance of spatial
variation in microsatellite allele frequencies between populations was assessed using
F-statistics (FST and RST) and their analogs ([
], respectively), using several computer programs
routinely used to analyze genetic data [49±51]. Significance of FST and RST values were based
on random permutation tests (n = 1,000); Critical α-values were adjusted using Bonferroni
corrections. Because populations are represented by unequal sample sizes , we also
estimated population differentiation based on the distribution of microsatellite alleles across
populations, using Fisher's combined probability test [
]; significance was judged based on
random permutations tests (n = 1,000) across loci between populations and adjusting critical
α-values using Bonferroni tests. We used the likelihood ratio test criterion in MODELTEST
3.06  to determine the evolutionary distance model that best fit the mtDNA control region
sequence data and applied those weighted distances to calculate population pairwise FST (ϕST;
]) and test for significance [
]. Population differentiation estimated from the distributions
of mtDNA haplotypes were based on the log-likelihood (G) test [
] with significance based
on random permutation tests (n = 1,000). Microsatellite data were also analyzed using
Bayesian cluster analyses [
] to detect the occurrence of population structure without a priori
knowledge of putative populations.
We used hierarchical analyses of molecular variance (AMOVA, [
]) to test for significant
geographic partitioning of a priori hypothesized regional units (subspecies), using ARLEQUIN
]. Calculations for mtDNA sequence data incorporated a distance matrix based on the
model of evolution determined by MODELTEST that best fit the data, and were also
conducted without applying a distance matrix. In addition to the a priori groups of testable
subspecies hypotheses applicable to peregrine falcons of North America, we experimented with
various a posteriori groups in AMOVA analyses for both marker types that tested relationships
of certain populations to named subspecies (e.g., the placement of SCCOA and SJI relative to
anatum and pealei). We assumed that the best geographic subdivisions were significantly
different from random distributions and had maximum among group variance (FCT values). We
also examined regional interrelationships by constructing and visualizing a network based on
7 / 25
Cavalli-Sforza and Edwards [
] genetic distances (CSE) among microsatellite loci and
neighbor-joining methods [63±65]. Two analyses were conducted: 1) populations were analyzed
separately, and 2) populations were pooled into assigned subspecies, except for F. p. pealei,
which was partitioned into the Aleutian group and the eastern group, based on results of
population differentiation analyses.
Estimation of levels and polarity in gene flow. We estimated the magnitude and polarity
of evolutionary dispersal (gene flow) among populations and subspecies, using the maximum
likelihood approach implemented in MIGRATE 3.0.3 [
]; details of the gene flow analyses
are provided in S1 Methods. Migrate uses a coalescent approach to estimate gene flow rates
(Nm) among populations, assuming a constant per-locus mutation rate (μ). Values calculated
(see S1 Methods) included number of migrants (m) among populations per generation
(4Nem) and number of female migrants per generation (Nfm) for nuclear microsatellites and
mtDNA sequences, respectively. Significance of asymmetry in gene flow was based on
nonoverlapping 95% confidence intervals generated from full models, in which composite
measures θ (4Neμ or Nfμ) and M (m/μ) were estimated individually from the data and allowed to
vary among populations. For this analysis, HB and MCV, which showed no significant
interpopulational structuring even though they have traditionally been ascribed to different
subspecies, were pooled to represent Arctic Canada; similarly, POR, TAN and YUK were pooled to
represent interior boreal Alaska and the different Aleutian Archipelago islands were pooled to
represent the Aleutian Islands as a single group. Arctic Alaska is represented by COL.
Genetic signals of changes in population demography. Genetic evidence for recent
demographic fluctuations was evaluated for the microsatellite loci using BOTTLENECK 1.2.02
], applying three models of mutation (the infinite allele model, IAM; [
]; the stepwise
mutation model, SMM [
], and two-phase model, TPM [
]); parameters for the TPM
followed published guidelines [
]. Significance was assessed using a Wilcoxon sign-rank test.
Significant heterozygosity deficit values relative to the number of alleles indicate recent
population growth, whereas heterozygosity excess relative to the number of alleles indicates a recent
population bottleneck . To assess recent population demography using sufficient numbers
of representative individuals, we pooled data from TAN, POR, and YUK into a single
population, and similarly pooled NEAR and COMM.
Evidence of more historical population change trends was evaluated using mtDNA
sequence data, employing FLUCTUATE [
], which estimates a population growth parameter, g,
incorporating coalescence theory. Positive values of g suggest population growth over deeper
time frames; negative values suggest population decline. Because this method may be upwardly
], we used g to indicate population growth if g > 3 SD(g). Fluctuations in historical
population size were also inferred from Tajima's D and Fu's FS, generated using ARLEQUIN
to test for locus neutrality. Additional details of these analyses are provided in S1 Methods.
Genetic diversity, population, and regional differentiation
Microsatellite data. Standard genetic diversity metrics for each sampling locale for
microsatellite data are summarized in Table 1. All loci were in HWE except two, NVHfp54
(χ2 = 46.24, df = 22, P = 0.002) and NVHfp92-1 (χ2 = 96.98, df = 28, P < 0.001), which differed
significantly from HWE expectations due to deviations in 2 and 6 of the 14 populations,
respectively. Analyses using MICRO-CHECKER 2.2.3 [
] suggested the homozygote excess
observed in NVHfp92-1 (in TAN, MVC, COL, SJI, and COMM) may be attributable to null
alleles, although homozygote excess was not observed in the remaining 11 populations. No
population deviated from HWE overall, and we detected no evidence of significant linkage
8 / 25
lowest in F. p. tundrius. Microsatellite genotype data are accessioned at the USGS-ASC data
Overall, significant differentiation at microsatellite loci was observed across traditional
metrics (FST = 0.101, P < 0.001; RST = 0.073, P < 0.003; χ2 = infinity, df = 22, P < 0.001). All
populations comprising F. p. pealei were significantly differentiated from those comprising both F.
p. anatum and F. p. tundrius (S1 Table). However, only 8 among 42 possible pairwise
comparisons among and within populations representing F. p. tundrius and anatum were significant,
and among these, significant differentiation between populations representing F. p. tundrius
and anatum involved HB and LKAT (S1 Table). By contrast, population pairwise
differentiation within and among populations from across the range of F. p. pealei was observed; all but 5
pairwise comparisons across analyses (all involving island groups of the Aleutian Island
segment or Commander Islands and one involving SJI) were significantly different from zero (S1
Table). SJI was significantly differentiated from all other populations except the single instance
of a non-significant FST value (SJI and NEAR).
Results of Bayesian cluster analyses using STRUCTURE 2.1 [
] indicated the two most
likely models assigning breeding individuals was 2 (ΔK = 757.8, LnPr(X|K) = ±4301.2) or 3
(ΔK = 5.3, LnPr(X|K) = ±4286.5) based on maximizing the ΔK or LnPr(X|K) values,
respectively (Fig 2). The 2-cluster model (Fig 2A) reflected 2 distinct clusters corresponding largely
to individuals sampled from locales within the range F. p. anatum/tundrius, and within F. p.
pealei. The 3-cluster model (Fig 2B) similarly clusters F.p. anatum/tundrius away from F. p.
pealei, but in addition clustered the Aleutian/Commander islands group away from more
easterly distributed populations within F. p. pealei; the latter showed substantial admixture with F.
p. anatum/tundrius populations.
