See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/280587601

Islands within islands: Diversification of

tailless whip spiders (Amblypygi, Phrynus) in
Caribbean caves

ARTICLE in MOLECULAR PHYLOGENETICS AND EVOLUTION · JULY 2015

Impact Factor: 3.92 · DOI: 10.1016/j.ympev.2015.07.005 · Source: PubMed

READS

123

8 AUTHORS, INCLUDING:

Trevor Bloom Laura J May-Collado

Western Washington University University of Vermont
2 PUBLICATIONS 1 CITATION 50 PUBLICATIONS 538 CITATIONS

SEE PROFILE SEE PROFILE

Greta J Binford Ingi Agnarsson

Lewis & Clark College University of Vermont
36 PUBLICATIONS 397 CITATIONS 105 PUBLICATIONS 2,150 CITATIONS

SEE PROFILE SEE PROFILE

Available from: Ingi Agnarsson

Retrieved on: 19 October 2015
 Molecular Phylogenetics and Evolution 93 (2015) 107–117

Contents lists available at ScienceDirect

Molecular Phylogenetics and Evolution

journal homepage: www.elsevier.com/locate/ympev

Islands within islands: Diversification of tailless whip spiders

(Amblypygi, Phrynus) in Caribbean caves q
Lauren A. Esposito a, Trevor Bloom b, Laura Caicedo-Quiroga c, Angela M. Alicea-Serrano c,
Jose A. Sánchez-Ruíz c, Laura J. May-Collado d, Greta J. Binford b, Ingi Agnarsson d,e,⇑
a
University of California at Berkeley, Essig Museum, 130 Mulford Hall, Berkeley, CA 94720-3114, United States
b
Lewis and Clark College, 0615 S.W. Palatine Hill Road, Portland, OR 97219, United States
c
University of Puerto Rico at Rio Piedras, San Juan, PR 00931, United States
d
University of Vermont, Department of Biology, 109 Carrigan Drive, Burlington, VT 05405-0086, United States
e
Department of Entomology, National Museum of Natural History, Smithsonian Institution, NHB-105, PO Box 37012, Washington, D.C. 20013-7012, United States

a r t i c l e i n f o a b s t r a c t

Article history: Islands have played a key role in understanding species formation ever since Darwin’s work on the
Received 4 March 2015 Galapagos and Wallace’s work in the Malay Archipelago. Like oceanic islands, habitat ‘islands’, such as
Revised 30 June 2015 mountaintops and caves similarly may drive diversification. Here we examine patterns of diversification
Accepted 11 July 2015
in the tailless whip spider genus Phrynus Larmarck, 1809 (Amblypygida: Phrynidae) a system that shows
Available online 26 July 2015
evidence of diversification under the influence of ‘islands within islands’. We estimate phylogeographic
history and measure genetic diversity among representatives of three nominal Phrynus species from epi-
Keywords:
gean and cave systems of Puerto Rico and nearby islands. Data from five loci (mitochondrial 12S, 16S,
Biodiversity hotspot
Cavernicolar
Cox1; nuclear H3, 28S) were used to generate phylogenetic hypotheses and to assess species monophyly
Cryptic radiation and phylogeographic relationships. Genetic divergences and population limits were estimated and
Island biogeography assessed using the Geneious barcoding plugin and the genealogical sorting index. We find that mtDNA
Speciation sequence divergences within each of the three Phrynus species range between 15% and 20%. Genetic
Sexually biased dispersal divergence is structured at three spatial scales: among islands in a manner consistent with the
GAARlandia hypothesis, among bedrock formations within Puerto Rico, and among caves within these
formations. Every isolated cave system contains a unique mtDNA genetic lineage of Phrynus, with diver-
gence among cave systems far exceeding that within. In some localities epigean specimens nest among
cave taxa, in others caves are monophyletic. Remarkably, clades that show up to 20% mtDNA sequence
divergence show little or no variation in the nuclear markers. We interpret this pattern as resulting from
extreme conservation of our nuclear markers rather than male sex-biased dispersal, based on high con-
servation of 28S and H3 between our individuals and other amblypygid genera that are restricted to
Africa. While this study includes but a tiny fraction of Caribbean caves, our findings suggest Phrynus
may be much more diverse than hitherto thought, at least in terms of mtDNA diversity, and that the
arthropod fauna of caves may represent a dimension of biodiversity that is yet to be discovered in the
Caribbean biodiversity hotspot.
Ó 2015 Elsevier Inc. All rights reserved.

1. Introduction Whittaker et al., 2001). The important roles of different kinds of

Detecting major influences on the formation of biodiversity is a
central goal in biology (Allen and Savage, 2007; Avise et al., 2008;
Colwell and Coddington, 1994; Etienne et al., 2007; Gavrilets,
2003; Gillespie et al., 2008; Ricklefs, 2006; Turner, 2004;
q
This paper was edited by the Associate Editor Alfried Vogler.
⇑ Corresponding author at: University of Vermont, Department of Biology, 109
Carrigan Drive, Burlington, VT 05405-0086, United States.
E-mail address: iagnarsson@gmail.com (I. Agnarsson).
barriers have long been noted in species formation (Darwin,
1859; Futuyma, 2009; Wallace, 1876). A panmictic population
can be divided by the formation of a barrier (vicariance model),
or the rare dispersal of a few individuals across an existing barrier
(dispersal model), which can immediately restrict gene flow
among the subdivided populations (Reno et al., 2009; Turelli
et al., 2001; Wiley, 1978). This can swiftly lead to genetic diver-
gence, both due to natural selection and genetic drift, and may
eventually result in sexual incompatibility if the populations were
to come in contact again; the classical model of allopatric

http://dx.doi.org/10.1016/j.ympev.2015.07.005
1055-7903/Ó 2015 Elsevier Inc. All rights reserved.
108 L.A. Esposito et al. / Molecular Phylogenetics and Evolution 93 (2015) 107–117

