Content uploaded by Jason Strickland
Author content
All content in this area was uploaded by Jason Strickland on Nov 20, 2014
Content may be subject to copyright.
Phylogeography of Agkistrodon piscivorus with Emphasis on the Western
Limit of Its Range
Jason L. Strickland
1,2
, Christopher L. Parkinson
2
, J. Kelly McCoy
1,3
, and
Loren K. Ammerman
1
The Cottonmouth, Agkistrodon piscivorus, is a semi-aquatic pitviper that occupies the southeastern U.S. west into
Texas. Several previous studies have investigated the biogeographic history of A. piscivorus. It has been hypothesized
that A. piscivorus was split into two separate populations during the last glacial maximum and climate change has
impacted its distribution. Additionally, a geographically isolated population of A. piscivorus occurs at the western
limit of the species’ range in the Concho Valley of Texas. To investigate biogeography and population structure
within A. piscivorus in Texas and throughout its range, we utilized amplified fragment length polymorphism (AFLP)
and sequence data from cytochrome b(cyt-b). The AFLP data indicate a lack of gene flow between the population of
A. piscivorus in the Concho Valley and other nearby populations. However, based on cyt-b, there is no genetic
differentiation. The AFLP data for the entire species show a signature of two historic populations that have recently
come into secondary contact. Finding two historic populations is consistent with previously published data based on
mitochondrial DNA analyses; however, due to the rapid evolution rate of AFLP data, we were able to detect a high
level of gene flow between these populations. We conclude that it is possible Texas and Florida served as refugia for
A. piscivorus during the last glacial maximum, and, as the glaciers receded, the two populations expanded, coming
into secondary contact. The subsequent gene flow has resulted in shared loci across the two populations. The
difference between the conclusions drawn between our two markers and previous research is due to the different
time scales that AFLP and cyt-bmarkers measure. The AFLP data provided a contemporary marker and cyt-bindicated
historic separation.
THE Cottonmouth, Agkistrodon piscivorus, is a semi-
aquatic pitviper that occurs in the southeastern
United States into Texas (Fig. 1). Within the last
500 years, the western limit of its distribution has been
contracting due to the drying and desertification of west
Texas (Brune, 1975; Werler and Dixon, 2000). This has led to
a geographically isolated peripheral population in the
Concho River Valley (Werler and Dixon, 2000). Historically,
the species distribution has expanded and contracted due to
the advancement and retreat of the Laurentide ice sheet
during Pleistocene glaciation (Gloyd and Conant, 1990;
Guiher and Burbrink, 2008; Douglas et al., 2009). Previous
research, using mitochondrial DNA (mtDNA), has suggested
that fluctuations in the distribution resulted in two distinct
lineages of Cottonmouths. This hypothesis does not reflect
the currently accepted taxonomy that there is one species
with three subspecies: Agkistrodon piscivorus piscivorus (Eastern
Cottonmouth), A. piscivorus leucostoma (Western Cotton-
mouth), and A. piscivorus conanti (Florida Cottonmouth;
Gloyd and Conant, 1990; Knight et al., 1992; Castoe and
Parkinson, 2006; Fig. 1). Guiher and Burbrink (2008) and
Douglas et al. (2009) concluded that additional genetic
markers, particularly nuclear loci, were needed before
taxonomy can be addressed. Using range-wide sampling with
emphasis on populations in Texas, we sought to determine if
the remote peripheral population in Texas is genetically
isolated and if there are two lineages within A. piscivorus using
amplified fragment length polymorphisms (AFLPs) supple-
mented with additional mtDNA cyt-bsequences.
AFLP markers are advantageous because they are repre-
sentative of the nuclear genome and it is easy generate a
large number of polymorphic loci with enough power to
differentiate populations (Bensch and Akesson, 2005). AFLPs
can be applied to populations sampled throughout the
range of a species to determine overall population structure.
Additionally, AFLPs have been used to differentiate sister
taxa and determine relationships among species (Creer et al.,
2004; Mendelson and Simons, 2006; Althoff et al., 2007;
Makowsky et al., 2009). Another marker that has been used
traditionally to distinguish species-level relationships in a
variety of taxa is mitochondrial DNA (mtDNA; Rosenberg et
al., 2002; Jacobsen et al., 2010). Results from AFLP markers
and mtDNA sequence data such as cyt-bhave been
compared for a variety of taxa. When analyzing small
portions of a species’ distribution or examining population
structure within an entire species, AFLP markers yield more
fine-scale information compared to cyt-bsequence data
(Mendelson and Simons, 2006; Egger et al., 2007; Phillips
et al., 2007; Garcı´a-Pereira et al., 2011).
Given the geographic isolation of the peripheral Concho
Valley population on the western edge of the species’ range,
we would expect genetic isolation. Thus, the initial goal of
our study was to evaluate the fine-scale population structure
in Texas, and we hypothesized that the population in the
Concho Valley is genetically distinct and would have a
lower level of genetic variation than other populations
throughout Texas. The second goal of our study was to test
the hypothesis proposed by Guiher and Burbrink (2008) and
Douglas et al. (2009) that the Florida population of A.
piscivorus is genetically isolated. We expected to see genetic
separation between the Florida population and other
populations in our sample consistent with the presence of
1
Department of Biology, Angelo State University, 2601 W. Avenue N., San Angelo, Texas 76909; E-mail: (LKA) Loren.Ammerman@
angelo.edu.
2
Department of Biology, University of Central Florida, 4000 Central Florida Blvd., Orlando, Florida 32816; E-mail: (JLS) jason.strickland@
knights.ucf.edu; and (CLP) parkinson@ucf.edu. Send reprint requests to JLS.
3
College of Arts and Sciences, Georgia Southwestern State University, 800 Georgia Southwestern State University Drive, Americus, Georgia
31709; E-mail: kelly.mccoy@gsw.edu.
Submitted: 5 October 2013. Accepted: 3 June 2014. Associate Editor: D. S. Siegel.
F2014 by the American Society of Ichthyologists and Herpetologists DOI: 10.1643/CG-13-123 Published online: November 19, 2014
Copeia 2014, No. 4, 639–649
two lineages. To test our predictions, we generated AFLP
data for the entire distribution of A. piscivorus and added
mtDNA sequence data to a previously published mtDNA
matrix (Guiher and Burbrink, 2008). We discuss our results
in the context of the biogeographic history of Texas and the
southeastern United States as well as to previous studies on
Agkistrodon.
MATERIALS AND METHODS
Taxon sampling and DNA extraction.—Through collection
and tissue loans, we sampled 75 A. piscivorus from 24
separate localities as our ingroup (Fig. 1; Appendix 1). We
also sampled 24 A. contortrix, two A. bilineatus, two A. taylori,
one Crotalus atrox, and one C. molossus for our analysis as
outgroups (Appendix 1; Castoe and Parkinson, 2006). For
samples we collected, blood was taken from the caudal vein
using an insulin syringe, and stored in modified Tris-EDTA
Longmire lysis buffer which increased DNA yield (Longmire
et al., 1997; removed NaCl and increased sodium dodecyl
sulfate from 0.5%to 1.0%.). We deposited voucher speci-
mens in the Angelo State Natural History Collection at
Angelo State University in San Angelo, Texas (Appendix 1).