Variance in the distribution of microsatellite alleles was not significantly partitioned at the
regional level in hierarchical AMOVA analyses based on the current subspecies hypotheses
(Model A, Table 2). The greatest regional differentiation was uncovered when populations
representing F. p. tundrius and anatum were pooled into a single group, and F. p. pealei was
partitioned into the eastern (including SJI) and western segments (FCTnuc = 0.093, P < 0.001; see
Model D, Table 2). The CSE tree generated from distances among all populations showed few
nodes with bootstrap support greater than 50% (Panel A in S1 Fig). However, the CSE tree
showed strong bootstrap support (100%) between nodes separating the Aleutian and southeast
and southcentral populations comprising F. p. pealei, and pooled populations comprising of F
p. anatum and tundrius (Panel B in S1 Fig).
MtDNA data. We collected control region sequence data for 169 individuals from
presumptive populations of F. p. tundrius, anatum, pealei, and SJI and observed 15 haplotypes
]. Haplotype CR2 occurred in the highest frequency within almost all populations and
representing 72% of haplotypes overall (Fig 3, S2 Table). Of the remaining 14 haplotypes, 2 were
unique to F. p. pealei, 4 to tundrius, and 1 to SJI (S2 Table). Haplotype diversity was lower in F.
p. pealei than in either tundrius or anatum (Table 1). Nucleotide diversity was higher in F. p.
pealei. Near and Andreanof island groups (F. p. pealei) and Porcupine River (F. p. anatum)
were characterized by a single haplotype, resulting in haplotype and nucleotide diversity values
of zero (Table 1). We observed no significantly positive values of D or FS in any population,
suggesting there is little deviation from selective neutrality in mtDNA (see S1 Methods), save
in one population: COMM showed a significantly positive value of D (P < 0.05; Table 1).
Although fewer significant pairwise population comparisons (based on either χ2 or ϕST)
were observed for the mtDNA relative to the microsatellite data, we nevertheless observed a
greater percentage of significant pairwise population comparisons between populations
representing F. p. pealei and populations representing F. p. anatum and tundrius, versus within (S3
Table). LKAT was involved in the majority of significant comparisons within F. p. anatum,
10 / 25
Fig 2. Genetic structure among Alaskan and western Canadian peregrine falcon populations,
estimated by Bayesian clustering using STRUCTURE, without prior population information. A) most
11 / 25
likely model (K = 2) estimated by maximizing ΔK (ΔK = 757.8, LnPr(X|K) = ±4301.2); B) most likely model
(K = 3), estimated by maximizing LnPr(X|K) (ΔK = 5.3, LnPr(X|K) = ±4286.5).
and SCCOA was involved in the majority of significant comparisons associated with F. p.
pealei (excluding SJI, which was also strongly differentiated from most other populations,
regardless of subspecies membership). Similar to microsatellite data, variance in the distribution of
haplotypes was not significantly partitioned at the regional level in hierarchical AMOVA
analyses based on the current subspecies hypotheses (Model A, Table 2), and similar to analyses
based on microsatellite data, the greatest regional differentiation was uncovered when
populations representing F. p. tundrius and anatum were pooled into a single region, and F. p. pealei
was partitioned into the eastern and western segments (FCTmit = 0.136, P < 0.004; see Model
D, Table 2).
Estimation of levels and polarity in gene flow. M and θ values along with confidence
limits estimated in MIGRATE from microsatellite genotypes and mtDNA were used to
calculate Nem and Nfm (θ M), respectively, and suggested levels of asymmetry in gene flow varied
among populations (Table 3, Table 4); confidence intervals around estimates of effective
number of female migrants were relatively wide (Table 4). Among Alaskan regions, Arctic and
[COL,HB] [TAN,YUK,POR,LKAT,MCV] [SJI, NPAC,SCCOA, ANDR,RAT,NEAR,
[COL,HB] [TAN,YUK,POR,LKAT,MCV, SJI] [NPAC,SCCOA, ANDR,RAT,NEAR,
[COL,HB] [TAN,YUK,POR,LKAT,MCV] [SJI,NPAC,SCCOA] [ANDR,RAT, NEAR,
[COL,HB,TAN,YUK, POR,LKAT,MCV] [SJI,NPAC,SCCOA] [ANDR RAT,NEAR,
[COL,TAN,YUK,POR] [SJI,NPAC,SCCOA] [ANDR,RAT, NEAR,COMM] [MCV,
MtDNA Control Region
[COL,HB] [TAN,YUK, POR,LKAT,MCV] [SJI,NPAC,SCCOA] [ANDR,RAT,NEAR,
[COL, HB, TAN, YUK, POR,LKAT,MCV] [SJI, NPAC, SCCOA] [ANDR, RAT,
[COL,TAN,YUK,POR] [SJI,NPAC,SCCOA] [ANDR,RAT, NEAR,COMM] [MCV,
[COL,TAN,YUK,POR,LKAT, MCV,HB] [SJI,NPAC,SCCOA,ANDR,RAT NEAR,
12 / 25
Fig 3. Unrooted reduced median network illustrating relationships among 15 mtDNA control region haplotypes observed among
169 peregrine falcons assayed from northern North American populations. F. p. tundrius populations are indicated in blue, F. p.
anatum in yellow, F. p. pealei in red (hatched red for NPAC, SOLID red for the Aleutian Island locales), and the population on SJI in orange.
The size of the circle node corresponds to the frequency of each haplotype.
interior boreal populations receive the most immigrants, based on microsatellite data (Table 3)
and mtDNA (Table 4), respectively. Results for both marker types suggest that, on average, the
maritime Alaskan peregrine (F. p. pealei) populations (Aleutian, SCCOA, NPAC) receive the
fewest immigrants relative to the other Alaskan population segments assayed. Gene flow
estimates based on the microsatellite data suggest that the western segment of F. p. pealei
(Aleutians) receives, on average, the lowest immigration levels, generally <1 per generation per
pairwise population comparison among all populations assayed (Table 3). The highest levels of
gene flow (based on Nem) into the Aleutians originated from the North Pacific populations
(NPAC), and effective female gene flow estimates (Nfm) also suggest substantial gene flow
13 / 25
[F. p. anatum]
Interior Alaska LKAT
originating from populations in the eastern portion of the distribution of F. p. pealei (SCCOA,
NPAC), though confidence limits overlap (Table 4). Analyses of microsatellite data suggest
there is asymmetrical gene flow (Nem) from SJI into populations within the range of all three
subspecies (F. p. pealei: NPAC; F. p. anatum: LKAT; and F. p. tundrius: Arctic Alaska and
Canada; see Table 3). Analyses of mtDNA data also suggest there is asymmetrical female gene flow
(Nfm) from SJI into interior Alaska (F. p. anatum; see Table 4). Average evolutionary dispersal
(gene flow) between unaugmented interior Alaskan populations and Arctic Alaska, typically
ascribed to F. p. anatum and F. p. tundrius, respectively, has been largely symmetrical (based
on microsatellite loci, Nem, Table 3) or biased toward Arctic Alaska (based on mtDNA, Nfm,
Table 4). Interior Alaskan populations receive evolutionary dispersal from the western
(Aleutian) segment of F. p. pealei (based on mtDNA, Nfm, only; see Table 4). Signals of significant
asymmetry in gene flow between augmented Canadian populations (represented in this
analysis by LKAT and Arctic Canada [HB and MCV]) and unaugmented Alaskan populations have
been from Arctic Canada into Arctic Alaska, as represented by COL (based on microsatellite
loci, Nem, Table 3), and from interior Alaska into Arctic Canada (based on mtDNA, Nfm,
Table 4). Although estimates of effective female gene flow between Arctic Canada and Arctic
Alaska are high and confidence intervals overlap, net gene flow is largely from Arctic Alaska
into Arctic Canada (Table 4).
