Fire and ice: genetic structure of the Uinta ground squirrel (Spermophilus armatus) across the Yellowstone hotspot.

Abstract

The range of the Uinta ground squirrel, Spermophilus armatus, is centred over one of the most tectonically active regions today, the Yellowstone hotspot. We document the role of Quaternary tectonic and climatic history on the genetic structure of this species by screening museum and extant individuals throughout its range. Phylogeographic, divergence time, and demographic analyses of partial mitochondrial cytochrome b and control region DNA sequences yield insight into the cadence of evolution across three spatiotemporal scales: (i) a relatively deep intraspecific divergence of S. armatus into three lineages coincident with the last major volcanic eruption in the region and maintained by the Snake River Plain; (ii) demographic expansion in two lineages corresponding to the time of last deglaciation of the region; and (iii) a recent (< 50 years) local extinction of the third lineage coincident with climatic change and conversion of habitat for agricultural purposes in eastern Idaho. Beyond these inferences, our study highlights the unique value of museum material to phylogeography, and shows that small mammal recolonization of previously glaciated montane 'islands' differs from northward postglacial expansion observed in areas previously covered by continental ice sheets. Montane 'islands' may harbour high genetic diversity because of admixture and recurrent expansion/extinction.

Full text

Page 1

Molecular Ecology (2008)doi: 10.1111/j.1365-294X.2008.03671.x
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
Blackwell Publishing Ltd
Fire and ice: genetic structure of the Uinta ground squirrel
(Spermophilus armatus) across the Yellowstone hotspot
MARCEL VAN TUINEN,§* KIM O’KEEFE,§† UMA RAMAKRISHNAN‡ and ELIZABETH A. HADLY†
*Department of Biology and Marine Biology, University of North Carolina at Wilmington, Wilmington, NC 28403, USA, †371 Serra
Mall, Department of Biological Sciences, Stanford University, Stanford, CA 94305-5020, USA, ‡National Centre for Biological
Sciences, GKVK Campus, Bellary Road, Bangalore 560065, India
Abstract
The range of the Uinta ground squirrel, Spermophilus armatus, is centred over one of the
most tectonically active regions today, the Yellowstone hotspot. We document the role of
Quaternary tectonic and climatic history on the genetic structure of this species by screening
museum and extant individuals throughout its range. Phylogeographic, divergence time,
and demographic analyses of partial mitochondrial cytochrome b and control region DNA
sequences yield insight into the cadence of evolution across three spatiotemporal scales: (i)
a relatively deep intraspecific divergence of S. armatus into three lineages coincident with
the last major volcanic eruption in the region and maintained by the Snake River Plain; (ii)
demographic expansion in two lineages corresponding to the time of last deglaciation of
the region; and (iii) a recent (< 50 years) local extinction of the third lineage coincident with
climatic change and conversion of habitat for agricultural purposes in eastern Idaho.
Beyond these inferences, our study highlights the unique value of museum material to
phylogeography, and shows that small mammal recolonization of previously glaciated
montane ‘islands’ differs from northward postglacial expansion observed in areas previously
covered by continental ice sheets. Montane ‘islands’ may harbour high genetic diversity
because of admixture and recurrent expansion/extinction.
Keywords: glaciation, ground squirrels, Lava Creek Caldera, phylogeography, postglacial recolon-
ization, Snake River Plain, Yellowstone hotspot
Received 16 June 2007; revision received 19 October 2007; accepted 4 December 2007
Introduction
A frequent observation in phylogeographic studies is
that the modern distribution of genetic diversity remains
associated with the geographic landscape in which species
evolved, thus giving credence to vicariance as an important
mechanism of species and population subdivision
(Avise et al. 1987; Riddle 1996). Palaeoclimatic, orogenic or
tectonic change is often correlated with species or sub-
species divergence through the inference of fossil-calibrated
molecular clocks (van Tuinen et al. 2004). A prime
example of palaeoclimatic change stimulating isolation
and divergence of populations involves the waxing and
waning of continental ice sheets throughout the
Pleistocene (Avise et al. 1987; Brunsfield et al. 2001; Lessa
et al. 2003; Hewitt 2004).
Small boreal mammals in particular have provided a
good test of the importance of glacial/interglacial cycling
to shallow genetic history with recent investigations
indicating a central role of dispersal in montane mammal
phylogeography (Brant & Orti 2003; Demboski & Sullivan
2003; Eddingsaas et al. 2004; Hewitt 2004; Runck & Cook
2005; Waltari & Cook 2005). In high-latitude species,
postglacial recovery generally has been rapid after popu-
lation expansion from individuals surviving in southern
refugia (Hewitt 2004). Because mutational buildup is
too slow to fix novel genetic diversity subsequent to the
Pleistocene, standing genetic diversity in previously
glaciated areas is often low.
In contrast, genetic diversity may be much higher in
areas previously covered with isolated ice sheets not
connected to the continental ice sheets, such as montane
§These authors contributed equally to this manuscript.
Correspondence: Kim O’Keefe, Fax: 650-725-822; E-mail:
kokeefe@stanford.edu

