Syst. Biol. 66(5):663–697, 2017
© The Author(s) 2017. Published by Oxford University Press, on behalf of the Society of Systematic Biologists. All rights reserved .
For Permissions, please email: journals.permissions@oup.com
DOI:10.1093/ sysbio/ syx029
Advance Access publication February 27, 2017
A Phylogenetic, Biogeographic, and Taxonomic study of all Extan t Species of Anolis
(Squamata; Iguanidae)
STEVEN POE1,∗, ADRIÁN NIETO-MONTES DE OCA2, OMAR TORRES-CARVAJAL3, KEVIN DE QUEIROZ4, JULIÁN A. VELASCO5,
BRAD TRUETT1, LEVI N. GRAY1, MASON J. RYAN1, GUNTHER KÖHLER6, FERNANDO AYALA-VARELA3, AND IAN LATELLA1
1Department of Biology and Museum of Southwestern Biology, University of New Mexico, Albuquerque, NM 87131, USA; 2Departamento de Biología
Evolutiva, Facultad de Ciencias, Universidad Nacional Autónoma deMéxico, México, D.F. 04510, México; 3Museo de Zoología, Escuela de Biología,
Pontificia Universidad Católica del Ecuador, Avenida 12 de Octubre y Roca, Apartado 17-01-2184, Quito, Ecuador; 4Smithsonian Institution, National
Museum of Natural History, Washington DC 20560-0162, USA; 5Laboratorio de Análisis Espaciales, Instituto de Biologia, Universidad Nacional Autónoma
deMéxico, México, D.F., México & Grupo de Investigación en Ecología Animal, Departamento de Biología, Facultad de Ciencias Naturales y Exactas,
Universidad del Valle, Cali-Colombia; 6Forschungsinstitut und Naturmuseum Senckenberg, Senckenberganlage 25, 60325 Frankfurt a.M., Germany
∗Correspondence to be sent to: Department of Biology and Museum of Southwestern Biology, University of New Mexico, Albuquerque, NM 87131, USA;
E-mail:anolis@unm.edu
Received 21 December 2015; reviews returned 9 October 2016; accepted 21 January 2017
Associate Editor: Robb Brumfield
Abstract.——Anolis lizards (anoles) are textbook study organisms in evolution and ecology. Although several topics in
evolutionary biology have been elucidated by the study of anoles, progress in some areas has been hampered by limited
phylogenetic information on this group. Here, we present a phylogenetic analysis of all 379 extant species of Anolis, with
new phylogenetic data for 139 species including new DNA data for 101 species. We use the resulting estimates as a basis
for defining anole clade names under the principles of phylogenetic nomenclature and to examine the biogeographic
history of anoles. Our new taxonomic treatment achieves the supposed advantages of recent subdivisions of anoles
that employed ranked Linnaean-based nomenclature while avoid ing the pitfalls of those approaches regard ing artificial
constraints imposed by ranks. Our biogeographic analyses demonstrate complexity in the dispersal history of anoles,
including multiple crossings of the Isthmus of Panama, two invasions of the Caribbean, single invasions to Jamaica and Cuba,
and a single evolutionary dispersal from the Caribbean to the mainland that resulted in substantial anole d iversity. Our
comprehensive phylogenetic estimate of anoles should prove useful for rigorous testing of many comparative evolutionary
hypotheses. [Anoles; biogeography; lizards; Neotropics; phylogeny; taxonomy.]
Anolis is a well-stud ied , ecologically d iverse, species-
rich clade of Neotropical lizards. Anatomically, Anolis
lizards (anoles) are characterized by expanded toepads
that facilitate an arboreal lifestyle and a throat fan,
or dewlap, used mainly in intraspecific signaling.
Anoles occupy a diverse range of microhabitats with
most species living on trees, bushes, or grasses, but
some specializing on rocks, streams, or leaf litter.
Communities of anoles range from up to 12 sympatric
species (e.g., at Parque Omar Torrijos in Panama; Poe
2012) to solitary species. Behaviorally, all species are
d iurnal except for a few Caribbean forms that may be
nocturnally active around artificial lighting (examples
in Schwartz and Henderson 1991). The over 379 species
of Anolis (see below) natively range from Florida south
through Central America and the Caribbean to Bolivia,
with naturalized populations as far as Asia.
Anolis lizards are model study organisms in ecology
and evolution. They have been subjects of classic
stud ies of community ecology (e.g., Williams 1983),
ecomorphology (e.g., Collette 1961), communication
(e.g., Rand and Williams 1970), character d isplacement
(e.g., Schoener 1970), biogeography (e.g., Lazell 1972),
adaptive rad iation (e.g.,Williams 1972), and competition
(e.g., Pacala and Roughgarden 1982), to name just a few
textbook examples. Recent authors have incorporated
comparative methods and anole phylogeny into studies
of these and other important topics in evolution and
ecology (e.g., Losos et al. 1998;Nicholson et al. 2005;Ord
and Martins 2006). Many evolutionary stud ies of Anolis
would have benefited from better phylogenies based
on more comprehensive taxon sampling, particularly
of mainland forms, and attempts at a comprehensive
taxonomy have also been hampered by limited sampling.
For instance, the best-sampled molecular phylogenetic
analysis of anoles to date (Gamble et al. 2014) included
216 (<57% of) species, the most recent comparison of
mainland and island evolution in Anolis (Pinto et al. 2008)
included 35 (<17% of) mainland species, and the most
recent attempt at a comprehensive taxonomy of anoles
(Nicholson et al. 2012) analyzed 240 (<63% of) species.
Etheridge’s (1959) landmark study of skeletal
morphology was the first large-scale phylogenetic
analysis of anoles. This work erected informal groups
ranked as “sections” and “series”,which were elaborated
upon (e.g., by the addition of “species groups”) in
Williams’ (1976a,b) influential taxonomic treatments
that were utilized by describers of species seeking pools
for taxonomic comparison and evolutionary biologists
seeking units for comparative study. Guyer and Savage
(1992; preceded by Guyer and Savage 1986) erected new
genera within Anolis based on a phylogenetic analysis
of 27 species. The advent of molecular data brought
reorganization of the Etheridge–Williams groups (e.g.,
Gorman 1973; Shochat and Dessauer 1981; Burnell
and Hedges 1990), as well as molecular phylogenetic
analyses of many subclades of Anolis (e.g., Gorman
et al. 1983; Hedges and Burnell 1990; Creer et al. 2001;
Schneider et al. 2001; Brandley and de Queiroz 2004;
Glor et al. 2004; Castañeda and de Queiroz 2011).
663
664 SYSTEMATIC BIOLOGY VOL. 66
Following the pioneering DNA sequence work of
Jackman et al. (1999) and the combined-data study of
Poe (2004), the most recent large-scale phylogenetic
work (Alföld i et al. 2011; Nicholson et al. 2012; Gamble
et al. 2014) has added taxa and data to build on the
Etheridge–Williams framework.
Progress in anole phylogeny sometimes has been
overshadowed by controversy regard ing the taxonomy
of anoles (e.g., Guyer and Savage 1986, 1992; Cannatella
and de Queiroz 1989; Williams 1989; Nicholson et al.
2012, 2014; Poe 2013). Disagreements in anole taxonomy
owe largely to differences among authors concerning the
clade or clades to which the Linnaean rank of genus
is to be assigned . Because trad itional nomenclature is
based on taxonomic ranks, those differences have created
major d iscrepancies in the names applied to various
anole clades despite considerable agreement regard ing
their composition and phylogenetic relationships.
Consequently, debates have tended to focus on the
scientifically meaningless question of how many genera
ought to be recognized , thus d iverting attention from
scientifically germane disagreements concerning the
relationships of anole species and the composition of
anole clades.
In an attempt to rectify this and similar counter-
productive situations in taxonomies throughout the tree
of life some systematic biologists have been developing
a tree-based approach to biological nomenclature in
which taxon names are tied explicitly and directly to
clades (e.g., de Queiroz and Gauthier 1990, 1992;Cantino
and de Queiroz 2014). By contrast, the trad itional,
rank-based system does not necessarily tie names to
clades, and even when it does, the connection is
ind irect and tenuous. The tree-focused approach also
has the advantage of producing taxonomies with higher
information content, because the named clades are not
restricted to a particular taxonomic level (in this case, the
genus). That is, instead of the names all being applied
to mutually exclusive clades (as would be the case with
genera), they can be applied to both nested and mutually
exclusive clades. Although the tree-based approach has
been adopted for some subclades of anoles (Nicholson
2002; Brandley and de Queiroz 2004; Castañeda and de
Queiroz 2013), it has not yet been applied across the
entire anole clade.
As with taxonomy, rigorous biogeographic treatments
of anoles mainly have been confined to subgroups of the
clade, with a focus on Caribbean forms (e.g., Brandley
and de Queiroz 2004; Glor et al. 2005; Klutsch et al.
2007; Rodríguez-Robles et al. 2007; see Phillips et al. 2015
for a mainland example). Larger-scale treatments have
examined general Caribbean patterns (Alföld i et al. 2011;
Helmus et al. 2014) or specific biogeographic hypotheses
(e.g., the “back-invasion” of the mainland; Nicholson
et al. 2005). The one quantitative attempt at describing
overall Anolis biogeographic history Nicholson et al.
(2012) likely suffers from gross overestimation of the
age of the Anolis clade (see Townsend et al. 2011;
Mulcahy et al. 2012; Prates et al. 2015; and below).
Nevertheless, that work erected testable hypotheses that
may be assessed with more realistic dating. Here, we
test several biogeographic hypotheses to explain the
present-day distribution of Anolis in the Neotropics. In
particular, we examine the following historical events:
timing and ancestral area of the most recent common
ancestor of anoles; timing and frequency of transitions
of anole lineages between areas (including mainland and
islands, and among islands); timing of biotic exchange
of anole lineages between Middle America and South
America; existence of “sources” or “sinks” for anole
d iversity.
The goals of this work are to estimate the phylogeny of
all 379 species of Anolis and use this estimate to describe
the biogeographic history of the clade and erect a new
phylogenetic taxonomy of anoles.
MATERIALS AND METHODS
Taxon Sampling
We endeavored to include all valid species
of Anolis as of 1 June 2014 in our analysis.
Supplementary Appendix S1 (available on Dryad
at http:/ / dx.doi.org/ 10.5061/ dryad .s80jq) lists our
judgements of species status for forms included here;
this approach resulted in 379 species included for
phylogenetic analysis. We included the following
outgroups: Basiliscus plumifrons, Polychrus marmoratus,
Pristidactylus scapulatus, Urostrophus gallardoi. These
were selected based on maximizing available data for
close relatives of Anolis (e.g., Pyron et al. 2013).
Data
We obtained DNA sequence data (varying coverage
of mitochondrial genes ND2 and COI and the nuclear
exon that codes for endothelin-converting enzyme-like
1 [ECEL1]) for 101 species of Anolis not previously
scored for DNA and combined these with published
DNA data (e.g., Jackman et al. 1999; Castañeda and
de Queiroz 2011; Alföld i et al. 2011) to produce a
matrix with varying taxonomic coverage of 24,879 sites
across 50 loci for 317 species. Appendix 1 shows gene
coverage for each species. Supplementary Appendix
S2 gives specimen vouchers and genes for DNA data
new to this article. Sanger sequencing was done in
the labs of SP, AN, GK, and OT, and by the Barcode
of Life Initiative (www.barcodeoflife.com). Alignments
of our newly generated sequence data (ECEL1, ND2,
COI) were performed using Muscle in Mega (Tamura
et al. 2011) and checked and improved with reference
to codon position, previous alignments of these genes
in Anolis (e.g., Jackman et al. 1999), and the published
Anolis genome (Alföld i et al. 2011). Alföld i et al.’s (2011)
alignment was used for their data, except that we aligned
2017 POE ET AL.—EVOLUTION OF ANOLES 665
16S ourselves after adding additional sequences from
Genbank.