Genetic signals of population demography. There was evidence of significant
heterozygosity excess, expected in the case of a recent bottleneck, only in SJI (P < 0.001) and only
14 / 25
under the IAM. Analysis under the SMM and TPM also gave evidence of significant
heterozygosity deficit expected in the case of recent population growth due to the influx of alleles,
in Alaskan F. p. anatum (P < 0.003), represented by pooling data from POR, YUK, and
TAN. After the application of Bonferroni correction, no significant heterozygosity excess
or deficit was observed in falcons sampled from Canada (MCV and LKAT, P > 0.006; HB,
P > 0.008), Arctic Alaska (COL, P > 0.100), or the eastern or Aleutian components of F. p.
pealei (P > 0.006, P > 0.016, respectively).
Significant fluctuations in historical population demography, inferred from tests of
neutrality of the mtDNA data (Table 1), were observed in RAT, YUK, TAN, HB, and COL, viz
significantly negative Fu's FS (P < 0.02) and Tajima's D (P < 0.05). SJI also showed a significantly
negative Fu's FS. FLUCTUATE analyses of mtDNA sequence (Table 1) uncovered a signal of
population increase in two locales within the distribution of F. p. tundrius (COL, HB) and in
peregrines in two locales within the range of anatum in Alaska (YUK and TAN).
15 / 25
Genetic diversity, differentiation, demography, and gene flow
Although inference of relationships among North American subspecies made by Brown et al.
] were based on neutral genetic data from populations that were augmented by introduction
of peregrines with ancestry from exotic locales, our analyses from non-augmented populations
in Alaska corroborate their finding, as well as that of White et al.  that there is no
discernable regional-level genetic differentiation between F. p. anatum and tundrius in high latitude
habitats in western North America. A close relationship among populations representing F. p.
anatum and tundrius is supported by lack of regionally significant variance in allele and
haplotype frequencies and estimates of substantial levels of gene flow between populations ascribed
to one or the other subspecies, the high percentage of shared common haplotypes, and general
lack of diagnostic genetic markers. These results highlight a signal of both historical and
contemporary gene flow between F. p. anatum and F. p. tundrius, even in Alaskan populations,
which were not augmented subsequent to the mid 20th century population decline.
By contrast, and also like Brown et al. [
], we found that F. p. pealei is differentiated at the
regional level from populations currently ascribed to the other two subspecies. Traditional
population differentiation metrics based on both marker types showed significant structuring
among most populations comprising F. p. pealei and those comprising the mainland and
Arctic Alaskan and Canadian populations, supporting a distinction of peregrines within the
distribution of F. p. pealei. As well, Bayesian analyses of population structure suggested peregrine
falcons in populations belonging to F. p. pealei cluster away from populations ascribed to the
other two subspecies.
A novel finding is that of substructuring within F. p. pealei; the Aleutian segment is
genetically differentiated from the North Pacific segment (the Alexander and Haida Gwaii
archipelagos) as well as from populations along south central coastal Alaska (Icy Bay, Kenai Fjords,
Kodiak Island), for which subspecies attribution is unclear [
]. Peregrine falcons occupying
the San Juan Islands of Washington are distinct from other populations occupying habitats
ascribed to F. p. pealei, and from mainland Alaskan and Canadian populations currently
ascribed to F. p. anatum; nevertheless, falcons occupying the San Juan Islands consistently
cluster with F. p. pealei in AMOVA tests. Coastal Alaskan populations comprising the eastern
arm of F. p. pealei, including locales along the northern Gulf of Alaska's Lost Coast (SCCOA:
Yakutat and Kenai Fjords) that were recently attributed to F. p. anatum based on morphology
and migratory behavior [
], appear to include individuals with mixed ancestry, as does SJI.
For SJI, this is possibly due to a trend toward female-biased gene flow from populations
occupying Arctic Alaska, although confidence intervals for immigration and emigration overlap.
For peregrines occupying the Lost Coast of Alaska (SCCOA), this may be due to female-biased
gene flow from interior boreal Alaska populations. Clarification of local relationships for SJI
will require additional sampling of temperate latitude peregrine falcons. Similarly, elucidation
of local relationships of south central Alaskan maritime populations may emerge upon
increased sampling. Given we have detected gene flow between potentially augmented
Canadian populations and unaugmented Alaskan populations, and possible admixture in certain
portions of the range of F. p. pealei (e.g., the Lost Coast of Alaska), we suggest a comprehensive
survey of North American populations be conducted to better understand the relationships
among contemporary North American peregrine falcons.
Surprisingly, despite the well-characterized DDT-induced decline of peregrine falcon
populations in North America during the mid 20th century, analyses using the approach of Cornuet
and Luikart [
] found no evidence for a recent population bottleneck in peregrines occupying
unaugmented habitats in Alaska, including peregrines sampled from boreal and Arctic locales
16 / 25
within the distribution of the two subspecies (F. p. anatum and tundrius, respectively) that
were listed for several decades as endangered or threatened under two Acts of the U. S.
Congress. However, heterozygosity deficit, a genetic signal of demographic increase attributed to
the influx of alleles [
], was observed in boreal (F. p. anatum) peregrines in Alaska. This is
consistent with data emerging from long-term field surveys that have documented an increase
in the number of occupied territories along the upper Yukon River from 1970 to 2012 [
Although average generation times for western high latitude peregrine falcons are not known,
given an average generation time estimate of 6.8 years for this species [
], the increase on the
Yukon River would have occurred across approximately 6 generations. We note also that
falcons occupying interior boreal and Arctic regions of Alaska show growth signals (g) consistent
with population expansion over more historical time periods (Table 1). It is unclear why the
analysis using BOTTLENECK failed to uncover a genetic signature (i.e., heterozygosity excess)
of recent population fluctuation in Colville River peregrines, which, like peregrine nesting
along the Yukon River, declined in the mid-1950s, then recovered over a 40-year period [
Hailer et al. [
] suggest that species with long generation times, such as raptors, may be
buffered from rapid loss of genetic diversity when their populations pass through a bottleneck. As
well, based on surveys conducted along the Colville River between 1981 and 2002, the number
of occupied nest sites and cliffs by nesting peregrine falcons was less than 30 for only a few
years, increasing to around 50 nest sites by 1988 [
]. This suggests a severe bottleneck (just
over 25 effective breeders) that nevertheless 1) was possibly of short duration, spanning only a
few generations at most; 2) occurred only a few generations prior to the collection of the
samples from the Colville population used in the BOTTLENECK analysis (mostly 2001±2002; see
]), and 3) occurred in a species that is thought to have never been abundant in North
America, even prior to the mid 20th century decline [
]. Under those circumstances, the
demographic bottleneck may have failed to generate a genetic signature that can be detected using
the methods employed by BOTTLENECK .