We extracted whole genomic DNA using a Qiagen DNA
extraction kit (Valencia, CA) following the kit protocol for
blood or tissue samples stored in lysis buffer or 95%ethanol.
Amplified fragment length polymorphism (AFLP) protocol.—We
followed the AFLP protocols of Phillips et al. (2007) and Lee
et al. (2010) based on modifications of Vos et al. (1995) who
initially described the method. Restriction enzymes (EcoRI,
AseI, and TaqI) were used to digest approximately 200 ng of
genomic DNA into fragments of different lengths. General-
ly, only two restriction enzymes are used and all primer
combinations are created based on those. In this study, each
sample underwent two separate protocols to increase the
number of polymorphic loci scored. We used a total of nine
primer combinations in the analyses of A. piscivorus and
eight in the analysis with all taxa (Table 1). All reactions
used 20 units of EcoRI (New England Biolabs [NEB], Ipswich,
MA). One treatment used 20 units of AseI (NEB) as the
second enzyme and the other treatment used 20 units of
TaqI (NEB) as the second enzyme. For all restriction
digestions, 1X enzyme buffer was added to the reaction
and the restriction digest was placed at 37uC for three hours.
Next, 75 pmoles of the appropriate enzyme adapter
(Table 1) were ligated to the ends of the fragments that
were created by the restriction digest using T4 DNA ligase
and 4 mL of 10X ligase buffer (NEB).
The pre-selective PCR decreased the number of fragments
because of an additional base pair on the primer (Table 1).
With the additional base, the number of fragments were
reduced to approximately 1/16 of those that were initially
created in the restriction digest (Meudt and Clarke, 2007).
The second PCR, the selective step, lowered the number of
fragments even more depending on how many bases were
added to the primer (Table 1). This step also attached a
fluorescent dye onto each fragment for detection by Beck-
man-Coulter CEQ 8000 Genetic Analysis System (Beckman-
Coulter, Inc., Fullerton, CA).
Nine primer combinations were used to yield a large
number of fragments, giving a representative measure of
polymorphic loci in the genome. Bonin et al. (2007)
suggested that at least 200 total fragments be used in an
AFLP analysis to show population structure and to get
enough polymorphic loci to differentiate populations. The
more total fragments that are scored, the higher the
resolution and the better statistical support for analysis
(Albertson et al., 1999; Ogden and Thorpe, 2002; Bensch
and Akesson, 2005). The fragments in the selective PCR
reactions were separated by loading 0.8 mL of the reaction
with 0.25 mL of 400 base pair (bp) size standard in the
CEQ8000. Fragments greater than 80 bp were scored as
present (1) or absent (0) using software available on the
CEQ8000, which created a binary matrix. Once the initial
scoring was complete, fragments were evaluated by eye to
ensure proper scoring. Any fragments scored inconsistently
or that were too close to other fragments were removed from
the analysis, leaving only unambiguous fragments. All
individuals were scored in random order to minimize bias
in results (Bonin et al., 2005, 2007).
Population genetic analysis.—GenAlEx ver. 6.41 was used to
analyze the data and visualize population structure (Peakall
and Smouse, 2006). GenAlEx initially created a genetic
distance matrix based on Nei-Li distances from the binary
matrix (Nei and Li, 1979). Both inter- and intraspecific Nei-
Li genetic distances were calculated for all four species of
Agkistrodon. That information was used in Principal Coordi-
nate Analysis (PCoA) to visualize population divergence
(Orlo´ ci, 1975). This analysis does not require groups
assigned a priori and makes it possible to examine relation-
ships in space (usually two or three dimensional) depending
on how many eigenvectors are used. PCoA was performed
first on all samples, then on only A. piscivorus, and finally on
populations of A. piscivorus from Texas to visualize the
pattern at different geographic scales. Average heterozygos-
ity was calculated for each population using Hickory ver. 1.1
which relaxes Hardy-Weinberg assumptions and uses Bayes-
ian statistics to calculate heterozygosity from dominant
markers (Holsinger et al., 2002). The genetic distance matrix
was analyzed via Analysis of Molecular Variance (AMOVA)
to compare variation between and within populations to
determine if there was population differentiation. We
calculated Fpt which is an analogous measure of F
st
Fig. 1. Distribution map of the Cottonmouth, Agkistrodon piscivorus,
showing the two current views on its taxonomy. The patterns designate
the subspecies view and the dashed line splits the Florida and
continental groups discussed in text. Shapes indicate sampling localities
and correspond to the shapes in Figure 4.
640 Copeia 2014, No. 4
specifically for binary data. Both are measures of genetic
differentiation between populations (Andrade et al., 2007).
The Fpt values were calculated based on 1000 replicates (a5
0.05). This estimate gave a statistical measure of gene flow
among the populations and made it possible to examine
variation in Agkistrodon.
To test for isolation by distance (IBD), a Mantel test,
which is a pairwise comparison to determine correlation
between geographic and genetic distances, was used to
examine population structure in all A. piscivorus sampled
(Mantel, 1967; Jensen et al., 2005). The data matrix was
formatted for the program STRUCTURE ver. 2.3.3 (Pritchard
et al., 2000; Falush et al., 2003, 2007) using the program
AFLP-SURV ver. 1.0 (Vekemans, 2002). STRUCTURE esti-
mated the highest degree of genetic structure between the
populations and calculated the number of populations (K)
in the entire sample based on genetic distance. For the
STRUCTURE analysis, the admixture model was used with a
burn in of 30,000 followed by 100,000 iterations. This
process was applied for K values of 1–10 with ten
replications at each K value. The resulting log likelihood
scores were averaged for each K. The admixture model was
chosen because we did not want to bias the results toward
finding a lack of gene flow. With the log likelihood scores,
we determined DK and then used that to find the true
number of groups, K* (Evanno et al., 2005).
To determine the phylogenetic position of A. piscivorus,
we created a neighbor joining phylogram from the Nei-Li
genetic distances (Saitou and Nei, 1987) in PAUP* (phylo-
genetic analysis using parsimony) ver. 4.0b10 (Swofford,
2003). The two rattlesnakes were used as outgroup taxa and
nodal support was calculated with 1000 bootstrap pseudo-
replicates. Parsimony methods were not used because they
are not appropriate for binary AFLP data according to
Robinson and Harris (1999) and Sullivan et al. (2004). For
AFLP analysis, bands may be lost independently in more
than one lineage and could result in poorly resolved trees if
parsimony is used (Dasmahapatra et al., 2009). Moreover,
analysis of discrete characters could result in the situation
where a few markers determine the phylogenetic pattern,
whereas the neighbor-joining analysis takes into account
overall similarity (Dasmahapatra et al., 2009).
DNA sequencing and analysis.—To test for isolation in west
Texas and to compare AFLP to mtDNA data, eight cyt-b
sequences from A. piscivorus in Texas were generated
(GenBank accession KC431019–KC431026) and added to
the previously published tree of Guiher and Burbrink (2008).