Page 2

2 M. VAN TUINEN ET AL.
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
glaciers or ‘islands’, but this concept has little empirical
support (Knowles 2000). Understanding the process of
recolonization of these isolated glaciated areas is of interest
to us for several reasons. First, these areas may be reposi-
tories of genetic diversity because of the shorter times
between glacial cycles and more frequent, multidirectional
recolonization events (Hadly et al. 2004). Second, in the face
of global warming, they are thought to be vulnerable to
population extinctions as cool climate species are forced
off these montane islands and are replaced by lower-
altitude species (Knowles 2000; Masta 2000; DeChaine
& Martin 2005). Montane glaciers are not uncommon,
particularly in western North America, where many are
found throughout the Rocky Mountains as far south as
Mexico (Halfter 1987).
Members of the squirrel family Sciuridae are an inform-
ative mammalian system for ascertaining the role that
climate plays in diversification among species (Lamb et al.
1997; Mercer & Roth 2003). For several reasons, ground
squirrels of the genus Spermophilus provide for valuable
phylogeographic study of mammals. Several species occur
today in intermontane regions that have been profoundly
influenced by extreme, nonseasonal environmental pertur-
bations such as repeated montane glaciation cycles and
volcanism during the Quaternary. In addition, previous
work on the phylogenetic relations among North American
Spermophilus found that closely related species are geo-
graphically proximate, suggesting that most species evolved
in or near their current ranges (Harrison et al. 2003). Thus,
geographic barriers that have contributed to allopatric
speciation in the genus are likely to be found within the
present distribution of the group. In addition, although
they have restricted ranges relative to other small herbivores,
spermophilines are among the most locally abundant
mammals in the north-central Rocky Mountains (Yensen &
Sherman 2003). To date, phylogeographic investigation
among ground squirrels has been limited to two North
American species, the arctic (Spermophilus parryi) and
antelope ground squirrels (Ammospermophilus leucurus).
These studies describe two opposing genetic scenarios: (i)
a northward postglacial expansion of antelope ground
squirrels from southern relict but more genetically diverse
populations (Whorley et al. 2004), and (ii) maintenance of
high genetic diversity through glacial vicariance combined
with a Beringian refugium in the arctic ground squirrels
(Eddingsaas et al. 2004). However, neither study involved
species recolonization of isolated glaciers.
Among the North American Spermophilus species, the
Uinta ground squirrel (Spermophilus armatus) is a montane
species with a relatively small species range (Fig. 1B)
centered over one of the most tectonically active regions in
the world today, the Yellowstone hotspot (Pierce & Morgan
1992). The species thus provides an important test for the
relative influence of dispersal among montane basins vs.
the vicariance caused by separation between basins during
glaciation and/or volcanism. The Yellowstone hotspot
is so named because it currently resides in Yellowstone
National Park and is causing the topographic uplift of the
Beartooth Plateau. The uplift increased the altitude of this
region enough that it is thought to have contributed to the
formation of montane ice caps, as much as a mile thick
(Smith & Siegel 2000), that formed during the last glaciation
and covered the region as recently as 12 000 bp (Pierce
1979). Thus, the animals present in Yellowstone National
Park today have only recently recolonized the region since
the last glaciation (Pinedale glaciation 40 000–12 000 bp).
The Yellowstone hotspot has had a long history of
influence on the region in which S. armatus evolved. Over
the last 17 million years, the continent has migrated over
the hotspot in a southwestern course resulting in a
swath of extreme tectonic instability from the southeast of
Oregon, through Idaho and into the Greater Yellowstone
ecosystem (GYE). As the hotspot ‘migrated’, it caused
topographic uplift. Repeated caldera eruptions, layers of
heavy basalt lava flows, and the gradual cooling and
shrinking of the earth’s crust that took place after the
hotspot moved on resulted in the subsequent topographic
collapse that formed the massive snake river plain (SRP)
(Smith & Siegel 2000). The SRP, a 500-mile long, 50-mile
wide swath of basalt plain (Pierce & Morgan 1992) (see
Fig. 1B), has been shown to be a geographic barrier for
some fish species (Johnson 2002) and has been hypothe-
sized to restrict movement of S. armatus (Davis 1939) and
other squirrels in the area (Zegers 1984; Demboski &
Sullivan 2003; Yensen & Sherman 2003).
Here, we investigate the recolonization of the previously
glaciated parts of the GYE and the influence of tectonic
activity in shaping contemporary patterns of genetic
diversity of S. armatus. We hypothesize that genetic diversity
in the montane island situated on the Yellowstone Plateau
will be high relative to what would be expected from
a unidirectional post-Pleistocene expansion from other
portions of the species range and that the SRP has served
as a barrier to gene flow. To study these questions, we
required a comprehensive sampling of the entire species
range. Comprehensive sampling of modern populations
across the entire range is challenging because most popu-
lations are found in montane regions and have extremely
short active seasons (3–4 months). Thus, we have utilized
a combination of mostly historic and limited modern
specimens from numerous mammal collections. Here, we
present results from a phylogeographic study of S. armatus
through genetic analysis of cytochrome b (cyt b) and control
region (D-loop) sequences obtained from museum, tissue
and raptor pellet samples that represent the entire species’
distribution. We report the presence of three intraspecific
genetic lineages structured by the dynamic tectonic and
climatic history of this region, and the historic loss of one

Page 3

GENETIC STRUCTURE OF THE UINTA GROUND SQUIRREL 3
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
of these lineages. Divergence times estimated for the faster
marker (D-loop) yield dates consistent with the timing of
tectonic and climatic events across the Yellowstone region.
These results reveal the value of museum specimens for
phylogeographic study over historic time and provide
unique insight into the direct influence of Quaternary
environmental change linked to the evolution of mammals
of the intermontane American west.
Materials and methods
Taxon sampling
Individuals were sampled from throughout the current
species range from a variety of sources: tissues from
field-trapped individuals (n = 7); dental material found
in raptor pellets from Yellowstone National Park (n = 5),
and museum skins from several North American museums
(n = 78). Sequences obtained from the sister species,
Spermophilus beldingi (Harrison et al. 2003), serve as an
outgroup (n = 2). Specimen number and sampling localities
are found in Table S1, Supplementary material. Our total
sampling equals 92 individuals from 81 localities, including
nine localities originally used in determination of the
species’ marginal records and three museum specimens
collected in 1936, 1940, and 1987 from outside those
established margins (Hall & Kelson 1959). Two of the 92
specimens were identified as Spermophilus elegans and
therefore not used in our analyses. Our total geographic
coverage spans 39 counties, and is distributed across the
major mountain ranges (Uinta, Wasatch, Wyoming, Wind
River, Teton, Absaroka, Lost River), plateaus (Beartooth,
Yellowstone, Wasatch), and other suitable habitat within
the current distribution of Spermophilus armatus. It also
Fig. 1 (A) Unrooted haplotype network of absolute distances between mitochondrial DNA haplotypes of control region (D-loop). Each
circle represents a unique haplotype, its size proportional to the haplotype frequency. Colours represent the three genetic lineages found
within the species range from analysis of all sequences in tcs: northwest lineage (purple), northeast lineage (blue), and southern lineage
(yellow and gold). The yellow within the southern lineage represents a ‘star’ phylogeny indicating recent expansion. Haplotypes from this star
phylogeny are colour-coded to show the location of the recent expansion within the geographic range. Lines connecting each haplotype
represent a single nucleotide substitution, and the hatch marks along those branches represent additional substitutions. Haplotypes
with more than one branch connecting them to other haplotypes represent alternative pathways of equal likelihood. (B) Distribution Map
of Spermophilus armatus. Gray shaded relief is species range derived from Hall & Kelson (1959), georeferenced in ArcGIS and shown with
top of the map orientated as due North. State lines and boundaries of the Snake River Plain are shown, representative geographic ranges
are labelled and the position of the Lava Creek Caldera in Yellowstone National Park (YNP) that erupted approximately 640 000 years ago
is drawn in black. The Yellowstone Plateau referred to in the text overlaps well with the YNP park boundaries. Sampling localities are
based on georeferenced collecting localities and are shown by coloured circles. The colours represent the three genetic lineages found within
the species’ range: northwest lineage (purple), northeast lineage (blue), and southern lineage (yellow and gold). The yellow colour
represents an area of recent population expansion and corresponds with the star phylogeny portion of the haplotype network also in
yellow. Note that two localities showed genetic identification to S. armatus outside of its accepted species range. This phenomenon is
likely real, particularly in Utah.