However, we note that in a population that has undergone a bottleneck, we would expect to
see certain signatures in the mtDNA, including low nucleotide diversity coupled with high
haplotype diversity, negative Tajima's D, and negative Fu's FS [
]. Haplotype diversity at the
mtDNA control region for the peregrines assayed here is generally low, with only three of the
14 populations with haplotype diversity values greater than 0.50 (Table 1). However, certain
populations, including those representing interior boreal (YUK, TAN) and Arctic Alaskan
(COL) peregrines, demonstrated significantly negative FS and D coupled with low nucleotide
diversity but also signatures of growth (g) (Table 1). Combined, these metrics provide a signal,
albeit weak, of population fluctuations, including possibly population bottlenecks. We
acknowledge, however, the difficulty in parsing the roles of selection and population demography as
inferred from widely-used neutrality metrics, including FS and D [
Although we uncovered moderate population-level structuring overall, gene flow rate
estimates among subspecies and populations, based on both marker types, suggest substantial
gene flow within and among regions. The Aleutian segment of F. p. pealei is notable for its
isolation: although they provide immigrants into the eastern segment of pealei, across generations
and, on average, Aleutian Island populations appear to receive the fewest effective immigrants
(Nem) among assayed populations. Augmentation of Aleutian populations appears to be
largely via female gene flow from other southcentral and southeast Alaskan maritime
populations. Other populations of F. p. pealei occupying habitats along the coast of Alaska, British
Columbia, and Washington are connected via varying levels of evolutionary dispersal with
boreal and Arctic Canadian and Alaskan populations. In particular, the population on the San
Juan Islands represents a source for evolutionary dispersal into interior boreal and Arctic
Canada and Alaska.
17 / 25
Of considerable interest is the finding of gene flow between the Canadian populations,
presumably augmented, and the non-augmented populations in interior boreal and Arctic Alaska,
including estimates of asymmetrical gene flow (Nem) from Arctic Canada into Arctic Alaska,
based on microsatellite data. However, significant signals of asymmetry in gene flow between
augmented Canadian populations and non-augmented Alaskan populations have been in the
form of effective female gene flow (Nfm) from interior boreal Alaska into Arctic Canada. We
note that the female peregrines typically disperse farther than males from natal areas to breed
], suggesting female bias in gene flow. Estimates of effective female gene flow between
Arctic Canada and Arctic Alaska are high, but with overlapping confidence intervals; still, net
gene flow appears to be out of the Arctic Alaska, suggesting that boreal and Arctic Alaska have
represented a source of dispersing females for Arctic Canada populations. We caution that
gene flow values provide a rough estimation of the rate of gene exchange among population
groupings, and should not be viewed as precise measures, but rather within the context of
whether certain regional groupings within northwestern North American peregrine falcons
are substantially isolated from other groupings, as observed elsewhere [
The three recognized subspecies of F. peregrinus examined here displayed similar levels of
genetic diversity overall, although F. p. pealei showed lower levels of mtDNA haplotype
diversity, driven largely by low number of haplotypes in the Aleutian segment of the subspecies.
Low genetic diversity cannot be attributed to a small population size; population size of
peregrines in the Aleutian Islands is thought to be similar to elsewhere within the range of F. p.
], and the population size of peregrines within the range of pealei is higher than found
within the geographic range of either anatum or tundrius. Although Murie [
] estimated that
100 breeding pairs occupied the Aleutian Island chain, densities in the Rat and Near island
groups and Buldir [
] and the eastern portion of the Aleutian Archipelago [
] led Ambrose
et al. [
] to estimate that approximately 300 pairs occupied the Aleutians. Based on
extrapolation from extensive surveys on the Rat Island group [
], more recent estimates of abundance
across the North American portion of the Aleutian chain range from 375±580 breeding pairs
]. The region between the Kenai Peninsula to southeast Alaska is estimated to hold more
than 140 territories, and White [
] tallied 69 eyries along the Alaska Peninsula and on Kodiak
Island. An axiom in conservation biology is that island populations are more prone to
extinction than mainland populations [88±91], and island populations often demonstrate lower
levels of genetic diversity and increased inbreeding [
]. Insular populations of peregrine falcons
on oceanic island archipelagos elsewhere are characterized by lower levels of genetic diversity
than mainland populations, and in at least one case these low levels of variability may have
fostered inbreeding depression [
]. While our results showed that average levels of genetic
diversity in populations in the Aleutian Island archipelago are lower than elsewhere, a significantly
positive inbreeding coefficient was not uncovered. Lower levels of genetic variability in the
Aleutian distribution of F. p. pealei are likely facilitated, at least in part, by the isolation of the
Aleutian Island archipelago and restricted augmentation via gene flow from mainland
populations and attendant relative effective population size.
Our analyses, similar to those of Brown et al. [
], fail to support the current subspecies
designations as applied to peregrine falcons occupying high latitude North America, particularly
with regard to falcons sampled from within the distribution currently ascribed to either
tundrius or anatum. The concept of `subspecies' remains controversial, particularly within
ornithology [2,3,93±96]. Presumably, an evolutionary basis for taxonomy requires the application
of genetic tools to questions involving subspecies designations. As pointed out by Brown et al.
], however, genetic criteria for subspecies uniqueness are not universal and depend upon the
investigator and metric used for subspecies diagnosis. For example, Zink [
] uses monophyly
of mtDNA haplotypes as a criterion for attributing subspecies. Applied in this case, all
18 / 25
peregrine falcons in the northwestern high latitudes of North America would belong to a single
subspecies. Under both Barrowclough's [
] criterion of `predictiveness,' which posits that
subspecies designations should allow the prediction of a suite of quantifiable character states,
and Crandall et al.'s [
] criteria of `ecological and genetic exchangeability', the eastern and
western distributions of F. p. pealei, which are genetically and to some degree morphologically
], would each qualify as separate subspecies, whereas F. p. anatum and
tundrius would be pooled into a single subspecies. We note that F. p. pealei in both segments
(Aleutian and eastern) occupy island archipelagos (albeit one more isolated from mainland
populations than the other two) and unlike the boreal and Arctic peregrines tend to show little
to no migratory behavior, a characteristic that may contribute to greater population and
regional structuring, as seen in other avian species [
] and particularly in insular
Our genetic survey provides insights into the status and delineation of natural populations as
well as a better understanding of gene flow among unaugmented high latitude peregrine
falcons. While we make no attempt at formal taxonomic revision, our data suggest that peregrine
falcons occupying habitats in Alaska and the North Pacific coast of North America belong to
two distinct regional groupings±a coastal grouping (F. p. pealei) and a boreal/Arctic grouping
(currently F. p. anatum and tundrius), each comprised of discrete populations that are
variously intra-regionally connected. Gene flow estimates indicating that distant high-latitude
populations in Canada provide immigrants into and receive immigrants from mainland
Alaskan populations ascribed to both previously listed subspecies. This suggests that management
prescriptions applied at a distance may have unforeseen consequences in locally-managed
populations of species with high gene flow potential, even in species characterized by
reasonably strong philopatry. This highlights the importance of inter-regional and international
cooperation in management and conservation in such species [
]. Differentiation among
populations within the distribution of F. p. pealei is stronger than among boreal/Arctic
populations, possibly due to decreased or absent migratory tendencies that restrict gene flow into the
Aleutian Island segment of F. p. pealei. Thus, peregrines residing on islands in the Aleutian
Archipelago likely remain the least genetically impacted by DDT-induced population
fluctuations that fostered the mid 20th century decline of peregrines in North America and
subsequent augmentation by non-native subspecies.