We collected new morphological data for 144 species
not previously scored for morphology and combined
these with published data to produce a morphological
phylogenetic matrix of 46 characters (Appendix 2) for
all 379 species of Anolis. Supplemental Appendix S3
describes our codings for species for which we were
unable to examine specimens.
Phylogenetics
Phylogenetic matrices such as ours that include
large numbers of terminals and diverse kinds of data
are not straightforward to analyze. In particular,
the intent to integrate ordered and unordered
multistate morphological data with GTR-modeled
DNA data greatly restricts the available approaches
and computer programs for analysis. Here, we use
Bayesian phylogenetic analysis of our combined matrix
implemented in MrBayes version 3.2.6 (Huelsenbeck
and Ronquist 2001; Ronquist et al. 2012; Suchard and
Huelsenbeck 2012). This approach allows integration of
complex morphological datasets with model-averaging
of GTR-class models for DNA datasets (Huelsenbeck
et al. 2004).
We used Partitionfinder (Lanfear et al. 2012) to select
an optimal partitioning scheme for the DNA data
accord ing to the Bayesian Information Criterion under
Partitionfinder’s “greedy” algorithm. We hypothesized
separate models for each codon position for the well-
sampled mitochondrial genes (ND2, COI) and for
each entire gene for the 48 other analyzed genes.
Because MrBayes allows model-averaging across the
entire GTR model space (“nst=mixed”), there is little
reason to designate particular GTR-class models in
comparison of partitioning schemes. Therefore, we
compared only GTR versus GTR+G models and
ignored GTR-class submodels (e.g., HKY, F81) for each
partition in Partitionfinder. We agree with previous
authors (e.g., Stamatakis 2006; Moyle et al. 2012)
that the assumed benefit of adding an invariant-
sites parameter (i.e., “accounting for” gene regions
that cannot change) does not outweigh the potential
downsides (i.e., duplication of the function of the
rate heterogeneity parameter; parameter interaction;
overparameterization) and therefore excluded invariant-
sites models from consideration.
The 46 morphological characters (Appendix 2) were
analyzed with the Standard model for informative
characters (“coding= informative”) including 42
ordered (“ctype:ordered”) and four unordered (the
default) characters and allowing gamma-distributed
rate variation with six categories (“rates= gamma
ngammacat= 6”). Topology and branch lengths were
linked across partitions and other parameters were
unlinked .
Some of our analyses require a timetree so we
employed a relaxed-clock approach allowing rate
variation across lineages accord ing to the independent
gamma rates model (“brlenspr= clock:uniform
clockvarpr= igr”) with Urostrophus gallardoi constrained
as the outgroup in MrBayes.
We experimented extensively with MCMC parameter
settings and settled on the following strategy: two
concurrent runs of one cold and five heated chains with
heating parameter T = 0.001, for 10 million generations,
sampling every 1000 trees. We examined parameter
estimates over generations using Tracer (Rambaut and
Drummond 2007).We discarded the first 50%of sampled
trees as burnin.MrBayes analyses were performed on the
computers of the Cyberinfrastructure for Phylogenetic
Research (CIPRES) Project. We present a majority-
rule consensus of post burnin trees for taxonomic
conclusions, and use two fully resolved trees selected
from the post burnin sample for biogeographic and
dating analyses: a maximum clade cred ibility tree
(hereafter, MCC tree; Rambaut et al. 2014) and the tree
with the minimal symmetric d istance (Robinson and
Foulds 1981) from the 50% majority rule consensus
tree (hereafter, MRC tree). The topology of the MRC
tree was also analyzed with BEAST (Drummond et al.
2012) to produce a third fully resolved tree for analysis
(see below). The MrBayes NEXUS file of DNA and
morphological data are in Supplemental Appendix S4.
Biogeography
Divergence times—Divergence-time estimates were
generated using a Bayesian approach in BEAST v. 1.8.1
(Drummond et al. 2012). We fixed the tree topology as
the MRC tree and pruned species not scored for at least
one gene (i.e., species scored only for morphology).
We applied an uncorrelated log-normal relaxed-
clock model to the DNA data using the same DNA
partitioning scheme discussed above and two fossil
calibrations. The root of our tree was calibrated with
the crown group pleurodont iguanian Saichangurvel
(Conrad et al. 2007) from the late Campanian (70.6
+ - 0.6 Mya) (Townsend et al. 2011). This fossil was
used by Townsend et al. (2011) to constrain the crown
of the Pleurodonta clade of the iguanian tree. We
assigned this fossil point calibration to the root of our
tree (anoles + outgroups) using a uniform distribution
prior (70–72 mya). The second fossil calibration point
was located in the most recent common ancestor
(MRCA) of the Anolis chlorocyanus group (A. aliniger, A.
chlorocyanus, A. coelestinus, A. singularis). We used a
Dominican amber anole fossil assigned to this group (de
Queiroz et al. 1998) with a minimum age of 23 mya.
We also used a uniform prior d istribution for this
node based on stratigraphic information from the fossil
(17–23 mya). Both fossil calibration points used in this
study were placed conservatively at the crown of each
clade. Analyses were performed on the CIPRES cluster,
with two independent runs for 50 million generations
sampling every 5000. We checked log files to assure
stationarity in likelihood values and convergence using
Tracer. We used 5 million generations as a burn-in
period and generated a maximum clade cred ibility time
666 SYSTEMATIC BIOLOGY VOL. 66
tree (hereafter, MRCT tree). We summarized posterior
d ivergence date estimates for the most recent common
ancestor of anoles and for regional trees in order to
associate particular historical events (e.g., the uplift of
the Andes) with anole d ivergence times.
Biogeographic regions.—We defined a set of 14 areas
for biogeographic analyses based on the present-day
distribution patterns of Anolis lizards and the geological
history of the Middle and South American mainland
and the Caribbean region (i.e., geological barriers and
areas of endemism; Gregory-Wodzicki 2000; Losos 2011;
Coates and Obando 1996; Supplementary Fig. S1). We
follow previous workers on Middle America, South
America and the Caribbean islands (Castoe et al. 2009;
Santos et al. 2009; Antonelli et al. 2009; Daza et al. 2010)
and use the following regions in our analysis: a) Lesser
Antilles; b) Puerto Rico and satellite islands and banks;
c) Cuba and satellite islands and banks plus Cayman
islands; d) Hispaniola and satellite islands and banks;
e) Jamaica; f) the Bahamas; g) small Caribbean islands
(i.e., San Andres and Providencia islands and Swan
islands); h) Nearctic from the Isthmus of Tehuantepec
to the United States; i) Upper Central America from the
Nicaraguan depression to the Isthmus of Tehuantepec;
j) Lower Central America from the Panama Isthmus to
the Nicaraguan depression; k) South American Chocó
region encompassing Pacific lowlands from Colombia
and Ecuador; l) Caribbean region and inter-Andean
valleys in Colombia and northwestern Venezuela; m)
Andes region from Venezuela to Bolivia, above 1000 m;
n) Amazonia, including Orinoco and Amazon river
basins. We assigned each species to one or more
regions based on distributional records compiled from
several sources (e.g., Nicholson et al. 2012; Velasco et al.
2015).
Statistical biogeographic methods.—Biogeographic
analyses were performed using the BioGeoBEARS
R package (Matzke 2013b) on the MCC tree and the
MRC tree (i.e., including all taxa) and on the MRCT tree
(i.e., on the dated tree including only those taxa scored
for molecular data).
We performed two biogeographic reconstructions on
the MRC tree and the MCC tree, one focused on the
mainland and the other focused on the Caribbean.
We performed separate reconstruction analyses rather
than a single large analysis due to the computational
complexity of performing likelihood reconstructions
with a large number of areas as in this case (Matzke 2014).
Because our exploratory analyses detected a low number
of d ispersal events between the mainland and Caribbean
islands (see below), this strategy seems unlikely to have
had a major effect on ancestral range estimates. For the
analysis on the mainland areas, all Caribbean islands
were merged into two discrete regions, a) Lesser Antilles
and b) Greater Antilles plus the Bahamas and Cayman
Islands. Thus, this first analysis was conducted on nine
regions, including seven mainland and two Caribbean
regions. For the analysis focused on the Caribbean
region, all mainland areas were merged into two regions,
Middle America and South America. Thus, this second
analysis included seven Caribbean and two mainland
regions.
We performed an additional set of time-calibrated
analyses on the MRCT tree. This tree is better sampled
for Caribbean forms relative to mainland forms, so we
used the Caribbean-focused coding dicussed above. We
performed an analysis where dispersal rate is assumed
constant between all areas. This scenario costitu tes a null
biogeographic model that ignores the geological history
of Caribbean landmasses and the land connections
between islands. We also performed a time-stratified
analysis based on the geological model from Iturralde-
Vinent and MacPhee (1999; see also Iturralde-Vinent
2006). For this scenario, we penalized strongly against
d ispersal across water assigning a very low probability of
traversal (almost zero; d = 0.001) when landmasses were
separated and a probability of 1 when landmasses were
connected . We built cost d ispersal matrices based on
paleogeographical scenarios hypothesized by Iturralde-
Vinent and MacPhee (1999; see also Iturralde-Vinent
2006) for five time periods as follows: (1) Early Eocene
(55 Ma): all Caribbean landmasses were hypothesized
to be separated at this time, so we penalized dispersal
between islands, assigning dispersal probability of
0.01 for all transitions between land masses; (2) Late
Eocene–Early Oligocene (35-33 Ma): during this narrow
time frame all Greater Antilles except Jamaica were
hypothesized to be connected as a single landmass
(GAARlandia), which was connected to South America
by the Aves Ridge; thus, we assigned a dispersal
probability of 1 for all transitions between regions
(Greater Antilles, Lesser Antilles, and South America)
except transitions involving Jamaica; (3) Late Oligocene
(27–25 Ma): Cuba was hypothesized to be fragmented in
three landmasses, Hispaniola was divided into northern
and southern islands with Southern Hispaniola being
connected to Puerto Rico, and the Lesser Antilles
were isolated ; for this period , we assigned a dispersal
probability of 1 to transitions between connected
landmasses and 0.01 between separated landmasses;
(4) Middle Miocene (16–14 Ma): Cuba remained
fragmented , Hispaniola merged again but remained
narrowly connected to Puerto Rico, and the Lesser
Antilles were isolated ; for this period , we assigned a
dispersal probability of 0.01 between all landmasses;
(5) Pliocene to present (5–0 Ma): current separation of
landmasses is assumed; we set a d ispersal probability to
0.01 between all islands. We compared the fit of null and
time-stratified biogeographic models using the Akaike
information criterion (AIC).
For each set of reconstructions, we conducted
Dispersal–Extinction–Cladogenesis (DEC) analyses
comparing models with and without a jump dispersal
parameter J (Matzke 2014). We used our ancestral area
reconstructions to estimate the number of d ispersal
events in and out of each biogeographic region. The
counts allowed us to establish whether an area was a
2017 POE ET AL.—EVOLUTION OF ANOLES 667
relative sink (i.e., areas receiving immigrant lineages
from other areas) or source (i.e., areas from which
lineages dispersed to other areas) (Sedano and Burns
2010; Castroviejo-Fisher et al. 2014). We use the terms
sink and source in a macroevolutionary context, as in
Goldberg et al. (2005).
Taxonomy
In the interest of developing a taxonomy of anoles that
is stable with respect to taxonomic ranks but allows for
appropriate changes in the hypothesized composition
of taxa under new hypotheses about phylogenetic
relationships, we apply the methods of phylogenetic
nomenclature to the major clades of anoles. Specifically,
we apply names to both nested and mutually exclusive
anole clades using explicit phylogenetic definitions
(e.g., de Queiroz and Gauthier 1990, 1992; Cantino
and de Queiroz 2014). Our current selection of anole
clades to name is subjective and based on trad itionally
recognized groups of anoles. We do not name every
clade due to space constraints, and we do not name only
well-suppored clades because some poorly supported
clades have recurred in several analyses and seem
likely to withstand further scrutiny. In the interest of
nomenclatural stability, we apply existing names to the
clades with which they have the longest associations.