Table 1. List of restriction enzymes, adapters, pre-selective primer, and selective primer sequences for the PCR used in the AFLP analysis of
Agkistrodon piscivorus and outgroup taxa. Asterisk (*) indicates the primer with the fluorescent label attached. Primers used in combination with
EcoRI-CAC are indicated with {and those used with EcoRI-CAT are indicated by {.TaqI-TTG was only used in the analysis of A. piscivorus.
Name Sequence
Restriction enzymes
EcoRI 59-- . . . G|AATTC . . . --39
AseI 59-- . . . AT|TAAT . . . --39
TaqI 59-- . . . T|CGA . . . --39
Adapters
EcoRI 59--CTCGTAGACTGCGTACC--39
39--CATCTGACGCATGGTTAA--59
AseI 59--GACGATGAGTCCTGA--39
39--TACTCAGGACTCAT--59
TaqI 59--CGGTCAGGACTCAT--39
39--AGTCCTGAGTAGCAG--59
Pre-selective primers
EcoRI 59--ACTGCGTACCAATTCC--39
AseI 59--GATGAGTCCTGAGTAATT--39
TaqI 59--ATGAGTCCTGACCGAT--39
Selective primers
EcoRI-CAC* 59--ACTGCGTACCAATTCCAC--39
EcoRI-CAT* 59--ACTGCGTACCAATTCCAT--39
AseI-TAG{59--GATGAGTCCTGAGTAATTAG--39
AseI-TCC{59--GATGAGTCCTGAGTAATTCC--39
AseI-TGA{59--GATGAGTCCTGAGTAATTGA--39
AseI-TGC{59--GATGAGTCCTGAGTAATTGC--39
AseI-TCT{59--GATGAGTCCTGAGTAATTCT--39
AseI-TAT{59--GATGAGTCCTGAGTAATTAT--39
TaqI-TCA{59--ATGAGTCCTGACCGATCA--39
TaqI-TTC{59--ATGAGTCCTGACCGATTC--39
TaqI-TTG{59--ATGAGTCCTGACCGATTG--39
Cyt-bsequencing primers (Burbrink et al., 2000)
L14910 (Forward) 59--GACCTGTGATMTGAAAAACCCAYCGTT--39
H16064 (Reverse) 59--CTTTGGTTTACAAGAACAATGCTTTA--39
Strickland et al.—Phylogeography of Cottonmouths 641
These samples included seven individuals from the isolated
population in the Concho Valley and one from south Texas.
One additional sequence of A. piscivorus from southern
Georgia and one from South Carolina were also added to
ensure accuracy of sequence comparison (GenBank acces-
sion KC431027 and KC431028). PCR amplification was
accomplished using primers L14910 and H16064 designed
by Burbrink et al. (2000; Table 1), following the protocol
described by Castoe and Parkinson (2006). PCR product was
sequenced in both directions at Arizona Research Laborato-
ries, Division of Biotechnology, University of Arizona
Genetics Core Facility (http://uagc.arl.arizona.edu/). Se-
quences were edited using Sequencher ver. 4.2 (Gene Codes)
and novel sequences were aligned with those from Guiher
and Burbrink (2008; GenBank accession EU483411–
EU483493) in GeneDoc ver. 2.7.0 (Nicholas et al., 1997).
To determine phylogenetic position of the samples from
Texas, we used the same parameters for Bayesian inference
(BI) as Guiher and Burbrink (2008). For BI, the GTR +I+C
model was used in MrBayes ver. 3.1.2 with default
parameters (Huelsenbeck and Ronquist, 2001). Chains were
run for 5310
6
generations with a burn-in period of 5310
5
generations and sampling every 1000
th
generation. Tracer v
1.4 (Ronquist and Huelsenbeck, 2003) was used to ensure
stationarity was reached during the burn-in period.
RESULTS
Eight primer combinations were used to create 622 AFLP
fragments for all taxa (Table 2). Of these, 498 (80%) were
polymorphic and between 59 and 102 fragments were
scored from each primer combination. In the PCoA for all
105 individuals, the first three axes explained 83.7%of the
variation (Fig. 2). In this PCoA, the two rattlesnake species
(Crotalus) were separated from Agkistrodon along the third
axis. Agkistrodon piscivorus had the largest number of
polymorphic loci (44.5%) which can be visualized by the
amount of spread in the A. piscivorus cluster. Agkistrodon
contortrix (32.53%) collected from throughout their range
and Texas A. piscivorus (31.36%) had a similar level of
polymorphism as indicated by the similarity in shape of
their clusters in Figure 2. Because there were only two
samples from each of the two cantils (A. bilineatus and A.
taylori), it was not possible to determine the amount of
variation seen in each of those species. For the neighbor-
joining analysis, a 50%majority rule consensus tree was
created (Fig. 3). Branches with over 70%support were
considered to be significantly supported (Felsenstein, 1985;
Hillis and Bull, 1993). When Nei-Li genetic distances were
calculated, A. piscivorus had the highest amount of intra-
specific variation (6.4%) and the two cantil species had the
closest genetic distance between any two species (9.5%).
After outgroups were removed, nine primer combinations
were used and all A. piscivorus were analyzed (Table 2). There
were 479 fragments used in the analysis with 44.9%(215)
polymorphic. The average amount of polymorphism within
a population was 10.0761.42%. The AMOVA indicated a
significant lack of gene flow among A. piscivorus across their
entire range (F
pt
50.466, P,0.001). For this PCoA, only
the first two axes were used (73.4%variation explained); this
ordination did not show any clustering and there was an
east to west pattern for all A. piscivorus (Fig. 4). The Mantel
Fig. 2. Three-dimensional principal coordinate analysis (PCoA) of all
samples used in the analysis. Rattlesnakes fall out on the third axis away
from the four species within Agkistrodon. Ellipses were used to help
delineate groups.
Table 2. Mean heterozygosity and descriptive statistics of Agkistrodon piscivorus based on 479 AFLP loci. Global F
st
=0.294±0.016.
Population nH
e
±SE No. private alleles Polymorphic within (%) Ave. no. bands±SE
Angelina Nat. Forest (TX) 4 0.15180860.0098 0 5.22 260.861.70
Knickerbocker, TX 2 0.15678560.0113 0 2.71 254.560.50
Florida 14 0.20812560.0071 5 15.45 256.862.24
Ft. Worth, TX 3 0.16507260.0106 2 5.01 262.760.33
Galveston, TX 5 0.1560460.0093 0 7.10 259.461.81
Georgia 3 0.18723360.0102 0 6.47 255.360.33
Huntsville, TX 5 0.15673160.0092 1 7.72 258.861.24
Junction, TX 3 0.11565860.0079 0 4.80 254.862.47
San Angelo, TX 12 0.15048360.0107 1 3.13 259.061.00
Menard, TX 1 — 0 — 264.060.00
Louisiana 1 — 0 — 258.060.00
Mississippi 5 0.18605760.0009 3 10.86 258.862.78
Palmetto State Park (TX) 5 0.160360.0095 2 7.52 260.061.30
South Carolina 4 0.19702260.0098 0 8.98 259.365.63
Tyler, TX 5 0.14124760.01 0 4.80 260.460.93
Welder WMA (TX) 3 0.15245260.0105 0 3.34 259.362.19
642 Copeia 2014, No. 4
IBD test indicated that genetic distance was significantly
correlated with geographic distance (R
2
50.731, P,
0.0001). A value of K 52 was determined using STRUCTURE
for the number of groups within A. piscivorus.There
appeared to be a geographic cline based on the pattern
observed in the STRUCTURE output (Fig. 5). Samples from
the middle of the distribution had some proportion of their
genes estimated to be from both of the populations. Within
A. piscivorus, there was significant support for two clades
from both markers but high gene flow made it difficult to
determine the geographic location of the divergence. The
neighbor joining tree based on the AFLP markers had a clade
that included Texas, Mississippi, and Louisiana individuals
and a clade that included the Florida, Georgia, and South
Carolina individuals (Fig. 3). We recovered a Florida clade
and a continental clade from our Bayesian tree based on cyt-
b, which was consistent with the tree from Guiher and
Burbrink (2008). Both our STRUCTURE and phylogenetic
analyses determined there were two populations of A.
piscivorus throughout their range (Fig. 5).