S1 Methods. Supplementary material showing details of laboratory methods and data
S1 Fig. Neighbor-joining network based on CSE distances and 11 microsatellite loci.
Illustrated are relationships among A) 14 populations of peregrine falcon in high latitudes habitats
in Canada, Alaska and the Commander Islands, Russia; and B) three subspecies (F. p. anatum,
F. p. tundrius, and F. p. pealei). Populations within the distribution of F. p. pealei are designated
with red branches, within the distribution of F. p. anatum with yellow branches, and within
the distribution of F. p. tundrius with blue branches. Values in nodes designate bootstrap
values, based on 1000 permutations (only values > 50% are shown).
19 / 25
S1 Table. Results of traditional population differentiation analyses, based on microsatellite
fragment data. Shown are values for pairwise χ2 (above diagonal) and FST (below diagonal)
comparisons among populations of F. peregrinus within Alaska, based on data from 11
microsatellite loci. See manuscript for locales associated with acronyms. Values in bold, and values
listed as 1 (= infinity) indicate significant differences in the distribution of alleles/haplotypes
or variance in allelic/haplotypic frequency, respectively, following Bonferroni corrections (α =
0.0045). The population on the San Juan Islands (SJI) is considered a contact zone between
continental populations (F. p. anatum) and maritime populations (F. p. pealei).
S2 Table. Distribution of mitochondrial DNA haplotypes among peregrine falcon
populations in high latitude western North America. Populations augmented with non-native
subspecies following the population decline are shaded.
S3 Table. Results of traditional population differentiation analyses, based on
mitochondrial DNA control region sequences. Shown are values for pairwise χ2 (above diagonal) and
ϕST (below diagonal) comparisons among populations of F. peregrinus within Alaska, based on
sequence data from the mitochondrial DNA control region. See manuscript for locales
associated with acronyms. Values in bold or (for χ2) designated as 1 (infinity) indicate significant
differences in the distribution of haplotypes or variance in haplotype frequency (P < 0.05),
respectively. The population on the San Juan Islands (SJI) is considered a contact zone between
continental populations (F. p. anatum) and coastal populations (F. p. pealei).
We thank Katie Sciuto and Devon Berthiaume for laboratory assistance, Lee Tibbitts and
Jason Fristensky for assistance with graphics, and Barbara Pierson for curation of data.
Professor Potopov of the Museum of the Zoological Institute in St. Petersburg, Russia, kindly allowed
sampling of museum specimens of peregrines of the Commander Islands. Technological
support was provided by the University of Alaska Life Science Informatics computer cluster (NIH
P20RR016466). We thank John Pearce and two anonymous reviewers for providing reviews of
earlier drafts. The findings and conclusions in this article are those of the authors and do not
necessarily represent the views of the Fish and Wildlife Service. Any use of trade, firm, or
product names is for descriptive purposes only and does not imply endorsement by the U. S.
Conceptualization: Sandra L. Talbot, Sarah A. Sonsthagen, Clayton M. White.
Data curation: Sandra L. Talbot, George K. Sage, Meg C. Gravley.
Formal analysis: Sandra L. Talbot, George K. Sage, Sarah A. Sonsthagen, Meg C. Gravley.
Funding acquisition: Sandra L. Talbot, Ted Swem, Jeffrey C. Williams, Jonathan L. Longmire,
Skip Ambrose, Melanie J. Flamme, Stephen B. Lewis, Laura Phillips, Clifford Anderson,
Clayton M. White.
Investigation: Sandra L. Talbot, George K. Sage, Sarah A. Sonsthagen, Jonathan L. Longmire.
Methodology: Sandra L. Talbot, George K. Sage, Sarah A. Sonsthagen.
20 / 25
Project administration: Sandra L. Talbot.
Resources: Sandra L. Talbot, Ted Swem, Jeffrey C. Williams, Jonathan L. Longmire, Skip
Ambrose, Melanie J. Flamme, Stephen B. Lewis, Laura Phillips, Clifford Anderson, Clayton
Supervision: Sandra L. Talbot, Clayton M. White.
Validation: Sandra L. Talbot, George K. Sage, Sarah A. Sonsthagen.
Visualization: Sandra L. Talbot, Sarah A. Sonsthagen, Meg C. Gravley.
Writing ± original draft: Sandra L. Talbot, Sarah A. Sonsthagen, Meg C. Gravley, Clayton M.
Writing ± review & editing: Sandra L. Talbot, George K. Sage, Sarah A. Sonsthagen, Meg C.
Gravley, Ted Swem, Jeffrey C. Williams, Jonathan L. Longmire, Skip Ambrose, Melanie J.
Flamme, Stephen B. Lewis, Laura Phillips, Clifford Anderson, Clayton M. White.
21 / 25
22 / 25
23 / 25
24 / 25
1. Lessa EP , Cook JA , Patton JL . Genetic footprints of demographic expansion in North America, but not Amazonia, during the Late Quaternary . Proc Natl Acad Sci U S A . 2003 ; 100 : 10331 ± 10334 . https:// doi.org/10.1073/pnas.1730921100 PMID: 12913123
2. Phillimore AB , Owens IPF . Are subspecies useful in evolutionary and conservation biology? Proc R Soc B Biol Sci . 2006 ; 273 : 1049 ± 1053 . https://doi.org/10.1098/rspb. 2005 .3425 PMID: 16600880
3. Zink RM . The role of subspecies in obscuring avian biological diversity and misleading conservation policy . Proc R Soc B Biol Sci . 2004 ; 271 : 561 ± 564 . https://doi.org/10.1098/rspb. 2003 .2617 PMID: 15156912
4. Johnson JA , Burnham KK , Burnham WA , Mindell DP . Genetic structure among continental and island populations of gyrfalcons . Mol Ecol . 2007 ; 16 : 3145 ± 3160 . https://doi.org/10.1111/j. 1365 - 294X . 2007 . 03373 . x PMID : 17651193
5. Hull JM , Mindell DP , Talbot SL , Kay EH , Hoekstra HE , Ernest HB . Population structure and plumage polymorphism: the intraspecific evolutionary relationships of a polymorphic raptor, Buteo jamaicensis harlani . BMC Evol Biol . 2010 ; 10 : 224 . https://doi.org/10.1186/ 1471 -2148-10-224 PMID: 20650000 6 . White CM , Sonsthagen SA , Sage GK , Anderson C , Talbot SL . Genetic relationships among some subspecies of the Peregrine Falcon (Falco peregrinus L.), inferred from mitochondrial DNA control-region sequences . Auk . 2013 ; 130 : 78 ± 87 . https://doi.org/10.1525/auk. 2012 .11173
7. Brown JW , van Coeverden de Groot PJ , Birt TP , Seutin G , Boag PT , Friesen VL . Appraisal of the consequences of the DDT-induced bottleneck on the level and geographic distribution of neutral genetic variation in Canadian peregrine falcons, Falco peregrinus . Mol Ecol . 2007 ; 16 : 327 ± 343 . https://doi. org/10.1111/j. 1365 - 294X . 2007 . 03151 . x PMID : 17217348