speciation (Gavrilets, 2003). Island archipelagos are systems where sampling to date. We also conduct a preliminary assessment of lar-
such processes have led to spectacular radiations resulting in high ger scale biogeographic influences on this lineage in the Caribbean.
richness and endemicity (Allen and Savage, 2007; Avise et al.,
2008; Colwell and Coddington, 1994; Emerson, 2002; Etienne
2. Materials and methods
et al., 2007; Gavrilets, 2003; Gillespie, 2004; Gillespie et al.,
2008; Losos et al., 1998; Ricklefs, 2006; Schluter, 1996, 2000;
2.1. Study organisms and taxon sampling strategy
Turner, 2004; Whittaker et al., 2001). Indeed, many of the world’s
biodiversity hotspots are archipelagos, including the Caribbean, the
Phrynus are cryptic, nocturnal organisms and can be found in
region of focus in this study. However, not all islands are oceanic.
epigean habitats, typically dwelling under logs or rocks, or caver-
For some groups of organisms terrestrial habitats are effective
niculous habitats where they predate on other arthropods on the
‘islands’ in being surrounded by a matrix of ‘non-habitat’ with
cavern walls (Quintero, 1981; Weygoldt, 2000). The genus is dis-
common examples including mountaintops, isolated humid or arid
tributed in the New World, from the southern United States to
patches, and cave systems (Cardoso, 2012; Gillespie and Clauge,
northern South America, and throughout the Caribbean, and pecu-
2009; Juan and Emerson, 2010; Juan et al., 2010; Snowman et al.,
liarly with a single known species from Indonesia (Harvey, 2002;
2010). If islands to some extent drive diversification, it should be
Rahmadi and Harvey, 2007). Phrynus is currently represented by
particularly interesting to look at systems that can be character-
24 extant species (de Armas, 2004; de Armas and Angarita Arias,
ized as ‘islands within islands’ such as the rich systems of caves
2008; de Armas and Perez Gonzalez, 2001; de Armas and Teruel,
found on many of the Caribbean islands.
2010; de Armas and Viquez, 2001; Harvey, 2003; Quintero, 1981)
Cave species often show limited dispersal among caves or cave
and 3 fossil species from Dominican and Chiapas amber (de
systems, resulting in relatively small geographic ranges and high
Armas, 2004; Poinar and Brown, 2004). The island of Puerto Rico
degrees of endemism in comparison to epigean species (Culver
contains three described species of Phrynus: P. longipes Pocock,
et al., 2006; Porter, 2007; Snowman et al., 2010). Mountain ranges
1893 (type locality Haiti, Hispañola), P. alejandroi Armas &
and other discontinuities in karst regions and lack of direct subter-
Teruel, 2010 (type locality southern PR including Isla Caja de
ranean connections among cave systems may act as barriers for
Muertos), and the Pan-Caribbean species P. marginemaculatus C.L.
cave fauna, resulting in genetic divergence among populations
Koch, 1841 (not included in present study). Phrynus eucharis
(Chiari et al., 2012; Howarth, 1996b). Interfaces between geologic
Armas and Perez-Gonzalez, 2010, type locality Samana peninsula
substrates may also act as barriers to gene flow. Thus large and
in Dominican Republic, is recorded in this study from Isla de
geologically complex islands rich in caves, such as the Greater
Mona for the first time.
Antilles, can be seen as harboring ‘islands within islands’ each con-
This study focuses on sampling Puerto Rican caves as a first
taining a large number of cave ‘islands’. In this system the potential
insight into the role of caves as ‘islands’ in the Caribbean. Puerto
for allopatric speciation occurs at three distinct spatial scales, (1)
Rico is one of the ancient Greater Antilles islands, all of which
among geological islands, (2) among geologically distinct regions
are thought to have been emergent during the last 40 my or so
within the islands, and (3) among cave systems within these
(Iturralde-Vinent and MacPhee, 1999). Various islands, especially
regions. As a result of these augmented opportunities for specia-
within the Greater Antilles, have been connected at some point,
tion, Caribbean caves may harbor remarkably large and unique
and the GAARlandia hypothesis is particularly important implying
pockets of biodiversity, as described in various lineages in other
connections among the Greater Antilles and mainland South
cave systems (Cardoso, 2012; Juan et al., 2010). Many examples
America 33–35 mya, and gradual breakup of the islands after they
of lineages with cave divergence and endemism are from arthro-
separated from South America (Iturralde-Vinent and MacPhee,
pods and arachnids in particular (Hedin and Thomas, 2010;
1999). We also sample some of its outlying islands, that are
Hedin, 1997; Snowman et al., 2010); these lineages may be excep-
thought to have been connected to Puerto Rico at some point in
tionally diverse within cave systems in the Caribbean. However,
time, such as Vieques, Caja de Muertos, but are relatively young
the arthropod fauna of Caribbean caves is extremely poorly known,
as islands. Of these islands, Mona stands out due to its unique his-
most caves have never been systematically surveyed, and the
tory. Isla de Mona is the second largest satellite island in the Puerto
majority of studies of arachnids are isolated taxonomical descrip-
Rico Commonwealth (after Vieques), situated in the center of the
tions based on morphology (e.g. Bloom et al., 2014).
Mona channel, midway between Puerto Rico and Hispaniola. The
Here we explore cave systems in Puerto Rico and its outlying
geological history of Isla de Mona is not clear. It is a tectonically
islands of Isla de Mona, Caja de Muertos, and Vieques, and in the
uplifted Mio-Pliocene carbonate island (Frank et al., 1998) that is
Dominican Republic (Hispaniola) the caves of Parque del Este and
thought to have been a part of the GAARlandia ridge, but potential
Seibo, to test their role in diversification of a lineage found in
connections with proto-Hispaniolan islands and/or Puerto Rico are
nearly every Caribbean cave: the tailless whip spider genus
unclear (Iturralde-Vinent and MacPhee, 1999), it may have been
Phrynus Larmarck, 1801 (Amblypygi: Phrynidae). There have been
submerged until the Mio Pliocene (23–5 mya) (Frank et al.,
no previous molecular studies examining Phrynus or other caver-
1998). Nevertheless, geological and biological evidence suggests a
niculous Caribbean arachnids. Amblypygids are of particular inter-
more recent connection between Isla de Mona and Hispaniola than
est for biogeographic studies as dispersal-limited lineage—in some
between Isla de Mona and Puerto Rico. Therefore we also sample a
species the same individual has been observed on the same tree
couple of localities on eastern Hispaniola as potential sources of
over the course of months or even years (Weygoldt, 2000), though
Isla de Mona populations.
such site fidelity has not been studied in Phrynus—that has both
epigean and cave taxa that tend to be restricted to single biogeo-
graphic regions (Harvey, 2003). We examine the genetic structure 2.2. Data
of whip spiders among caves, to begin to explore how cave systems
may promote diversification of island taxa. Specifically we use This study includes 176 total specimens including two out-
molecular phylogenetics to analyze the genetic structure and groups. Of the ingroup, 142 specimens are morphologically consis-
patterns of relationships of Phrynus and assess monophyly and tent with the nominal species P. alejandroi and P. longipes and were
genetic depth of cave and epigean lineages. With these patterns collected from epigean and caverniculous habitats across the main-
we estimate the number of cryptic taxa contained within our land of Puerto Rico (Table S1, Fig. 1). Thirty-two additional
 L.A. Esposito et al. / Molecular Phylogenetics and Evolution 93 (2015) 107–117
Los Guayuyos
DOMINICAN REPUBLIC
longipes
Seibo
PN del Este*
MONA
alejanderoi and eucharis
Fig. 1. A diagrammatic map of the sampling region. Orange oval outlines the distribution of the alejandroi and eucharis lineages, in this study – both lineages are known to
also occur elsewhere on Hispaniola – while the longipes lineage is widespread outside that highlighted area. (For interpretation of the references to color in this figure legend,
the reader is referred to the web version of this article.)
individuals are from eastern Hispaniola, and from islands sur-
rounding Puerto Rico, including Isla Caja de Muertos, Vieques,
and Isla de Mona (Fig. 1); localities from which individuals were
collected that are consistent with all three nominal species.
Whole whip spiders were collected and preserved in 95% etha-
nol. Genomic DNA was extracted from a single leg using a Quiagen
DNeasy Blood and Tissue Kit. DNA was eluted in 200 ll of
de-ionized H20 buffer and stored at 20 °C. Five loci were selected
for their utility in phylogenetic and phylogeographic studies of
arachnids: mitochondrial 12S, 16S and CO1; nuclear H3 and 28S
(Bidegaray-Batista et al., 2015; McHugh et al., 2014; Opatova and
Arnedo, 2014; Ruch et al., 2015), for a total of 2740 aligned charac-
ters. Selected loci were amplified using universal primers
(Table S2) following protocols of prior studies in our labs
(Agnarsson and Rayor, 2013; Kuntner et al., 2013; Prendini et al.,
2005). Amplified fragments were sequenced in both directions by
the University of Arizona Genetic Core and then assembled using
the Chromaseq module (Maddison and Maddison, 2011a) in
Mesquite 3.02 (Maddison and Maddison, 2011b) through Phred
and Phrap (Green, 1999; Green and Ewing, 2002), and then proof-
read. For a portion of the specimens, notably all individuals from
Vieques and epigean samples from Mata de Platano (kindly pro-
vided by Kenny Chapin) we only have COI sequences. Thus our
main phylogenetic analyses focus on the multi-gene dataset with
additional analyses run to accommodate the additional
single-gene specimens. These datasets will be referred to as ‘focal’
and ‘all data’ henceforth.
Alignments were performed for each locus in MAFFT v7.036
(Katoh, 2013) using the FFT-NS-i strategy with a 1PAM/k = 2
parameter, and a gap opening penalty of 1.53. Resulting align-
ments were tested for models of substitution in jModelTest2
(Darriba et al., 2012). Phylogenetic analyses were done on parti-
tioned by gene, concatenated alignments including separate analy-
ses for mtDNA only, nucDNA only and the full dataset. Additional
analyses were done including adding partitioning by codon for pro-
tein coding genes, and analyzing a matrix where taxa with the
most missing data were eliminated (excluding taxa with >60%