Fig. 4. Two-dimensional PCoA of Agkistrodon piscivorus from their entire range based on AFLP data. The first two axes explain 73.40%of the
variation, and the pattern indicates a west to east trend in genetic variation.
Fig. 3. Neighbor-joining phylogram with terminal branches condensed
from all samples using AFLP data. One thousand bootstrap pseudo-
replicates were performed, and those with support over 50 percent
are shown.
Fig. 5. Agkistrodon piscivorus posterior mean estimates of the
proportion of each individual’s genome that belongs to each of the
two estimated populations from STRUCTURE.
Strickland et al.—Phylogeography of Cottonmouths 643
For the final analysis, only A. piscivorus from Texas were
used. Once again, all nine primer combinations were used
and yielded a total of 440 fragments with 31.36%of them
being polymorphic (138). The average population level of
polymorphism was 8.2360.95%with the Concho Valley
population at 3.13%. The number of fragments decreased
from the previous analysis because fragments that were all 0
were removed. The AMOVA analysis indicated significant
lack of gene flow between populations (F
pt
50.348, P,
0.001). The PCoA indicated that there was a cluster of
individuals from the Concho Valley that was separated from
the other populations. Individuals from the Llano River in
Junction, Texas (Kimble Co.) were also isolated (Fig. 6).
STRUCTURE analysis indicated four groups. One of the
groups was comprised of solely Concho Valley individuals
and the other three groups were split up throughout the
remaining distribution in Texas. All groups had mixing of
genes from other populations.
Using the mtDNA sequence data we were not able to
distinguish population-level separation between the isolated
Concho Valley population and the rest of Texas. The
individuals from the Concho Valley used in the analysis
were recovered throughout the topology of the continental
clade as presented in Guiher and Burbrink (2008). The cyt-b
phylogeny (not shown) recovered using BI did not differ
from the tree published in Guiher and Burbrink (2008) or
the NJ tree generated based on AFLP data (Fig. 3). There was
support for the currently recognized relationships within
Agkistrodon as well as two lineages within A. piscivorus that
generally correspond to the continental and Florida clades
(Guiher and Burbrink, 2008).
DISCUSSION
The population structure of A. piscivorus based on AFLP and
mtDNA sequence data indicates a complex history in the
southeastern U.S. We were able to determine the population
structure in Texas and the entire distribution of A. piscivorus
to address the goals of our study. In Texas, the AFLP markers
indicated that the geographically isolated Concho Valley
population is also genetically isolated. The PCoA (Fig. 6)
indicates distinct separation from the remaining individuals
in Texas, whereas BI based on cyt-bdid not resolve distinct
lineages in Texas. Amplified fragment length polymorphism
markers work at finer taxonomic and temporal scales than
cyt-band are able to detect genetic changes in populations
sooner (Bensch and Akesson, 2005; Meudt and Clarke,
2007). These AFLPs are predominately neutral and can
accrue changes much more quickly than cyt-b, which makes
it possible to use them to determine if gene flow is occurring
between populations in close proximity (Andrade et al.,
2007). Even with just 15 individuals from the Concho
Valley, the analyses demonstrated that this population had
lower genetic variation. Most of the other populations
sampled in Texas were from a much smaller geographic area
than the area sampled in the Concho Valley but still had
higher levels of genetic variation based on the visualization
of populations in the PCoA (Fig. 6).
Approximately 50–100 years ago, there were at least eight
springs south of the headwater springs of the South Concho
River that have since run dry (Brune, 1975). These springs
provided habitat corridors for A. piscivorus between the
Concho Valley and the San Saba River near Menard, Texas
(Brune, 1975). It is possible that the drying period occurring
over the last 200 years has slowly shrunk the western
population of A. piscivorus leaving populations isolated in
the Concho Valley and at the head of the San Saba River.
Our STRUCTURE results are consistent with an isolated
Concho Valley population. The three eastern populations
inferred by STRUCTURE have more characters in common
with each other than any has with the Concho Valley
populations. This, along with the significant genetic
structure demonstrated by the Fpt value, indicates that
the population has recently become genetically isolated. To
fully understand the history of A. piscivorus in the Concho
Valley, a more thorough sampling effort will be needed from
areas directly surrounding the valley.
The overall genetic variation in Texas was not different
from the rest of the range of A. piscivorus. The neighbor
joining (Fig. 3) and Bayesian phylograms show species
Fig. 6. Two-dimensional PCoA of only Texas Agkistrodon piscivorus based on AFLP data. The first two axes explain 56.88%of the variation, and the
pattern indicates that the Concho Valley population is less genetically variable than other populations in Texas.
644 Copeia 2014, No. 4
relationships consistent with those of Parkinson et al.
(2000). In agreement with recent work, there was significant
support for the monophyly of Agkistrodon as well as support
for each of the four species currently recognized within
Agkistrodon (Knight et al., 1992; Castoe and Parkinson,
2006). Also in agreement, we found strong support for the
sister relationship of A. contortrix to the A. piscivorus and
cantil clade. These groupings were supported by PCoA for all
individuals based on AFLP markers (Fig. 2).
Our results indicate that there are two lineages within A.
piscivorus, and there was no evidence for three subspecies as
presented in Gloyd and Conant (1990). The AMOVA
indicated significant genetic structure throughout the range
which showed A. piscivorus does not form a panmictic
population. The PCoA for all samples of A. piscivorus showed
an east to west pattern in population structure (Fig. 4). Our
STRUCTURE analysis demonstrates that there are many
markers shared throughout the range from east Texas to
north Florida of the two proposed populations, likely
indicating high levels of gene flow, but it is not possible to
rule out incomplete lineage sorting (Fontenot et al., 2011).
Incomplete lineage sorting is unlikely because the large
number of randomly distributed markers throughout the
genome effectively neutralize differential lineage sorting
(Avise, 2004; Koblmuller et al., 2010). The Mantel test
showed a significant, positive correlation between geo-
graphic and genetic distance. Because A. piscivorus had this
pattern, STRUCTURE should be interpreted with caution
(Pritchard et al., 2000). STRUCTURE has a tendency to
return a greater number of populations than are actually
present when the species shows isolation by distance (Frantz
et al., 2009). We conclude that there are two lineages within
A. piscivorus, but there is a high amount of gene flow
between the two. Our inferred groupings are similar to those
presented in previous studies with a few exceptions.
