MEC653.fm Page 1093 Wednesday, June 30, 1999 2:13 PM
Molecular Ecology (1999) 8, 1093 –1104
Effects of natural habitat fragmentation on an endemic
scrub lizard (Sceloporus woodi): an historical perspective
based on a mitochondrial DNA gene genealogy
Blackwell Science, Ltd
A . M . C L A R K , * B . W. B O W E N , † and L . C . B R A N C H ‡
*BEECS Genetic Analysis Core, University of Florida, 421 Carr Hall, Gainesville, FL 32611, USA, †Department of Fisheries and
Aquatic Sciences, University of Florida, 7922 NW 71st St., Gainesville, FL 32653, USA, ‡Department of Wildlife Ecology and
Conservation, University of Florida, Box 110430, Gainesville, FL 32611, USA
Abstract
The Florida scrub lizard, Sceloporus woodi, is endemic to scrub habitat patches along
the central portion of the Florida peninsula and xeric coastal regions. Scrub ecosystems
are the patchily distributed remnants of previously widespread habitats formed during
the Pleiocene and early Pleistocene. Scrub lizards appear to have limited dispersal capabilities due to high habitat specificity and low mobility. To assess the population structure
and phylogeography of S. woodi, 135 samples were collected from 16 patches on five
major ridges in Florida, USA. Analysis of 273 bp of mitochondrial DNA (mtDNA) cytochrome b reveals a very strong geographic distribution of genetic diversity. Haplotype
frequencies are significantly different in 63 of 66 comparisons between patches. With one
exception, samples from the five major ridges are characterized by fixed differences in
haplotype distribution and deep evolutionary separations (3–10%). Fixed genetic differences were also observed between northern and southern segments of several ridges.
Analysis of molecular variance (amova) shows an estimated 10.4% total genetic variation within patches, 17.5% among patches (within ridges), and 72.1% among ridges. This
strong population structure among patches within ridges indicates that the distribution
of S. woodi is tightly linked to sandy scrub habitat and that the discontinuous distribution of scrub habitats significantly inhibits dispersal and gene flow. Phylogeographic
analyses indicate a pattern of dispersal down the Florida peninsula during the late
Pliocene–early Pleistocene, followed by habitat fragmentation and vicariant isolation
events. Therefore, the deep genetic structuring among scrub lizard populations on separate ridges is attributed to ancient isolation events induced by a shift from dry (xeric) to
wet (mesic) conditions on the Florida peninsula. These findings indicate that some scrub
lizard populations have persisted in isolation for time frames in excess of 1 Myr, providing
a case history on the genetic consequences of habitat fragmentation.
Keywords: biogeography, conservation genetics, endangered ecosystems, Florida scrub, habitat
fragmentation, mitochondrial DNA
Received 26 July 1998; revision received 23 December 1998; accepted 23 December 1998
Introduction
The Florida scrub lizard, Sceloporus woodi, is a small, spinyscaled species restricted to dry, sandy habitats associated
Correspondence: A. M. Clark. Fax: +1-352-392-3704; E-mail:
[email protected]
Representative sequences from this project are deposited in
Genbank under Accession nos. AF144631–AF144635.
© 1999 Blackwell Science Ltd
with ancient shorelines of the Florida peninsula, USA.
These xeric uplands, known as scrub, are characterized
by a high level of endemism and distinctive floral assemblages (Myers 1990). A natural archipelago of older scrub
ridges occurs along the central part of the Florida peninsula; geologically younger scrubs are found on relict
dunes behind contemporary shorelines. This scrub system is part of a complex landscape mosaic. Sandy ridges
MEC653.fm Page 1094 Wednesday, June 30, 1999 2:13 PM
1094 A . M . C L A R K , B . W. B O W E N and L . C . B R A N C H
dominated by scrub are typically separated by scores of
kilometres ( Jackson 1973); scrub habitats within ridges
are fragmented into islands separated by more mesic and
hydric habitats and, more recently, by human development.
The scrub lizard is conspicuously absent from some
scrub ridges ( Jackson 1973) and has a patchy distribution
among scrub islands within each ridge (Hokit et al. 1999).
This pattern suggests that the scrub lizard has limited
dispersal ability (Tiebout & Anderson 1997; Hokit et al.
1999), yet this species occurs on isolated, sand ridges
across the central and southern Florida peninsula, and
some of the geologically youngest habitats along the
southern Florida coastline contain S. woodi populations.
As noted by Jackson (1973), ‘the disjunct nature of the
habitat of S. woodi offers interesting problems of geographical distribution and variation.’
The history of S. woodi can be best understood in terms
of the historical biogeography of the Florida peninsula.
The genus Sceloporus is widespread and speciose in xeric
habitats of Mexico and the southwestern USA, prompting suggestions that S. woodi was derived from a western
ancestor during the Pliocene or early Pleistocene (Jackson
1973). Several other scrub taxa (e.g. the Florida scrub jay,
Aphelocoma coerulescens coerulescens) are affiliated with
sister taxa in the western USA. These faunal affinities
have been explained by the presence of semiarid terrain
along the northern rim of the Gulf of Mexico in the late
Pliocene and early Pleistocene, providing a corridor of
suitable habitat between Florida and western North
America during glacial maxima. Meylan (1982) reported
fossil evidence of several ‘western’ species in a glacialage, early Pleistocene site at Inglis, Florida. The subsequent formation of extensive wetlands along the Gulf
coast, beginning in the middle-Pleistocene and continuing to the present, effectively divided the formerly continuous scrub habitat into southwestern and southeastern
assemblages. Consistent with this biogeographic scenario,
several researchers have suggested that S. woodi is derived
from ancestors closely related to western Sceloporus forms
(e.g. S. virgatus or S. undulatus consobrinus) rather than
from S. undulatus undulatus, a form widespread in northern Florida and the southeastern USA (Jackson 1973;
Larsen & Tanner 1975; Smith et al. 1992).
During the Pliocene and Pleistocene, the Florida peninsula contracted and expanded repeatedly with glacial
cycles and corresponding changes in sea level (Webb
1990, fig. 4.8). The central Florida ridges were isolated
from the North American mainland during highest sea
levels. Hence, changes in sea level and corresponding
inundations have been a major influence on the distributions of Florida biota and may explain high levels of
endemism for Florida plants, insects, birds, and lizards
(Deyrup 1989; Huck et al. 1989; Christman & Judd 1990).