However, because not all of the clades have existing
names, and because some names have previously been
applied to more than one clade, we resurrect four long-
unused name and coin two new ones. For the purpose
of assessing trad itional use, we adopt a criterion similar
to that adopted by the ICZN (1999) for maintaining
prevailing use (Art. 23.9.1.2). That is, the name must
have been applied to the taxon in question (judged on
the basis of composition and diagnostic characters) in at
least 25 works, published by at least 10 authors, in the
immediately preceding 50 years, and encompassing a
span of not less than 10 years. This approach corresponds
roughly with treating the (formal) names recognized
by Etheridge (1959), who is commonly considered to
have initiated the modern era of anole systematics, as
those having trad itional use. The main exception is
Tropidodactylus, which was recognized by Etheridge but
does not meet all the criteria for trad itional use.
RESULTS
The Partitionfinder analysis suggested a 15-partition
scheme for the 50 gene, 54 potential partition molecular
data (Table 1). The MrBayes analysis took 138 h on the
CIPRESXSEDE cluster (maximum allowed is 168 h). The
average standard deviation of split frequencies (ASDSF)
was 0.059. The two runs appear to have plateaued and
converged on similar likelihoods (Supplementary Fig.
S2), and MCC trees calculated separately for each run
are nearly identical. These results are consistent with the
idea that MCMC mixing was sufficient.
TABLE 1. Partitioning scheme estimated in Partitionfinder analysis
Partition Model Genes
1 GTR+G COI position 1, KCNV2
2 GTR+G COI position 2, unnamed
# 146
3 GTR+G COI position 3
4 GTR+G ECEL1, HOXB1, KIF24
5 GTR+G ND2 position 1
6 GTR+G ND2 position 2
7 GTR+G ND2 position 3
8 GTR+G SOCS5 12, PCDH10,
FNIP2, KIAA2018, IRS2,
RAPGEF2, unnamed # 50,
unnamed # 57, TMTC4,
RAG1
9 GTR+G 16S
10 GTR+G unnamed # 127,
ENSACAG00000014694,
STRN4, PDE4D
11 GTR+G ENSACAG00000011799,
PDS5A, unnamed #183,
PPP2R5C, SFRS18,
unnamed # 59
12 GTR+G FRYL, DHX15, KIAA,
TLK2, TJP2, ATP2B1,
unnamed # 53, TRPA1
13 GTR unnamed #60
14 GTR+G ELAVL2, unnamed #10,
FBXW7, unnamed #152,
GLRB, BNC2, NFIB,
PTPRD
15 GTR+G RXF3, EXPH5, "C10 or
F71", BRCA1, unnamed
#177
Note: “Unnamed” gene sequences refer to regions listed in Alfold i et al.
(2011: Supplementary Table 20).
Relationships
A majority-rule consensus (including additional
compatible groupings) of post-burnin trees from the
MrBayes analysis is shown in Figures 1–4. This tree
depicts many relationships inferred in previous analyses
(e.g., Etheridge 1959; Jackman et al. 1999; Alföld i
et al. 2011). In particular, the overall structure of
two predominantly mainland radiations, two Lesser
Antillean clades, and multiple Greater Antillean
rad iations is evident. We consider our estimate
to supersede previous attempts at reconstructing
anole phylogeny due to our greater character and
species sampling. Furthermore, comparison with earlier
analyses is not straightforward in cases where taxonomic
coverage differs. For these reasons, we do not make
group-by-group comparisons with all previous stud ies,
but instead mainly discuss previously recognized
groups that bear on our formally named clades d iscussed
below.We note that several well-supported small groups
recognized by Etheridge (1959) and Williams (1976a,b)
and subsequent authors were mostly or completely
monophyletic (e.g., Beta anoles, bimaculatus series, roquet
series, cristatellus series, sagrei series, grahami series), and
that five of the eight genera of Nicholson et al. (2012)
were corroborated as monophyletic. In the summary
668 SYSTEMATIC BIOLOGY VOL. 66
FIGURE 1. Consensus phylogenetic estimate for the Dactyloa clade of Anolis based on Bayesian analysis of morphological and DNA data.
Numbers on clades are posterior probabilities × 100.
below, we refer to named groups in Figures 1–4. The
names of these groups are defined phylogenetically in
Appendix 3.
A large clade of predominantly South American
species (approximately equal to Etheridge’s [1959]
latifrons series and Nicholson et al. [2012] Dactyloa) is
2017 POE ET AL.—EVOLUTION OF ANOLES 669
FIGURE 2. Consensus phylogenetic estimate for Digilimbus (minus Ctenonotus and Norops) clade of Anolis based on Bayesian analysis of
morphological and DNA data. Numbers on clades are posterior probabilities × 100.
670 SYSTEMATIC BIOLOGY VOL. 66
FIGURE 3. Consensus phylogenetic estimate for Norops (minus Draconura) and Ctenonotus clades of Anolis based on Bayesian analysis of
morphological and DNA data. Numbers on clades are posterior probabilities × 100.
sister to the rest of Anolis (Fig. 1). This Dactyloa clade
includes lineages that extend north into the southern
Lesser Antilles (members of the roquet series) and into
Central America. Many of these species are large-bodied
with high numbers of toe lamellae. The clade includes
great variation in head scale size with some species
(e.g., species formerly recognized as Phenacosaurus)
d isplaying as few as two scales across the snout whereas
2017 POE ET AL.—EVOLUTION OF ANOLES 671
FIGURE 4. Consensus phylogenetic estimate for Draconura clade of Anolis based on Bayesian analysis of morphological and DNA data.
Numbers on clades are posterior probabilities × 100.
672 SYSTEMATIC BIOLOGY VOL. 66
TABLE 2. Comparison of DEC (Dispersal–Extinction–Cladogenesis) and DEC+ J (Dispersal–Extinction–Cladogenesis plus jump dispersal)
models of historical biogeography of Anolis lizards
Tree Region Model LnL P d e j AIC dAICc Weights
MCC tree
Mainland DEC − 597.3 2.0 1.2 0.0 0.0 1198.6 38.1 0.0DEC+ J − 577.3 3.0 1.0 0.0 0.0 1160.5 0.0 1.0
Caribbean DEC − 335.6 2.0 0.5 0.0 0.0 675.1 91.0 0.0DEC+ J − 289.1 3.0 0.3 0.0 0.0 584.1 0.0 1.0
MRC tree
Mainland DEC − 591.1 2.0 1.3 0.3 0.0 1186.1 36.2 0.0DEC+ J − 572.0 3.0 1.0 0.2 0.0 1149.9 0.0 1.0
Caribbean DEC − 347.4 2.0 0.3 0.0 0.0 698.8 108.5 0.0DEC+ J − 292.1 3.0 0.3 0.1 0.0 590.3 0.0 1.0
Notes: Ln: Ln likelihood; P: number of parameters; d : d ispersal; e: extinction; j: jump-dispersal; AIC: Akaike Information Criterion; dAICc: delta
AICc; weights: model weights. MCC tree:Maximum Clade Credibility tree from MrBayes analysis;MRC tree: Consensus-like tree from MrBayes
analysis (see text). Mainland: refers to biogeographic analysis focused on continental areas. Caribbean: refers to biogeographic analysis focused
on Caribbean areas. Best model in each comparison is highlighted in bold .
others (e.g., Williams’ 1976b aequatorialis group) have
more than 20.
Sister to the Dactyloa clade is the Digilimbus clade,
which includes Deiroptyx as sister to the remainder of
Anolis (Fig. 2). Deiroptyx is composed of the Cuban
crown-giant anoles (e.g., A. equestris) as sister to a
clade of Hispaniolan forms including A. darlingtoni,
the Hispaniolan green anoles, and the hendersoni and
monticola series of Williams (1976a). The dewlapless
Cuban species A. bartschi and A. vermiculatus are sister
to this clade, and the unusual Puerto Rican twig anole
A. occultus is sister to the rest of Deiroptyx.
Also within Digilimbus, a weakly supported clade
includes five named groups as sister to a clade composed
of Ctenocercus, Ctenonotus, and Norops (Fig. 2). This clade
includes the cybotoid anoles (Audantia), the d istinctive
terrestrial form Anolis (Chamaelinorops) barbouri, and the
grass-bush anoles (Schmidtanolis) from Hispaniola, as
well as a clade of mostly large-bodied Greater Antillean
anoles (Xiphosurus). The Cuban chamaeleon-like anoles
(Chamaeleolis) form a strongly supported clade within
Xiphosurus.
Ctenocercus is an ecomorphologically d iverse clade of
predominantly Cuban species that is sister to Ctenonotus
and the beta anoles (Norops) of Etheridge (1959) (Fig. 2).
Ctenocercus includes mini-rad iations of grass anoles (e.g.,
Anolis alutaceus), twig anoles (e.g., A. angusticeps), and
green trunk-crown anoles (e.g., A. carolinensis), as well as
the weakly supported placement (posterior probability
0.41) of Cuban A. argenteolus and A. lucius as sister to the
remaining Ctenocercus.
Ctenonotus is a chromosomally d iverse clade (Gorman
1973) that includes well supported Lesser Antillean
(Williams’ [1976a] bimaculatus series), Hispaniolan
(distichus series), and Puerto Rican Bank (cristatellus
series) subclades (Fig. 3). Ctenonotus anoles tend to be
abundant and highly visible, and some are among the
best-stud ied anole species (e.g., Rand 1964; Pacala and
Roughgarden 1982; Losos 1990;Dobson et al. 1992;Hertz
1992; Fleishman et al. 1997).
Species of the Norops clade of “beta” anoles (Etheridge
1959) share the anatomical trait of anteriorly d irected
transverse processes on the posterior caudal vertebrae
(Figs. 3 and 4). This well-established clade includes
three geographically coherent subclades. Trachypilus is
a Cuban clade of species mainly belonging to the trunk-
ground ecomorph. Placopsis is the Jamaican rad iation of
multiple ecomorphs. Draconura is a mainland rad iation,
with several forms that have become established on
offshore islands (e.g., A. townsendi, A. concolor, A.
villai). Trachypilus and Placopsis are very well-stud ied
clades (e.g. Underwood and Williams 1959; Ruibal 1961;
Bundy et al. 1987; Losos 1990; Jackman et al. 2002;
Vanhooydonck et al. 2005; Knouft et al. 2006; Cádiz et al.
2013), whereas Draconura remains the proportionately
least-known anole clade (but see, e.g., Andrews 1971;
Fitch et al. 1976; Nicholson 2002; Vitt et al. 2002).
Biogeography
Biogeographic model selection and anole diversification.—
In all analyses, DEC+ J models were favored over DEC
models accord ing to both AIC and model weights,
suggesting that founder-event d iversification has
been prevalent during the anole rad iation (Table 2).
Furthermore, biogeographic models incorporating
paleogeographic information were favored (Table 3).
These biogeographic models incorporate information
about historical connections between Caribbean
landmasses by reducing dispersal probabilities between
areas that were not connected over time. Several
instances of founder-event d iversification were inferred
during the colonization of the Caribbean (Fig. 5;
Supplementary Figs. S3–6). Divergence date estimates
in our BEAST analysis (MRCT tree; Supplementary Fig.