Page 4

4 M. VAN TUINEN ET AL.
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
includes historic individuals from the Lost River Range in
the northwestern part of the species range, in localities
presently devoid of S. armatus. Initial sampling regime
included limited sampling across the southern part of the
range; hence, our expanded sampling included mostly
individuals from the southern end of the species range,
allowing us to test whether expansion into the northern
montane islands could have been from the south alone.
Genetic sampling
In accordance with parallel work in our laboratory on
rodents from the GYE, we used a genetic marker (mito-
chondrial cytochrome b) identified previously as phylo-
geographically informative for Spermophilus (Eddingsaas
et al. 2004), Ammospermophilus (Whorley et al. 2004), and
other squirrel taxa (Arbogast et al. 2001; Demboski & Sullivan
2003). We also screened a faster marker (mitochondrial
D-loop), which yielded additional resolution for shallow
branches. Our initial data set included cytochrome b (381 bp)
and D-loop (262–263 bp) sequences for 34 individuals.
To better test the pattern and directionality of postglacial
recolonization, we expanded our sampling to compre-
hensively cover localities within the southern portion of
the species range, focusing this time on the faster evolving
D-loop gene. Our final data set includes D-loop sequences
for 90 individuals spanning 262–263 bp, seven additional
shorter (152 bp) D-loop sequences from individuals showing
lower amplification success as well as the 34 cytochrome
b sequences (381 bp) for a representative subset. Phylo-
genetic analyses were performed on the cyt b and D-loop
subset data sets (n = 34) separately (see below, Figs 2 and
3, respectively) as well as the two mtDNA segments
combined (see below, Fig. 4). A final phylogenetic analysis
(see below, Fig. 5) and all of the demographic analyses
for S. armatus are based on the ‘expanded’ D-loop data set
(n = 90; 262 or 263 bp). We refer to Table S1 for the GenBank
Accession numbers and Table 1 for primer sequences.
Laboratory methods
DNA extraction, amplification and contamination control
of modern and historic specimens followed the published
protocol for ancient DNA by Hadly et al. (2004). Sequencing
of polymerase chain reaction (PCR) products was conducted
by Cogenics formerly Genaissance Pharmaceutical, New
Haven, Connecticut. We treated museum skins and pellets
as specimens like ‘ancient DNA’ with low-copy mtDNA
and high probability for contamination. Additional
processing for museum skins included the removal of
hair before extraction, which we found to reduce the
number of PCR inhibitors. Authentication of mitochondrial
sequences included: (i) use of multiple contamination
controls, (ii) spatial separation of modern and historic
DNA extraction and amplification in separate buildings
by different personnel, (iii) sequencing of both forward
and reverse directions for multiple overlapping fragments,
(iv) repeated extraction and amplification for historic
individuals showing unique haplotypes, (v) use of two
linked markers for a representative taxon subset to confirm
topological consistency, and (vi) cloning (TOPO TA) from
PCR products of two representatives from each of the two
most divergent lineages.
Phylogenetic analyses
Phylogenetic analyses were run in paup 4.0 beta (Swofford
2003) using the NJ and ML algorithm and with model
parameters established from the Akaike information
criterion in modeltest version 3 (Posada & Crandall 1998).
We performed an analysis using unweighted MP with
indels specified as a fifth state. To establish the phylo-
geographic information content and consistency of the
observed indels, we compared the resulting MP tree to
the unweighted MP, ML/NJ trees in which indels were
excluded from analyses. Nodal support was determined
by NJ bootstrapping (BS) and ML quartet puzzling with
1000 iterations. Because multiple MP trees were found for
every data set, we determined nodal consistency in the MP
analysis by estimating the majority rule consensus values
with a 50% cutoff and setting the Maxtrees option to 500.
Consistency between different methods (NJ, ML and MP)
was determined through comparison of nodal support
values in the presented mtDNA gene trees (Figs 2–5).
Genetic structure
We used tcs (Clement 2000), which does not assume a
bifurcating pattern of genetic divergence, to identify
unique haplotypes and to generate an unrooted minimum-
spanning network with a 95% plausibility of each linkage
between haplotypes. We performed a series of analyses of
molecular variance (amova) in arlequin 3.1 (Excoffier &
Table 1 Primer pairs for cytochrome b and D-loop fragments.
Fragments for each gene are overlapping
Primer namePrimer sequence (5′–3′)Gene
Sarm1
Sarm2
Sarm3
Sarm4
Sarm10
Sarm11
SarmD10
SarmD11
SarmD12
SarmD13
AAACCCCCTAAGCACCCCACCTC
TAGAATTCAGAATATGCATTGAC
CCATCTATCTAAACAACGAAGCA
TTTTCGATTAGGCTGACGGTTGG
ACTATCAAAGATATCCTTGGAGTCC
CACACCTCCCAGTTTATTAGGG
ACTCCTATGTAAATCGTGCATT
TCCACGGTCATGYTGACGGGTRG
CGTYCATAATACTAACATAGTAC
GAGACCAAATTTGGTAGGGGATAGTC
Cytochrome b
Cytochrome b
Cytochrome b
Cytochrome b
Cytochrome b
Cytochrome b
D-loop
D-loop
D-loop
D-loop

Page 5

GENETIC STRUCTURE OF THE UINTA GROUND SQUIRREL 5
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
Fig. 2 mtDNA gene tree of modern and historic individuals of Spermophilus armatus based on cytochrome b (n = 34, bp = 381). Depicted is
the maximum-likelihood tree using the HKY model (T ratio = 22.6) as specified by modeltest. Numbers shown above nodes are bootstrap
values using neighbour-joining and 1000 iterations (left), and bootstrap values using maximum likelihood in combination with quartet-
puzzling and 500 iterations (right). Numbers below nodes identify the frequency of branching support from equally likely maximum-
parsimony trees.