During the last half of the Pleistocene (or the last million
years), xeric habitat in southern Florida gradually gave
way to the wet subtropical habitat that predominates
today. During this interval, formerly widespread and
semicontinuous scrub habitats were reduced to the fragmented islands that persist today.
The biogeographic history of the Florida peninsula provides the context for our analysis of mitochondrial DNA
(mtDNA) cytochrome b diversity in Florida scrub lizards.
Range-wide mtDNA surveys have proven useful for
resolving biogeographic patterns in other terrestrial and
aquatic faunas of the southeastern USA (Avise 1992;
Osentoski & Lamb 1995; Walker & Avise 1998). Here, we
compare S. woodi populations from five major ridges that
span the range of this species (Fig. 1). Relationships
among lizard populations on separate ridges are used to
elucidate historical patterns of colonization and isolation.
Multiple sample sites within ridges are used to assess
fine-scale population structure.
Phylogeographic data from S. woodi and other endemic
faunas can reveal the geographical distribution of genetic
diversity among endangered scrub islands, substantially
enhancing the scientific foundations of reserve design
and management (Dizon et al. 1992; Vogler & Desalle
1994; Moritz & Faith 1998). The conservation context for
this study is especially compelling because of the precarious status of Florida scrub (McCoy & Mushinsky 1992).
As a result of burgeoning urbanization and agriculture in
the central peninsula and the coasts, Florida scrub is one
of the most endangered ecosystems in the southeastern
Unites States. Thirteen species of scrub plants and five
vertebrates are listed as endangered or threatened under
the U.S. Endangered Species Act. The Florida Committee
on Rare and Endangered Plants and Animals recognizes
the rapid loss of scrub habitat as a strong conservation
concern, and classifies S. woodi as threatened based on
these considerations (DeMarco 1992). Major conservation efforts are being directed toward establishing an
archipelago of scrub reserves to protect endemic flora
and fauna (The Nature Conservancy 1991; U.S. Fish &
Wildlife Service 1991).
Materials and methods
Scrub lizard samples (n = 135) were collected from five
ridges covering the historic distribution of the species:
Mt Dora Ridge, with one collection site in the north,
two in the centre, and two sites on the southern portion;
Atlantic Coastal Ridge, with one central and one southern
collection site; Lake Wales Ridge, with two sites on the
northern portion and two sites on the southern portion;
Bombing Range Ridge (Avon Park Air Force Range), with
four sites; and Gulf Coast Ridge, with a single collection
site (Fig. 1). Sample sizes and locations are described in
Table 1.
© 1999 Blackwell Science Ltd, Molecular Ecology, 8, 1093 –1104
MEC653.fm Page 1095 Wednesday, June 30, 1999 2:13 PM
P H Y L O G E O G R A P H Y O F S C E L O P O R U S W O O D I 1095
Fig. 1 Map of Florida showing locations
of the major ridges sampled for Sceloporus
woodi. Collection sites are indicated by
an ‘X’ adjacent to the ridge. S. Ocala
includes the Nicotoon collection site.
Ridges are redrawn from Deyrup
(1996). (RRB/SSR = Railroad Bed/Shirley
Shores, APAFR = Avon Park Air Force
Range, ARB = Arbuckle, CC = Carter Creek,
ARCH = Archbold).
Ridge
Mt Dora
North segment
Central segment
Central segment
South segment
South segment
Lake Wales
North segment
North segment
South segment
South segment
Bombing Range
N. of Kissimmee Rd.
N. of Kissimmee Rd.
S. of Kissimmee Rd.
S. of Kissimmee Rd.
Gulf Coast
Atlantic Coastal
Site
North Ocala
South Ocala
Nicotoon
Shirley Shores
Railroad Bed
Arbuckle
Carter Creek
Archbold
Venus
Patch 27
Patch 31
Patch 79
Patch 84
Naples
Titusville
JDSP
All ridges
N
36
9
10
3
12
2
39
10
10
8
11
29
10
5
12
2
11
20
9
11
135
Π
h
0.005
0.71
Haplotypes
A(2), B(3), C(4)
C(9), G(1)
C(3)
I(11), dd(1)
H(2)
0.016
0.69
D(9), T(1)
D(10), L(3)
J(6), K(1), Q(1)
J(9), R(1), S(1)
0.004
0.000
0.019
0.76
0.00
0.78
D(7), M(3)
D(2), U(2), V(2)
D(5), E(3), M(1),
N(2), O(1)
E(1), F(1)
P(11)
bb(9)
W(3), X(1), Y(1),
Z(4), aa(1), cc(1)
0.009
© 1999 Blackwell Science Ltd, Molecular Ecology, 8, 1093–1104
0.91
Table 1 Haplotypes of Sceloporus woodi
and number of samples from each ridge
and collection locality. N is the number
of individuals, Π is nucleotide diversity,
and h is haplotype diversity. The number
in parentheses indicates the number of
samples with each haplotype ( JDSP =
Jonathan Dickinson State Park)
MEC653.fm Page 1096 Wednesday, June 30, 1999 2:13 PM
1096 A . M . C L A R K , B . W. B O W E N and L . C . B R A N C H
Scrub lizards were captured by noosing or by hand, a
portion of the tail was removed, and, in most cases, animals were released promptly at the capture site. Tissues
were preserved in a salt buffer (saturated NaCl; 25 mm
EDTA pH 7.5; 20% DMSO) (protocol modified from
Amos & Hoelzel 1991). Samples of Sceloporus undulatus
undulatus (n = 3), S.u. consobrinus (n = 1), S. virgatus
(n = 1), and S. occidentalis (n = 1) were obtained to
determine whether the scrub lizard is derived from a
southeastern or southwestern ancestor, and to evaluate
divergences within S. woodi.
DNA isolations were accomplished with standard
phenol– chloroform methodology (Hillis et al. 1996).
mtDNA cytochrome b sequences were amplified with
polymerase chain reaction (PCR) technology using both
biotinylated and nonbiotinylated versions of primers
described by Kessing et al. (1989) (see also Kocher et al.
1989). Primer sequences (5′–AAAAAGCTTCCATCCAACATCTCAGCATGATGAAA-3′ and 5′–AAACTGCAGCCCCTCAGAATGATATTTGTCCTCA-3′) framed a
308-bp fragment. An 18-bp ‘universal’ M13 sequence
was added to the 5′-end of biotinylated primers to
facilitate automated sequencing (see below). The thermal
cycling parameters used were: 1 cycle at 94 °C (3 min)
followed by 35 cycles at 94 °C (1 min), 52 °C (1 min), and
72 °C (1 min). Standard precautions, including negative
controls (template-free PCR reactions), were used to
test for contamination and to ensure the fidelity of PCR
reactions.