S6; Fig. 5) provided evidence of an origin of the anole
rad iation at the Paleocene–Eocene boundary (64.4–46.3
Ma). Posterior density plots summarizing divergence
date estimates (Fig. 6) show that most cladogenetic
events occurred during the Miocene (20–5 Ma). For
South American clades, the majority of cladogenetic
events seem not to be associated with Andean uplift
(main Andean uplift events occurred between 10–3 Ma
2017 POE ET AL.—EVOLUTION OF ANOLES 673
TABLE 3. Comparison of DEC (Dispersal–Extinction–Cladogenesis) and DEC+ J (Dispersal–Extinction–Cladogenesis plus jump dispersal)
models of historical biogeography of Anolis using MRCT tree (BEAST version of MRC, consensus-like tree; see text) with focus on Caribbean
areas
Type Model LnL P d e j AIC dAICc Weights
Time stratified DEC − 290.9 2.0 0.0 0.0 0.0 585.9 78.1 0.0
DEC+ J − 250.9 3.0 0.0 0.0 0.0 507.8 0.0 1.0
Null DEC − 298.6 2.0 0.0 0.0 0.0 601.3 93.5 0.0
DEC+ J − 260.5 3.0 0.0 0.0 0.0 527.0 19.2 0.0
Notes: Null model assumes equal d ispersal probability between islands or landmasses. Time stratified model allows differences in d ispersal rates
in the model formulation based on Caribbean paleogeographic models (Iturralde-Vinent and MacPhee 1999; Iturralde-Vinent 2006; see main
text for details). LnL: Ln likelihood; P: number of parameters; d : d ispersal; e: extinction; j: jump-dispersal; AIC: Akaike Information Criterion;
dAICc: delta AICc; weights: model weights. Best model in each comparison is highlighted in bold .
[Antonelli et al. 2009; Horn et al. 2010]). In contrast,
most cladogenetic events for Middle American clades
appear correlated with intense tectonic activity during
the mid-Miocene (15–10 Ma) (Fig. 6; Castoe et al. 2009;
Daza et al. 2010).
Early evolution and mainland-island transitions.—
Analysis of the MRCT tree (Fig. 5) estimated a
composite ancestral area for all Anolis. This estimate
may be interpreted either as ambiguity or that
the ancestor of all Anolis occupied a large area
including Caribbean and South American regions.
Other reconstructions using the full complement of
taxa (MCC, MRC trees; Supplementary Figs. S3-6)
identify South America as the origin of Anolis. If the
anole ancestor was only present in South America,
particularly the Amazonia region (Supplementary Figs.
S4, S6), at least two dispersal events are necessary to
explain the current d istribution of Caribbean clades.
The first d ispersal event, to the Northern Caribbean,
likely occurred during the Paleocene–Eocene boundary
(42.4– 61.7 Ma; Supplementary Fig. S7; Fig. 5). The
timing of this event predates the emergence of the Aves
ridge landbridge (Iturralde-Vinent and MacPhee 1999;
Iturralde-Vinent 2006). Thus, under this scenario, an
overwater d ispersal event likely explains the distribution
of all Northern Caribbean clades (i.e., a jump dispersal
event promoting founder-event d iversification in the
Caribbean Digilimbus clade). The second dispersal event
is the invasion of the roquet series to the Lesser Antilles,
which likely occurred near the Eocene–Oligocene
boundary (23.9–40.1 Ma; Supplementary Fig. S7; Fig. 5)
when the Aves Ridge is hypothesized to have been
present.
All biogeographic reconstructions indicate a West
Indian ancestry for the Draconura invasion of Middle
America (Nicholson et al. 2005) with later d ispersal to
South America. One of the ancestral range estimates
for Draconura inferred a scenario involving a dispersal
event from Jamaica to Middle America, which would
have occurred during the Eocene–Oligocene boundary
(29.9–41.Ma; Fig. 5; Supplementary Fig. S7).
Caribbean dispersal events.—In the biogeographic models
that included maximal taxonomic coverage and did not
incorporate geological information (i.e., MCC and MRC
analyses) the ancestral range estimated for Caribbean
Anolis was Hispaniola (Supplementary Figs. S3, S5).
However, in the MRCT tree analysis (Fig. 5), the
ancestral area was composite including Cuba and
Hispaniola. This latter inference is consistent with
ancestral Caribbean anole occupation of the composite
area known as GAARlandia (Iturralde-Vinent and
MacPhee 1999; Iturralde-Vinent 2006). The MRCT tree
indicated at least 18 transitions among Caribbean
islands.Many biogeographicmovements occurred when
islands were connected as either GAARlandia or when
Hispaniola was connected to Puerto Rico. Both Cuba and
Hispaniola were major sources of Northern Caribbean
anole lineages, with at least 12 dispersal events to other
islands. Some Caribbean transitions were from Greater
Antilles to small Caribbean islands (e.g., Bahamas and
Cayman islands) and likely occurred across water. At
least three dispersals are inferred to explain the anole
d iversity in each of Cuba, Hispaniola, and Puerto Rico.
Excluding A. sagrei, which may not be native to Jamaica
(Underwood and Williams 1959), a single overwater
d ispersal event explains current anole d iversity in
Jamaica (Supplementary Table S1).
Faunal exchange through the Isthmus of Panama.—From
18-20 crossings of the Isthmus of Panama were
reconstructed (Supplementary Table S2). Depending on
which tree is analyzed ,we inferred more dispersal events
from Middle America (MA) to South America (SA) than
the reverse (MRC, MRCT trees), or an approximately
equal number of MA to SA and SA to MA dispersals
(MCC tree; Supplementary Table S2). Based on the
MRCT tree, at least two dispersal events from MA to SA
were very early (~ 30 Ma; Supplementary Fig. S7; Fig. 5),
one for the smallest clade containing Anolis auratus
and A. brasiliensis and another for the smallest clade
containing A. notopholis and A. gracilipes. Accord ing to
recent geological evidence, the emergence of the South-
Middle American landbridge started between 25 and 23
Ma (Farris et al. 2011), and the final closure occurred by
10Ma (Farris et al. 2011;Bacon et al. 2015).Other d ispersal
events inferred in the trees involved range expansions
during the Miocene.
674 SYSTEMATIC BIOLOGY VOL. 66
FIGURE 5. Biogeographical reconstruction for Anolis lizards with emphasis in Caribbean areas using the dated MRCT tree from MrBayes
and BEAST analyses (see text). Vertical dotted lines represents the timing of paleogeographical reconstructions based on Iturralde-Vinent and
MacPhee (1999; see also Iturralde-Vinent 2006). Scale is millions of years. SLA = Southern Lesser Antilles; SA = South America; CA = Central
America; C = Cuba; H = Hispaniola; NLA = Northern Lesser Antilles; PR = Puerto Rico; J = Jamaica.
Dispersals among other mainland areas.—Our mainland
analyses estimated at least 77 dispersal events
either as range expansions or long-d istance events
between mainland areas (Supplementary Table S3;
Supplementary Figs. S4, S6). Dispersal events were
estimated to occur evenly across the timespan of anole
history. This result may indicate some constancy
of d ispersal rates, but we note that some very
ancient d ispersal events may not be inferred due
to lineage extinction, which could bias the overall
pattern to underestimate the relative rate of earlier
d ispersals. Mainland areas with the highest number of
immigrations were Nearctic, Upper Central America,
Lower Central America and the Chocó. Invasions to
these areas occurred in several instances and involved
multiple clades. Upper Central America, Lower Central
America, the Chocó and the Andes exhibited the
highest number of lineage emigrations. Finally, the
Caribbean coast of Colombia and Amazonia exhibited
the highest number of d ispersals into versus out of
the area. In other words, anole d iversity in these two
areas is a combination of immigration of lineages
from nearby regions and in situ speciation. In general,
mainland areas can be characterized as both source
and sink areas of faunal d iversity (Supplementary
Table S3).
2017 POE ET AL.—EVOLUTION OF ANOLES 675
FIGURE 6. Posterior density plots of d ivergence times for Anolis based on the dated MRCT tree from MrBayes and BEAST analysis (see text). a)
All species; b) only Caribbean species. The vertical lines represent the timing of the formation of the Aves Ridge; c) South American species. The
vertical lines represent the timing of the Andean uplift in South America; d) Middle American species. The vertical lines represent the timing
of the formation of the Isthmus of Panama.
DISCUSSION
Anolis lizards are classic study organisms in
evolution, physiology, and ecology. The phylogenetic
estimate presented here should enable novel and
more comprehensive comparative analyses of this well-
stud ied clade. Many subjects that could be addressed
only weakly or partially with limited sampling,
such as mainland-Caribbean comparisons, comparative
community evolution, and rates of speciation, may now
be tested rigorously.
The outstanding aspect of our phylogenetic estimates
is their completeness. Their principal fault is the weak
support for many nodes, especially deep in the trees.
Sixty-three percent of clades are supported at less than
95% probability in the comprehensive estimate (Figs. 1-
4). We suggest this weak support is due to two factors.
First, appropriately evolving nuclear genes have not
yet been sufficiently taxonomically sampled to provide
support for the deep splits in the anole tree (e.g.,
Appendix 1). Second, the matrix includes several taxa
scored for only a few characters of external morphology
(e.g., Anolis vicarius, A. pseudotigrinus) that are likely to be
weakly placed in the tree. It would be possible to improve
support values by removing these taxa, as is sometimes
done (e.g., Sanderson and Shaffer 2002; Moyle et al.
2012). For example, a RaxML (v 1.5, Stamatakis 2006)
analysis of all DNA data including the 294 species
scored for ND2 had only 41% of clades supported at
less than 95% bootstrap (same partitioning scheme as
above, “ML + thorough bootstrap” command; results
not shown). But this practice obviously would result
in a less comprehensive estimate of anole phylogeny
and taxonomy, and accuracy might be reduced as
well (see e.g., Gauthier et al. 1988). Most importantly,
such an approach would guarantee an incomplete (i.e.,
inaccurate) biogeographic reconstruction for the anole
clade, as transitions to areas of missing species might
not be represented . For example, removal of poorly
scored species A. concolor and A. pinchoti would preclude
estimation of an important biogeographic event—the
dispersal of the Draconura (i.e., mainland Norops) clade to
oceanic Caribbean islands San Andres and Providencia,
where these anoles are solitary endemic species.
Many inferred relationships in Figures 1–4 make
sense in light of previous stud ies and expectations
676 SYSTEMATIC BIOLOGY VOL. 66
based on morphology. Our phylogenetic estimates for
much of the anole clade are heavily determined by
previously published data, and our results for these well-
stud ied forms are largely congruent with previous work
on Caribbean (e.g., Jackman et al. 1999) and Dactyloa
(e.g., Castañeda and de Queiroz 2011) anoles. But
relationships of mainland beta anoles (Draconura; Fig. 4)
were largely unknown before this study (see Nicholson
[2002], Poe [2004], and Nicholson et al. [2012] for analyses
including a few Draconura). Draconura relationships
that are unsurprising include the monophyly of anoles
similar to Anolis fuscoauratus (the clade spanning A.
tenorioensis to A. bocourtii in Fig. 4), the monophyly
of tropidolepis-like anoles (clade spanning A. pachypus
to A. pseudopachypus), the geographic coherence of the
Mexican forms (the A. cobanensis to A. pygmaeus clade and
the A. dunni to A. subocularis clade), and the monophyly
of anole species previously referred to or associated
with A. limifrons (clade spanning A. apletophallus to
A. zeus) and A. lemurinus (clade spanning A. bicaorum
to A. vittigerus). Numerous smaller clades likewise
reassuringly align with expectation (e.g., geographically
proximal island forms A. concolor–A. pinchoti; South
American semiaquatic anoles A. macrolepis–A. rivalis–A.
lynchi; Central American semiaquatic anoles A. lionotus–
A. oxylophus–A. poecilopus; formerly conspecific species
pairs like A. biporcatus–A. parvauritus, A. tropidogaster–
A. gaigei, and A. cupreus–A. macrophallus). In spite of
the limited data used to reconstruct these relationships,
these groupings seem likely to withstand further
scrutiny.
On the other hand , some novel Draconura
results seem questionable given the degree of
morphological convergence they entail. For example,
the nonmonophyly of the humilis group (Anolis humilis,
A. compressicauda, A. marsupialis, A. notopholis, A.
quaggulus, A. tropidonotus, A. uniformis, A. wampuensis) is
surprising. These species share a deep axillary pocket,
strongly keeled dorsal and ventral scales, an enlarged
band of middorsal scales, and leaf-litter habitat niche.