Page 6

6 M. VAN TUINEN ET AL.
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
Schneider 2005) to determine how much of the observed
genetic structure could be explained by distinct geographic
divisions. Standard estimates of genetic distance were
calculated in arlequin 3.1 using the best model of
evolution for our data set as specified by modeltest.
Demographic history
We tested for past changes in population size using
several methods. We used Fu’s neutrality test (Fu 1997) in
arlequin 3.1, which is a particularly powerful test for
detecting sudden and recent expansions in populations
with an excess of rare alleles. We also examined the
distribution of pairwise differences (mismatch distributions)
for each group to look for evidence of past expansion
(Rogers & Harpending 1992). In general, populations
undergoing recent and sudden expansion exhibit a
Poisson-shaped mismatch distribution while populations
in equilibrium tend to have ragged distributions (Slatkin
& Hudson 1991).
Fig. 3 mtDNA gene tree from modern and
historic individuals of Spermophilus armatus
based on D-loop (n = 34, bp = 262–263)
sequences. maximum-likelihood (ML) tree
is shown here using the HKY + G model (T
ratio = 63.21, alpha = 0.08) as specified by
modeltest. Numbers shown at nodes are
bootstrap values using NJ and ML (above
nodes), and MP consensus values.

Page 7

GENETIC STRUCTURE OF THE UINTA GROUND SQUIRREL 7
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
Divergence times
Approximate sublineage ages were estimated from the
faster marker (D-loop) because of its increased resolution
for the shallow history of S. armatus and our interest in
dating population expansions and recent admixture.
The fossil record of S. armatus in the Pleistocene is too
discontinuous to be of use for internal calibration
(Eshelman & Sonnemann 2000), which prevented direct
calibration of the D-loop molecular rate. We therefore
estimated the observed rate of the D-loop data set relative
to the cyt b rate of molecular evolution and converted this
to an absolute rate based on a published cyt b rate for
Spermophilus (1.52% per million years, Eddingsaas et al.
2004). This rate was originally calibrated from fossil and
geologic evidence suggesting a divergence time of 5.16
million years ago (Ma) between S. beecheyi and S. parryi.
Likelihood models were run with and without a molecular
clock constraint using the likelihood ratio test under a
chi-square probability distribution to assess whether the
data were significantly clock-like. We used mega 3.0 to
estimate sublineage divergence times from ‘net’ distances
after correction for within-lineage polymorphism. We
approximated the modeltest-preferred model of TrN +G+ I
using a TrN + G model with application of a stringent
correction for rate heterogeneity across sites (G = 0.19)
estimated from the Akaike information criterion-averaged
alpha (G), [not alpha (IG)] output.
Results
Overall amplification success and resulting
phylogeographic structure
DNA was readily amplified from skins when hair is
removed and amplification is limited to small length
(100–150 bp) fragments. Our protocols resulted in a 92%
amplification rate based on 85 skins attempted, five primer
pairs (Table 1), and two mitochondrial markers spanning
~650 bp. Genetic analyses reveal that the genetic diversity
throughout the entire modern range of Spermophilus
armatus is relatively low and can be traced to three lineages.
The most widespread and abundant lineage is present
throughout the south and southwest of the modern dis-
tribution (southern lineage, ‘SL’) and is found exclusively
south or east of the SRP (Fig. 1B, yellow and gold: the SRP
on this map runs in a north–northeast direction along the
Yellowstone hotspot). A second lineage is predominantly
found north of the SRP in the northeast of the current
distribution (northeast lineage, ‘NEL’) (Fig. 1B, blue). A
third lineage is very rare (represented by only three
individuals out of the 88 S. armatus that we sampled) and
only found historically north of the SRP in the northwest of
the species range (northwest lineage ‘NWL’) (Fig.1B, purple).
Phylogenetic analyses
Cyt b. Our observed substitution pattern for cytochrome
b (cyt b) is consistent with the average mitochondrial
pattern among mammals (Irwin et al. 1989), with
abundance of substitutions at third codon positions, low
Fig. 4 mtDNA tree from modern and historic individuals of
Spermophilus armatus based on a combination of cytochrome b
and D-loop sequences (n = 34, bp = 624). Depicted is a maximum-
likelihood tree using a TiM + G + I model and parameters
Rmat = 1, 20.2, 0, 0, 36.3, alpha = 0.5151, I = 0.5824 as specified by
modeltest. Asterisks denote nodes with significant (P = 0.95)
support. Nodes with hatch-bars have additional indel support
(see text for discussion).

Page 8

8 M. VAN TUINEN ET AL.
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
Fig. 5 mtDNA gene tree of D-loop sequences
from modern and historic individuals of
Spermophilus armatus sampled throughout
the species range (n = 88, bp = 263). Depicted
is the maximum-likelihood tree using a
TN + G (alpha = 0.19) model specified by
modeltest. Numbers shown above nodes
are: bootstrap values using neighbour-joining
and 1000 iterations (left), and bootstrap
values using maximum likelihood in com-
bination with quartet-puzzling and 500
iterations (right). Numbers below nodes
identify the frequency of branching support
from equally likely maximum-parsimony
trees. Support < 50% is not shown. Dotted
line depicts a genetically distinct lineage
only found in historic individuals.