Streptavidin-coated magnetic beads (Dynabeads M280
streptavidin, Dynal, Sweden) were used to purify some of
the PCR products. Single-stranded PCR products were
generated by denaturing the double-stranded DNA with
fresh 0.2 m NaOH, and using the nonbiotinylated strand
(in solution) as a template for sequencing reactions. Some
of the samples were processed as double-stranded PCR
products and purified with 30 000 MW Millipore filters.
Single-stranded and double-stranded sequencing reactions were conducted with fluorescently labelled M13
primers in a robotic work station (Applied Biosystems
model 800), and the labelled extension products were
analysed with an automated DNA sequencer (Applied
Biosystems model 373A) in the DNA Sequencing Core at
the University of Florida. Raw data from the sequencer
were edited and aligned using Sequencher 3.0 software
(Gene Codes Corp.). Those mtDNA sequences that
matched known haplotypes were collated for analysis,
whereas new haplotypes were re-amplified and sequenced
from the opposite direction to ensure the accuracy of
nucleotide sequence designations.
Sequencher 3.0 was used to identify the open reading frame and both the nucleotide and the amino acid
sequences were compared to registered sequences in
GenBank and confirmed as a fragment of cytochrome b.
Genetic distances between haplotypes were determined
with the Kimura 2-parameter model (Kimura 1980) and
an empirically derived 3:1 transition/transversion ratio.
To estimate relationships among haplotypes, dendrograms
were generated with the parsimony approach of paup*
4.0b1 (Swofford 1998) and the neighbour-joining algorithm
(Saitou & Nei 1987) in the program phylip version 3.57a
(Felsenstein 1990). Support for nodes in the mtDNA
dendrograms was assessed with bootstrap resampling of
the neighbour-joining tree using 100 replicates. In addition, haplotypes were linked in an unrooted parsimony
network and imposed on a map of Florida to resolve
phylogeographic patterns.
Haplotype diversity (h; eqn 8.4 in Nei 1987) and nucleotide diversity (π; eqn 10.5 in Nei 1987) within each
ridge and overall were calculated with the software
package reap version 4.0 (McElroy et al. 1992). The proportion of genetic diversity distributed among scrub
patches within a ridge, and among ridges across the
range of S. woodi, was estimated with an analysis of
molecular variance (amova version 1.55, Excoffier et al.
1992). In cases where sample locations shared haplotypes, a chi-squared test for differences in haplotype
frequencies was conducted with the program chirxc
(Zaykin & Pudovkin 1993) using 1000 randomizations of
the original data matrix to estimate a probability distribution for each test (see Roff & Bentzen 1989). Sites with
sample sizes of less than eight were excluded from
amova and chi-squared analyses, but were included in
overall estimates of nucleotide and haplotype diversity.
Results
A cytochrome b fragment of 273 bp was resolved in all
135 samples, and subsequent analyses refer to this
fragment. Third position transitions constituted 73% of
the polymorphic sites. Sequence comparisons revealed 44
variable sites containing 35 transitions and 14 transversions (Table 2). Thirty haplotypes were resolved, differing by an average sequence divergence of P = 0.048
and a maximum of P = 0.099. Haplotype and nucleotide diversities for each ridge and overall are presented
in Table 1. Among the younger coastal scrubs, Jonathan
Dickinson State Park (JDSP, southern Atlantic Coastal
Ridge) had relatively high haplotype diversity, with six
haplotypes in 11 individuals, whereas the populations
in Titusville (central Atlantic Coastal Ridge; n = 9) and
Naples (Gulf Coast Ridge; n = 11) each had a single
haplotype (Table 1).
Each of the older central ridges is characterized by a
pool of closely related haplotypes that cluster around
haplotypes ‘D’ and ‘J’ within the Lake Wales Ridge and
Bombing Range Ridge and haplotypes ‘C’ and ‘I’ on the
Mt Dora Ridge (Fig. 2). In cases where haplotypes are
© 1999 Blackwell Science Ltd, Molecular Ecology, 8, 1093 –1104
MEC653.fm Page 1097 Wednesday, June 30, 1999 2:13 PM
© 1999 Blackwell Science Ltd, Molecular Ecology, 8, 1093–1104
Table 2 Variable sites observed in cytochrome b sequences of Sceloporus woodi. The thirty haplotypes are indicated as letters A to dd. The numbers at the top indicate the locations of
polymorphic sites within the 273-base sequence
3
3
0
3
6
A T
G
G
G
A
G
G
G
T T
C
C
C
G
G
4
9
G C
G C
5
7
8
4
8
7
9
2
1
0
2
1
1
4
T A C A G T
C
A
C
C
A
A
A
C
1
1
5
1
2
3
1
2
9
1
5
0
1
5
9
A C
C
T
G T
A
A
A
A
A
T
G
C
C
C
C
C
C
C
G C
A
A
A
A
A
A
A
C
C
C
C
C
C
C
1
6
5
1
6
8
1
7
4
A A C
G
1
8
6
1
8
9
2
0
5
2
1
0
2
1
3
2
1
4
2
1
6
G A A C A C
A
A
A
2
2
2
2
2
3
2
2
5
2
2
8
A G A T
2
3
1
2
4
0
2
4
3
2
4
9
2
5
1
2
5
5
2
5
8
2
6
7
2
7
0
2
7
2
C A
G
A
A
A
C
C C G C
T
T
T
C
C
T
C
C
C
A
A
A
A
A
A
A
A
A
A
C
C
C
C
C
2
0
4
G
T
T
T
T
G
1
9
7
G A A T
C
C
C
A
A
A
A
A
A
A
C
C
C
C
C
C
A C
A
A
A
C
C
C
T
T
T
T
T
G
G
G
G
G
C G
T
A
G A
A A
G
T
A
A
A
A
A
G
A
T
T
T
T
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
G
G
G
G
G
C
1
9
3
G
T
T
A
A
1
6
4
A
A
A
A
A
G
A
C C
A
A
A
A
A
A
A
A
T
T
T
T
T
T
T
G T
G
G
G
G T
G
G
C
C
C
C
C
A C
C
C
C
C
C
C
C
C
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
A
T
T
T
T
T
T
T
T
T
T
T
T
T
P H Y L O G E O G R A P H Y O F S C E L O P O R U S W O O D I 1097
CONCENSUS
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
P
Q
R
S
T
U
V
W
X
Y
Z
aa
bb
cc
dd
2
3
MEC653.fm Page 1098 Wednesday, June 30, 1999 2:13 PM
1098 A . M . C L A R K , B . W. B O W E N and L . C . B R A N C H
Fig. 2 Parsimony network for Sceloporus woodi indicating the relationship
of the five major ridges and collection
localities on those ridges. Hash marks
between the haplotypes indicate the
number of base differences (if more
than one). Thin lines without hash
marks represent single base changes.