Some members of this group that were found to be
nonmonophyletic previously have been considered
conspecific (e.g., A. humilis and A. marsupialis). Our data
for most of these forms is mainly mitochondrial and
morphological, and it is tempting to suspect a mislead ing
mitochondrial signal or mishandled tissue. However,
Philllips et al. (2015) found nonmonophyly of this group
using greater sampling of ind ividuals and a nuclear
gene (ITS-1), so perhaps our estimate is correct. Another
seemingly questionable result is the nonmonophyly
of the pentaprion group (A. beckeri, A. charlesmyersi, A.
cristifer, A. fungosus, A. ortonii, A.pentaprion, A. salvini, A.
sulcifrons, A. utilensis). These species are d istinguished
from each other only subtly (e.g., Köhler 2010) and share
an unusual pale lichenous coloration, short limbs, and
large smooth headscales. The separate monophyly of
South (A. sulcifrons, A. ortonii) and Central American
pentaprion anoles shown in the tree is geographically
reasonable, and A. fungosus (which seems out of place,
if poorly supported , in a clade of nondescript brown
mainland anoles like A. trachyderma and A. tropidogaster
[Fig. 4]) is unusual enough that few placements within
Draconura seem completely implausible for this species.
But the separation of A. salvini, which seems essentially
to be a high elevation version of A. pentaprion (Myers
1971), from the other Central American pentaprion
group anoles strains cred ibility. Similar to the pentaprion
and humilis group anoles, the anoles similar to A.
laeviventris (A. laeviventris, A. cusuco, A. kreutzi) are
nearly ind istinguishable from each other but they
are not monophyletic in our trees. In this case, the
result is likely due to limited character information for
some “problem” taxa that are similar to the species
of the laeviventris group. We lack molecular data and
possess only scant morphological data for species
such as A. wermuthi that d isrupt the monophyly of
the laeviventris-like forms, and we expect that more
comprehensive scoring of these species will render
the laeviventris-like species monophyletic. Additional
DNA data will illuminate all the unexpected results
noted above. It will be interesting to see which of the
surprising results in Figures 1–4 are “corrected” with
additional data and which, if any, ind icate convergence
to the degree seen in the anoles of the Greater Antilles
(see Losos et al. 1998).
Biogeography
In this study, we have provided the most
comprehensive biogeographic analysis of anoles to
date. Inclusion of all known Anolis species allowed
elucidation of a complex biogeographic history that
involved multiple vicariance and dispersal events. As
in other neotropical lineages (Miller et al. 2008; Smith
et al. 2014), both simple range expansions and long
distance dispersals were found to be important aspects
of d iversification in anoles. The best supported DEC+ J
(d ispersal–extinction–cladogenesis plus jump dispersal)
models incorporated a series of range evolution
models (Matzke 2013a, 2014) that allow distinction of
biogeographic scenarios based on maximum likelihood .
Our analyses allowed us to corroborate or contrad ict
some previous biogeographic hypotheses regard ing
the present-day distribution of anoles. Below we
discuss the origin of the anole clade. More detailed
discussion of our results with regard to dispersal within
Caribbean and mainland regions and the role of geologic
events in anole biogeography is in our Supplementary
Appendix 5.
The origin of anoles.—Based on DEC+ J analyses using all
anole species (Supplementary Figs. S3–S6), we were able
to provide an unambiguous ancestral area of the most
recent common ancestor (MRCA) of Anolis (analysis of
the molecular-scored taxa alone produced an ambiguous
root state; Fig. 5).We inferred a South American ancestor
for Anolis. Our hypothesized timing of the origin of
Anolis (46.3–64.4 Ma; Supplementary Fig. S7, Fig. 5)
contrad icts previous studies (e.g., Nicholson et al. 2012)
2017 POE ET AL.—EVOLUTION OF ANOLES 677
that suggested anoles originated approximately 95 Ma.
The results of our study are concordant with recent work
by Prates et al. (2015)who found similar d ivergence dates
for the MRCA of Anolis using more fossil calibration
points.We suspect that Nicholson et al.’s (2012) estimates
of d ivergence times were biased to older dates by an
incorrect assignment of the Anolis electrum amber fossil
(Lazell 1965) to the fuscoauratus clade. As Castañeda
et al. (2014) showed, this fossil lacks synapomorphies
that would allow it to be assigned to any anole clade
with confidence. Several stud ies have found evidence
that incorrect fossil assignment may dramatically affect
estimates of dating (Magallón 2004, 2010; Mello and
Schrago 2014).
Taxonomy
Based on our results concerning the phylogeny
of the Anolis clade (Figs. 1–4), we here propose a
revised taxonomy of anoles. To promote stability in the
associations between names and clades by dissociating
the references of names from considerations about
taxonomic ranks, clade names are defined following the
methods of phylogenetic nomenclature (Cantino and
de Queiroz 2014). Listed synonyms are all considered
approximate (as they are not phylogenetically defined)
and are inferred primarily on the basis of composition.
Inferred composition is stated in terms of crown
subclades and known, extant species only, although
it also includes extinct members of the corresponding
total clades. Our phylogenetic taxonomy of anoles is
described in Appendix 3.
SUPPLEMENTARYMATERIAL
Supplementary material, includ ing additional
d iscussion, MrBayes matrix, and online appendices,
figures and tables, can be found in the Dryad data
repository: http:/ / dx.doi.org/ 10.5061/ dryad .s80jq.
ACKNOWLEDGMENTS
MrBayes analyses were undertaken on the XSEDE
cluster of CIPRES (www.phylo.org). Thanks to the
International Barcode of Life Project at the Biodiversity
Institu te of Ontario for performing some of the mtDNA
sequencing.
For help in the field and/ or the lab we thank Eric
Schaad , Norma L. Manríquez-Morán, Uri O. García-
Vázquez, Heather MacInnes, Julian Davis, Jenny Hollis,
Erik Hulebak, Carlos Vásquez Almazán, Sofia Nuñez,
Gustavo Cruz, Federico Bolaños,Roberto Ibañez,Martha
Calderon, Andrew Crawford , Andrés Quintero-Angel,
Juan Carlos Chaparro, Christian Yañez-Miranda, Carlos
Pavón, Devon Graham, Julie Ray, Alvaro Aguilar, James
Aparicio, and Liliana Jaramillo.
For loan of specimens, we thank Jose Rosado
(MCZ), Jonathan Losos (MCZ), Joe Martinez (MCZ),
Alan Resetar (FMNH), Chris Phillips (INHS), Dan
Wylie (INHS), Nefti Camacho (LACM), Greg Pauly
(LACM), Toby Hibbits (TCWC), Lee Fitzgerald (TCWC),
Rafe Brown (KU), Rob Wilson (USNM), Darrel Frost
(AMNH), Margaret Arnold (AMNH), David Kizirian
(AMNH), James McCranie, QCAZ, and MSB. We
also thank Juan Diego Palacio Mejia and Institu to
Alexander von Humbold t and Andrew Crawford and
the University of the Andes for use of space and
laboratory equipment to conduct molecular work in
Colombia.
Permits were provided by the Secretaría de Medio
Ambiente y Recursos Naturales, Dirección General
de Vida Silvestre (Mexico); Institu to Nacional de
Conservación y Desarrolo Forestal, Áreas Protegidas
y Vida Silvestre (Honduras); MinisterioMinistereo de
Ambiente y Energía (Costa Rica); Autoridad Nacional
del Ambiente (Panama); Institu to Nacional de Recursos
Naturales (Peru); Ministerio de Ambiente (Ecuador);
Corporación Autónoma Regional de Risaralda –
CARDER (Colombia)
FUNDING
Funding was provided by the National Science
Foundation (DEB-0844624 to S.P.); SENESCYT and
Pontificia Universidad Católica del Ecuador (to O.T.
and F.A.); DGAPA, UNAM (PAPIIT no. 224009) and
CONACYT (no. 154093) (to A.N.M.O.).
APPENDIX 1: SPECIES LIST AND GENE COVERAGE FOR DNA
DATA
See Alföld i et al. (2011) for varying coverage of 46
genes (the difference in number of sites between our
paper and Alfold i et al.’s [2012]—19,878 versus 19,987–
is due to our use of a shorter segment of the 16S gene).
Number in parentheses is numbers of species scored for
that gene/ dataset. og = number of outgroup species.
All species were scored for some or all characters of
morphology.
APPENDIX 2. MORPHOLOGICAL CHARACTERS
Continuous quantitative characters were coded using
the approach of Thiele (1993). Wiens’s (1995) frequency
coding was used in cases wherein there appeared to be a
morphological break between recognizable states. States
were “ordered” if change between morphologically
adjacent states seemed evolutionarily more likely than
change between nonadjacent states.
1. Maximum snout to vent length (SVL; mm;
ordered). 0: <61; 1: 61–86; 2: 87–112; 3: 113–138; 4:
139–164; 5: >165.
2. Maximum female SVL/ maximum male SVL
(ordered). 0: <0.60; 1: 0.60–0.69; 2: 0.70–79; 3: 0.80–
0.89; 4: 0.90–0.99; 5: >1.00.
678 SYSTEMATIC BIOLOGY VOL. 66
Species CO1
(142 + 1 og)
734 sites
ND2 (294 +
4 og)
1039 sites
ECEL1
(111)
474 sites
RAG1
(62 + 4 og)
2754 sites
Alfold i et al.
(89 + 1 og)
19878 sites
acutus X X
aeneus X X X X
aequatorialis X X X
agassizi X X X
agueroi X
ahli X X
alayoni X
alfaroi X
aliniger X X
allisoni X X X X
allogus X X
altae X X X X
altavelensis
altitudinalis X
alumina X
alutaceus X X
alvarezdeltoroi X X X
amplisquamosus X X
anatoloros X X X
anchicayae X X
anfiloquiae
angusticeps X X
annectens X
anoriensis X X X
antioquiae X
antonii X X
apletophallus X
apollinaris X X
aquaticus X X X
argenteolus X X
argillaceus X
armouri X X
auratus X X X X
aurifer X
bahorucoensis X X X
baleatus X X
baracoae X
barahonae X X
barbatus X
barbouri X X
barkeri X X X
bartschi X X
beckeri X X X X
bellipeniculus
benedikti X X
bicaorum X X X
bimaculatus X X X
binotatus X X
biporcatus X X X
birama
biscutiger X
blanquillanus
bocourtii X
boettgeri
bombiceps X X
bonairensis X
(continued)
2017 POE ET AL.—EVOLUTION OF ANOLES 679
Species CO1
(142 + 1 og)
734 sites
ND2 (294 +
4 og)
1039 sites
ECEL1
(111)
474 sites
RAG1
(62 + 4 og)
2754 sites
Alfold i et al.
(89 + 1 og)
19878 sites
boulengerianus X X X
brasiliensis X
bremeri X X
breslini X
brevirostris X X
brunneus X
calimae X X X
campbelli X
capito X X X X
caquetae
carlostoddi
carolinensis X X X X X
carpenteri X
casildae X X X X
caudalis X
centralis X X
chamaeleonides X X
charlesmyersi X X X
chloris X X X X
chlorocyanus X X
chocorum X X X X
christophei X X
chrysolepis X
chrysops
clivicola X
cobanensis X
coelestinus X X
compressicauda X
concolor
confusus X
conspersus X
cooki X X
crassulus X X X
cristatellus X X X X X
cristifer X X X
cryptolimifrons X X X
cupeyalensis X
cupreus X X X X
cuprinus X
cuscoensis
cusuco X X
cuvieri X X X X X
cyanopleurus X
cybotes X X
cymbops X
danieli X X X
darlingtoni X
datzorum X X
desechensis X X
desiradei
dissimilis
distichus X X X
dolichocephalus X
dollfusianus X X X
dominicensis X
duellmani X
(continued)
680 SYSTEMATIC BIOLOGY VOL. 66
Species CO1
(142 + 1 og)
734 sites
ND2 (294 +
4 og)
1039 sites
ECEL1
(111)
474 sites
RAG1
(62 + 4 og)
2754 sites
Alfold i et al.