In this study, individuals from South Carolina were in the
same lineage as the individuals from Florida and Georgia
based on AFLP data, whereas they clustered with individuals
from Texas and Mississippi based on cyt-b. It is possible that
the proposed species boundary could be farther north than
presented in either Guiher and Burbrink (2008) or Douglas
et al. (2009), and the Florida lineage presented in those two
papers should include the samples from South Carolina used
in this study. The difference in inferred relationships when
comparing AFLP with sequence data is likely caused by the
difference in temporal scale reflected by the two markers
(Egger et al., 2007; Fontenot et al., 2011). Cyt-bis expected
to retain the signature from when the continental and
southern populations were isolated during Pleistocene
glacial cycles (Douglas et al., 2009). Once in Florida, A.
piscivorus colonized southern Florida and during interglacial
periods, it moved into the southeastern United States.
Florida could have served as a glacial refuge for A. piscivorus
that was stable over long periods of time and may have
allowed for the mtDNA lineage differentiation. It is likely
that Texas served as the continental refuge for A. piscivorus
during glacial periods (Swenson and Howard, 2005). In his
unpublished dissertation, Guiher (2011) used additional
sequence data from the nuclear genome, ecological niche
modeling, and morphological data and found evidence for a
large hybrid zone. The markers used by Guiher were
biparentally inherited which allowed him to detect the area
of gene flow. AFLP markers detect the finest scale which
explains why these data exhibit the largest area of gene flow
of all studies of A. piscivorus. The results from this study are
consistent with the explanation that Texas and Florida
refuge populations expanded into the southeastern United
States eventually coming into secondary contact resulting in
the current continuous distribution (Barrowclough et al.,
2011). After secondary contact, gene flow occurred and
allowed for AFLP markers to move between Texas and
Florida.
Since colonizing the southeastern U.S., A. piscivorus has
been influenced by glacial and interglacial periods. This
species carries the genetic signature of historic separation,
but contemporary markers indicate that gene flow is
occurring between formerly isolated lineages. Guiher
(2011) proposed that speciation has occurred and there are
two species within A. piscivorus (above and below the dashed
line in Fig. 1). He detected an area of gene flow that spread
from Mississippi along the Florida/Georgia border and into
South Carolina. Our data confirm the hybrid zone but
indicate it is larger than was proposed by Guiher. The
incongruence between these studies is likely the result of
difficulty in defining when speciation has occurred (de
Queiroz, 2007; Nosil, 2008). It is difficult to determine when
a speciation event has occurred and exactly how much gene
flow is allowable to recognize distinct species (Nosil, 2008).
The incongruence is also influenced by the historical
biogeography of the southeastern United States since the
last glacial cycle in the Pleistocene (Hewitt, 1996, 2000;
Swenson and Howard, 2005; Soltis et al., 2006; Fontanella
et al., 2008) and by the natural history of the A. piscivorus
lineage (Gloyd and Conant, 1990). Future work with A.
piscivorus will be able to use genetic structure to better
understand its biogeography. Agkistrodon piscivorus is a good
model species to test finer scale markers such as single
nucleotide polymorphisms to determine the extent of gene
flow throughout their range and to determine if secondary
contact will remove the evidence of two lineages.
ACKNOWLEDGMENTS
We thank D. Palmer, A. Osmanski, T. Fisher, R. Heischman,
H. Romo, B. Hubbell, J. Chamberlain, K. Chamberlain, R.
Swanson, E. Smith, J. Streicher, C. Roelke, T. Wood, and the
Texas Herpetology Society for assistance with field collec-
tion; B. Lutterschmidt, L. Densmore, and N. Ford for
assistance in finding field sites and help working with A.
piscivorus;M.Tipps,D.Lee,M.McDonough,andA.
Ferguson for lab assistance; G. Territo for sequencing cyt-b;
B. Sims, D. Sykes, S. Tweedy, C. Denise, and R. Howard for
access to private lands; S. Crook at Old Sabine Bottom WMA,
R. Denkhaus and S. Tuttle at Ft. Worth Nature Center and
Refuge, A. Byboth at Camp Tyler, T. Blankenship at Welder
Wildlife Foundation, C. Strobel at Aransas National Wildlife
Refuge, J. Engle at Angelina National Forest, and Staff at
Palmetto State Park for access to public lands; E. Sanchez
and J. Perez at the National Natural Toxins Research Center,
T. Laduc and C. Cannatella at Texas Natural Science Center,
D. Dittmann and R. Brumfield at LSUMZ, K. Neil for
Oklocknee, GA samples, M. Gaston and M. Forstner at
Texas State University, B. Greene at Missouri State Univer-
sity, T. Guiher and F. Burbrink at The City University of New
York, B. Stuart at NCMNS, and K. Wray and E. Lemmon at
Florida State University for sending tissue; N. Strenth, R.
Dowler, J. Osterhout, N. Flynn, R. Phau, and T. Maxwell for
helpful discussion of this research; R. Howard and the Head
of the River Research Grant, the East Texas Herpetology
Strickland et al.—Phylogeography of Cottonmouths 645
Society, CARR Research Grant, Tri-Beta Research Grant and
the Southwestern Association of Naturalists Howard McCar-
ley Student Research Award for funding to JLS.; T. Guiher for
providing his dissertation; E. Hoffman, A. Fenwick, G.
Territo, and the Parkinson and Hoffman labs at UCF for
helpful comments and discussion of this manuscript. Snakes
collected in Texas were under permit number SPR-0390-029
to R. Dowler and subpermitted to JLS.
LITERATURE CITED
Albertson, R., J. Markert, P. Danley, and T. Kocher. 1999.
Phylogeny of a rapidly evolving clade: the cichlid fishes of
Lake Malawi, East Africa. Proceedings of the National
Academy of Sciences of the United States of America
96:5107–5110.
Althoff, D. M., M. A. Gitzendanner, and K. A. Segraves.
2007. The utility of amplified fragment length polymor-
phisms in phylogenetics: a comparison of homology within
and between genomes. Systematic Biology 56:477–484.
Andrade, I., S. Mayo, C. Van Den Berg, M. Fay, M. Chester,
C. Lexer, and D. Kirkup. 2007. A preliminary study of
genetic variation in populations of Monstera adansonii var.
klotzschiana (Araceae) from north-east Brazil, estimated
with AFLP molecular markers. Annals of Botany 100:
1143–1154.
Avise, J. C. 2004. Molecular Markers, Natural History, and
Evolution. Second edition. Sinauer Associates, Sunder-
land, Massachusetts.
Barrowclough, G. E., J. G. Groth, K. J. Odom, and J. E. Lai.
2011. Phylogeography of the Barred Owl (Strix varia):
species limits, multiple refugia, and range expansion. The
Auk 28:696–706.
Bensch, S., and M. Akesson. 2005. Ten years of AFLP in
ecology and evolution: why so few animals? Molecular
Ecology 14:2899–2914.
Bonin, A., D. Ehrich, and S. Manel. 2007. Statistical
analysis of amplified fragment length polymorphism data:
a toolbox for molecular ecologists and evolutionists.