Thick lines delineate the three major
clusters of haplotypes, with number of
nucleotide differences described in base
pairs (bp).
Ridge
Patches
Distance (km)
χ2
P
Mt Dora
N Lake Wales
S Lake Wales
Bombing Range
Bombing Range
North ONF vs. South ONF
Arbuckle vs. Carter Creek
Archbold vs. Venus
Patch 27 vs. Patch 79
Arbuckle vs. Patch 27
Arbuckle vs. Patch 79
Carter Creek vs. Patch 27
Carter Creek vs. Patch 79
41.8
17.5
10.5
12.6
9.5
16.5
24.8
19.3
7.8
4.3
4.2
12.1
4.3
9.0
6.0
10.2
0.03
0.21*
0.58*
0.03
0.24*
0.01
0.05
0.03
Lake Wales Ridge
Table 3 Chi-squared tests for differences
in haplotype frequencies in Sceloporus
woodi populations that share mtDNA
haplotypes. P is the probability that the
observed χ2-value will be exceeded in
1000 Monte Carlo simulations. Distance
equals the number of km between patches
ONF = Ocala National Forest.
Asterisks indicate the three nonsignificant P-values.
shared between patches (within ridges), each patch typically contains one of these ‘central’ haplotypes and one or
two closely related haplotypes (Table 1; Fig. 2). In the
cases of fixed differences between patches within a ridge,
the endemic haplotypes are distinguished from a common haplotype by 1– 3 nucleotide substitutions.
Samples from the five major ridges are characterized by
fixed differences in mtDNA sequences with the exception
of haplotype ‘D’, which is shared by lizard populations
in the northern portion of the Lake Wales Ridge and the
adjacent Bombing Range Ridge. All patches that share
haplotype ‘D’ on the Lake Wales Ridge and the Bombing
Range Ridge have significantly different haplotype frequencies except the two patches in closest proximity to
the northern region of the two ridges (Arbuckle and
Bombing Range Patch 27; Table 3). Fixed differences were
observed between populations on the northern and
southern segments of the Lake Wales Ridge, and between
the southernmost populations on the Mt Dora Ridge
(Shirley Shores and Railroad Bed) and the more northern
© 1999 Blackwell Science Ltd, Molecular Ecology, 8, 1093 –1104
MEC653.fm Page 1099 Wednesday, June 30, 1999 2:13 PM
P H Y L O G E O G R A P H Y O F S C E L O P O R U S W O O D I 1099
Table 4 Hierarchical analysis of variance across ridges for
Sceloporus woodi. The percentage of variance (%), probability
estimated from permutation tests (P), and the Φ-statistic are
given at each hierarchical level (see Excoffier et al. 1992). Ridges
are defined as in Table 1
Variance component
%
P
Φ
Among ridges
Within ridges
Within patches
72.11
17.50
10.39
< 0.01
< 0.01
≤ 0.01
0.896
0.627
0.721
populations (north and south Ocala National Forest
[ONF] and Nicotoon; Table 1). In addition, there are significant haplotype frequency differences between northern and southern populations on the Bombing Range
Ridge and within the Mt Dora Ridge (north and south
ONF, Table 3). Among 66 possible pairwise comparisons
between scrub patches, 58 were characterized by fixed
differences in haplotype composition, five were characterized by significant haplotype frequency shifts, and only
three pairs of adjacent patches were not significantly
different (Table 3). This strong genetic structuring is
apparent in a hierarchical analysis of genetic variance. An
estimated 10.4% of total variation was observed within
patches, 17.5% among patches within ridges, and 72.1%
among ridges (Table 4). These values are all significant,
indicating strong structuring at the finest scale assayed in
this study, as well as relatively deep evolutionary separations among ridges.
A hand-constructed parsimony network (Fig. 2) demonstrates the presence of three clusters of haplotypes within
Sceloporus woodi, corresponding to: (1) the northern central ridge (Mt Dora Ridge); (2) the southern central ridges
(Bombing Range Ridge and the Lake Wales Ridge) and
the Gulf Coast Ridge; and (3) the Atlantic Coastal Ridge.
The three lineages differ from each other by 11–21 base
substitutions. Lineages representing the northern and
southern portions of the southern central Florida ridges
differ by about 4% sequence divergence. Populations on
the Atlantic Coastal Ridge (Titusville and JDSP) exhibit
3–4% sequence divergence from each other and 7– 9%
from all other localities (Fig. 3).
A neighbour-joining tree of the most common haplotypes from each location (Fig. 3) has the same topology
as the most parsimonious tree recovered with the
branch and bound option in paup* 4.0b1. All nodes in the
neighbour-joining tree are in consensus with the parsimony tree and the three major branches are supported by
bootstrap values of 87–100% (Fig. 3). The lineage containing
Fig. 3 Neighbour-joining tree for Sceloporus woodi based on the most common
haplotypes in each location. The ridges
and some of the corresponding collection
sites are indicated at the right of the figure.
Numbers at the nodes indicate bootstrap
support.
© 1999 Blackwell Science Ltd, Molecular Ecology, 8, 1093–1104
MEC653.fm Page 1100 Wednesday, June 30, 1999 2:13 PM
1100 A . M . C L A R K , B . W. B O W E N and L . C . B R A N C H
the Lake Wales and Bombing Range ridges is further
differentiated into northern and southern components
with 90% bootstrap support, and Atlantic Coastal Ridge
sites (Titusville and JDSP) are affiliated with the Mt Dora
Ridge with 87% bootstrap support. The neighbour-joining
tree also demonstrates the relationships of the out-groups
indicating that S. woodi is the sister taxon to eastern
S. undulatus undulatus rather than one of the western
species, with 77% bootstrap support. The minimum
divergences between S. woodi and the out-groups are
9.4% for S.u. undulatus, 12% for S.u. consobrinus, 17% for
S. virgatus, and 13% for S. occidentalis as compared with
the maximum divergence within S. woodi at 9.9%.