missing data). Each dataset was analyzed under Bayesian
(MrBayes 3.2.1) (Huelsenbeck and Ronquist, 2001; Ronquist and
Huelsenbeck, 2003), and likelihood (RaxML) (Stamatakis, 2006)
criteria, employing the models selected by jModelTest 2, and
109
PUERTO RICO
Vieques
Guanica
Isla Caja de Muertos
parsimony (TNT) under equal weights. Bayesian analyses were
run on the CIPRES cluster (Miller et al., 2010): 4 runs of 60 million
generations. Convergence was assessed with AWTY (Nylander
et al., 2008) and the burnin discarded. A majority-rule consensus
tree was constructed from the post-burnin distribution of trees.
The concatenated dataset, partitioned by gene and codon position,
was analyzed in raxmlHPC v7.0.4 (Stamatakis, 2006). Each parti-
tion was analyzed under the GTR + C model.
2.3. Species delineation and population assignment
Populations were assigned using the species delimitation plugin
in Geneious (Kearse et al., 2012; Masters et al., 2011) using the ‘all
data’ dataset. The majority-rule consensus tree from the 95% pos-
terior probability of trees was imported into Geneious. Working
from the tip of the tree downward, each monophyletic clade of
two or more specimens, or single taxon branches, were treated
as a population. The Geneious plugin provides a number of metrics
for assessing species or population delimitation including
Rosenberg’s reciprocal monophyly P(AB) (Rosenberg, 2007) and
Rodrigo’s P(RD) (Rodrigo et al., 2008) measures; as well as values
for the probability of population identification of a hypothetical
sample based on the groups being tested (P ID(Strict) and P ID
(Liberal)). The genealogical sorting index (gsi) statistic
(Cummings et al., 2008) was calculated using the gsi webserver
(http://genealogicalsorting.org) on the estimated tree and an
assignment file that contained the same user specified groups
identified in the Geneious plugin.
An additional species delimitation analysis used a Bayesian
implementation (bPTP) of the Poisson tree processes model
(Zhang et al., 2013) to infer putative species boundaries on a given
single locus phylogenetic input tree. The input tree used in the
bPTP analysis was the resulting maximum clade credibility tree
from our Bayesian phylogenetic analysis. The analysis was run as
a rooted tree with outgroups removed for 100,000 generations
with 10% burnin removed.
2.4. Genetic distances and isolation
Estimates of the number of base substitutions per site were
conducted in MEGA5 (Tamura et al., 2011) using uncorrected
110 L.A. Esposito et al. / Molecular Phylogenetics and Evolution 93 (2015) 107–117
p-distances and the Kimura 2-parameter model (Kimura, 1980)
utilizing the ‘all data’ dataset. These were from averaging over all
mtDNA (COI, 16S, 12S) sequence pairs between reciprocally mono-
phyletic population/species level groups (identified in the resulting
phylogenies). Estimates of net evolutionary divergence (Tamura
et al., 2011) were calculated using the Kimura 2-parameter model.
Within group distances were estimated using p-distances and the
Kimura-2 model. For all calculations, ambiguous positions were
removed for each sequence pair.
The Isolation By Distance Web Service v3.23 (Jensen et al., 2005)
was used to test for the role of isolation by distance (ibd) in the
population structure of Phrynus. The ibd webserver assesses the
role of isolation by distance through the use of a Mantel test for
matrix correlation between a genetic similarity matrix (fst) and a
geographic distance matrix (in meters). To do this, the average lat-
itude and longitude for each clade was calculated and the pairwise
geographic distances were then computed using the Geographic
Distance Matrix Calculator v1.2.3 (Ersts, 2006). A file containing
the pairwise geographic distances, population assignments and
concatenated DNA sequences was then uploaded into the web ser-
ver, 10,000 randomizations were ran using PhiST, missing data
were ignored, and ibd was analyzed using Slatkin’s Similarity index
(M).
3. Results
3.1. Data, phylogenetic, and population structure
The resulting concatenated alignment of the five loci was
2740 bp in length with 838 variable sites (720 parsimony informa-
tive), and had an average nucleotide composition of 33.5% thymine,
19.8% cytosine, 30.3% adenine, and 16.5% guanine. Individual parti-
tions were 382 bp (12S), 514 bp (16S), 540 bp (28S), 975 bp (Cox1),
and 329 bp (histone 3) in length, and had 1%, 3%, 27%, 30%, and 8%
of missing data respectively. Model testing in jModelTest identified
the best model for each partition, however in those cases where
models other than GTR + gamma had the highest likelihood and
AIC scores, the score was not statistically significantly different
from GTR + gamma. As a result, GTR + gamma was selected and
applied separately to each partition for analyses.
Phylogenetic analyses of the focal dataset under Bayesian, like-
lihood, and parsimony criteria produced nearly identical topologies
(Figs. 2, S1), differing only in terminal rearrangements of samples
within population clades. A total of 13, well-supported, recipro-
cally monophyletic population/species-level clades were recovered
in all combined analysis and all analyses including individual
mtDNA partitions. The nucDNA had limited phylogenetic signal
for populations, and while supporting the monophyly of the long-
ipes clade, did not resolve relationships within it. Analyses of the
focal dataset recover two major groups, one containing P. eucharis
and P. alejandroi from Hispaniola, Isla de Mona, Puerto Rico, Isla
Caja de Muertos, and one containing all P. longipes specimens from
Hispaniola and Puerto Rico. Fourteen clades represent individual
caves or cave systems, four are epigean (Figs. 2 and 3). In several
cases epigean taxa are sister to cave taxa, in one case forming a
grade leading to a cave clade in the Guanica area. Support values
for higher level clades in the focal analysis are universally high,
but are somewhat lower in the full data analysis, presumably
because the only difference between the two is the addition of taxa
with few data. Notably, in the full data analysis specimens from
Vieques islands are highly distinct and unique and fall out as sister
to the remaining taxa in the analysis (Fig. 4).
Within each of the three nominal species we found mtDNA
sequence divergences were high, up to 20% (Table 1). The analy-
ses provide weak support of the monophyly of individuals we
identified as P. longipes with a sister relationship between Puerto
Rican and Hispaniolan taxa (Fig. 2). However, in the full dataset
analysis the P. longipes lineage from Hispaniola no longer nest with
its Puerto Rican ‘conspecifics’, a pattern that is possibly an artifact
of missing data in that analysis. Within the P. longipes group, nine
monophyletic populations were identified phylogenetically, and
supported by various species delimitation methods. Three popula-
tions include both epigean and cavernicolar specimens, two are
singletons collected from epigean localities, and the remaining four
were collected from within caves and cave systems. Of the mixed
locality clades, one (Aguas Buenas + El Yunque) is an epigean spec-
imen sister to a cave clade. Furthermore, in the full dataset analy-
ses individuals from epigean habitats in Mata de Platano nest
among individuals from the Mata de Platano and Mario Soto caves
(Fig. 4).
P. alejandroi plus P. eucharis form a clade but they are not each
monophyletic (Fig. 2). The P. eucharis clade 1 unites individuals
from Parque del Este in southeastern Dominican Republic (DR)
with individuals from three caves in Isla de Mona, with about 2%
mtDNA sequence divergence between these. Another set of taxa
morphologically consistent with P. eucharis (clade 2) is placed
within a clade of P. alejandroi rendering the latter species para-
phyletic. In a pattern that mirrors clade 1, a single individual P.
eucharis from Parque Nacional del Este (Dominican Republic- DR)
is placed in lineage 2 and sister to all other samples (individuals
from Isla de Mona and Vieques), also with 2% sequence diver-
gence. One Isla de Mona cave, Cueva Lirios, is represented by indi-
viduals in both clades 1 and 2.
All individuals we identified morphologically as P. alejandroi
were collected from mainland southwestern Puerto Rico (PR) or
Isla Caja de Muertos, both characterized by very dry climate.
There are high levels of genetic divergence among these individu-
als (max > 15%) with the distinct clades 1 and 2 (Table 1). Both
clades include individuals from Guanica, southwestern Puerto
Rico. Clade 2 includes a lineage from a cave and epigean individu-
als from the area immediately surrounding the cave. In clade 1, the
sister lineage to individuals from Isla Caja de Muertos (a locality
from the type series) comes from the southern tip of Guanica,
the coastal Punta Ballena.
The thirteen unique clades identified as putative species (Figs. 2
and 4, circled numbers), were all monophyletic, and among them
had inter-clade distances ranging from 5.6% to 26% divergence
(Table 1). Using the ‘‘strict’’ criterion described by Ross et al.
(2008) the probabilities of correctly identifying a hypothetical
unknown ranged from 0.83 for the P. longipes Aguas Buenas + El
Yunque clade, to 0.97 for the P. longipes Mapancha clade. Results
of the bPTP analysis for species delimitation indicate an estimated
31–42 ‘species’ with a mean of 36.4. This reflects the highly geo-
graphically structured populations within some of these putative
species level clades (Fig. 2, e.g. clade 11). However, employing mul-
tiple approaches, only nine putative species were statistically sig-
nificant for all measures. Two are represented by singletons and
so could not be assessed (P. longipes from Cambaleche and P. long-
ipes from Quebradillas). Two populations – P. eucharis clade 1 and
P. longipes from Aguas Buenas + El Yunque had non-significant
P(RD) measures, but were statistically significant for P(AB) and gsi.
3.2. Genetic structure and diversity at three spatial scales
Both deeper (nominal species and species groups) and shal-
lower (genetic ‘species’ or populations) clades are nearly com-
pletely allopatric (Figs. 1–3). At the broadest geographic scale all
islands explored (Hispaniola, Puerto Rico, Isla de Mona, Vieques,
and Isla Caja de Muertos) contained some single-island mono-
phyletic lineages. At a finer scale, within the most densely sampled
island of Puerto Rico, lineages are resolved into clades that seem to
 L.A. Esposito et al. / Molecular Phylogenetics and Evolution 93 (2015) 107–117 111