Page 9

GENETIC STRUCTURE OF THE UINTA GROUND SQUIRREL 9
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
GC content across all sites and especially at third codon
positions. All observed substitutions in cyt b were tran-
sitional, suggestive of a recent species history. Phylogenetic
results show tentative geographic clustering among S.
armatus, low genetic diversity (HKY model, T ratio = 22:
2.3% overall pairwise divergence = 2.3%; mean pairwise
divergence = 0.95%; see Johns & Avise 1998 for comparison
to other species) and limited bootstrap support (Fig. 2).
Intermediate support is found for a north and south of
the SRP division (75% BS). Three historic individuals
north of the SRP and at the western edge of the species
range in the Lost River Range form their own grouping
(91% BS). A second lineage found primarily north of the
Snake River has strong support (89% BS), and includes
individuals from northeast Idaho, Montana, northern
Wyoming, and the Wind River Mountains. Cytochrome
b variation was too low (< 1%) to detect structure in the
southern lineage (Fig. 2).
D-loop. Use of the faster-evolving D-loop mitochon-
drial marker for these individuals added considerable
resolution (Fig. 3). The HKY model again best described
this data set, but with an additional correction for rate
heterogeneity among sites (T ratio = 63.21, G = 0.08). The
unrooted D-loop phylogeny and that of the cytochrome
b for the same individuals lack conflict, although the
two genes differ in the placement of the root. In species
with shallow lineage coalescence, this difference is not
unexpected because of a long branch leading from the
in-group to the sister-group (Demboski & Sullivan 2003).
The D-loop data support the grouping of a distinct
northeast lineage (85% BS), and a southern lineage
subgrouping in Idaho (90% BS), both of which have
support from several unique substitutions and a 1-bp
deletion or insertion (Fig. 3).
Cyt b and D-loop. As expected from the observed congruence,
analysis of the combined fragments (under model TiM+G+ I
and parameters Rmat = 1, 20.2, 0, 0, 36.3, G = 0.5151,
I = 0.5824) increases support for both lineages to 96–100%
and again supports a distinct lineage in the northwest of
the species range near the Lost River Range (100% BS), as well
as a clustering of these Lost River individuals within an
expanded northeast lineage grouping (73% BS) (Fig. 4). The
combined data still provides little information on S. armatus
from Utah, which is likely due to the limited sampling from
the southern range relative to the diversity found there.
Thus, we expanded our sampling to include individuals
from the entire range, focusing our sequencing efforts on
the faster evolving D-loop which is better suited for shallow
phylogeographic analysis of the Uinta ground squirrel.
Expanded D-loop. The expanded D-loop phylogeny is
consistent with that based on cytochrome b, but with
added resolution from both variable sites and informative
indels. The expanded D-loop data set again shows a 100%
consistency index for the two indels. Furthermore, it
confirms the Snake River Plain as a phylogeographic
break, and indicates two areas of admixture (east of the
Snake River Plain and Teton Mountains in the Wind River
Range, and at the confluence of the Bear River Range and
Wasatch Mountains) (Fig. 5).
Genetic structure and diversity
Significant FST values were found between samples taken
from north of the SRP and those existing south of the
SRP (D-loop: FST= 0.454, P < 0.001). Variation between
these two groups (ΦSC) explained more than 45% of the
variation across the entire species range. Grouping
individuals by the four clades represented in Fig. 5
(northeast lineage, northwest lineage, southern lineage
south of the SRP and southern lineage east of the SRP
also resulted in a significant FST value (D-loop: FST= 0.482,
P < 0.001), increasing the ΦSC from 45% to 48%. The un-
rooted haplotype cladogram generated by TCS with a
0.95 probability however, depicts only three haplogroups
(Fig. 1A) that correspond with the three best-supported
lineages depicted in the mtDNA gene tree in Fig. 5
(northeast, northwest, southern).
The southern lineage (SL) is the most widespread and
abundant lineage across the range of the species and is
present throughout the south and southwest of the modern
distribution. It is the only lineage represented immedi-
ately south of the Snake River Plain and west of the Teton
and Salt River Mountains (Fig. 1B, yellow and gold).
The southern lineage also contains the greatest amount
of genetic diversity both in terms of the number of unique
haplotypes (31) as well as the mean pairwise distance
(6.83 ± 3.26). The northeast lineage (NEL) is found pre-
dominantly to the north and northeast part of the current
range with only two of 27 individuals from that lineage
found in the southern region (Morgan County and Rich
County, Utah) of the current range (Fig. 1B, blue). This
NEL contains 21 unique haplotypes from 27 individuals
and a mean pairwise distance of 5.62 ± 2.78. The northwest
lineage (NWL) was found historically north of the SRP in
the Lost Rivers Mountains (Fig.1B, purple) and is represented
by only three individuals with two unique haplotypes
and a mean pairwise distance of 0.76 ± 0.67.
Demographic analyses
The demographic analyses showed evidence for recent
expansion in the SL and the NEL. For example, the
haplotype network exhibited a star phylogeny indicating
recent population expansion in the SL (Fig. 1A, yellow).
Fu’s test of neutrality also produced significant FS values

Page 10

10 M. VAN TUINEN ET AL.
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
(FS= –1.92, P < 0.001) further supporting the finding of
recent expansion within the SL. While recent expansion
is much less apparent from visual inspection of the NEL
haplotype network (there is no obvious star phylogeny),
Fu’s FS were also significant for the NEL (FS= –1.53,
P < 0.001). The mismatch distributions for the NEL and
SL were both unimodal and fall well within the 95%
confidence limits showing recent population expansion
in each lineage (Fig. 6A). As expected, the mismatch
distribution of all haplotypes divided into groups either
north or south/southeast of the SRP based entirely on
geography (vs. by lineage) show a bimodal distribution
because of the presence of gene flow to the east of the
Salt River and Teton Mountains along the Idaho–Wyoming
border (Fig. 6B). This also explains why Fu’s FS is significant
for the haplotypes found east of the SRP, suggesting
expansion, but the mismatch distribution is very ‘ragged’,
suggesting stasis. The individuals sampled southeast of
the SRP are not reciprocally monophyletic but made up
of several rare haplotypes derived from both the SL and
NEL, indicative of admixture.
Divergence times
A likelihood ratio test on the D-loop data set did not
reject clock-like behaviour (2*deltaLnL = 88.02; P > 0.10).
Comparison of distances based on the D-loop and
cytochrome b for the same pairwise groupings indicates
that the intraspecific D-loop distances increase linearly
with time, at a 7.5 times faster rate. Thus, we estimated an
11.5% per million-year divergence rate from a previously
proposed Spermophilus cytochrome b rate (see Methods).
Application of this rate puts the deepest divergence
within all extant S. armatus lineages at 0.58 million years.
The separation of the two northern lineages (NWL and
NEL) from the southern (SL) lineage occurred relatively
early in the history of S. armatus (0.43 Ma), as did the
initial divergence within the northern Snake River between
the NWL and NEL (0.33 Ma). We estimated a relatively
recent age for the common ancestor of the majority of
haplotypes found in the Yellowstone Plateau (0.11 Ma),
which was completely covered with ice at the end of the
Pleistocene. Expansion of an Idaho population south of the
Snake River Plain (12 000 years ago) and northward
expansion of southern haplotypes (13 500 years ago) both
are estimated to have occurred near the end of the Pleistocene.
The faster marker lacks resolution for evidence of Holocene
expansion but the presence of identical haplotypes across
the Bear River Range and Wasatch Range, parts of which
were glaciated in the Pleistocene (Laabs et al. 2006), does
not conflict with this notion.
Discussion
Our analyses of two linked mitochondrial markers provided
us with a multifaceted look into the evolution of Spermophilus
armatus. The deepest genetic divergence within S. armatus
is approximately 580 000 years ago. This follows the
cataclysmic explosion of the Lava Creek caldera at ~640 000
years ago (Lanphere et al. 2002) which formed a massive
caldera in what is now the middle of the species range and
may explain the low maximum diversity within this species.
The Lava Creek volcanic eruption and ash-fall dumped
between 1 m and 10 m of ash over most of the western USA
(Izett & Wilcox 1982; Perkins & Nash 2002). Smaller ash-fall
than the Lava Creek eruption is known to have caused extreme
population fragmentation and loss of genetic diversity in
some modern rodent popu-lations (Gallardo et al. 1995).
Fig. 6 Mismatch distributions by lineage
association (A) and by geography (B). Black
continuous lines correspond to observed
frequencies of pairwise nucleotide dif-
ferences, broken lines represent upper and
lower bounds for the 95% confidence
intervals around the observed distribution
(α = 0.05), and the grey solid lines represent
the expected frequencies under a sudden
expansion model.