Discussion
Population structure
The distribution of mtDNA diversity in Sceloporus woodi
is highly structured and clearly reflects the historical
isolation of scrub islands at multiple scales. Fixed differences
were documented between patches within the southern
segment of the Mt Dora ridge and between northern and
southern segments of the Lake Wales Ridge. These partitions coincide with previously recognized biogeographic
breaks near Josephine Creek on the Lake Wales Ridge and
the southern end of the Mt Dora Ridge (Deyrup 1996; M.
Deyrup, personal communication). Populations within
Ocala National Forest separated by 41.8 km (North and
South ONF) are more similar than populations that are
separated by a shorter distance on either side of the
southern Mt Dora Ridge break (e.g. South ONF and
Shirley Shores, 23.5 km; Table 3).
Fine-scale population structure also is demonstrated by
significant differences in haplotype frequencies among all
but three pairs of patches (Table 3). Strong population
structure within ridges is consistent with field studies
indicating that the distribution of S. woodi is tightly linked
to open, sandy scrub habitat and that the habitat matrix
around scrub islands significantly inhibits interpatch
movements (Tiebout & Anderson 1997; Hokit et al. 1999).
Dispersal between patches is unlikely unless patches are
very close together (e.g. under 400 m) or connected by
corridors of suitable habitat.
The only haplotypes shared between ridges occurred on
the northern Lake Wales Ridge and the nearby Bombing
Range Ridge, and corresponding haplotype frequencies
are not significantly different in one pair of patches.
Although these two patches are on ridges with different
geological histories, numerous scrub islands, currently
unoccupied by scrub lizards, lie between these two patches
(Hokit et al. 1999). These islands may have enhanced
landscape connectivity by serving as ‘stepping stones’ for
infrequent dispersers.
Differences in current and historical patterns of landscape connectivity may explain why significant differences in haplotype frequencies occur between some
lizard populations but not others (Table 3). Scrub patches
within the northern and southern parts of the Lake Wales
Ridge (Arbuckle and Carter Creek, Archbold and Venus)
once formed part of a very large, highly connected scrub
ridge bisected by Josephine Creek (Lohrer & Commings
1993). During the last century, more than 85% of the xeric
uplands of the Lake Wales Ridge has been converted to
human use resulting in large-scale habitat fragmentation
(McDonald & Hamrick 1996). The genetic structure of
scrub lizards on the Lake Wales Ridge evidently reflects
the past connectivity among patches, rather than current
fragmentation. In contrast, aerial photographs spanning
six decades on Bombing Range Ridge suggest that the
patchiness of this scrub was generated by natural landscape processes rather than anthropogenic factors. The
significant differences in haplotype frequencies reported
for patches on the northern (Patch 27) and southern
(Patch 79) parts of this ridge complement field studies
that demonstrate demographic independence among
these populations and indicate that multiple metapopulations may occur within a ridge (Hokit et al. 1999).
The population-level separations between patches
within ridges, as resolved with mtDNA sequences, are
consistent with a metapopulation model. In a companion
study, Hokit et al. (1999) concluded that extinction and
recolonization processes are important influences on the
local population structure of the Florida scrub lizard. This
model is supported by our finding of pools of closely
related haplotypes within each ridge, indicating recent
contact (in an evolutionary sense) between patches. Both
genetic data and field studies show that migration is low
between naturally isolated patches, but also indicate a
role for rare dispersal and recolonization events.
Phylogeography
In contrast to patterns observed within ridges, the genetic
differentiation among major ridges indicates separations
over a large evolutionary time frame. The mtDNA divergences among S. woodi populations on different ridges
range from 1 to 3% between scrubs in southwest Florida
to 7–10% between ridges of the Atlantic Coast and central
Florida. These latter values are at the high end of divergences
reported for intraspecific partitions in other taxonomic
groups (Grant & Bowen 1998; Walker & Avise 1998).
The history of S. woodi populations is, of necessity,
linked to the history of scrub habitats derived from
ancient coastlines. At least six ancient shorelines were
created during the 25-Myr history of the Florida peninsula, and the sand ridges along the central portion of the
Florida peninsula are considered to be the oldest of
© 1999 Blackwell Science Ltd, Molecular Ecology, 8, 1093 –1104
MEC653.fm Page 1101 Wednesday, June 30, 1999 2:13 PM
P H Y L O G E O G R A P H Y O F S C E L O P O R U S W O O D I 1101
these (Webb 1990). Thus, the present Florida landscape is
composed of older ridges on a north–south axis along the
centre of the state, followed by younger ridges towards
the coastlines. Sediment cores from the Lake Wales scrub
have been dated to late Miocene (9 Myr bp) (Ketner &
McGreevy 1959; Pirkle et al. 1970). However, the oldest
contemporary ridges (such as the Archbold site) date to
the Pliocene (3 –5 Myr bp) (Pirkle & Yoho 1970; Opdyke et al.
1984) and younger coastal scrubs are believed to have a
Pleistocene origin (0.5 –2 Myr; Jackson 1973; Webb 1990).
The accumulated geological evidence indicates that the
formation of contemporary scrub habitats began during
the Pliocene and Pleistocene. This time frame compares
favourably to the observed molecular evolutionary separations in our mtDNA genealogy. Estimates of molecular
evolutionary rates for cytochrome b range from 0.2%
per million years in sea turtles (Bowen et al. 1993) to a
more conventional 1–3% per million years for ungulate
mammals and other vertebrates (Irwin et al. 1991). The
lower rate can probably be dismissed, because a divergence
of P = 0.099 (the deepest separation within S. woodi) would
correspond to 50 million years, an age that predates the
Florida peninsula. The widely applied rate of 2% per
million years would correspond to approximately 5 Myr
for the oldest separations within S. woodi. This time frame
overlaps the proposed Pliocene origin of extant scrub
habitats of the central Florida ridge. Under the same time
estimation, the separations of 1–4% between northern
and southern sites on the Bombing Range and Lake Wales
ridges, and between the central ridge and southwest
Florida, correspond to about 0.5 –2 Myr. These values are
consistent with the estimated Pleistocene age of coastal
scrub habitats. While clock estimates must be interpreted
with caution, the molecular data indicate a scale of
millions of years for the isolation of S. woodi populations
on the major ridges of Florida.