Fig. 2. Majority rule consensus phylogram of the combined multi-locus dataset including 161 specimens of the whip spider genus Phrynus collected from caves and epigean
habitats (the latter indicated with an asterisk) primarily from Puerto Rico, with additional samples from Mona, Isla Caja de Muertos, and the Dominican Republic (DR). For
specimen detail see Fig. S1. Three morphological species (eucharis, alejandroi, longipes), were sampled, the analyses reveal strong genetic structure at three spatial scales:
among islands, among bedrock formations within PR, and among caves within these. 9–42 Putative species are delimited by the various methods used (see Table 1,
supplementary material), with a conservative estimate of minimally 11 putative species, all of which show at least 4% mtDNA divergence. Color coding refers to single
collection localities or cave systems representing nearby clades that are likely interconnected. Numbers at nodes are posterior probability values, all clades containing such
numbers were also recovered in both likelihood and parsimony analyses of these data. Circled numbers identify conservatively estimated putative species level clades. Scale
bar shows expected substitutions per site. For individual specimen codes see Fig. 4. (For interpretation of the references to color in this figure legend, the reader is referred to
the web version of this article.)

be separated by a barrier formed by the interface between karst
limestone and igneous bedrock running along the central northern
and central southern part of the island (Fig. 3). Phrynus longipes, for
example, forms two main lineages within Puerto Rico, one
restricted to the northern limestone and the other to the central
igneous bedrock. Even within these, there is apparent geographic
structure. Within the northern limestone clade, animals from the
central north form a grade leading to a western clade (Matos,
Maria Soto, Agrodel, Mata de Platano).
Finally, at the finest scale of cave ‘islands within islands’, our
data show clear patterns where animals from nearby and presum-
ably interconnected caves form monophyletic groups, such as
Cueva Cangrejos and Cueva Cabachuelas in Morovis and Cueva
Vientos and Cueva Balcones in Florida municipal (Figs. 2 and 3).
112 L.A. Esposito et al. / Molecular Phylogenetics and Evolution 93 (2015) 107–117
Mata de Platano 1+2
Bajura del Cerezal*
Lirios
Lozeta
Caballo
Murcielagos
Guanica*
Fig. 3. A map of sampling sites in Puerto Rico, with colors reflecting clades in Fig. 1. The map also depicts the division of Puerto Rico based on fundamental types of bedrock
(modified from Briggs and Akers, 1965, Monroe, 1980). The shaded area shows karst limestone, while the un-shaded area contains various igneous/metamorphic rock. The
geological division of Puerto Rico is consistent with the three main clades found there, in complete allopatry. (For interpretation of the references to color in this figure legend,
the reader is referred to the web version of this article.)
However, each genetic lineage is completely restricted to such a
system of interconnected caves, or a single cave. For example,
Puerto Rican regional clades include cave systems of Arecibo and
southern Florida, Camuy, Maunabo, Morovis, northern Florida,
Guayanilla, Aguas Buenas, and Guanica (Fig. 3). In some cases
genetic structure is extremely fine-scaled with some neighboring
caves having two distinct genetic lineages. This is represented,
for example, between the Aguas Buenas oscuro and Aguas
Buenas claro caves, caves that are adjacent and partially intercon-
nected. Caves over 10 km apart never shared genetic lineages.
A test of isolation by distance was performed to assess the sig-
nificance of geographical distance on the genetic structure of
Phrynus populations. Uncorrected, log geographic distance, log
genetic distance, and both datasets log-scaled were
non-significant for isolation by distance (0.25 > p > 0.75), showing
no correlation between geographic and genetic distance (r = 0.1
to 0.07).
4. Discussion
4.1. Data, phylogenetic, and population structure
With strong support of clades and general congruence among
analyses, we uncover intriguing evidence for the role of isolation
among Caribbean islands, and cave systems within them, in driving
genetic diversification in Phrynus tailless whip spiders. Oceanic
islands have provided numerous examples of spectacular evolu-
tionary radiations and the patterns we uncover in this analysis sug-
gest Phrynus have diversified on the Caribbean islands to a much
greater degree than previously thought based on morphological
taxonomy (Figs. 1–3) (de Armas, 2004, 2010; de Armas and Perez
Gonzalez, 2001). This typically epigean and troglobitic lineage
appears to be diversifying under the hierarchical influence of isola-
tion due to island vicariance, habitat differences based on geology,
and genetic isolation among cave ‘islands within islands’
(Allegrucci et al., 2011; Cardoso, 2012; Juan et al., 2010;
Snowman et al., 2010).
Individuals that are morphologically consistent with P. eucharis
are resolved in two deeply divergent clades that are not sister lin-
eages. The large genetic divergence within and between these sug-
gests they are two species groups, each composed of multiple
species. Interestingly individuals from Cueva Lirio (Mona) and
Parque Nacional del Este (DR) are placed in both of those clades
further evidencing independently evolving lineages within the
nominal P. eucharis. Phrynus alejandroi in these analyses have at
least two deeply divergent genetic lineages that are also not
resolved as monophyletic. In fact, only P. longipes is resolved as
monophyletic in our focal analysis (Fig. 1), though not in the full
Cambalache *
Agrodel
Matos Cabachuelas
Balcones
Maria Soto
Cangrejos El Yunque, El Verde*
Vientos
and Gemelos
Aguas Buenas
El Convento
and Mapancha Monte Pirata Cayo de Tierra
Guajonales Matuyas Alto
Guajonales Matuyas Bajo
Guajonales Sector los Sánches
Punta Ballena*
Isla Caja de Muertos*
data analysis (Fig. 4). Within the P. longipes clade genetic diver-
gence is structured as a function of island, geographic area, and
cave locality. With mitochondrial genetic distances ranging up to
20% among P. longipes clades there is little doubt that lineages have
been isolated for long periods of time without exchange of genes.
Given very limited exploration of the caves of the Caribbean
islands, we expect that the actual number of species within ‘P.
longipes’ might be great. We find average intra-population dis-
tances between 0.2% and 0.4% (Table 1) depending on whether
the Aguas Buenas caves and El Yunque specimen are treated as
one or two groups. Using a simple ‘barcoding gap’ method that is
based on 10 more divergence among than within groups, up to
16 species would be implied among the sampled specimens
(Hebert et al., 2003) The single locus bPTP approach (Zhang et al.,
2013) suggests an even higher diversity of 31–42 species, but
should not be overinterpreted as evidence for species numbers
(Hedin, 2015). More conservatively looking at agreement among
all employed methods they all suggest at least nine putative spe-
cies, plus two highly divergent singletons for a relatively conserva-
tive estimate of 11+ putative species (Figs. 2 and 4, Table 1,
supplementary materials). Thus our limited sampling reveals
unappreciated diversity. Considering that the vast majority of
caves in Puerto Rico, and especially on the large islands of
Hispaniola and Cuba, are as yet unsampled, further effort is certain
to discover much greater diversity still within this lineage. These
findings are consistent with high species turnover (beta diversity)
that characterizes many cave systems (Cardoso, 2012; Juan et al.,
2010).
An intriguing pattern in our data is high mtDNA diversity (up to
20% within nominal species) with little variation in our nuclear
markers H3 and 28S (Fig. S2). Given this pattern, a possible alterna-
tive explanation to high species diversity is extreme sexually
biased dispersal. If this reflects sex-biased dispersal the implication
is that Phrynus females show extreme site fidelity but males dis-
perse widely, and may readily cross oceanic barriers. It would also
imply that P. eucharis and P. alejandroi are synonyms as they share
identical nuclear sequences. That hypothesis does not seem plausi-
ble given (1) current understanding of the biology of these animals
suggests they have limited dispersal abilities and high site fidelity
(Weygoldt, 2000), (2) there are clear morphological distinctions
between P. eucharis and P. alejandroi, and (3) lineages (e.g. genera)
tend to be restricted to single biogeographic regions (Harvey,
2003). Instead we prefer the hypothesis that the nuclear markers
employed here are highly conserved in the group. Notably, homol-
ogy searching in Genbank of Caribbean Phrynus nuclear sequences
finds little divergence across the order. For example, sequences of
genera restricted to Africa (Damon, Xenophrynus) show up to 98%
sequence identity to Caribbean Phrynus. Nevertheless, our current
 L.A. Esposito et al. / Molecular Phylogenetics and Evolution 93 (2015) 107–117 113