(89 + 1 og)
19878 sites
dunni X X X
equestris X X X
ernestwilliamsi X X
etheridgei X X
eugenegrahami X
eulaemus X X
euskalerriari X X X
evermanni X X X X
extremus X X X X X
fairchildi
fasciatus X X X
favillarum X
ferreus X
festae X X X
fitchi X X X X
forbesi X
forresti
fortunensis X X
fowleri X X
fraseri X X X X
frenatus X X X X X
fugitivus
fungosus X
fuscoauratus X X X X
gadovii X X X
gaigei X X X
garmani X X X
garridoi X
gemmosus X X X
ginaelisae X X
gingivinus X X
gorgonae X
gracilipes X X
grahami X X
granuliceps X X
griseus X X X
gruuo X X X
guafe X X
guamuhaya X
guazuma X
gundlachi X X X X
haetianus X
hendersoni X
heterodermus X X X
heteropholidotus
hobartsmithi X
homolechis X X
huilae X X X
humilis X X X X
ibanezi X X
ignigularis X
imias X X
inderenae X X X
inexpectatus X
insignis
(continued)
2017 POE ET AL.—EVOLUTION OF ANOLES 681
Species CO1
(142 + 1 og)
734 sites
ND2 (294 +
4 og)
1039 sites
ECEL1
(111)
474 sites
RAG1
(62 + 4 og)
2754 sites
Alfold i et al.
(89 + 1 og)
19878 sites
insolitus X X
isolepis X
jacare X X X
johnmeyeri X X
juangundlachi
jubar X X
kahouannensis
kemptoni X X
koopmani X
kreutzi
krugi X X X X X
kunayalae X X X
laevis
laeviventris X X X
lamari
latifrons X X
leachii X X
lemurinus X X X X
limifrons X X X X
limon
lineatopus X X
lineatus X X
liogaster X X X
lionotus X X X X
litoralis
lividus X
longiceps X
longitibialis X
loveridgei X
loysianus X X
luciae X X X
lucius X X X
luteogularis X
luteosignifer
lynchi X X
lyra X X
macilentus X
macrinii X X
macrolepis X X
macrophallus X X X
maculigula X X X
maculiventris X X
magnaphallus X
marcanoi X X X X X
mariarum X X
marmoratus X X
marron X
marsupialis X X X
matudai X
maynardi X X
medemi X
megalopithecus X
megapholidotus X X X
menta
(continued)
682 SYSTEMATIC BIOLOGY VOL. 66
Species CO1
(142 + 1 og)
734 sites
ND2 (294 +
4 og)
1039 sites
ECEL1
(111)
474 sites
RAG1
(62 + 4 og)
2754 sites
Alfold i et al.
(89 + 1 og)
19878 sites
meridionalis X
mestrei X X
microlepidotus X X X
microtus X
milleri X X
mirus
monensis X X
monteverde X X
monticola X
morazani
muralla
nasofrontalis
naufragus X
neblininus X X X
nebuloides X X X
nebulosus X X
nelsoni
nicefori X
noblei X
notopholis X X
nubilus X
occultus X X X
ocelloscapularis X X X
oculatus X
oligaspis
olssoni X X
omiltemanus X X
onca X
opalinus X X
ophiolepis X X
oporinus X
orcesi X X
ortonii X X X
otongae X X X
oxylophus X X
pachypus X X X
paravertebralis
parilis X X X
parvauritus X X X
parvicirculatus X X
paternus X X X
pentaprion X X X
peraccae X X X X
petersii X X X
peucephilus X
philopunctatus
phyllorhinus
pigmaequestris
pijolense
pinchoti
placidus X X
planiceps X
podocarpus X X
poecilopus X
poei X X X
pogus X X X
(continued)
2017 POE ET AL.—EVOLUTION OF ANOLES 683
Species CO1
(142 + 1 og)
734 sites
ND2 (294 +
4 og)
1039 sites
ECEL1
(111)
474 sites
RAG1
(62 + 4 og)
2754 sites
Alfold i et al.
(89 + 1 og)
19878 sites
polylepis X X X
poncensis X X
porcatus X X X X
porcus X
princeps X X X X
proboscis X X X
properus X
propinquus
pseudokemptoni X X
pseudopachypus X
pseudotigrinus
pulchellus X X X X
pumilus X X
punctatus X X X X
purpurgularis X
pygmaeus X
quadriocellifer X X
quaggulus X X
quercorum X X X
ravitergum X
reconditus X X
rejectus X
richardii X X X
ricordii X X
rimarum
rivalis
roatanensis X X
rodriguezii X X X
roosevelti
roquet X X X X
rubiginosus X X
rubribarbaris
rubribarbus X X
ruibali
ruizii
rupinae
sabanus X X
sagrei X X X X
salvini X X
santamartae
schiedii X
schwartzi X
scriptus X X
scypheus X X
semilineatus X X
sericeus X
serranoi X X X
sheplani X
shrevei X X
singularis X X
smallwoodi X
smaragdinus X X X X
sminthus X X X
soinii X X X
solitarius
spectrum
(continued)
684 SYSTEMATIC BIOLOGY VOL. 66
Species CO1
(142 + 1 og)
734 sites
ND2 (294 +
4 og)
1039 sites
ECEL1
(111)
474 sites
RAG1
(62 + 4 og)
2754 sites
Alfold i et al.
(89 + 1 og)
19878 sites
squamulatus
strahmi X X
stratulus X X X
subocularis X X X
sulcifrons X X
tandai X X
taylori X X X
tenorioensis
terraealtae
terueli
tetarii
tigrinus X X X
toldo
tolimensis X X
townsendi X
trachyderma X
transversalis X X X X
trinitatis X X X
tropidogaster X
tropidolepis X X
tropidonotus X X X X
umbrivagus
uniformis X X X
unilobatus X X X
utilensis X X X
valencienni X X
vanidicus X X
vanzolinii X X
vaupesianus
ventrimaculatus X X X
vermiculatus X
vescus
vicarius
villai
vinosus X
vittigerus X X X
wampuensis
wattsii X X
websteri X
wellbornae X X
wermuthi
whitemani X X
williamsmittermeierorum X X X
woodi X X X
yoroensis X X X
zeus X X X
B. plumifrons X X
P. marmoratus X X X X
P. scapulatus X X
U. gallardoi X X
2017 POE ET AL.—EVOLUTION OF ANOLES 685
3. Femoral length/ SVL (ordered). 0: <0.20; 1: 0.20–
0.22; 2: 0.23–0.25; 3: 0.26–0.28; 4: 0.29–0.31; 5: >0.32.
4. Head length/ SVL (ordered). 0: <0.23; 1: 0.23–0.24;
2: 0.25–.26; 3: 0.27–0.28; 4: 0.29–0.30; 5: 0.31+ .
5. Ear height/ SVL (ordered). 0: <.018; 1: 0.018–0.025;
2:0.026–0.033; 3: 0.034–0.041; 4: 0.042–0.048; >5:
0.049.
6. Toe length/ SVL (ordered). 0: <0.14; 1: 0.14–0.16;
2:0.17–0.19; 3: 0.20–0.22; 4: 0.23–0.25; 5: >0.26.
7. Tail length/ SVL (ordered). 0: <1.30; 1: 1.30–1.59; 2:
1.60–1.89; 3: 1.90–2.19; 4: 2.20–2.49; 5: >2.50.
8. Mean number of longitud inal ventral scales in 5%
of SVL (ordered). 0: <4.5; 1: 4.5–5.9; 2: 6.0–7.4; 3:
7.5–8.9; 4: 9.0–10.4; 5: >10.5.
9. Mean number of longitud inal dorsal scales in 5%
of SVL (ordered). 0: <5.0; 1: 5.0–7.4; 2: 7.5–9.9; 3:
10.0–12.4; 4: 12.5–14.9; 5: >15.0.
10. Mean number of expanded lamellae on toe IV
(ordered). 0: <15.0; 1: 15.0–20.9; 2: 21.0–26.9; 3:
27.0–32.9; 4: 33.0–38.9; 5: >39.0.
11. Male dewlap (ordered). 0: extends posteriorly past
arms; 1: to arms or shorter; 2: absent.
12. Female dewlap (ordered). 0: extends posteriorly
past arms; 1: to arms or shorter; 2: absent.
13. Head scales (frequency-coded). 0: keeled ; 5:
smooth.
14. Subocular scales (frequency-coded). 0: in contact
with supralabials; 5: separated from supralabials
by a row of scales.
15. Mean number of scales across the snout at the
second canthals (ordered). 0: <5; 1: 5–7; 2: 8–10; 3:
11–13; 4: 14–16; 5:>17.
16. Mean number of supralabial scales from rostral to
center of eye (ordered). 0: <6; 1: 6; 2: 7; 3: 8; 4: 9; 5:
>10.
17. Supraorbital semicircles (frequency-coded). 0:
separated by one or more rows of scales; 5: in
contact.
18. Interparietal scale (frequency-coded). 0: separated
from supraorbital semicircles by at least one scale;
1: in contact with supraorbital semicircles.
19. Length of interparietal scale/ length of scale lateral
to interparietal (ordered). 0: <1.25; 1: 1.25–2.24; 2:
2.25–3.24; 3: 3.25–4.24; 4: 4.25–5.24; 5: >5.25.
20. Modal number of elongate superciliary scales
(ordered). 0: none; 1: one; 2: two; 3: three.
21. Scales in supraocular d isc (frequency-coded). 0:
some enlarged , gradually decreasing in size, or all
scale equal; 5: 2–4 abruptly enlarged , at least 2 ×
larger than other supraocular scales.
22. Differentiated scales in lower eyelid (frequency-
coded). 0: absent; 5: present.
23. Mental (frequency-coded). 0: partially d ivided; 5:
completely d ivided .
24. Mental (frequency coded). 0: extends along
mouth posteriorly past rostral; 5: rostral extends
posteriorly past mental.
25. Mean number of postmental scales (ordered). 0:
<4.5; 1: 4.5–5.4; 2: 5.5–6.4; 3: 6.5–7.4; 4: 7.5–8.4; 5:
>8.5.
26. Posterior border of mental (frequency-coded). 0:
convex or straight; 5: concave.
27. Dorsal surface of rostral (frequency-coded). 0:
smooth; 5: cleft.
28. Preocciptal scale (frequency-coded). 0: absent; 5:
present.
29. Dorsal snout scales (frequency-coded). 0: not in
regular rows; 5: in longitud inal parallel rows.
30. Scales around naris (unordered). 0: anterior nasal
in contact with rostral; 1: circumnasal separated
from rostral by one scale, not in contact with
supralabial; 2: external naris separated from rostral
by two scales, not in contact with supralabial; 3:
external naris separated from rostral by three or
more scales, not in contact with supralabial; 4:
circumnasal in contact with rostral; 5: circumnasal
in contact with supralabial, separated from rostral
by 1–2 scales.
31. Modal number of abruptly enlarged sublabial
scales (ordered). 0: zero; 1: one; 2: two or more.
32. Ventral scales (frequency coded). 0: keeled ; 5:
smooth.
33. Middorsal scales (frequency-coded). 0: 0-4
enlarged; 5: >5 enlarged .
34. Middorsal crest (frequency coded). 0: absent; 5:
present.
35. Deep tubelike axillary pocket (frequency-coded) 0:
absent; 5: present.
36. Lateral scales (frequency-coded). 0: homogeneous;
5: heterogeneous.
37. Middorsal caudal scales (frequency-coded). 0:
single row; 5: double row.
38. Tail fin (frequency-coded). 0: absent in large males;
5: present in large males.
686 SYSTEMATIC BIOLOGY VOL. 66
39. Scales on dewlap (frequency-coded). 0: in rows
of single scales; 5: in rows of multiple scales or
scattered .
40. Enlarged postcloacal scales (frequency coded). 0:
present in males; 5: absent in males.
41. Discrete expanded toepad on toe IV (frequency-
coded). 0: present; 5: absent.
42. Modal dominant dorsal color when sleeping
(unordered). 0: brown; 1: green; 2: gray/ white; 3:
blue.