8. GalvaÂn I , Jorge A . The rusty plumage coloration of juvenile gyrfalcons is produced by pheomelanin and its expression is affected by an intracellular antioxidant . J Raptor Res . 2015 ; 49 : 59 ± 65 . https:// doi.org/10.3356/jrr-14 -00018 . 1
9. Marshall HD , Ritland K. Genetic diversity and differentiation of Kermode bear populations . Mol Ecol . 2002 ; 11 : 685 ± 697 . PMID: 11972757
10. Gray SM , McKinnon JS . Linking color polymorphism maintenance and speciation . Trends Ecol Evol . 2007 ; 22 : 71 ± 79 . https://doi.org/10.1016/j.tree. 2006 . 10 .005 PMID: 17055107
11. Gangoso L , Roulin A , Ducrest A-L , Grande JM , Figuerola J . Morph-specific genetic and environmental variation in innate and acquired immune response in a color polymorphic raptor . Oecologia . 2015 ; 178 : 1113 ± 1123 . https://doi.org/10.1007/s00442-015 -3306-6 PMID: 25834999
12. Mundy NI . A window on the genetics of evolution: MC1R and plumage colouration in birds . Proc R Soc B Biol Sci . 2005 ; 272 : 1633 ± 1640 . https://doi.org/10.1098/rspb. 2005 .3107 PMID: 16087416
13. Cooch G . Ecological aspects of the blue-snow goose complex . Auk . 1961 ; 78 : 72 ± 89 . https://doi.org/ 10.2307/4082236
14. Zhan XJ , Dixon A , Fox NC , Bruford MW . Missense SNP of the MC1R gene is associated with plumage variation in the Gyrfalcon (Falco rusticolus) . Anim Genet . 2012 ; 43 : 460 ± 462 . https://doi.org/10.1111/j. 1365- 2052 . 2011 . 02263 . x PMID : 22497627
15. Johnson JA , Ambers AD , Burnham KK . Genetics of plumage color in the gyrfalcon (Falco rusticolus): analysis of the melanocortin-1 receptor gene . J Hered . 2012 ; 103 : 315 ± 321 . https://doi.org/10.1093/ jhered/ess023 PMID: 22504110
16. American Ornithologists' Union. Checklist of North American Birds. 5th ed. American Ornithologists Union; 1957 .
17. Cramp S , Simmons KEL . The birds of the western Palearctic, volume II . Oxford: Oxford University Press; 1980 .
18. Potapov E , Sale R. The gyrfalcon . Yale University Press; 2005 .
19. White CM , Clum NJ , Cade TJ , Hunt WG . Peregrine falcon (Falco peregrinus) [Internet] . Poole A , editor. Birds of North America Online. Ithaca ; 2002 . https://doi.org/10.2173/bna.660
20. Enderson JH , Heinrich W , Kiff L , White CM . Population changes in North American peregrines . Trans North Am Wildl Nat Resour Conf . 1995 ; 60 : 142 ± 161 .
21. Heinrich W. Peregrine Falcon recovery in the continental United States, 1974 ±1999, with notes on related programs of The Peregrine Fund . In: Siclicki J , Mizera T , editors. Peregrine Falcon populations - status and perspectives in the 21st century . Warsaw, Poland: Tural/Poznan University of Life Sciences Press; 2009 . pp. 431 ± 444 .
22. White CM , Boyce Jr. DA. An overview of peregrine falcon subspecies . In: Cade TJ, Enderson JH , Thelander CG , White CM , editors. Peregrine Falcon populations: their management and recovery. Boise , Idaho: The Peregrine Fund; 1988 . pp. 789 ± 810 .
23. White CM . Diagnosis and relationships of the North American tundra-inhabiting peregrine falcons . Auk . 1968 ; 85 : 179 ± 191 .
24. Kiff LF. CommentaryÐchanges in status of the peregrine in North America . In: Cade TJ, Enderson JH , Thelander CG , White CM , editors. Peregrine Falcon populations: their management and recovery. Boise , Idaho: The Peregrine Fund; 1988 . pp. 123 ± 139 .
25. Cade TJ , Lincer JL , White CM , Roseneau DG , Swartz LG . DDE residues and eggshell changes in Alaskan falcons and hawks . Science . 1971 ; 172 : 955 ± 957 . PMID: 5573569
26. Cade TJ , Enderson JH , Thelander CG , White CM , editors. Peregrine falcon populations: their management and recovery . Boise, Idaho: The Peregrine Fund; 1988 .
27. Ambrose S , Florian C , Ritchie RJ , Payer D , O'Brien RM . Recovery of American peregrine falcons along the Upper Yukon River , Alaska, 1977 ± 2014 . J Wildl Manage . 2016 ; 80 : 1± 12 . https://doi.org/10. 1002/jwmg.1058
28. Haugh J . Population and reproductive changes in Alaskan Arctic peregrines . Trends Genet . 1976 ; 90 : 359 ± 361 . https://doi.org/10.1016/j.tig. 2014 . 09 .010
29. Cade TJ , Burnham WA , editors. Return of the peregrine: a North American saga of tenacity and teamwork . Boise, Idaho: The Peregrine Fund; 2003 .
30. Fish US and Service Wildlife . Endangered and threatened wildlife and plants; final rule to remove the American Peregrine falcon from the federal list of endangered and threatened wildlife . Federal Register . 1999 . pp. 46542 ± 46558 .
31. Tordoff HB , Redig PT . Role of genetic background in the success of reintroduced peregrine falcons . Conserv Biol . 2001 ; 15 : 528 ± 532 . https://doi.org/10.1046/j.1523- 1739 . 2001 . 015002528 .x
32. Bruggeman JE , Swem T , Andersen DE , Kennedy PL , Nigro D. Multi-season occupancy models identify biotic and abiotic factors influencing a recovering Arctic Peregrine Falcon Falco peregrinus tundrius population . Ibis (Lond 1859 ). 2016 ; 158 : 61 ± 74 . https://doi.org/10.1111/ibi.12313
33. White CM , Cade TJ. Long-term trends in Peregrine populations in Alaska . In: Chancellor RD, editor. Proceedings of the First World Conference on Birds of Prey . Vienna 1975: International Council for Bird Preservation , London; 1976 . pp. 63 ± 72 .
34. Mesta R , Swem T , Lawrence S. Endangered and threatened wildlife and plants; advance notice of a proposal to remove the American Peregrine Falcon from the list of endangered and threatened wildlife . Fed Regist . 1995 ; 60 : 34406 ± 34409 .
35. Jacobsen F , Nesje M , Bachmann L , Lifjeld JT . Significant genetic admixture after reintroduction of Peregrine Falcon (Falco peregrinus) in Southern Scandinavia . Conserv Genet . 2008 ; 9 : 581 ± 591 . https:// doi.org/10.1007/s10592-007-9373-4
36. Ambrose S. Peregrine falcon surveys in interior and northern Alaska , 1990 . U.S. Fish and Wildlife Service . Fairbanks, Alaska; 1990 .
37. Shank CC , Bromley RG , Poole KG . Increase in breeding population of tundra peregrine falcons in the central Canadian arctic . Wilson Bull . 1993 ; 105 : 188 ± 190 .
38. Holroyd GL , Banasch U. The 1990 Canadian peregrine falcon (Falco peregrinus) survey . J Raptor Res . 1996 ; 30 : 145 ± 156 .