Genetic separations seem to be approximately the same
age as the extant scrub habitats. One notable exception to
this pattern is that the younger scrubs on the Atlantic coast
contain some of the deepest lineages in the mtDNA phylogeny. This can also be understood in terms of sea levels
and habitat changes over the last 2 Myr. Through the late
Pliocene (2–3 Myr bp), Florida’s landmass was about twice
the current size due to lower sea levels, and xeric conditions (typical S. woodi habitat) were widespread in
the southern Florida peninsula (Watts & Hansen 1988). By
the mid-Pleistocene (about 1 Myr bp), sea level rise had
reduced the Florida landmass, and mesic conditions
extended into the southern peninsula. Broad areas of
xeric habitat persisted into the mid-Holocene (5000 years
bp) but were gradually replaced by mesic habitat as a result
of increased precipitation and/or increased water tables
(Watts & Hansen 1988). As mesic habitat expanded, xeric
habitats were inevitably fragmented, producing the mosaic
© 1999 Blackwell Science Ltd, Molecular Ecology, 8, 1093–1104
Fig. 4 Parsimony network for Sceloporus woodi showing the
most common haplotypes from each ridge connected by lines
representing the probable sequence of colonization from north to
south on the Florida peninsula.
of S. woodi habitats that exists today. Therefore, the scrubs
that exist behind current coastlines may represent the
remnants of more extensive habitats. The deep differentiation of S. woodi samples in young Atlantic coastal scrubs
probably predates contemporary coastal geography.
Sceloporus is believed to have invaded Florida during
the Pliocene (Jackson 1973), and the genetic data are consistent with a Pliocene radiation of Sceloporus lineages
through Florida, followed by fragmentation of habitat
into scrub islands during the Pleistocene. A sequence of
colonization events can be inferred from the relationships
among mtDNA haplotypes. The parsimony network
(Fig. 2) indicates three primary branches corresponding
to: (1) the northern central ridge of Florida (Mt Dora);
(2) the Atlantic Coastal Ridge; and (3) the southern
central ridges (Lakes Wales and Bombing Range, plus
Gulf Coast Ridge). When this network is imposed on
a map of contemporary Florida (using only the most
abundant haplotype from each location for simplicity),
the branch leading towards the Atlantic coastline suggests colonization events originating from the northern
central ridge (Mt Dora Ridge, Fig. 4) and proceeding in
a southerly direction along the coastline. Similar genetic
affiliations between the Mt Dora and Atlantic Coastal
ridges have been found in scrub plants (McDonald &
Hamrick 1996) and in Florida scrub jays (S. Edwards,
personal communication). In contrast, the low level of genetic differentiation of the Gulf coast samples indicates more
recent divergence from the southern Lake Wales Ridge.
MEC653.fm Page 1102 Wednesday, June 30, 1999 2:13 PM
1102 A . M . C L A R K , B . W. B O W E N and L . C . B R A N C H
The major lineages observed in the mtDNA genealogy
align well with a partition based on morphological differentiation. Jackson (1973) surveyed 19 morphological
characters to resolve biogeographic patterns in S. woodi
and reported a fundamental split between the southern
central ridge (Lake Wales Ridge and adjacent areas)
and the northern central ridge (Mt Dora Ridge), with
Atlantic coastal populations affiliated with the northern
central ridge. Both the parsimony network (Fig. 2) and
the neighbour-joining tree support the differentiation
of northern and southern ridges, and Fig. 3 indicates
that Atlantic coastal populations (Titusville and JDSP)
are affiliated with the northern central ridge (bootstrap
value 87%). In this respect, the mtDNA gene genealogy is consistent with the distinction of two morphotypes in S. woodi. However, the mtDNA data also
indicate that the Atlantic coastal populations are well
differentiated, perhaps indicating an additional evolutionary partition.
The centre of Sceloporus species diversity lies in western North America and Central America, and it has been
suggested that S. woodi is derived from a western population of the S. virgatus group ( Jackson 1973; Smith et al.
1992). However, recent genetic evidence from 12S and 16S
ribosomal RNA genes unites S. woodi with S.u. undulatus,
the only other Sceloporus species in the southeastern
United States, rather than with the western S.u. consobrinus
or S. virgatus (Weins & Reeder 1997). Our findings with
cytochrome b reinforce this conclusion (Fig. 3). Notably,
the observed differentiation of S.u. undulatus (the eastern
subspecies) and S. woodi (P = 0.098 – 0.138) overlaps the
higher levels of divergence within S. woodi (Pmax = 0.099).
These findings prompt the conclusion that the primary
mtDNA lineage separations within S. woodi occurred
early in the evolutionary history of this species.
In many organisms surveyed with mtDNA sequences,
the branch lengths within species are an order of magnitude shorter than the differentiation between species.
This phenomenon is typically attributed to lineage sorting,
population bottlenecks, metapopulation processes on
a species-wide scale, or other demographic processes
(Grant & Bowen 1998). In contrast, almost the entire
evolutionary history of S. woodi seems to be preserved in
extant mtDNA lineages. We attribute this to dispersal
through the Florida peninsula followed by habitat
fragmentation and isolation that occurred before natural
lineage sorting could erase the early biogeographic
history of S. woodi.
Conservation, genetic diversity,
and wildlife management
Our genetic data indicate that S. woodi populations in isolated scrub patches are demographically independent,
and that lizard populations on the major scrub archipelagos have been disjunct for several million years.
These findings have three implications for the conservation of S. woodi. First, these data corroborate field studies
and landscape modelling (Tiebout & Anderson 1997;
Hokit et al. 1999), which indicate that scrub lizards are not
likely to recolonize distant patches following local extinctions, particularly if intervening scrub patches have been
removed. Thus, scrub populations separated by more
than a few hundred metres should be treated as separate
management units. Second, management practices such
as translocation may have profound consequences for the
maintenance of genetic diversity in the Florida scrub
lizard. Translocations have become a prominent strategy
in efforts to conserve threatened and endangered species
(Griffith et al. 1989), but the strong genetic partitioning in
the Florida scrub lizards indicates that translocations
could compromise the integrity of genetic differences that
have accumulated over thousands to a few million years.
Third, these findings invoke concern that the taxonomy
of this species does not adequately reflect evolutionary
partitions. The extensive intraspecific divergence of mtDNA
lineages in S. woodi is equivalent to that spanning distinct
species pairs in other settings. At least one other scrubassociated reptile, the mole skink (Eumeces egregius), is
divided into subspecies corresponding to different parts
of the Florida peninsula.
Although our research reveals deep genetic subdivisions in S. woodi, we do not advocate new taxonomic
designations at this time, based on the observation
by Jackson (1973) that the morphological differences
between lizards on northern and southern ridges are not
sufficient to warrant species or subspecies assignments.