Page 11

GENETIC STRUCTURE OF THE UINTA GROUND SQUIRREL 11
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
A second, relatively deep divergence (0.43 Ma) between
the NWL, NEL, and SL falls within local glacial cycles
beginning with the Sacajawea Ridge (0.6 Ma) (Pierce 2004)
and ending with the initiation of Bull Lake glaciation
(Fullerton et al. 2004). These glacial cycles caused repeated
separation of intermontane basins within the present range
of S. armatus. Late Pleistocene glaciation (18 000 years ago)
of the Yellowstone plateau (Pierce 2004) and other high
montane regions within the species range did not cause
appreciable divergence within the lineages. However,
movement since deglaciation, about 14 000 years ago, has
caused admixture along the eastern front of the continental
divide near the Idaho–Wyoming border. Recolonization of
the previously glaciated montane terrains came from two
genetically distinct lineages (NEL and SL) and admixture
between these lineages occurred exclusively east or south
of the SRP at the southern edge of the Yellowstone Plateau
leading into the Wind River and Salt River Range and at
the confluence of the Bear River Range and Wasatch
Mountains. The NEL shows evidence of recent (13 000 years
ago) population expansion, consistent with population
size increase since the terminal Pleistocene. Wholesale
postglacial recolonization of the northern range of the GYE
in the Yellowstone Plateau is demonstrated here to be from
both the northeast and southern lineages, with coalescence
times for both lineages significantly preceding the terminal
Pleistocene. Admixture of distinct genetic lineages in the
Yellowstone Plateau and Wind River Range has been found
(Hadly et al. 2004; Hadly et al. unpublished findings) also
in montane and long-tailed voles (Microtus montanus and
M. longicaudus). Gene flow of S. armatus into Idaho imme-
diately south of the SRP has been restricted to populations
solely derived from the SL lineage, which shows evidence
of the most recent population expansion. The SRP thus is
an obvious barrier to north–south migration in the Uinta
ground squirrel and is presently unoccupied by the species.
Finally, very recently (< 50 bp), the populations repre-
senting the NWL, found only in eastern Idaho north of
the SRP in the Lost River drainage, have become locally
extirpated (Madison et al. 2004). The NWL is not repre-
sented in any extant populations that we have sampled
from throughout the species range today; thus, it is likely
extinct. Local extirpation north of the SRP near Craters of
the Moon National Park has also been observed in
other species. Spermophilus mollis artemisiae and S. elegans
aureus, both of which occupied the same region, are
hypothesized to have been extirpated due to habitat
degradation and agricultural conversion (Zegers 1984;
Yensen & Sherman 2003). It is possible that the NWL of
S. armatus was once more widespread north of the Snake
River Plain. We aim to further test this hypothesis by using
a phylochronologic approach from subfossils described
from two caves at the northern edge of the species range
(Waterfall Locality, Lamar Cave), as well as by increased
modern field sampling from other localities within the
current species range.
Cautionary notes
The consistency of tracing shallow genealogical events to
Quaternary environmental events in the Central Rocky
Mountains is convincing, especially when considering
extensive evidence for genetic subdivisions linked to
Pleistocene climate changes elsewhere (the Pacific Northwest,
Northern Rocky Mountains) (e.g. Demboski & Sullivan
2003; Eddingsaas et al. 2004). Nonetheless, we point out
that the scenario depicted above is the maternal side of
the total story that is likely more complex given strong
male-biased dispersal in S. armatus. A nuclear gene
phylogeographic investigation will prove more difficult,
but not impossible (Zhang & Hewitt 2005), if based mostly
on museum specimens. Also, the divergence times
presented here are current best estimates, yet still appro-
ximate because of the paucity of ground squirrel fossils
that can be linked to extant species. Additionally, the
estimates of the tectonic and climatic events we describe
are not necessarily precise because few are within the time
period spanned by radiocarbon dating. Caution also
must be advised because we pooled samples from different
years (1920–2003) to represent a single phylogeographic
story. If gene flow is extremely rapid in species with a small
range, phylogeographic structure could theoretically vary
over short time periods. Over periods spanning thousands
of years, recent ancient DNA research indicated that
current phylogeography does not always paint an entire
picture of species history, and that ancient gene flow could
have left different traces in the past not reflected today
(Barnes et al. 2002; Hadly et al. 2004; Hofreiter et al. 2004).
Although over a period of less than a century we have
observed extirpation of an entire lineage, phylogeographic
turnover is not readily apparent in our sampling of the
other lineages from 1920 to 2003. Our data instead support
a long-term geographic separation of haplotypes across
the SRP, with recent admixture of the two largest lineages
(NEL, SL) occurring to the east of the SRP where the
mountains run in north–south direction.
Conclusions
We report low levels of maximum intraspecific genetic
diversity in the cytochrome b sequences of Uinta ground
squirrels, suggesting significant loss of diversity consistent
with the last cataclysmic eruption of the Yellowstone
hotspot. Using a faster-evolving marker, we recovered
shallow-time phylogeographic resolution and found that
the deepest split within the species is consistent with
advance of the Sacagawea Ridge glaciation. Screening
of this faster marker across a reasonably large set of