Fig. 4. Majority rule consensus phylogram of the all taxon dataset including 176 specimens. Here specimens from the island of Vieques and from several epigean habitats—for
which we only have COI sequences—have been added to the core matrix. The addition of character-sparse specimens alter some of the fundamental relationships at the base
of the phylogeny and must be considered tentative. However, these added specimens clearly show that (1) Vieques contains a unique and highly genetically distinct lineage of
Phrynus, and (2) epigean specimens from Mato de Platano in some cases nest within cave clades suggesting movement of individuals and/or genes among at least some of the
epigean/cave dyads. Circled numbers identify putative species level clades that were not sampled in the main analysis (Fig. 2).

data are too limited to fully resolve these questions and future 4.2. Genetic structure and diversity at three spatial scales
work will emphasize use of next generation sequencing to obtain
a broad spectrum of nuclear markers necessary for a conclusive The phylogenetic lineages that are resolved are generally allo-
test of these hypotheses. patric at scales of islands, zoogeographical regions and caves. The
 114
Table 1
Genetic distances among and within the 13 population/species level clades highlighted in Fig. 2 (Column 1 in parenthesis are the number of individuals in each clade, asterisks indicate epigean samples included in clade, double asterisks
indicate an epigean singleton sample). Identified monophyletic clades were tested for species/population distinctiveness as measured by the Geneious species delimitation plugin and the genealogical sorting index (gsi).