43. Modal lateral pattern when sleeping (unordered).
0: solid ; 1: lateral stripe along body; 2: bands; 3:
ocelli/ spots; 4: speckled ; 5: jumbled , lichenous.
44. Interorbital bar (frequency-coded). 0: absent; 5:
present.
45. Throat color (frequency-coded). 0: light; 5: dark.
46. Color of iris (unordered). 0: brown; 1: yellow; 2:
blue or gray; 3: green; 4: red .
APPENDIX 3. PHYLOGENETIC TAXONOMYOF ANOLES
Anolis Daudin 1802 [nobis], converted clade name
Synonyms: Anolius of Cuvier (1817), Anolidae? of Cope
(1864), Anolidae of O’Shaughnessy (1875), Anolinae
(except for the inclusion of Polychrus) of Cope (1900),
Anolinae of Varnoa (1985), and Dactyloidae of Townsend
et al. (2011). Dactyloa of Wagler (1830) and Fitzinger
(1843), Dactyloae of Fitzinger (1843), Anolini and Anolina
of Varona (1985), alpha section (informal) of Etheridge
(1959) and punctatus subsection (informal) of Williams
(1976b) are partial synonyms that refer to paraphyletic
groups originating in approximately the same ancestor.
Definition : The crown clade for which both adhesive toe
pads and an extensible throat fan (dewlap), as inherited
by Anolis carolinensis Voigt 1832, are apomorphies
relative to other crown clades.
Reference phylogeny: Figure 1 of this study.
Comments: Daudin (1802) originally coined the name
Anolis for the group of saurian reptiles d iagnosed by
toe pads and an extensible dewlap. Later 19th century
authors, most notably Wagler (1830, 1833) and Fitzinger
(1826, 1843) named several mutually exclusive taxa
(ranked as genera and subgenera) for species possessing
those characters; however, Boulenger (1885) treated
all but three of those names as synonyms of Anolis.
Additional “genera” were named during the first half
of the 20th century (Schmidt 1919; Barbour 1920, 1923;
Cochran 1934; Dunn 1939). The modern era of anole
systematics is commonly considered to have begun
with the work of Etheridge (1959), who recognized five
“genera” of anoles: Anolis, including the vast majority
of the species, and four other small, segregate ‘genera.”
The last of these, Tropidodactylus, which differed from
Anolis in having lost the toepads, was eliminated
when a morphologically intermediate species was
discovered (Williams 1974). Subsequent phylogenetic
analyses revealed that the other three “genera”—
Chamaeleolis, Chamaelinorops, and Phenacosaurus—were
also derived from within Anolis, lead ing the authors
of those studies either to reject those taxa (Hass et al.
1993; Poe 2004) or to treat them as unranked subclades
of Anolis (Jackman et al. 1999). Thus, the name Anolis
was applied to the smallest clade containing all species
possessing adhesive toe pads and an extensible dewlap,
including some species lacking one or the other of those
characters (through secondary loss) that were inferred
to be part of that clade (e.g., Poe 2004; Nicholson et al.
2005; Losos 2011).
Etheridge (1959; see also Williams 1976a,b) also
recognized various informally named subgroups of
anoles associated with the ranks of section and series.
The two sections that he recognized , designated the
alpha and beta sections, were recognized formally as
the “genera” Anolis and Norops by Savage and Talbot
(1978). However, the finding that the alpha section
(Anolis sensu Savage and Talbot) is paraphyletic (Shochat
and Dessauer 1981; Gorman et al. 1984; Guyer and
Savage 1986), led authors following Savage and Talbot
(1978) to partition their paraphyletic version of Anolis
into multiple genera and to shift the name Anolis to
smaller and smaller clades (Guyer and Savage 1986, 1992;
Nicholson et al. 2012).
In the interest of historical continuity, the name Anolis
is here applied to the crown clade for which adhesive
toe pads and an extensible dewlap are apomorphies
relative to other non-nested crown clades. Applying the
name to this clade associates it with the most recent
common ancestor of the species originally included
in Anolis by Daudin (1802); except that he included
a gecko), O’Shaughnessy (1875), Boulenger (1885),
Schmidt (1919), Cochran (1934), Dunn (1939), Barbour
(1920, 1923), Etheridge (1959), and Williams (1976a,b),
whose paraphyletic versions of Anolis excluded various
small, deeply nested , segregate “genera,” as well as by
Savage and Talbot (1978; see also Savage (1980, 1982),
whose paraphyletic version of Anolis also excluded an
expanded version of Norops (= beta section of Etheridge
1959). It also associates the name Anolis with the clade
to which it has been applied by authors who did
not recognize Etheridge’s beta section as a separate
“genus” subsequent to the finding that Chamaeleolis,
Chamaelinorops, and Phenacosaurus are nested within
that clade (e.g., Hass et al. 1993; Jackman et al. 1999; Poe
2004; Nicholson et al. 2005; Losos 2011).
Inferred composition : Dactyloa and Digilimbus (see
below).
2017 POE ET AL.—EVOLUTION OF ANOLES 687
Etymology: Accord ing to Daudin (1802), the name is
that given in the French colonies in the Americas to
lizards of this kind .
Dactyloa Wagler 1830 [Castañeda and de Queiroz 2013]
Synonyms: latifrons series (informal) (Etheridge 1959).
Definition : The most inclusive crown clade containing
Anolis punctatus Daudin 1802 but not A. bimaculatus
(Sparrman 1784), A. cuvieri Merrem 1820, A. equestris
Merrem 1820, A. occultus (Williams and Rivero 1965),
and A. sagrei Duméril and Bibron 1837 (Castañeda and
de Queiroz 2013).
Reference Phylogeny: Figure 1 of this study.
Comments: Early d ivergence between the members of
this clade and all other anoles, with the exception of
some distinctive groups that were treated as separate
“genera”, was inferred in several early stud ies of anole
phylogeny (Etheridge 1959;Guyer and Savage 1986, 1992;
Cannatella and de Queiroz 1989), but evidence for the
monophyly of this group, including species formerly
placed in the “genus” Phenacosaurus, emerged later (e.g.,
Jackman et al. 1999; Poe 2004; Nicholson et al. 2005,
2012; Alföld i et al. 2011; Castañeda and de Queiroz
2011, 2013, this study). Although originally proposed
as a substitu te name for Anolis Daudin 1802, the name
Dactyloa Wagler 1830 has been applied to the clade of
mainland alpha anoles (latifrons series of Etheridge 1960)
by Guyer and Savage (1986) and various subsequent
authors (e.g., Savage and Guyer 1989; Guyer and Savage
1992; Castañeda and de Queiroz 2011; Castañeda and de
Queiroz 2013;Nicholson et al. 2012;Prates et al. 2015), and
was defined phylogenetically as applying to that clade by
Castañeda and de Queiroz (2013). We have adopted an
equivalent definition with a more concise wording.
Inferred Composition : The following five (informally
named) non-nested crown clades: aequatorialis series,
latifrons series, punctatus series, heterodermus series,
roquet series (Castañeda and de Queiroz 2013). The
compositions of these clades as inferred in the present
study are largely congruent with those proposed by
Castañeda and de Queiroz (2013), with the following
exceptions: Anolis bellipeniculus, A. calimae, A. carlostoddi,
and A. neblininus were considered incertae sedis within
Dactyloa, and A. dissimilis was tentatively referred
to the punctatus series; all of these species are here
referred to the heterodermus series. Anolis laevis and
A. phyllorhinus were considered incertae sedis within
Dactyloa, and A. philopunctatus was tentatively referred
to the latifrons series; all of these species are referred
to the punctatus series. Anolis parilis and A. mirus were
tentatively referred to the aequatorialis series, and A.
limon had not been described; all of these species are
referred to the latifrons series. Anolis cuscoensis was
considered incertae sedis within Dactyloa, A. soinii and
A. gorgonae were tentatively referred to the punctatus
series, and A. poei had not been described; all of these
species are referred to the aequatorialis series. The
series assignments (including tentative ones) of all other
Dactyloa species d iscussed by Castañeda and de Queiroz
(2013) that were included in the present study are
corroborated by our results. In addition, Castañeda and
de Queiroz (2013) defined the (formal) names Megaloa
Castañeda and de Queiroz 2013 and Phenacosaurus
Barbour 1920 for clades corresponding roughly to,
but potentially less inclusive than, the latifrons series
and the heterodermus series, respectively. Accord ing to
the results of the present study, Megaloa corresponds
precisely to the latifrons series in terms of known
composition, but Phenacosaurus may correspond to the
largest subclade of the heterodermus series that includes
A. heterodermus but not A. neblininus and A. calimae,
which are not twig anoles (Castañeda et al., manuscript
in preparation), although the ecomorph assignments of
some critical species are currently unknown. Poe et al.
(2015) defined Continenteloa to include the non-roquet
series Dactyloa, a clade that is weakly corroborated
here.
Etymology: Derived from the Greek dactyl (finger) +
oa (hem, border), presumably referring to the toepads
of the lizards in this clade (the name was originally
proposed as a substitu te name for Anolis).
Digilimbus nobis, new clade name
Synonyms: None.
Definition : The most inclusive crown clade containing
Anolis carolinensis Voigt 1832 but not Anolis punctatus
Daudin 1802.
Reference Phylogeny: Figure 2 of this study.
Comments: Although a clade composed of all anoles
except the “mainland” alpha anoles has been inferred
repeated ly and consistently (e.g., Gorman et al. 1984;
Hass et al. 1993; Jackman et al. 1999 [with the exception of
A. occultus]; Poe 2004 [with the exception of A. occultus];
Nicholson et al. 2005, 2012; Alföld i et al. 2011; Castañeda
and de Queiroz 2011, 2013; this study), and despite
the naming of its sister group (Dactyloa; see above), an
emphasis on ranks has left this highly corroborated clade
unnamed.We therefore take this opportunity to name it.
Inferred Composition : The following mutually
exclusive crown clades: Deiroptyx, Audantia,
Schmidtanolis, Xiphosurus, Ctenocercus, Ctenonotus,
and Norops (see below).
Etymology: Derived from the Latin digitus (finger, toe),
truncated for the sake of euphony, and limbus (edge,
border), referring to the toepads. Digilimbus is the
Latin equivalent of the Greek Dactyloa and thus seems
appropriate as the name of the sister group of Dactyloa,
688 SYSTEMATIC BIOLOGY VOL. 66
given that toepads are present in the vast majority of
the lizards in both clades.
Deiroptyx Fitzinger 1843 [nobis], converted clade name
Synonyms: None.
Definition : The most inclusive crown clade containing
Anolis vermiculatus Duméril and Bibron 1837 but not
A. auratus Daudin 1802, A. bimaculatus (Sparrman 1784),
A. armouri (Cochran 1934), A. carolinensis Voigt 1832, A.
cuvieri Merrem 1820, A. semilineatus Cope 1864, and A.
punctatus Daudin 1802.
Reference Phylogeny: Figure 2 of this study.
Comments: Jackman et al. (1999) inferred a close
relationship between the Anolis vermiculatus species
group and the A. chlorocyanus species group (both
sensu Williams 1976a). Poe (2004; see also Alföld i et al.
[2011]) inferred a larger clade composed of those two
species groups plus the A. equestris species group,
the A. hendersoni species group, and the A. monticola
species group (all sensu Williams 1976a), and Nicholson
et al. (2005) added A. occultus and A. darlingtoni,
which was corroborated by Nicholson et al. (2012)
and this study. Nicholson et al. (2012) applied the
name Deiroptyx Fitzinger (1843) to this clade under
rank-based nomenclature. Although Deiroptyx was
previously applied to a smaller clade composed only
of A. vermiculatus and A. bartschi (e.g., Cochran 1928),
with the exception of Varona (1985), that name has not
been so applied for more than 50 years and therefore
is here considered available to be applied to the larger
clade, following Nicholson et al. (2012). However, in the
interest of maintaining the association the name with
that clade (here conceptualized as the largest crown
clade containing A. vermiculatus but not certain other
species, including A. auratus, which seems consistent
with the concept of Nicholson et al. 2012), we have
provided it with a formal phylogenetic definition.