Molecular Ecology 16:3737–3758.
Bonin, A., F. Pompanon, and P. Taberlet. 2005. Use of
amplified fragment length polymorphism (AFLP) markers
in surveys of vertebrate diversity. Methods in Enzymology
395:145–161.
Brune, G. 1975. Major and Historical Springs of Texas. Board,
Technical Report #189. Texas Water Development, Austin,
Texas. Available at http://www.twdb.texas.gov/publications/
reports/numbered_reports/doc/R189/Report189.asp
Burbrink, F. T., R. Lawson, and J. B. Slowinski. 2000.
Mitochondrial DNA phylogeography of the polytypic
North American rat snake (Elaphe obsoleta): a critique of
the subspecies concept. Evolution 54:2107–2118.
Castoe, T. A., and C. L. Parkinson. 2006. Bayesian mixed
models and the phylogeny of pitvipers (Viperidae:
Serpentes). Molecular Phylogenetics and Evolution
39:91–110.
Creer, S., R. Thorpe, A. Malhotra, W. H. Chou, and A.
Stenson. 2004. The utility of AFLPs for supporting
mitochondrial DNA phylogeographical analyses in the
Taiwanese Bamboo Viper, Trimeresurus stejnegeri. Journal
of Evolutionary Biology 17:100–107.
Dasmahapatra, K., J. Hoffman, and W. Amos. 2009.
Pinniped phylogenetic relationships inferred using AFLP
markers. Heredity 103:168–177.
de Queiroz, K. 2007. Species concepts and species delimi-
tation. Systematic Biology 56:879–886.
Douglas, M. E., M. R. Douglas, G. W. Schuett, and L. W.
Porras. 2009. Climate change and evolution of the New
World pitviper genus Agkistrodon (Viperidae). Journal of
Biogeography 36:1164–1180.
Egger, B., S. Koblmu¨ller, C. Sturmbauer, and K. Sefc. 2007.
Nuclear and mitochondrial data reveal different evolu-
tionary processes in the Lake Tanganyika cichlid genus
Tropheus. BMC Evolutionary Biology 7:137.
Evanno, G., S. Regnaut, and J. Goudet. 2005. Detecting the
number of clusters of individuals using the software
STRUCTURE: a simulation study. Molecular Ecology
14:2611–2620.
Falush, D., M. Stephens, and J. K. Pritchard. 2003.
Inference of population structure using multilocus geno-
type data: linked loci and correlated allele frequencies.
Genetics 164:1567–1587.
Falush, D., M. Stephens, and J. K. Pritchard. 2007.
Inference of population structure using multilocus geno-
type data: dominant markers and null alleles. Molecular
Ecology Notes 7:574–578.
Felsenstein, J. 1985. Confidence limits on phylogenies: an
approach using the bootstrap. Evolution 39:783–791.
Fontanella, F. M., C. R. Feldman, M. E. Siddall, and F. T.
Burbrink. 2008. Phylogeography of Diadophis punctatus:
extensive lineage diversity and repeated patterns of
historical demography in a trans-continental snake.
Molecular Phylogenetics and Evolution 46:1049–1070.
Fontenot, B. E., R. Makowsky, and P. T. Chippindale.
2011. Nuclear-mitochondrial discordance and gene flow
in a recent radiation of toads. Molecular Phylogenetics
and Evolution 59:66–80.
Frantz, A., S. Cellina, A. Krier, L. Schley, and T. Burke.
2009. Using spatial Bayesian methods to determine the
genetic structure of a continuously distributed popula-
tion: clusters or isolation by distance? Journal of Applied
Ecology 46:493–505.
Garcı´a-Pereira, M. J., A. Caballero, and H. Quesada. 2011.
The relative contribution of band number to phylogenetic
accuracy in AFLP data sets. Journal of Evolutionary
Biology 24:2346–2356.
Gloyd, H. K., and R. Conant. 1990. Snakes of the
Agkistrodon complex: a monographic review. Society for
the Study of Amphibians and Reptiles, Oxford, Ohio.
Guiher, T. J. 2011. Systematics and historical biogeogra-
phy of Agkistrodon contortrix and Agkistrodon piscivorus.
Unpubl. Ph.D. diss., The City University of New York,
New York.
Guiher, T. J., and F. T. Burbrink. 2008. Demographic and
phylogeographic histories of two venomous North Amer-
ican snakes of the genus Agkistrodon. Molecular Phyloge-
netics and Evolution 48:543–553.
Hewitt, G. M. 1996. Some genetic consequences of ice ages,
and their role in divergence and speciation. Biological
Journal of the Linnean Society 58:247–276.
Hewitt, G. M. 2000. The genetic legacy of the Quaternary
ice ages. Nature 405:907–913.
Hillis, D. M., and J. J. Bull. 1993. An empirical test of
bootstrapping as a method for assessing confidence in
phylogenetic analysis. Systematics Biology 42:182–192.
Holsinger, K. E., P. O. Lewis, and D. K. Dey. 2002. A
Bayesian approach to inferring population structure from
dominant markers. Molecular Ecology 11:1157–1164.
646 Copeia 2014, No. 4
Huelsenbeck, J. P., and F. Ronquist. 2001. MRBAYES:
Bayesian inference of phylogenetic trees. Bioinformatics
17:754–755.
Jacobsen, F., N. R. Friedman, and K. E. Omland. 2010.
Congruence between nuclear and mitochondrial DNA:
combination of multiple nuclear introns resolves a well-
supported phylogeny of New World Orioles (Icterus).
Molecular Phylogenetics and Evolution 56:419–427.
Jensen, J. L., A. J. Bohonak, and S. T. Kelley. 2005.
Isolation by distance, web service. BMC Genetics 6:13 ver.
3.15 Available at http://ibdws.sdsu.edu/. Archived by
WebCite at http://www.webcitation.org/6E07xm29k on
28 January 2013.
Knight, A., L. D. Densmore, and E. D. Rael. 1992.
Molecular systematics of the Agkistrodon complex,
p. 49–69. In: Biology of the Pitvipers. J. A.Campbell and
E. D. Brodie, Jr. (eds.). Selva, Tyler, Texas.
Koblmuller, S., B. Egger, C. Sturmbauer, and K. M. Sefc.
2010. Rapid radiation, ancient incomplete lineage sorting
and ancient hybridization in the endemic Lake Tangan-
yika cichlid tribe Tropheini. Molecular Phylogenetics and
Evolution 55:318–334.
Lee, D. N., R. S. Pfau, and L. K. Ammerman. 2010.
Taxonomic status of the Davis Mountains Cottontail,
Sylvilagus robustus, revealed by amplified fragment length
polymorphism. Journal of Mammalogy 91:1473–1483.
Longmire, J. L., M. Maltbie, and R. J. Baker. 1997. Use of
‘‘lysis buffer’’ in DNA isolation and its implications for
museum collections. Occasional Papers 163, Museum of
Texas Tech University.
Makowsky, R., J. Chesser, and L. J. Rissler. 2009. A striking
lack of genetic diversity across the wide-ranging amphib-
ian Gastrophryne carolinensis (Anura: Microhylidae). Ge-
netica 135:169–183.