L.A. Esposito et al. / Molecular Phylogenetics and Evolution 93 (2015) 107–117

Putative species/population Monophyletic? Intra Dist K2P Intra/ P ID(Strict) P ID(Liberal) Av(MRCA- P(RD) P(AB) gsi P-value Average lat Average long
Inter tips)
Species delimitation results
1: P. eucharis clade 1 yes 0.229 4.374 0.05 0.95 (0.87, 1.0) 0.99 (0.94, 1.0) 0.216 0.99 2.00E 11 0.861 1.00E 04 18.392614 68.444385
2: P. eucharis clade 2⁄ yes 0.623 4.138 0.15 0.94 (0.89, 0.99) 0.98 (0.95, 1.0) 0.6507 0.05 5.70E 13 1 1.00E 04 18.521095 67.786054
3: P. alejandroi South Guanica⁄ yes 0.457 4.138 0.11 0.90 (0.79, 1.0) 0.97 (0.91, 1.0) 0.3352 0.05 1.80E 07 0.597 1.00E 04 17.921554 66.691763
4: P. alejandroi North Guanica⁄ yes 1.059 5.584 0.19 0.91 (0.85, 0.98) 0.97 (0.93, 1.0) 1.5918 0.05 1.80E 07 1 1.00E 04 17.971102 66.868821
5: P. longipes Cambalache⁄⁄ yes N/A 0.26 N/A N/A 0.96 (0.83, 1.0) N/A NA 1.30E 04 NA NA 18.452574 66.597221
6: P. longipes Florida yes 0.033 0.26 0.13 0.95 (0.90, 1.0) 0.98 (0.96, 1.0) 0.0174 0.05 3.90E 25 1 1.00E 04 18.346764 66.542715
7: P. longipes Hispaniola yes 1.71 5.031 0.34 0.80 (0.72, 0.89) 0.94 (0.88, 0.99) 1.525 0.05 1.00E 11 1 1.00E 04 18.283148 68.64057
8: P. longipes Guajonales⁄ yes 0.101 1.333 0.08 0.96 (0.89, 1.0) 0.99 (0.95, 1.0) 0.0728 0.05 2.40E 11 1 1.00E 04 18.07234 65.866262
9: P. longipes Aguas Buenas + El Yunque⁄ yes 0.204 0.738 0.28 0.83 (0.75, 0.92) 0.94 (0.89, 1.00) 0.208 0.15 7.90E 08 1 1.00E 04 18.288624 65.971759
10: P. longipes Mapancha yes 0.037 0.738 0.05 0.97 (0.90, 1.0) 0.99 (0.95, 1.0) 0.0243 0.05 7.90E 08 1 1.00E 04 18.0275 66.549764
11: P. longipes Morovis yes 0.05 0.427 0.12 0.95 (0.90, 1.0) 0.98 (0.96, 1.0) 0.0846 0.05 4.40E 17 1 1.00E 04 18.345404 66.449981
12: P. longipes Matas⁄ yes 0.081 0.373 0.22 0.92 (0.87, 0.97) 0.97 (0.95, 1.0) 0.0433 0.05 4.40E 17 1 1.00E 04 18.407288 66.739994
13: P. longipes Quebradillas⁄⁄ yes N/A 0.528 N/A N/A 0.96 (0.83, 1.0) N/A NA 9.20E 04 NA NA 18.474182 66.949052
IBD
r For {null r P 0}, p< For {null r 6 0},
p>
Analysis 1: raw data 0.077487 p = 0.7479. 7479 randomizations p = 0.2521
below, 0 tied, 2521 above
Analysis 2: log(geo. dist.) 0.006533 p = 0.5515. 5515 randomizations p = 0.4485
below, 0 tied, 4485 above
Analysis 3: log(genetic dist.) 0.055934 p = 0.4180. 4180 randomizations p = 0.5820
below, 0 tied, 5820 above
Analysis 4: log(both) 0.102233 p = 0.2858. 2858 randomizations p = 0.7142
below, 0 tied, 7142 above
 L.A. Esposito et al. / Molecular Phylogenetics and Evolution 93 (2015) 107–117
relationships among lineages show several consistent biogeo-
graphical patterns. Each tight genetic cluster, that could be a
genetic population or species, is an island endemic and these in
turn form endemic ‘cave island’ lineages. The relationships within
and among lineages in Puerto Rico reflect zoogeographic regions.
Moreover, the sampling we have, though not yet covering the
entire Caribbean, is sufficient to suggest relationships between
taxa on Puerto Rico, Hispaniola and Isla de Mona consistent with
those predicted by historical connections between these islands
(Iturralde-Vinent and MacPhee, 1999). Geological records, within
the framework of the GAARlandia hypothesis, suggest these land-
masses were connected up until approximately 8 mya
(Iturralde-Vinent and MacPhee, 1999), and based on biological evi-
dence from a range of taxa, Isla de Mona seems to have shared a
more recent connection with Hispaniola than it did with Puerto
Rico (Franz, 2010; Franz et al., 2009; Iturralde-Vinent, 1998;
Malone et al., 2000; Yamaguti, 2011). The tight relationship
between P. eucharis from Isla de Mona and Hispaniola similarly
suggests either recent colonization from DR to Isla de Mona, or
an historical contiguous connection between these two landmass.
The islands of Caja de Muertos and Vieques were likely submerged
until relatively recently, but nonetheless also contain endemic
forms. This indicates that short distance over-water dispersal does
episodically occur in this lineage.
The deepest genetic divergences among lineages are consistent
with isolation by oceanic barriers on one hand and by geological
barriers on the other. In particular, the three major lineages within
Puerto Rico are allopatric and found on each side of the geological
interface between karst and igneous bedrock dividing the island,
effectively, into three zoogeographic zones (Fig. 3). Phrynus alejan-
droi is found in the southern Puerto Rican limestone, the P. longipes
clade occurs in the northern limestone (Quaternary, Miocene), and
the central P. longipes clade is distributed on metamorphic and
igneous rock (Eocene, Paleocene, Cretaceous) covering most of cen-
tral Puerto Rico. These three major clades within Puerto Rico are
furthermore subdivided and each subclade corresponds to a single
geographic area or cave system (Figs. 1–3). Worldwide, the two
prevailing types of caves are karst limestone caves and igneous
caves (Christman and Culver, 2001). Despite rather significant
morphological differences between limestone and igneous caves,
prior to this study no patterns have been observed between bed-
rock of the study area and the species composition of trogloditic
arthropods (Caccone, 1985; Howarth, 1996a). However, arachnids
have been known to show substrate preference in epigean habitats
(Prendini, 2001; Kraus and Morse, 2005; Novak et al., 2009).
Furthermore, difference in patterns of rainfall and forest cover
may contribute to the distribution of these lineages, but more
detailed sampling is necessary to assess these ideas.
The striking patterns we found of cave monophyly, or regional
monophyly including cave and epigean specimens, are consistent
with reduced gene flow among caves that are within close proxim-
ity. At the finest scale this is represented by the Aguas Buenas
oscuro and claro, caves that are adjacent and partially intercon-
nected through a small passage. Also, in Guanica, a balanced sam-
pling of cave and epigean P. alejandroi still found evidence for cave
monophyly (Fig. 1). This pattern is consistent with the potential for
an ‘island within island’ turbo engine of diversification where
caves offer additional opportunities for formation of genetic varia-
tion, layered upon the opportunities offered by the islands on
which they are found. Nevertheless, preliminary data from Mata
de Platano indicate that at least in some systems animals may
move between epigean and cave habitats (full dataset, Fig. 4).
Further and more balanced sampling of neighboring cave and epi-
gean areas is necessary to gain a deeper understanding of exchange
of genes and individuals among these habitats. Cave ecosystems
can exert strong selection pressures for traits advantageous in
115
subterranean environment, which combined with limited dispersal
among caves may expedite the rate of population genetic diver-
gence compared to epigean taxa (Juan et al., 2010). At this point,
however, we cannot evaluate if Phrynus has diversified more on
the islands than on the surrounding mainland, as a similar survey
and analysis of mainland taxa would be necessary for comparison.
5. Conclusions
We perform the first phylogenetic analysis of the order
Amblypygi in the Caribbean with a focus on the role of caves as
potential drivers of diversification. We recover deep mtDNA diver-
gences among and within nominal species, genetic variation that is
strongly geographically structured among islands, geological for-
mations, and notably among caves. We conclude that the diversity
of Caribbean Phrynus is likely dramatically underestimated by cur-
rent taxonomy. With this early assessment, we predict the patterns
in Phrynus will be replicated in other cave lineages in the region –
the ‘islands within islands’ hypothesis of general increase in diver-
sification rates in such systems. If this hypothesis is supported, the
‘cave islands’ of the Caribbean likely harbor major underappreci-
ated pockets of biodiversity as they number in the thousands and
are generally unexplored. This study is among the first to system-