However, we note that lizard populations from isolated
scrub archipelagos, distinguished by 3 –10% sequence
divergence in cytochrome b, probably qualify as evolutionarily significant units (Moritz 1994), and therefore
may qualify for recognition under the U.S. Endangered
Species Act. To the extent that management plans serve
to protect genetic diversity, individual populations of
S. woodi merit protection and major scrub archipelagos
merit a high conservation priority.
Some of the surveyed areas are protected (e.g. Avon
Park Air Force Range, Jonathan Dickinson State Park,
Archbold Biological Station, and Ocala National Forest),
but many others are disappearing under an ongoing wave
of urban expansion and agricultural development. Indeed,
an effort to enhance sample size from one study site (at
Naples) failed because the area had been paved over. The
loss of this site probably signals the loss of one branch in
the mtDNA tree, and more branches are likely to follow
under the current pace of development. Many vertebrates
that inhabit Florida scrub are believed to have low dispersal
capability (Jackson 1973), and the plant communities on
© 1999 Blackwell Science Ltd, Molecular Ecology, 8, 1093 –1104
MEC653.fm Page 1103 Wednesday, June 30, 1999 2:13 PM
P H Y L O G E O G R A P H Y O F S C E L O P O R U S W O O D I 1103
separate scrub ridges are isolated as well (McDonald &
Hamrick 1996). It is likely that strong genetic partitions,
as a result of vicariant separations over the last few million years, are a key feature of Florida scrub inhabitants.
Under this scenario, the extirpation of scrub habitats
will carry an enormous toll in terms of loss of genetic
diversity. Phylogeographic data on other scrub taxa are
necessary to test this conclusion. If the relatively deep
evolutionary separations observed in S. woodi are indicative of a general pattern in scrub species, then an aggressive strategy for conservation of scrub habitat is required
to conserve the genetic diversity in this unique ecosystem.
Acknowledgements
This project was made possible by the outstanding contributions
of Grant Hokit, Paul Moler, Brad Stith, and David Cook. For
field assistance and tissue collections, we thank Ab Abercrombie,
John Arnett, Scott Berish, Steve Godley, Katie Greenberg, Kerry
Griffis, John Jensen, Steve Johnson, Krisann Kosel, Kevin Long,
Barry Mansell, Carol May, John Matter, Pam Mikula, Chris
O’Brien, Carl Qualls, Perran Ross, Wayne VanDevender, and
Kenny Wray. Wayne VanDevender collected and identified
the S. consubrinus, S. virgatus, and S. occidentalis used in the
analysis. For logistic support we are indebted to Paul Ebersbach,
Karla Hokit, Ruth Klinger, John Fitzpatrick, Julie Owens, Bob
Progulske, Hilary Swain, and Pat Walsh. Field studies were
facilitated by staff at Avon Park Air Force Range, Archbold
Biological Station, Arbuckle State Forest, and the Division of
Forestry. For assistance with genetic analyses, we thank Ernesto
Almira, Savita Shanker, Sandra Encalada, Angela Garcia, Anna
Bass, Daniel Brazeau, and the DNA Sequencing Core at University
of Florida. We thank Mark Deyrup, Dick Franz, Scott Edwards,
David Webb and the many Florida naturalists who contributed
ideas and expertise. The manuscript was improved by reviews
from Dave Webb, Trip Lamb, and Paul Moler. This project was
funded primarily by the Florida Game and Fresh Water Fish
Commission and the U.S. Department of Defense, with additional
support from the Biological Resources Division of the USGS, and
the Species at Risk Initiative of the National Biological Service.
This is Florida Agricultural Series No. R-06764.
References
Amos B, Hoelzel AR (1991) Long-term preservation of whale
skin for DNA analysis. Report to the International Whaling Commission, Special Issue, 13, 99 –103.
Avise JC (1992) Molecular population structure and the biogeographic history of a regional fauna: a case history with lessons
for conservation biology. Oikos, 63, 62–76.
Bowen BW, Nelson WS, Avise JC (1993) A molecular phylogeny for marine turtles: Trait mapping, rate assessment, and
conservation relevance. Proceedings of the National Academy of
Sciences of the USA, 90, 5574 – 5577.
Christman SP, Judd WS (1990) Notes on plants endemic to
Florida scrub. Florida Scientist, 47, 52–73.
DeMarco V (1992) Florida scrub litard, Sceloporus woodi. In: Rare
and Endangered Biota of Florida, Vol. III. Amphibians and Reptiles
© 1999 Blackwell Science Ltd, Molecular Ecology, 8, 1093–1104
(ed. Moler PE), pp. 141–145. University Press of Florida,
Gainesville, FL.
Deyrup M (1989) Arthropods endemic to Florida scrub. Florida
Scientist, 52, 254–270.
Deyrup M (1996) Two new grasshoppers from relict uplands of
Florida (Orthoptera: acrididae). Transactions of the American
Entomological Society, 122, 199–211.
Dizon AE, Lockyer C, Perrin WF, Demaster DP, Sisson J (1992)
Rethinking the stock concept: a phylogeographic approach.
Conservation Biology, 6, 24–36.
Excoffier L, Smouse PE, Quattro JM (1992) Analysis of
molecular variance inferred from metric distances among
DNA haplotypes: application to human mitochondrial
restriction data. Genetics, 131, 479–491.
Felsenstein J (1990) PHYLIP manual, Version 3.57a. University of
Washington, Seattle.
Grant WS, Bowen BW (1998) Shallow population histories in
deep evolutionary lineages of marine fishes: insights from the
sardines and anchovies and lessons for conservation. Journal of
Heredity, 89, 415–426.
Griffith B, Scott JM, Carpenter JW, Reed C (1989) Translocation
as a species conservation tool: status and strategy. Science,
245, 477–480.
Hillis DM, Mable BK, Larson A, Davis SK, Zimmer EA (1996)
Nucleic Acids IV: sequencing and cloning. In: Molecular
Systematics, 2nd edn (eds Hillis DM, Moritz C, Mable BK),
pp. 321–381. Sinauer Associates, Sunderland, MA, USA.
Hokit DG, Stith BM, Branch LC (1999) Effects of landscape
structure in Florida scrub: a population perspective. Ecological
Applications, 9, 124–134.
Huck RB, Judd WS, Whitten WM (1989) A new Dicerandra
(Labiatae) from the Lake Wales Ridge of Florida, with a
cladistic analysis and discussion of endemism. Systematic
Botany, 14, 197–213.
Irwin DM, Kocher TD, Wilson AC (1991) Evolution of the
cytochrome b gene of animals. Journal of Molecular Ecology, 32,
128–144.