39. MacCluskie M , Ambrose S , Florian C , Flamme M . The American Peregrine Falcons of Yukon-Charley Rivers National Preserve, Alaska . Alaska Park Sci. 2010 ; 9 : 28 ± 31 .
40. Rowell P , Stepnisky DP . Status of the peregrine falcon (Falco peregrinus anatum) in Alberta . Wildlife Status Report . 1997 .
41. Fair JM , Paul E , Jones J , editors. Guidelines to the use of wild birds in research . 3rd ed. The Ornithological Council . Washington D.C.: The Ornithological Council; 2010 .
42. Pierson BJ . Peregrine Falcon (Falco peregrinus) genetic data, Alaska, Canada and Russia ( 1880 ± 2012 ) [Internet]. U.S. Geological Survey; 2017 . Available: https://dx.doi.org/10.5066/F7F18WV0
43. Longmire JL , Lewis AK , Brown NC , Buckingham JM , Clark LM , Jones MD , et al. Isolation and molecular characterization of a highly polymorphic centromeric tandem repeat in the family Falconidae . Genomics . 1988 ; 2 : 14 ± 24 . PMID: 3384438
44. Handel CM , Pajot LM , Talbot SL , Sage GK . Use of buccal swabs for sampling DNA from nestling and adult birds . Wildl Soc Bull . 2006 ; 34 : 1094 ± 1100 .
45. Nesje M , Røed KH , Lifjeld JT , Linberg P , Steen O . Genetic relationships in the peregrine falcon (Falco peregrinus) analysed by microsatellite DNA markers . Mol Ecol . 2000 ; 9 : 53 ± 60 . PMID: 10652075
46. Johnson JA , Talbot SL , Sage GK , Burnham KK , Brown JW , Maechtle TL , et al. The use of genetics for the management of a recovering population: temporal assessment of migratory peregrine falcons in North America . PLoS One . 2010 ; 5: e14042 . https://doi.org/10.1371/journal.pone. 0014042 PMID: 21124969
47. Talbot SL , Palmer AG , Sage GK , Sonsthagen SA , Swem T , Brimm DJ , et al. Lack of genetic polymorphism among peregrine falcons Falco peregrinus of Fiji . J Avian Biol . 2011 ; 42 : 415 ± 428 . https://doi. org/10.1111/j. 1600 - 048X . 2011 . 05280 .x
48. Bandelt H-J , Forster P , Sykes BC , Richards MB . Mitochondrial portraits of human populations using median networks . Genetics . 1995 ; 141 : 743 ± 753 . PMID: 8647407
49. Excoffier L , Laval G , Schneider S. Arlequin (version 3.0): an integrated software package for population genetics data analysis . Evol Bioinform Online . 2005 ; 1 : 47 ± 50 . https://doi.org/10.1111/j.1755- 0998 . 2010 . 02847 .x
50. Rousset F. GENEPOP '007: a complete re-implementation of the GENEPOP software for Windows and Linux . Mol Ecol Resour . 2008 ; 8 : 103 ± 106 . https://doi.org/10.1111/j.1471- 8286 . 2007 . 01931 .x PMID: 21585727
51. Goudet J. FSTAT : a program to estimate and test gene diversities and fixation indices [Internet]. 2001 . Available: http://www2.unil.ch/popgen/softwares/fstat.htm
52. Fu Y-X. Statistical tests of neutrality of mutations against population growth, hitchhiking and background selection . Genetics . 1997 ; 147 : 915 ± 925 . PMID: 9335623
53. Tajima F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism . Genetics . 1989 ; 123 : 585 ± 595 . doi:PMC1203831 PMID: 2513255
54. Weir BS , Cockerham CC . Estimating F-statistics for the analysis of population structure . Evolution (N Y) . 1984 ; 38 : 1358 ± 1370 .
55. Slatkin M. A measure of population subdivision based on microsatellite allele frequencies . Genetics . 1995 ; 139 : 457 ± 462 . PMID: 7705646
56. Goudet J , Raymond M , De MeeuÈs T , Rousset F. Testing differentiation in diploid populations . Genetics . 1996 ; 144 : 1933 ± 1940 . https://doi.org/10.1111/j.1471- 8286 . 2007 . 01769 . x PMID : 8978076
57. Fisher RA . Statistical methods for research workers . 14th ed. Edinburgh: Oliver and Boyd; 1970 .
58. Manly BFJ . The statistics of natural selection on animal populations. London: Chapman and Hall; 1985 .
59. Posada D , Crandall KA . MODELTEST: testing the model of DNA substitution . Bioinformatics . 1998 ; 14 : 817 ± 818 . https://doi.org/10.1063/1.2218048 PMID: 9918953
60. Excoffier L , Smouse PE , Quattro JM . Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data . Genetics . 1992 ; 131 : 479 ± 491 . https://doi.org/10.1007/s00424-009 -0730-7 PMID: 1644282
61. Pritchard JK , Stephens M , Donnelly P . Inference of population structure using multilocus genotype data . Genetics . 2000 ; 155 : 945 ± 959 . https://doi.org/10.1111/j.1471- 8286 . 2007 . 01758 . x PMID : 10835412
62. Evanno G , Regnaut S , Goudet J . Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study . Mol Ecol . 2005 ; 14 : 2611 ± 2620 . https://doi.org/10.1111/j. 1365 - 294X . 2005 . 02553 . x PMID : 15969739
63. Cavalli-Sforza LL , Edwards AWF . Phylogenetic analysis. Models and estimation procedures . Am J Hum Genet . 1967 ; 19 : 233 ± 257 . Available: http://www.pubmedcentral.nih.gov/articlerender.fcgi? artid=1706274&tool=pmcentrez&rendertype=abstract PMID: 6026583
64. Saitou N , Nei M. The neighbor-joining method: a new method for reconstructing phylogenetic trees . Mol Biol Evol . 1987 ; 4 : 406 ± 425 . PMID: 3447015
65. Langella O . Populations 1.2 .30: population genetic software [Internet]. Gif-sur- Yvette , France: Laboratoire Evolution, GeÂnomes et SpeÂciation; 1999 . Available: http://bioinformatics.org/~tryphon/ populations/
66. Beerli P , Felsenstein J . Maximum-likelihood estimation of migration rates and effective population numbers in two populations using a coalescent approach . Genetics . 1999 ; 152 : 763 ± 773 . https://doi. org/10.1073/pnas.081068098 PMID: 10353916
67. Beerli P , Felsenstein J . Maximum likelihood estimation of a migration matrix and effective population sizes in n subpopulations by using a coalescent approach . Proc Natl Acad Sci U S A . 2001 ; 98 : 4563 ± 4568 . https://doi.org/10.1073/pnas.081068098 PMID: 11287657
68. Cornuet JM , Luikart G. Description and power analysis of two tests for detecting recent population bottlenecks from allele frequency data . Genetics . 1996 ; 144 : 2001 ± 2014 . PMID: 8978083
69. Maruyama T , Fuerst PA . Population bottlenecks and nonequilibrium models in population genetics. II. Number of alleles in a small population that was formed by a recent bottleneck . Genetics . 1985 ; 111 : 675 ± 689 . PMID: 4054612
70. Ohta J , Kimura M. A model of mutation appropriate to estimate the number of electrophoretically detectable alleles in a finite population . Genet Res . 1973 ; 22 : 201 ± 204 . PMID: 4777279
71. Di Rienzo A , Peterson AC , Garza JC , Valdes AM , Slatkin M , Freimer NB . Mutational processes of simple-sequence repeat loci in human populations . Proc Natl Acad Sci U S A . 1994 ; 91 : 3166 ± 3170 . https://doi.org/10.1073/pnas.91.8.3166 PMID: 8159720
72. Garza JC , Williamson EG . Detection of reduction in population size using data from microsatellite loci . Mol Ecol . 2001 ; 10 : 305 ± 318 . PMID: 11298947
73. Piry S , Luikart G , Cornuet J-M. BOTTLENECK : a computer program for detecting recent reductions in the effective population size using allele frequency data . J Hered . 1999 ; 90 : 502 ± 503 . https://doi.org/ 10.1093/jhered/90.4. 502
74. Kuhner MK , Yamato J , Felsenstein J . Maximum likelihood estimation of population growth rates based on the coalescent . Genetics . 1998 ; 149 : 429 ± 434 . PMID: 9584114
75. Van Oosterhout C , Hutchinson WF , Wills DPM , Shipley P . MICRO-CHECKER: software for identifying and correcting genotyping errors in microsatellite data . Mol Ecol Notes . 2004 ; 4 : 535 ± 538 . https://doi. org/10.1111/j.1471- 8286 . 2004 . 00684 .x
76. Petit RJ , el Mousadik A , Pons O . Identifying populations for conservation on the basis of genetic markers . Conserv Biol . 1998 ; 12 : 844 ± 855 .