Page 12

12 M. VAN TUINEN ET AL.
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
specimens allowed us to further recover a demographic
signal of expansion in two of the three lineages consistent
with Late Pleistocene deglaciation and subsequent gene
flow into some areas but not others. Our two hypotheses were
confirmed: (i) the SRP was shown to be a major geographic
barrier, and (ii) we found significant genetic diversity in
several central montane ‘islands’ (the Yellowstone Plateau,
Wind River Range, Wasatch-Uinta Range), which argues
against a unidirectional postglacial expansion from the south.
Finally, use of museum skins in a diachronic way allowed
us to document the historic extirpation of a third, divergent
lineage during the past several decades, probably due to
agricultural conversion of ground squirrel habitat.
Acknowledgements
We thank several museums and curators for their generous
loans of historic squirrel specimens. In particular, we thank
C. Conroy and J. Patton at the UC Berkeley Museum of Vertebrate
Zoology, I. Lovette and K. Bostwick at the Cornell University
Museum of Vertebrates, J. Phelps at the Field Museum of Natural
History Division of Mammals, G. Shugart at the University of
Puget Sound Slater Museum of Natural History, D. Lunde, T.
Pacheco and E. Westwig at the American Museum of Natural
History Department of Mammalogy, J. Eger at the Royal Ontario
Museum Department of Mammalogy, S. Hinshaw and P. Tucker
at the University of Michigan Museum of Zoology Mammal
Division, N. Slade at the Kansas University Mammal Division,
D. Rogers at the Brigham Young University Monte L. Bean Life
Science Museum, E. Rickart at the Utah University Utah Museum
of Natural History, and K. Zyskowski at the Yale Peabody
Museum. Rebecca Rowe provided tissue specimens from Wasatch
Plateau. Jill Mateo provided tissue specimens of Spermophilus
beldingii, the sister taxon and outgroup for our phylogenetic
analyses. Y. Chan and three anonymous reviewers provided
valuable suggestions and edits for the manuscript. We thank Jan
Zinck for early contributions to developing laboratory protocols
and Judsen Bruzgul for assisting us with a base map in ArcGIS.
We gratefully acknowledge the support of John Varley and the
staff of the Yellowstone Division of Research. For financial sup-
port, we thank the National Center for Environmental Research
(NCER) STAR Program, EPA (no. U915979012 to K. O’Keefe), and
the US National Science Foundation (DEB no. 0108541 to E.
Hadly). The research conducted for this study complies with
the current laws in the USA.
References
Arbogast B, Browne R, Weigl P (2001) Evolutionary genetics and
Pleistocene biogeography of North American tree squirrels
(Tamiasciurus). Journal of Mammalogy, 82, 302–319.
Avise JC, Arnold J, Ball RM et al. (1987) Intraspecific phylogeo-
graphy: the mitochondrial DNA bridge between population
genetics and systematics. Annual Review of Ecological Systematics,
18, 489–522.
Barnes I, Matheus P, Shapiro B, Jensen D, Cooper A (2002) Dynamics
of Pleistocene population extinctions in Beringian brown bears.
Science, 295, 2267.
Brant SV, Orti G (2003) Phylogeography of the Northern short-tailed
shrew, Blarina brevicauda (Insectivora: Soricidae): past frag-
mentation and postglacial recolonization. Molecular Ecology, 12,
1435–1449.
Brunsfield SJ, Sullivan J, Soltis DE, Soltis PS (2001) Comparative
phylogeography of northwestern North America: a synthesis.
In: Integrating ecology and evolution in a spatial context (eds Silver-
town J, Antonovics J), pp. 319–339. Blackwell Science Ltd, Oxford.
Clement M, Posada D, Crandall K (2000) TCS: a computer program
to estimate gene genealogies. Molecular Ecology, 9, 1657–1660.
DeChaine EG, Martin AP (2005) Marked genetic divergence among
sky island populations of Sedum lanceolatum (Crassulaceae) in
the Rocky Mountains. American Journal of Botany, 92, 477–486.
Demboski J, Sullivan J (2003) Extensive mtDNA variation within
the yellow-pine chipmunk, Tamias amoenus (Rodentia: Sciuridae),
and phylogeographic inferences for northwest North America.
Molecular Phylogenetics and Evolution, 26, 389–408.
Eddingsaas AA, Jacobsen BK, Lessa EP, Cook JA (2004) Evolutionary
history of the arctic ground squirrel (Spermophilus parryi) in
Nearctic Beringia. Journal of Mammalogy, 85, 591–600.
Eshelman BD, Sonnemann CS (2000) Spermophilus armatus.
Mammalian Species, 637, 1–6. American Society of Mammalogists.
Excoffier LGL, Schneider S (2005) arlequin, version 3.1: an
integrated software package for population genetics data
analysis. Evolutionary Bioinformatics Online, 1, 47–50.
Fu X (1997) Statistical neutrality of mutations against polyphyly:
frequency, causes, and consequences, with insights from animal
mitochondrial DNA. Annual Review of Ecology and Systematics,
34, 397–423.
Fullerton DS, Colton RB, Bush CA (2004) Limits of mountain and
continental glaciation east of the Continental Divide in northern
Montana and north-western North Dakota, USA. In: Quaternary
Glaciations — Extent and Chronology (eds Ehlers J, Gibbard PL). Part
II North America, Developments in Quaternary Science, 2, 131–145.
Gallardo M, Kohler N, Araneda C (1995) Bottleneck effects in local
populations of fossorial Ctenomys (Rodentia, Ctenomyidae)
affected by vulcanism. Heredity, 74, 638–646.
Hadly EA, Ramakrishnan U, Chan YL et al. (2004) Genetic
response to climatic change: insights from ancient DNA and
phylochronology. Plos Biology, 2, 302–319.
Halffter G (1987) Biogeography of the Montane Entomofauna of
Mexico and Central America. Annual Review of Entomology, 32,
95–114.
Hall ER, Kelson KR (1959) The Mammals of North America. The Ronald
Press Co, New York.
Harrison RG, Bogdanowicz SM, Hoffmann RS, Yensen E, Sherman
PW (2003) Phylogeny and evolutionary history of the ground
squirrels (Rodentia: Marmotinae). Journal of Mammalian Evolution,
10, 249–276.
Hewitt GW (2004) Genetic consequences of climatic oscillations in
the Quaternary. Philosophical Transactions of the Royal Society of
London B, 359, 183–195.
Hofreiter M, Serre D, Rohland N et al. (2004) Lack of phylogeography
in European mammals before the last glaciation. Proceedings of
National Academy of Sciences, USA, 101, 12963–12968.
Irwin DM, Kocher TD, Wilson AC (1989) Evolution of the cytochrome
b gene of mammals. Journal of Molecular Evolution, 32, 128–144.
Izett GA, Wilcox RE (1982) Map showing localities and inferred
distribution of the Huckleberry Ridge, Mesa Falls, and Lava
Creek ash beds (Pearlette family ash beds) of Pliocene and
Pleistocene age in the western United States and southern
Canada: 000. US Geological Survey Miscellaneous Investigations
Series Map I-1325, scale 1: 4 000 000.