Jackson JF (1973) Distribution and population phenetics of the
Florida scrub lizard, Sceloporus woodi. Copeia, 4, 746 –761.
Kessing B, Croom H, Martin A, McIntosh C, McMillan WO,
Palumbi S (1989) The Simple Fool's Guide to PCR. Dept. Zoology,
University of Hawaii, Honolulu, HI 96822.
Ketner KB, McGreevy LJ (1959) Stratigraphy of the area between
Hernando and Hardee counties, Florida. U.S. Geological Survey
Bulletin 1074-C.
Kimura M (1980) A simple method for estimating evolutionary
rate of base substitutions through comparative studies of
nucleotide sequences. Journal of Molecular Evolution, 16, 111–
129.
Kocher TD, Thomas WK, Meyer A et al. (1989) Dynamics
of mitochondrial DNA evolution in animals: amplification
and sequencing with conserved primers. Proceedings of the
National Academy of Sciences of the USA, 86, 6196 – 6200.
Larsen KR, Tanner WW (1975) Evolution of the sceloporine lizards
(Iguanidae). Great Basin Naturalist, 34, 1– 41.
Lohrer F, Commings D (1993) Fiftieth Anniversary Biennial Report
199 1 –9 2. Archbold Biological Station, Lake Placid, Florida.
McCoy ED, Mushinsky HR (1992) Rarity of organisms in the sand
pine scrub habitat of Florida. Conservation Biology, 6, 537– 548.
McDonald DB, Hamrick JL (1996) Genetic variation in some
plants of Florida scrub. American Journal of Botany, 83, 21–27.
MEC653.fm Page 1104 Wednesday, June 30, 1999 2:13 PM
1104 A . M . C L A R K , B . W. B O W E N and L . C . B R A N C H
McElroy D, Moran P, Bermingham E, Kornfield I (1992) reap:
an integrated environment for the manipulation and phylogenetic analysis of restriction data. Journal of Heredity, 83, 157–
158.
Meylan PA (1982) The squamate reptiles of the Inglis IA fauna
(Irvingtonian: Citrus County, Florida). Bulletin of the Florida
State Museum of Natural History, 27, 1–85.
Moritz C (1994) Applications of mitochondrial DNA analysis
in conservation: a critical review. Molecular Ecology, 3, 403 –
413.
Moritz C, Faith D (1998) Comparative phylogeography and the
identification of genetically divergent areas for conservation.
Molecular Ecology, 7, 419 – 429.
Myers RL (1990) Scrub and high pine. In: Ecosystems of Florida
(eds Myers RL, Ewel JJ), pp. 150 –193. University of Central
Florida Press, Orlando, FL.
Nei M (1987) Molecular Evolutionary Genetics. Columbia University Press, New York.
Opdyke ND, Spangler DP, Smith DL, Jones DS, Lindquis RC
(1984) Origin of the epeirogenic uplift of Pliocene-Pleistocene
beach ridges in Florida and development of the Florida karst.
Geology, 12, 226 –228.
Osentoski MF, Lamb T (1995) Intraspecific phylogeography of
the gopher tortoise, Gopherus polyphemus: RFLP analysis of
amplified mtDNA segments. Molecular Ecology, 4, 709–718.
Pirkle EC, Yoho WH (1970) The heavy mineral ore body of Trail
Ridge, Florida. Economic Geology, 65, 17–30.
Pirkle EC, Yoho WH, Hendry CW Jr (1970) Ancient Sea Level Stands
in Florida. Bureau of Geology, Geological Bulletin no. 52.
Roff DA, Bentzen P (1989) The statistical analysis of mitochondrial DNA polymorphisms: chi-square and the problem
of small samples. Molecular Biology and Evolution, 6, 539–545.
Saitou N, Nei M (1987) The neighbor-joining method: a new
method for reconstructing phylogenetic trees. Molecular Biology
and Evolution, 4, 406 – 425.
Smith HM, Bell EL, Applegarth JS, Chiszar D (1992) Adaptive
convergence in the lizard superspecies Sceloporus undulatus.
Bulletin of the Maryland Herpetological Society, 28, 123–149.
Swofford DL (1998) PAUP*: Phylogenetic Analysis Using Parsimony,
Version 4.0b1. Sinauer, Sunderland, MA.
The Nature Conservancy (1991) Lake Wales/Highland Ridge
Ecosystem Carl Project Proposal. Unpublished report. Archbold
Biological Station, Florida Natural Areas Inventory.
Tiebout HM, Anderson RA (1997) A comparison of corridors
and intrinsic connectivity to promote dispersal in transient
successional landscapes. Conservation Biology, 11, 620 – 627.
US Fish and Wildlife Service (1991) Preliminary project proposal:
Proposed establishment of Lake Wales Ridge National Wildlife
Refuge. USFWS, Atlanta, GA.
Vogler AP, Desalle R (1994) Diagnosing units of conservation
management. Conservation Biology, 8, 354 –363.
Walker D, Avise JC (1998) Principles of phylogeography as
illustrated by freshwater and terrestrial turtles in the southeastern United States. Annual Review of Ecology and Systematics,
29, 23–58.
Watts WA, Hansen BCS (1988) Environments of Florida in the
Late Wisconsin and Holocene. In: Wet Site Archaeology (ed.
Purdy BA), pp. 307–323. The Telford Press, Caldwell, NJ.
Webb SD (1990) Historical biogeography. In: Ecosystems of Florida
(eds Myers RL, Ewel JJ), pp. 70–102. University of Central
Florida Press, Orlando, FL.
Weins JJ, Reeder TW (1997) Phylogeny of the spiny lizards
(Sceloporus) based on molecular and morphological evidence.
Herpetological Monographs, 11, 1–101.
Zaykin DV, Pudovkin AI (1993) Two programs to estimate significance of X2 values using pseudo-probability tests. Journal of
Heredity, 84, 152.
This paper is the result of a multidisciplinary initiative
developed by L. C. Branch (University of Florida) to resolve
conservation problems of species with patchy distributions.
Dr Branch is studying the effects of landscape structure on the
distribution and population dynamics of animals using field
studies, modelling, genetics, and remote sensing. A.M. ‘Ginger’
Clark is laboratory coordinator for the BEECS Genetic Analysis
Core (University of Florida) and is interested in the phylogeography and conservation genetics of reptiles. Brian W. Bowen
has a long-standing interest in the phylogeography of reptiles
and marine vertebrates and is a recognized authority on premium
tequilas.
© 1999 Blackwell Science Ltd, Molecular Ecology, 8, 1093 –1104