77. Lewis SB , Kissling ML . Clarifying subspecies of peregrine falcons along the lost coast of Alaska . J Raptor Res . 2015 ; 49 : 367 ± 375 .
78. Bird Life International. Falco peregrinus . In: The IUCN Red List of Threatened Species 2015 : e. T45354964A80472203 [Internet]. 2015 . Available: http://dx.doi.org/10.2305/IUCN.UK. 2015 - 4 .RLTS. T45354964A80472203.en
79. Hailer F , Helander B , Folkestad AO , Ganusevich S a , Garstad S , Hauff P , et al. Bottlenecked but longlived: high genetic diversity retained in white-tailed eagles upon recovery from population decline . Biol Lett . 2006 ; 2 : 316 ± 319 . https://doi.org/10.1098/rsbl. 2006 .0453 PMID: 17148392
80. Tajima F. The effect of change in population size on DNA polymorphism . Genetics . 1989 ; 123 : 597 ± 601 . PMID: 2599369
81. Zeng K , Fu Y-X , Shi S , Wu C -I. Statistical tests for detecting positive selection by utilizing high-frequency variants . Genetics . 2006 ; 174 : 1431 ± 1439 . https://doi.org/10.1534/genetics.106.061432 PMID: 16951063
82. Ambrose RE , Ritchie RJ , White CM , Schempf PF , Swem T , Dittrick R . Changes in the status of Peregrine Falcon populations in Alaska . In: Cade TJ, Enderson JH , Thelander CG , White CM , editors. Peregrine Falcon populations: their management and recovery. Boise , Idaho: The Peregrine Fund; 1988 . pp. 73 ± 82 .
83. Murie OJ . Fauna of the Aleutian Islands and Alaska Peninsula . North Am Fauna. Washington; 1959 ; 61 : 1± 364 .
84. White CM . The 1975 North American peregrine falcon survey: Islands Aleutian . Fyfe RW , Temple SA , Cade TJ , editors. Can F Nat . 1976 ; 90 : 262 ± 264 .
85. Early TJ . Abundance and distribution of breeding raptors in the Aleutian Islands, Alaska . In: Ladd WN , Schempf PF , editors. Raptor management and biology in Alaska and western Canada . Anchorage, AK: U. S. Fish and Wildlife Service; 1982 . pp. 99 ± 111 .
86. White CM . Studies on peregrine falcons in the Aleutian Islands . Raptor Res Rep . 1975 ; 3 : 33 ± 50 .
87. White CM , Cade TJ , Enderson JH . Peregrine falcons of the world . Barcelona, Spain: Lynx Edicions; 2013 .
88. Myers N. The sinking ark: a new look at the problem of disappearing species . 1st ed. Pergamon Press; 1979 .
89. Diamond JM . Historic extinctions: their mechanisms and lessons for understanding prehistoric extinctions . Quaternary Extinctions. Tucson , AZ: University of Arizona Press; 1984 . pp. 824 ± 862 .
90. Smith FDM , May RM , Pellew R , Johnson TH , Walter KR . How much do we know about the current extinction rate? Trends Ecol Evol . 1993 ; 8 : 375 ± 378 . https://doi.org/10.1016/ 0169 - 5347 ( 93 ) 90223 -C PMID: 21236199
91. Frankham R . Do island populations have less genetic variation than mainland populations? Heredity (Edinb) . 1997 ; 78 : 311 ± 327 . https://doi.org/10.1038/sj.hdy.6880980
92. Frankham R . Inbreeding and extinction: island populations . Conserv Biol . 1998 ; 12 : 665 ± 675 .
93. Wilson EO , Brown WL Jr. The subspecies concepts and its taxonomic application . Syst Zool . 1953 ; 2 : 97 ± 111 .
94. Barrowclough GF . Geographic variation, predictiveness, and subspecies . Auk . 1982 ; 99 : 601 ± 603 .
95. Mayr E . Of what use are subspecies? Auk . 1982 ; 99 : 593 ± 595 .
96. Haig SM , Winker K. Avian subspecies: summary and prospectus . Ornithol Monogr . 2010 ; 67 : 172 ± 175 . https://doi.org/10.1525/om. 2010 . 67 .1.172.172
97. Crandall KA , Bininda-Emonds ORP , Mace GM , Wayne RK . Considering evolutionary processes in conservation biology . Trends Ecol Evol . 2000 ; 15 : 290 ± 295 . PMID: 10856956
98. Stangel PW . Genetic differentiation among avian populations . Acta XX Congr Int Ornithol. Christchurch , New Zealand: New Zealand Ornighological Congress Trust Board; 1991 ; 4 : 2442 ± 2453 .
99. Miller MP , Mullins TD , Parrish JW , Walters JR , Haig SM . Variation in migratory behavior influences regional genetic diversity and structure among American Kestrel populations (Falco sparverius ) in North America. J Hered . 2012 ; 103 : 503 ± 514 . https://doi.org/10.1093/jhered/ess024 PMID: 22563129
100. Baker AJ , Dennison MD , Lynch A , le Grand G. Genetic divergence in peripherally isolated populations of chaffinches in the Atlantic islands . Evolution (N Y) . 1990 ; 44 : 981 ± 999 .
101. Tarr CL , Conant S , Fleischer RC . Founder events and variation at microsatellite loci in an insular passerine bird, the Laysan finch (Telespiza cantans) . Mol Ecol . 1998 ; 7 : 719 ± 731 . https://doi.org/10.1046/ j. 1365 - 294x . 1998 . 00385 .x
102. Weckworth B V , Dawson NG , Talbot SL , Flamme MJ , Cook JA . Going coastal: shared evolutionary history between coastal British Columbia and Southeast Alaska wolves (Canis lupus) . PLoS One . 2011 ; 6: e19582 . https://doi.org/10.1371/journal.pone. 0019582 PMID: 21573241