Mantel, N. 1967. The detection of disease clustering and a
generalized regression approach. Cancer Research 27:209.
Mendelson, T. C., and J. N. Simons. 2006. AFLPs resolve
cytonuclear discordance and increase resolution among
Barcheek Darters (Percidae: Etheostoma: Catonotus). Mo-
lecular Phylogenetics and Evolution 41:445–453.
Meudt, H. M., and A. C. Clarke. 2007. Almost forgotten or
latest practice? AFLP applications, analyses and advances.
Trends in Plant Science 12:106–117.
Nei, M., and W. H. Li. 1979. Mathematical model for
studying genetic variation in terms of restriction endonu-
cleases. Proceedings of the National Academy of Sciences
of the United States of America 76:5269–5273.
Nicholas, K. B., H. B. Nicholas, Jr., and D. W. Deerfield, II.
1997. GeneDoc: analysis and visualization of genetic
variation. EMBNEW. NEWS 4:14. Available at http://
www.nrbsc.org/gfx/genedoc/ Archived by WebCite at
http://www.webcitation.org/6DNViNfqT on 2 January
2013.
Nosil, P. 2008. Speciation with gene flow could be common.
Molecular Ecology 17:2103–2106.
Ogden, R., and R. Thorpe. 2002. The usefulness of
amplified fragment length polymorphism markers for
taxon discrimination across graduated fine evolutionary
levels in Caribbean Anolis lizards. Molecular Ecology
11:437–445.
Orlo´ci, L. 1975. Multivariate Analysis in Vegetation Re-
search. Second edition. Junk, The Hague.
Parkinson, C. L., K. R. Zamudio, and H. W. Greene. 2000.
Phylogeography of the pitviper clade Agkistrodon: histor-
ical ecology, species status, and conservation of Cantils.
Molecular Ecology 9:411–420.
Peakall, R., and P. E. Smouse. 2006. GENALEX 6: genetic
analysis in Excel. Population genetic software for teaching
and research. Molecular Ecology Notes 6:288–295.
Phillips, C. D., C. A. Henard, and R. S. Pfau. 2007.
Amplified fragment length polymorphism and mitochon-
drial DNA analyses reveal patterns of divergence and
hybridization in the Hispid Cotton Rat (Sigmodon hispi-
dus). Journal of Mammalogy 88:351–359.
Pritchard, J. K., M. Stephens, and P. Donnelly. 2000.
Inference of population structure using multilocus geno-
type data. Genetics 155:945.
Robinson, J. P., and S. A. Harris. 1999. Amplified fragment
length polymorphisms and microsatellites: a phylogenetic
perspective. In: Which DNA Marker for Which Purpose?
Final Compendium of the Research Project Development,
optimisation and validation of molecular tools for
assessment of biodiversity in forest trees in the European
Union DGXII Biotechnology FW IV Research Programme
Molecular Tools for Biodiversity. E. M. Gillet (ed.).
Germany. Available at http://webdoc.sub.gwdg.de/ebook/
y/1999/whichmarker/index.htm. Archived by WebCite at
http://www.webcitation.org/6DNWI71ED.
Ronquist, F., and J. P. Huelsenbeck. 2003. MrBayes 3:
Bayesian phylogenetic inference under mixed models.
Bioinformatics 19:1572–1574.
Rosenberg, N. A., J. K. Pritchard, J. L. Weber, H. M. Cann,
K. K. Kidd, L. A. Zhivotovsky, and M. W. Feldman. 2002.
Genetic structure of human populations. Science 298:
2381–2385.
Saitou, N., and M. Nei. 1987. The neighbor-joining method:
a new method for reconstructing phylogenetic trees.
Molecular Biology and Evolution 4:406–425.
Soltis, D. E., A. B. Morris, J. S. Mclachlan, P. S. Manos,
and P. S. Soltis. 2006. Comparative phylogeography of
unglaciated eastern North America. Molecular Ecology
15:4261–4293.
Sullivan, J. P., S. Lavoue´, M. E. Arnegard, and C. D.
Hopkins. 2004. AFLPs resolve phylogeny and reveal
mitochondrial introgression within a species flock of
African electric fish (Mormyroidea: Teleostei). Evolution
58:825–841.
Swenson, N. G., and D. J. Howard. 2005. Clustering of
contact zones, hybrid zones, and phylogeographic breaks
in North America. The American Naturalist 166:581–591.
Swofford, D. L. 2003. PAUP*. Phylogenetic analysis using
parsimony (*and other methods). Version 4. Sinauer
Associates, Sunderland, Massachusetts.
Vekemans, X. 2002. AFLP-surv version 1.0. Distributed by
the author. Laboratoire de Ge´ne´tique et Ecologie Ve´ge´tale,
Universite´ Libre de Bruxelles, Belgium. Available at http://
www.ulb.ac.be/sciences/lagev/aflp-surv.html.
Vos, P., R. Hogers, M. Bleeker, M. Reijans, T. Van De Lee,
M. Hornes, A. Freiters, J. Pot, J. Peleman, M. Kuiper,
and M. Zabeau. 1995. AFLP: a new concept for DNA
fingerprinting. Nucleic Acids Research 23:4407–4414.
Werler, J. E., and J. R. Dixon. 2000. Texas Snakes:
Identification, Distribution, and Natural History. Univer-
sity of Texas Press, Austin, Texas.
Strickland et al.—Phylogeography of Cottonmouths 647
APPENDIX 1. Locality information for all individuals used in the AFLP portion of this study including outgroups. Museum number and tissue
collection abbreviations are as follows: ASK (Angelo State Karyotype), ASNHC (Angelo State Natural History Collection), CLP (Christopher L.
Parkinson), JLS (Jason L. Strickland), KW (Kenneth Wray), LSUMZ (Louisiana State University Museum of Zoology), NNTRC (National Natural Toxins
Research Center), PIT (Passive Integrated Transponder), TJL (Travis J. Laduc), and TNHC (Texas Natural History Collection). All individuals with
GenBank accession numbers were added to the individuals used by Guiher and Burbrink (2008) to build the cyt-bphylogeny. The four individuals at
the end with an * were only used to generate cyt-bsequence data and were not in the AFLP portion of this study.
Tissue ID Snake ID Museum ID Species Country State County Accession no.