Page 13

GENETIC STRUCTURE OF THE UINTA GROUND SQUIRREL 13
© 2008 The Authors
Journal compilation © 2008 Blackwell Publishing Ltd
Johns GC, Avise JC (1998) A comparative summary of genetic dis-
tances in the vertebrates from the mitochondrial cytochrome b
gene. Molecular Biology and Evolution, 15, 1481–1490.
Johnson JB (2002) Evolution after the flood: phylogeography of the
desert fish Utah chub. Evolution, 56, 948–960.
Knowles LL (2000) Tests of Pleistocene speciation in montane
grasshoppers from the sky islands of western North America
(Genus Melanoplus). Evolution, 54, 1337–1348.
Laabs BJC, Plummer MA, Mickelson DM (2006) Climate during the
Last Glacial Maximum in the Wasatch and southern Uinta Moun-
tains inferred from glacier modeling. Geomorphology, 75, 300–317.
Lamb T, Jones TR, Wettstein PJ (1997) Evolutionary genetics and
phylogeography of tassel-eared squirrels (Sciurus aberti). Journal
of Mammalogy, 78, 117–133.
Lanphere MA, Champion DE, Christiansen RL (2002) Revised
ages for tuffs of the Yellowstone Plateau volcanic field: assignment
of the Huckleberry Ridge Tuff to a new geomagnetic polarity
event. GSA Bulletin, 114, 559–568.
Lessa EP, Cook JA, Patton JL (2003) Genetic footprints of demo-
graphic expansion in North America, but not Amazonia, during
the Late Quaternary. Proceeding of National Academy of Sciences,
USA, 100, 10331–10334.
Madison E, Rodhouse T, Garret L (2004) 2003 Mammal Inventory,
Craters of the Moon National Monument and Preserve. Report
for subagreement no. 20 to cooperative agreement no. CA9000-
95–018. University of Idaho and National Park Service, Upper
Columbia Basin Network, University of Idaho and Department
of Wildlife and Fisheries Resources, Moscow, Idaho.
Masta SE (2000) Phylogeography of the jumping spider Habronattus
pugillis (Araneae: Salticidae): recent vicariance of sky island
populations? Evolution, 54, 1699–1711.
Mercer JM, Roth VL (2003) The effect of Cenozoic global change on
squirrel evolution. Science, 299, 1568–1572.
Perkins ME, Nash BP (2002) Explosive silicic volcanism of the Yellow-
stone hotspot: the ash fall tuff record. GSA Bulletin, 114, 367–381.
Pierce KL (1979) History and dynamics of glaciation in the
northern Yellowstone National Park area. US Geological Survey
Professional Paper, 729F, 91.
Pierce KL (2004) Pleistocene glaciations of the Rocky Mountains.
In: The Quaternary Period in the United States (eds Gillespie AR,
Porter SC, Atwater BF), pp. 63–76. Elsevier, Amsterdam.
Pierce KL, Morgan LA (1992) The track of the Yellowstone hop
spot — volcanism, faulting and uplift. Regional Geology of Eastern
Idaho and Western Wyoming, pp. 1–53. Geological Society of
America Memoir 179.
Posada D, Crandall KA (1998) modeltest: testing the model of
DNA substitution. Bioinformatics, 14, 817–818.
Riddle BR (1996) The molecular phylogeographic bridge between
deep and shallow history in continental biotas. Trends in Ecology
and Evolution, 11, 207–211.
Rogers AR, Harpending HC (1992) Population growth makes
waves in the distribution of pairwise genetic differences.
Molecular Biology and Evolution, 9, 552–569.
Runck A, Cook JA (2005) Postglacial expansion of the southern
red-backed vole (Clethrionomys gapperi) in North America.
Molecular Ecology, 14, 1445–1456.
Slatkin M, Hudson RR (1991) Pairwise comparisons of mitochondrial
DNA sequences in stable and exponentially growing popu-
lations. Genetics, 129, 555–562.
Smith RB, Siegel LJ (2000) Windows into the Earth. Oxford University
Press, New York.
Swofford DL (2003) PAUP*. Phylogenetic Analysis Using Parsimony (*and
Other Methods). Sinauer Associates, Sunderland, Massachusetts.
van Tuinen M, Ramakrishnan U, Hadly EA (2004) Investigating
the effect of environmental change on biotic evolution:
genetic contributions, current work and future developments,
Philosophical Transactions of the Royal Society of London: Physical,
Mathematical and Engineering Sciences, 362, 2795–2820.
Waltari E, Cook JA (2005) Historical demographics and phylo-
geography of arctic hares (Lepus): genetic signatures test glacial
refugia hypotheses. Molecular Ecology, 14, 3005–3016.
Whorley J, Alvarez-Castaneda ST, Kenagy J (2004) Genetic
structure of desert ground squirrels over a 20-degree-latitude
transect from Oregon through the Baja California peninsula.
Molecular Ecology, 13, 2709–2720.
Yensen E, Sherman PW (2003) Ground-Dwelling Squirrels of the
Pacific Northwest. US Fish and Wildlife Service, Boise, Idaho.
Zegers DA (1984) Spermophilus elegans. Mammalian Species, 214,
1–7. American Society of Mammalogists.
Zhang DX, Hewitt GM (2005) Nuclear DNA analyses in genetic
studies of populations: practice, problems and prospects.
Molecular Ecology, 12, 563–584.
Kim O’Keefe is interested in how the ecological and evolutionary
characteristic of a species affect its response to climate at multiple
spatial and temporal scales. She is currently investigating how
climate influences population dynamics and genetic structure of
ground squirrels. Marcel van Tuinen is assistant professor and
curator of the animal frozen tissue bank at UNC Wilmington. His
research program focuses broadly on birds and mammals and
encompasses the use of ancient DNA for historical population
genetic inquiries, fossils for molecular clock calibration, and
phylogenomic tools to investigate the tempo and mode of aquatic
bird evolution. Uma Ramakrishnan is interested in how genetic
variation is partitioned through space and time, and how such
variation allows us to understand past demographic processes.
She is currently investigating genetic variation and its partitioning
for mammals in the Indian subcontinent. Elizabeth A. Hadly
studies the ecology and evolution of vertebrates using both fossil
and modern assemblages from North America, South America
and most recently, India. Her investigations focus on the response
of animals to climatic change using genetic, morphological,
community, and geochemical analyses.
Supplementary material
The following supplementary material is available for this article:
Table S1 Museum and specimen numbers, sampling localities,
and GenBank Accession numbers for the control region (D-loop)
and cytochrome b (cyt b)
This material is available as part of the online article from:
http:/ /www.blackwell-synergy.com/doi/abs/10.1111/
j.1365-294X.2008.03671.x
(This link will take you to the article abstract).
Please note: Blackwell Publishing are not responsible for the
content or functionality of any supplementary materials supplied
by the authors. Any queries (other than missing material) should
be directed to the corresponding author for the article.