atically examine genetic divergences among Caribbean caves for
any arthropods. Given that caves contain many additional lineages,
including several spider genera, cockroaches, pseudoscorpions,
crickets, and a variety of microorganisms associated with those,
arthropods in caves may represent an undocumented dimension
of diversity within the Caribbean biodiversity hotspot.
Acknowledgments
Special thanks to the Sociedad Espeleológica de Puerto Rico and
Sociedad Espeleológica Unida del Sur for assistance with fieldwork
and various logistics. Also thanks to Departamento de Recursos
Naturales for arranging permits and to Isla de Mona park rangers,
and especially Tony Nieves for valuable help in the field. Additional
logistic support was provided by Fideicomiso de Conservación de
Puerto Rico, Universidad Interamericana de Puerto Rico, and Casa
Verde, Maunabo. Kenneth J. Chapin kindly provided COI sequences
of several specimens from Puerto Rico. Yadira Ortiz-Ruiz helped
with DNA extraction and PCRs of a portion of the specimens here
included. We are grateful to Dr Alfried Vogler and two anonymous
reviewers for comments that helped improve the manuscript, and
to Marshal Hedin for helpful comments on an earlier version of the
manuscript. This research is supported by NSF DEB-1050187,
1050253, 1314749 to IA and GB, NSF DBI-1349205 to D.
Barrington, I. Agnarsson and CW Kilpatrick, and NSF
DBI-1003087 to LE. This work represents the collective effort of
team CarBio and we are thankful to all the participants in this
project.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ympev.2015.07.
005.
References
Agnarsson, I., Rayor, L.S., 2013. A molecular phylogeny of the Australian huntsman
spiders (Sparassidae, Deleninae): implications for taxonomy and social
behaviour. Mol. Phylogenet. Evol. 69, 895–905.
Allegrucci, G., Trucchi, E., Sbordoni, V., 2011. Tempo and mode of species
diversification in Dolichopoda cave crickets (Orthoptera, Rhaphidophoridae).
Mol. Phylogenet. Evol. 60, 108–121.
116 L.A. Esposito et al. / Molecular Phylogenetics and Evolution 93 (2015) 107–117
Allen, A.P., Savage, V.M., 2007. Setting the absolute tempo of biodiversity dynamics.
Ecol. Lett. 10, 637–646.
Avise, J.C., Hubbell, S.P., Ayala, F.J., 2008. In the light of evolution II: biodiversity and
extinction. Proc. Natl. Acad. Sci. U.S.A. 105, 11453–11457.
Bidegaray-Batista, L., Taiti, S., Lopez, H., Ribera, C., Arnedo, M.A., 2015. Endemism
and evolution in the littoral woodlouse Halophiloscia Verhoeff, 1908
(Crustacea, Isopoda, Oniscidea) from the Canary Islands: implications for
conservation policies. Insect Conserv. Divers. 8, 17–30.
Bloom, T., Binford, G., Alayon, G., Esposito, L., Peterson, I., Nishida, A., Loubet-Senear,
K., Agnarsson, I., 2014. Discovery of two new species of eyeless spiders within a
single Hispaniola cave. J. Arachnol. 42, 148–154.
Briggs, R.P., Akers, J.P., 1965. Hydrogeologic map of Puerto Rico and adjacent
islands. U.S. Geological Survey Hydrologic Investigations Atlas HA-197.
Caccone, A., 1985. Gene flow in cave arthropods: a qualitative and quantitative
approach. Evolution (Lawrence) 39, 1223–1235.
Cardoso, P., 2012. Diversity and community assembly patterns of epigean vs.
troglobiont spiders in the Iberian Peninsula. Int. J. Speleol. 41, 83–94.
Chiari, Y., van der Meijden, A., Mucedda, M., Lourenço, J., Hochkirch, A., Veith, M.,
2012. Phylogeography of Sardinian cave salamanders (Genus Hydromantes) is
mainly determined by geomorphology. PLoS One 7 (3), e32332.
Christman, M., Culver, D., 2001. The relationship between cave biodiversity and
available habitat. J. Biogeogr. 28, 367–380.
Colwell, R.K., Coddington, J.A., 1994. Estimating terrestrial biodiversity through
extrapolation. Philos. Trans. Roy. Soc. Lond. Ser. B – Biol. Sci. 345, 101–118.
Culver, D., Deharveng, L., Bedos, A., Lewis, J., Madden, M., Reddell, J., Sket, B.,
Trontelj, P., White, D., 2006. The mid-latitude biodiversity ridge in terrestrial
cave fauna. Ecography, 29.
Cummings, M.P., Neel, M.C., Shaw, K.L., 2008. A genealogical approach to
quantifying lineage divergence. Evolution 62, 2411–2422.
Darriba, D., Taboada, G.L., Doallo, R., Posada, D., 2012. JModelTest 2: more models,
new heuristics and parallel computing. Nat. Meth. 9, 772.
Darwin, C., 1859. On the Origin of Species by means of Natural Selection, or the
Preservation of Favoured Races in the Struggle for Life, first ed. John Murray,
London.
de Armas, L.F., 2004. Arachnids of the Dominican Republic. Palpigradi, Schizomida,
Solifugae, and Thelyphonida (Chelicerata: Arachnida). Rev. Iber. Aracnol.
Volumen Especial Monografico 2, 1–63.
de Armas, L.F., 2010. New arachnids from Puerto Rico (Arachnida: Amblypygi,
Araneae, Opiliones, Parasitiformes, Schizomida, Scorpiones). Boletin de la SEA
47, 55–64.
de Armas, L.F., Angarita Arias, A.A., 2008. A new species Phrynus Lamarck, 1801
(Amblypygi: Phrynidae) from Colombia. Boletin de la SEA 43, 25–28.
de Armas, L.F., Perez Gonzalez, A., 2001. Amblypygids of the Dominican Republic
(Arachnida: Ambypygi). Rev. Iber. Aracnol. 3, 47–66.
de Armas, L.F., Teruel, R., 2010. A new species of Phrynus (Amblypygi: Phrynidae)
from Puerto Rico. Boletin de la SEA 47, 127–130.
de Armas, L.F., Viquez, C., 2001. A new species of Phrynus (Amblypygi: Phrynidae)
from Costa Rica. Rev. Iber. Aracnol. 4, 11–15.
Emerson, B.C., 2002. Evolution on oceanic islands: molecular phylogenetic
approaches to understanding pattern and process. Mol. Ecol. 11, 951–966.
Ersts, P.J., 2006. Geographic distance matrix generator (version 1.2.3.) American
Museum of Natural History. Center Biodivers. Conserv.
Etienne, R.S., Apol, M.E.F., Olff, H., Weissing, F.J., 2007. Modes of speciation and the
neutral theory of biodiversity. Oikos 116, 241–258.
Frank, E., Wicks, C., Mylroie, J., Troester, J., Alexander, E.j., Carew, J., 1998. Geology of
Isla de Mona, Puerto Rico. J. Cave Karst Stud. 60, 69–72.
Franz, N.M., 2010. Revision and phylogeny of the Caribbean weevil genus
Apotomoderes Dejean, 1834 (Coleoptera, Curculionidae, Entiminae). ZooKeys,
33–75.
Franz, N.M., O’Brien, C.W., Ruiz Nunez, D., 2009. New records of weevils
(Coleoptera: Curculionoidea) from Mona Island, Puerto Rico. Solenodon 8, 82–
98.
Futuyma, D.J., 2009. Evolution, second ed. Sinauer Associates, Sunderland,
Massachusetts.
Gavrilets, S., 2003. Perspective: models of speciation: what have we learned in 40
years? Evolution 57, 2197–2215.
Gillespie, R., Clauge, D., 2009. Encyclopedia of Islands. University of California Press,
Berkeley.
Gillespie, R.G., 2004. Community assembly through adaptive radiation in Hawaiian
spiders. Science 303, 356–359.
Gillespie, R.G., Claridge, E.M., Roderick, G.K., 2008. Biodiversity dynamics in isolated
island communities: interaction between natural and human-mediated
processes. Mol. Ecol. 17, 45–57.
Green, P., 1999. PHRAP. Available at <http://phrap.org/>.
Green, P., Ewing, B., 2002. PHRED. Available at <http://phrap.org/>.
Harvey, M.S., 2002. The first old world species of Phrynidae (Amblypygi): Phrynus
exsul from Indonesia. J. Arachnol. 30, 470–474.
Harvey, M.S., 2003. Catalogue of the smaller arachnid orders of the world:
Amblypygi, Uropygi, Schizomida, Palpigradi, Ricinulei and Solifugae.
Hebert, P.D.N., Cywinska, A., Ball, S.L., DeWaard, J.R., 2003. Biological identifications
through DNA barcodes. Proc. Roy. Soc. Lond. Ser. B – Biol. Sci. 270, 313–321.
Hedin, M., 2015. High stakes species delimitation in eyeless cave spiders (Cicurina,
Dictynidae, Araneae) from central Texas. Mol. Ecol. 24, 346–361.
Hedin, M., Thomas, S.M., 2010. Molecular systematics of eastern North American
Phalangodidae (Arachnida: Opiliones: Laniatores), demonstrating convergent
morphological evolution in caves. Mol. Phylogenet. Evol. 54, 107–121.
Hedin, M.C., 1997. Molecular phylogenetics at the population/species interface in
cave spiders of the southern Appalachians (Araneae: Nesticidae: Nesticus). Mol.
Biol. Evol. 14, 309–324.
Howarth, F., 1996. A comparison of volcanic and karstic cave communities. In:
Oromi, P. (Ed.), Proceedings of the Seventh International Symposium of
Vulcanspeleology, Canary Is., November 1994, S.A., Barcelona, pp. 63–68.
Howarth, F., 1996. A comparison of volcanic and karstic cave communities. In:
Oromi, P. (Ed.), Proceedings of Seventh International Symposiom of
Vulcanspeleology, Canary Is., November 1994, S.A., Barcelona, pp. 63–68.
Huelsenbeck, J.P., Ronquist, R., 2001. MRBAYES: Bayesian inference of phylogenetic
trees. Bioinformatics 17, 754–755.
Iturralde-Vinent, M., 1998. Proposed IGCP project on Caribbean plate tectonics.
Episodes 21, 201.
Iturralde-Vinent, M., MacPhee, R., 1999. Paleogeography of the Caribbean region:
implications for cenozoic biogeography. Bull. Am. Mus. Nat. Hist. 238, 1–95.
Jensen, J.L., Bohonak, A.J., Kelley, S.T., 2005. Isolation by distance, web service. BMC
Genet. 6.
Juan, C., Emerson, B.C., 2010. Evolution underground: shedding light on the
diversification of subterranean insects. J. Biol. (Lond.) 9, 17. Article No.: 17.
Juan, C., Guzik, M.T., Jaume, D., Cooper, S.J.B., 2010. Evolution in caves: Darwin’s
’wrecks of ancient life’ in the molecular era. Mol. Ecol. 19, 3865–3880.
Katoh, S., 2013. MAFFT multiple sequence alignment software version 7:
improvements in performance and usability. Mol. Biol. Evol. 30, 772–780.
Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton, S.,
Cooper, A., Markowitz, S., Duran, C., Thierer, T., Ashton, B., Meintjes, P.,
Drummond, A., 2012. Geneious Basic: an integrated and extendable desktop
software platform for the organization and analysis of sequence data.
Bioinformatics 28, 1647–1649.
Kimura, M., 1980. A simple method for estimating evolutionary rate of base
substitutions through comparative studies of nucleotide sequences. J. Mol. Evol.
16, 111–120.
Kraus, J.-M., Morse, D.-H., 2005. Seasonal habitat shift in an intertidal wolf spider:
proximal cues associated with migration and substrate preference. J. Arachnol.
33 (1), 110–123, 110 123 3.
Kuntner, M., Arnedo, M.A., Trontelj, P., Lokovsek, T., Agnarsson, I., 2013. A molecular
phylogeny of nephilid spiders: evolutionary history of a model lineage. Mol.
Phylogenet. Evol. 69, 961–979.
Losos, J.B., Jackman, T.R., Larson, A., de Queiroz, K., Rodriguez-Schettino, L., 1998.
Contingency and determinism in replicated adaptive radiations of island
lizards. Science 279, 2115–2118.
Maddison, D.R., Maddison, W.P., 2011. Chromaseq: a Mesquite module for
analyzing sequence chromatograms, version 1.0.
Maddison, W., Maddison, D., 2011. Mesquite: a modular system for evolutionary
analysis. <http://mesquiteproject.org>.
Malone, C.L., Wheeler, T., Taylor, J.F., Davis, S.K., 2000. Phylogeography of the
Caribbean rock iguana (Cyclura): implications for conservation and insights on
the biogeographic history of the West Indies. Mol. Phylogenet. Evol. 17, 269–
279.
Masters, B.C., Fan, V., Ross, H.A., 2011. Species delimitation – a geneious plugin for
the exploration of species boundaries. Mol. Ecol. Resour. 11, 154–157.
McHugh, A., Yablonsky, C., Binford, G., Agnarsson, I., 2014. Molecular
phylogenetics of Caribbean Micrathena (Araneae: Araneidae) suggests
multiple colonisation events and single island endemism. Invertebr. Syst.
28, 337–349.
Miller, M., Pfeiffer, W., Schwartz, T., 2010. Creating the CIPRES Science Gateway for
inference of large phylogenetic trees. In: Proceedings of the Gateway
Computing Environments Workshop (GCE), 14 November 2010, pp. 1–8.
Novak, T., Klokocovnik, V., Delakorda, S., Devetak, D., Janzelpvoc, F., 2009.
Preferences for different substrates in Phalangium opilio (Opiliones:
Phalangiidae) in natural environment. Acta Biol. Slovenica 52, 29–35.
Nylander, J., Wilgenbusch, J., Warren, D., Swofford, D., 2008. AWTY (are we there
yet?): a system for graphical exploration of MCMC convergence in Bayesian
phylogenetics. Bioinformatics 24, 581–583.
Opatova, V., Arnedo, M.A., 2014. Spiders on a hot volcanic roof: colonisation
pathways and phylogeography of the Canary Islands Endemic Trap-Door Spider
Titanidiops canariensis (Araneae, Idiopidae). PLoS One 9.
Poinar Jr., G., Brown, A.E., 2004. A new whip spider (Arachnida: Amblypygi),
Phrynus mexicana, is described from Mexican amber. Beitraege zur Araneol. 3B,
1881–1885.
Porter, M., 2007. Subterranean biogeography: what have we learned from
molecular techniques? J. Cave Karst Stud. 69, 179–186.
Prendini, L., 2001. Substratum specialization and speciation in southern African
scorpions: the Effect Hypothesis revisited. In: Fet, V., Selden, P. (Eds.), Scorpions
2001. In Memoriam Gary A. Polis. British Arachnological Society, Burnham
Beeches, Bucks, UK, pp. 113–138.
Prendini, L., Weygoldt, P., Wheeler, W.C., 2005. Systematics of the Damon
variegatus group of African whip spiders (Chelicerata: Amblypygi):
evidence from behaviour, morphology and DNA. Org. Divers. Evol. 5,
203–236.
Quintero, D., 1981. The amblypygid genus Phrynus in the Americas (Amblypygi,
Phrynidae). J. Arachnol. 9, 117–166.
Rahmadi, C., Harvey, M.S., 2007. The female of Phrynus exsul (Amblypygi,
Phrynidae) from Indonesia. J. Arachnol. 35, 137–142.
Reno, M.L., Held, N.L., Fields, C.J., Burke, P.V., Whitaker, R.J., 2009. Biogeography of
the Sulfolobus islandicus pan-genome. Proc. Natl. Acad. Sci. U.S.A. 106, 8605–
8610.
 L.A. Esposito et al. / Molecular Phylogenetics and Evolution 93 (2015) 107–117
Ricklefs, R.E., 2006. The unified neutral theory of biodiversity: do the numbers add
up? Ecology 87, 1424–1431.
Rodrigo, A., Bertels, F., Heled, J., Noder, R., Shearman, H., Tsai, P., 2008. The perils of
plenty: what are we going to do with all these genes? Philos. Trans. Roy. Soc. B –
Biol. Sci. 363, 3893–3902.
Ronquist, F., Huelsenbeck, J.P., 2003. MrBayes 3: Bayesian phylogenetic inference
under mixed models. Bioinformatics 19, 1572–1574.
Rosenberg, N.A., 2007. Statistical tests for taxonomic distinctiveness from
observations of monophyly. Evolution 61, 317–323.
Ross, H.A., Murugan, S., Li, W.L.S., 2008. Testing the reliability of genetic methods of
species identification via simulation. Syst. Biol. 57, 216–230.
Ruch, J., Riehl, T., May-Collado, L.J., Agnarsson, I., 2015. Multiple origins of
subsociality in crab spiders (Thomisidae). Mol. Phylogenet. Evol. 82, 330–340.
Schluter, D., 1996. Ecological causes of adaptive radiation. Am. Nat. 148, S40–S64.
Schluter, D., 2000. The Ecology of Adaptive Radiation. Oxford University Press, New
York.
Snowman, C., Zigler, K., Hedin, M., 2010. Caves as islands: mitochondrial
phylogeography of cave-obligate spider species Nesticus barri (Araneae:
Nesticidae). J. Arachnol. 38, 49–56.
Stamatakis, A., 2006. RAxML-VI-HPC: maximum likelihood-based phylogenetic
analyses with thousands of taxa and mixed models. Bioinformatics 22, 2688–
2690.
117
Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2011. MEGA5:
molecular evolutionary genetics analysis using maximum likelihood,
evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28,
2731–2739.
Turelli, M., Barton, N.H., Coyne, J.A., 2001. Theory and speciation. Trends Ecol. Evol.
16, 330–343.
Turner, J.R.G., 2004. Explaining the global biodiversity gradient: energy, area,
history and natural selection. Basic Appl. Ecol. 5, 435–448.
Wallace, A.R., 1876. The geographical distribution of animals: with a study of the
relations of living and extinct faunas as elucidating the past changes of the
earth’s surface. Harper and brothers.
Weygoldt, P., 2000. Whip spiders (Chelicerata: Amblypygi): their biology,
morphology and systematics.
Whittaker, R.J., Willis, K.J., Field, R., 2001. Scale and species richness: towards a
general, hierarchical theory of species diversity. J. Biogeogr. 28, 453–470.
Wiley, E.O., 1978. Evolutionary species concept reconsidered. Syst. Zool. 27, 17–26.
Yamaguti, H., 2011. Análise filogenética e biogeográfica do gênero Rhopalurus
Thorell, 1876 (Arachnida: Scorpiones. Instituto de Biociências. Université de São
Paulo, São Paulo.
Zhang, J., Kapli, P., Pavlidis, P., Stamatakis, A., 2013. A general species delimitation
method with applications to phylogenetic placements. Bioinformatics 29 (22),
2869–2876.