LSUMZ H-20951 A. bilineatus Mexico
LSUMZ H-6416 A. bilineatus Mexico
ASK 9056 JLS 25 A. contortrix USA TX Walker
ASK 9058 JLS 27 A. contortrix USA TX Walker
ASK 9059 JLS 28 A. contortrix USA TX Walker
ASK 9061 JLS 30 A. contortrix USA TX Walker
ASK 9077 JLS 49 A. contortrix USA TX Angelina
ASK 9088 JLS 60 A. contortrix USA TX Angelina
ASK 9100 JLS 73 A. contortrix USA TX Smith
ASK 9112 JLS 85 A. contortrix USA TX Pecos
ASK 9115 JLS 88 A. contortrix USA TX Pecos
ASK 9116 JLS 89 A. contortrix USA TX Pecos
ASK 9119 JLS 92 A. contortrix USA TX Brewster
ASK 9123 JLS 96 A. contortrix USA TX Jeff Davis
ASK 9124 JLS 97 A. contortrix USA TX Brown
LSUMZ H-18959 A. contortrix USA KY Hart
LSUMZ H-2240 A. contortrix USA MS Forrest
LSUMZ H-9234 A. contortrix USA IL Jersey
TNHC 58828 A. contortrix USA TX Edwards
TJL 922 TNHC 61851 A. contortrix USA TX Lee
TJL 1953 TNHC 84300 A. contortrix USA TX Travis
011-310-839 NNTRC A. contortrix USA TX Tarrant
011-367-560 NNTRC A. contortrix USA TX Midland
058-375-116 NNTRC A. contortrix USA MO Boone
058-557-565 NNTRC A. contortrix USA KY Wolf
058-594-037 NNTRC A. contortrix USA MO Cole
058-843-771 NNTRC A. contortrix USA TX Colorado
ASK 9044 JLS 11 ASNHC 14264 A. piscivorus USA TX Tom Green
ASK 9048 JLS 17 A. piscivorus USA TX Walker
ASK 9049 JLS 18 A. piscivorus USA TX Walker
ASK 9050 JLS 19 A. piscivorus USA TX Walker
ASK 9053 JLS 22 A. piscivorus USA TX Walker
ASK 9055 JLS 24 A. piscivorus USA TX Walker
ASK 9062 JLS 33 A. piscivorus USA TX Gonzales
ASK 9063 JLS 34 A. piscivorus USA TX Gonzales
ASK 9065 JLS 36 A. piscivorus USA TX Gonzales
ASK 9066 JLS 37 ASNHC 14284 A. piscivorus USA TX Gonzales
ASK 9067 JLS 38 ASNHC 14286 A. piscivorus USA TX Gonzales
ASK 9068 JLS 39 ASNHC 14279 A. piscivorus USA TX San Patricio
ASK 9071 JLS 42 A. piscivorus USA TX San Patricio
ASK 9072 JLS 43 A. piscivorus USA TX San Patricio
ASK 9082 JLS 54 A. piscivorus USA TX Angelina
ASK 9083 JLS 55 ASNHC 14276 A. piscivorus USA TX Angelina
ASK 9084 JLS 56 A. piscivorus USA TX Jasper
ASK 9085 JLS 57 A. piscivorus USA TX San
Augustine
ASK 9091 JLS 63 A. piscivorus USA TX Kimble
ASK 9092 JLS 64 A. piscivorus USA TX Kimble
ASK 9093 JLS 65 A. piscivorus USA TX Kimble
ASK 9095 JLS 67 A. piscivorus USA TX Smith
ASK 9098 JLS 70 A. piscivorus USA TX Smith
ASK 9101 JLS 74 ASNHC 14275 A. piscivorus USA TX Smith
ASK 9103 JLS 76 A. piscivorus USA TX Smith
ASK 9105 JLS 78 A. piscivorus USA TX Smith
ASK 9108 JLS 81 A. piscivorus USA TX Tarrant
ASK 9109 JLS 82 A. piscivorus USA TX Tarrant
648 Copeia 2014, No. 4
Tissue ID Snake ID Museum ID Species Country State County Accession no.
ASK 9110 JLS 83 A. piscivorus USA TX Tarrant
ASK 9121 JLS 94 A. piscivorus USA TX Tom Green KC431020
ASK 9127 JLS 100 A. piscivorus USA TX Menard
ASK 9131 PIT 114938716A A. piscivorus USA TX Tom Green
ASK 9137 PIT 115222097A A. piscivorus USA TX Tom Green
ASK 9143 PIT 114967277A A. piscivorus USA TX Tom Green
ASK 9144 PIT 115317467A A. piscivorus USA TX Tom Green
ASK 9145 PIT 114952455A A. piscivorus USA TX Tom Green
ASK 9150 PIT 114979652A A. piscivorus USA TX Tom Green
ASK 9152 PIT 114954121A A. piscivorus USA TX Tom Green KC431023
ASK 9154 PIT 114616122A A. piscivorus USA TX Tom Green
ASK 9160 PIT 114631146A ASNHC 14289 A. piscivorus USA TX Tom Green
ASK 9161 PIT 114633364A A. piscivorus USA TX Tom Green
ASK 9164 PIT 114949391A A. piscivorus USA TX Tom Green KC431025
ASK 9169 PIT 115322477A A. piscivorus USA TX Tom Green
LSUMZ H-2020 A. piscivorus USA MS Perry EU483465
LSUMZ H-2367 A. piscivorus USA MS Wilkinson EU483466
LSUMZ H-2368 A. piscivorus USA MS Wilkinson EU483467
LSUMZ H-19042 A. piscivorus USA LA East Baton
Rouge
TJL 1487 TNHC 65313 A. piscivorus USA TX Fort Bend EU483474
TJL 1476 TNHC 65358 A. piscivorus USA TX Jefferson EU483473
TJL 990 TNHC 66514 A. piscivorus USA TX Chambers
010-325-361 NNTRC A. piscivorus USA TX Harris
010-820-563 NNTRC A. piscivorus USA TX Galveston
011-311-367 NNTRC A. piscivorus USA FL
CLP 159 A. piscivorus USA FL Collier
CLP 984 A. piscivorus USA GA Grady
CLP 986 A. piscivorus USA GA Thomas KC431026
CLP 989 A. piscivorus USA GA Grady KC431027
KW 0548 A. piscivorus USA FL Glades
KW 0549 A. piscivorus USA FL Glades
KW 0579 A. piscivorus USA FL Jefferson
KW 0602 A. piscivorus USA FL Madison
KW 0623 A. piscivorus USA SC Aiken
KW 0629 A. piscivorus USA SC Barnwell
KW 0631 A. piscivorus USA SC Barnwell
KW 0648 A. piscivorus USA SC Jasper
KW 0655 A. piscivorus USA FL Columbia
KW 0660 A. piscivorus USA FL Liberty
KW 0661 A. piscivorus USA FL Columbia
KW 0679 A. piscivorus USA FL Levy
KW 0727 A. piscivorus USA FL Jefferson
KW 0728 A. piscivorus USA FL Baker
KW 0759 A. piscivorus USA MS Lafayette
KW 0769 A. piscivorus USA MS Lafayette
KW 0791 A. piscivorus USA FL Wakulla
KW 0805 A. piscivorus USA FL Wakulla
CLP 140 A. taylori Mexico Tamaulipas AY223613
ASK 9111 JLS 84 A. taylori Mexico Tamaulipas
ASK 9117 JLS 90 C. atrox USA TX Pecos
ASK 9118 JLS 91 C. molossus USA TX Pecos
ASK 9070* JLS 41 A. piscivorus USA TX San Patricio KC431019
ASK 9132* PIT 114409631A A. piscivorus USA TX Tom Green KC431021
ASK 9133* PIT 114625792A A. piscivorus USA TX Tom Green KC431022
ASK 9153* PIT 113932567A A. piscivorus USA TX Tom Green KC431024
APPENDIX 1. Continued.
Strickland et al.—Phylogeography of Cottonmouths 649