Inferred composition : Anolis equestris species group
(Williams 1976a; see also Schwartz and Garrido 1972),
Anolis chlorocyanus species group (Williams 1965, 1976a),
Anolis monticola species group (Williams 1976a) minus
A. etheridgei, Anolis hendersoni species group (Williams
1976a), Anolis vermiculatus species group (Williams
1976a), Anolis darlingtoni (Cochran 1935), Anolis occultus
Williams and Rivero 1965.
Etymology: Derived from the Greek deire (neck, throat)
and ptyx (a fold), possibly in reference to the transverse
gular fold of A. vermiculatus, upon which the name was
based .
Audant ia Cochran (1934) [nobis], converted clade name
Synonyms: cybotes subseries, cybotes species group, and
cybotes superspecies (all informal) of Williams (1976a);
cybotes series (informal) of Gorman et al. (1980; see also
Burnell and Hedges 1990).
Definition : The most inclusive crown clade containing
Anolis armouri (Cochran 1934) but not A. auratus Daudin
1802, A. bimaculatus (Sparrman 1784), A. carolinensis
Voigt 1832, A. cuvieri Merrem 1820, A. semilineatus Cope
1864, A. vermiculatus Duméril and Bibron 1837 and
A. punctatus Daudin 1802.
Reference phylogeny: Figure 2 of this study.
Comments: Anolis cybotes and its relatives were
considered close to Anolis cristatellus and its relatives by
Etheridge (1959), but this relationship was challenged
by early molecular stud ies (e.g., Gorman et al. 1980;
Wyles and Gorman 1980). Subsequently, Poe (2004)
inferred the cybotoids in a relatively isolated position,
as sister to the beta anoles (Norops). The relatively
isolated position of the cybotoids, but not necessarily a
close relationship to the beta anoles, was corroborated
by subsequent stud ies (e.g., Nicholson et al. 2005, 2012;
Alföld i et al. 2011; this study). The name Audantia
Cochran 1934 (type species A. armouri) was originally
proposed for Hispaniolan anoles with a transverse, as
well as a longitud inal, gular fold and came to include
A. armouri and A. shrevei (Cochran 1934, Cochran 1939,
1941). Audantia was not recognized by Etheridge (1959),
who noted that a transverse gular fold was also present
in Anolis cybotes, which was not included in Audantia
but which he considered closely related to the included
species. Nicholson et al. (2012) resurrected the name
Audantia for the cybotoid anoles (cybotes subseries of
Williams [1976a]; cybotes series of Gorman et al. [1980]
and Burnell and Hedges [1990]). Because that name
was not in use during the previous 50 years, and
because the transverse gular fold does not appear to be
diagnostic of the clade composed of A. armouri and A.
shrevei, we accept Audantia as the name of the cybotoid
clade and here provide it with a formal phylogenetic
definition.
Inferred composition : cybotes subseries, species
group, and superspecies of Williams (1976a) = cybotes
series of Gorman et al. (1980) and Burnell and Hedges
(1990).
Etymology: Named for the collector of the type
specimen of the type species, André Audant, zoologist
at the Government Agricultural School at Damien, Haiti
(Cochran 1934).
Schmidtanolis nobis, new clade name
Synonyms: Chamaelinorops of Nicholson et al. (2012).
Definition : The most inclusive crown clade containing
Anolis semilineatus Cope 1864 but not A. auratus Daudin
1802, A. bimaculatus (Sparrman 1784), A. armouri
(Cochran 1934), A. carolinensis Voigt 1832, A. cuvieri
2017 POE ET AL.—EVOLUTION OF ANOLES 689
Merrem 1820, A. vermiculatus Duméril and Bibron 1837,
and A. punctatus Daudin 1802.
Reference phylogeny: Figure 2 of this study.
Comments: A monophyletic group approximating this
clade was first inferred by Jackman et al. (1999), except
that it d id not include Anolis barbouri, and by Poe (2004),
except that it included some now-excluded Cuban grass
anoles (A. cyanopleurus, A. spectrum). A monophyletic
group matching this clade more precisely in composition
was inferred by Nicholson et al. (2005, 2012: Fig. 4a)
and Alföld i et al. (2011). Nicholson et al. (2012) applied
the name Chamaelinorops Schmidt 1919 to this clade;
however, that name was originally proposed for only
one of the included species, Anolis barbouri, based
on its d istinctive morphology (Schmidt 1919), a use
adopted by Etheridge (1959) and numerous subsequent
authors, whether as a “genus” name (e.g., Thomas 1966;
Williams 1976a; Schwartz and Insháustegui 1980; Wyles
and Gorman 1980; Case and Williams 1987; Guyer and
Savage 1986, 1992; Burnell and Hedges 1990; Autumn
and Losos 1997) or simply as a clade name (Jackman
et al. 1999). Therefore, we have preserved the trad itional
use of the name Chamaelinorops (see below) and propose
the name Schmidtanolis for the larger clade.
Inferred composition : semilineatus species group (Hertz
1976; Williams 1976a), Anolis etheridgei Williams (1962),
A. insolitusWilliams and Rand (1969), A. fowleri Schwartz
(1974), and A. barbouri Schmidt (1919).
Etymology: Named in honor of Karl P. Schmidt
(1890–1957), who made important contributions to
West Indian herpetology (Schmidtanolis is endemic to
Hispaniola), including the naming of species in both of
the primary subclades of the named clade. The name is
a combination of his surname with Anolis, the name of
a more inclusive clade.
Chamaelinorops Schmidt 1919 [nobis], converted clade
name
Synonyms: None.
Definition : The crown clade for which the presence
of both laterally extending zygapophysial plates
connecting the pre- and poszygapophyses of the
thoracolumbar vertebrae and laterally expanded
transverse process of the caudal vertebrae, as inherited
by Anolis barbouri (Schmidt 1919), are apomorphies
relative to other crown clades.
Reference phylogeny: Figure 2 of this study.
Comments: The name Chamaelinorops was proposed
by Schmidt (1919) for the single species C. barbouri
and distinguished from other then-recognized
anole “genera” (Anolis, Norops, Tropidodactylus, and
Chamaeleolis) based on relatively minor d ifferences.
However, the taxon was retained by Etheridge (1960)
on the basis of the unique zygapophysial plates of
the thoracolumbar vertebrae (possibly related to the
extreme compression of the body noted by Schmidt
1919) and the laterally expanded transverse processes
of the caudal vertebrae (both characters described in
detail by Forsgaard 1983) and was recognized by a
number of subsequent authors (e.g., Thomas 1966;
Williams 1976a; Schwartz and Incháustegui 1980; Wyles
and Gorman 1980; Case and Williams 1987; Guyer and
Savage 1986, 1992; Burnell and Hedges 1990; Autumn
and Losos 1997). The finding that Chamaelinorops was
derived from within Anolis led Hass et al. (1993)
to “synonymize” Chamaelinorops with Anolis under
rank-based nomenclature; however, Jackman et al.
(1999) noted that Chamaelinorops could be retained for
a subclade of Anolis under phylogenetic nomenclature.
More recently, Nicholson et al. (2012) applied the name
Chamaelinorops under rank-based nomenclature to a
clade including A. barbouri and several inferred close
relatives, thus changing the reference of that name to
a more inclusive clade that is not d iagnosed by the
distinctive morphological features with which the name
had previously been associated . Because of the long-
standing association of the name Chamaelinorops with
A. barbouri and its d istinctive morphological characters,
we here formalize that association by proposing a
phylogenetic definition based on the distinctive features
(apomorphies) with which that name has come to be
associated , and we propose a new name for the clade
called Chamaelinorops by Nicholson et al. (2012) (see
Schmidtanolis, above).
Inferred composition : Anolis barbouri
(Schmidt 1919).
Etymology: Derived from the Greek chamai (ground),
leon (lion), and norops (bright, flashing, gleaming), in
reference to “its apparent relations with Chamaeleolis
and Norops” (Schmidt 1919: 523).
Xiphosurus Fitzinger 1826 [nobis], converted clade
name
Synonyms: cuvieri series (informal) of Williams (1976a)
and Xiphosurus cuvieri species group of Nicholson et al.
(2012) are partial synonyms referring to a paraphyletic
group originating in approximately the same ancestor
in the context of our inferred phylogeny.
Definition : The most inclusive crown clade containing
Anolis cuvieri Merrem 1820 but not A. auratus Daudin
1802, A. bimaculatus (Sparrman 1784), A. armouri
(Cochran 1934), A. carolinensis Voigt 1832, Anolis
semilineatus Cope 1864, A. vermiculatus Duméril and
Bibron 1837, and A. punctatus Daudin 1802.
Reference phylogeny: Figure 2 of this study.
690 SYSTEMATIC BIOLOGY VOL. 66
Comments: A clade approximating the one here named
Xiphosurus was first inferred by Jackman et al. (1999)
and has been fully or partially corroborated , with
the addition of Anolis eugenegrahami, by Poe (2004),
Nicholson et al. (2005, 2012), Alföld i et al. (2011), and
the present study. The name Xiphosurus was proposed
by Fitzinger (1826) for A. cuvieri but was seldom used
after Boulenger (1885) treated it as a synonym of Anolis
Daudin (1802). Guyer and Savage (1986) applied the
name Semiurus Fitzinger (1843), a younger name also
based on A. cuvieri, to the cuvieri series of Williams
(1976a), a group composed of the giant anoles of
Hispaniola and the Puerto Rico Bank, which now
appears to be paraphyletic, but they later (Savage and
Guyer 1991) replaced Semiurus with the older name
Xiphosurus. Nicholson et al. (2012) applied the name
Xiphosurus to a larger clade including, in addition
to the members of Williams’s (1976a) cuvieri series, A.
christophei, A. eugenegrahami and the species trad itionally
included in Chamaeleolis. Because the name Xiphosurus
was not used by Boulenger (1885) and subsequent
authors, with the exception of Varona (1985), until
it was resurrected by Savage and Guyer (1991), and
because the clade to which this name was applied by
Nicholson et al. (2012) does not have another name, we
here formalize the association of the name Xiphosurus
with that clade by provid ing it with a phylogenetic
definition. The taxon to which Varona (1985) applied the
name Xiphosurus (composed of the equestris, ricordii, and
cuvieri species groups of Williams (1976a) appears to be
polyphyletic.
Inferred composition : ricordii species group (Schwartz
1974; Williams 1976a), Anolis christophei Williams 1960,
A. eugenegrahami Schwartz 1978, Chamaeleolis (Garrido
and Schwartz 1967; Rodríguez-Schettino 1999), A. cuvieri
Merrem 1820.
Etymology: Derived from the Greek xiphos (sword)
and oura (tail), presumably in reference to the crested
tail of adult Anolis cuvieri, upon which the name was
based .
Chamaeleolis Cochran 1838 [nobis], converted clade
name
Synonyms: Pseudochamaeleon Fitzinger 1843, Xiphosurus
chamaeleonides species group of Nicholson et al. (2012).
Definition : The crown clade for which both assignment
of its members to the twig giant ecomorph (including
short limbs and tail and a maximum body size
> 100 mm SVL) and possession of a head casque
formed by posterolateral extensions of the parietal
roof over the upper temporal fenestrae, as inherited
by Anolis chamaeleonides Duméril and Bibron 1837, are
apomorphies relative to other crown clades.
Reference phylogeny: Figure 2 of this study.
Comments: Recognition of Chamaeleolis, a taxon
composed of the distinctive Cuban twig giant
anoles, has a long history (e.g., Cochran 1838;
Cope 1864; O’Shaughnessy 1875; Boulenger 1885;
Barbour and Ramsden 1919; Etheridge 1959; Garrido
and Schwartz 1968; Williams 1976a; Garrido 1982;
Rodriguez-Schettino 1999). Although originally
described for a single species, Anolis chamaeleonides,
that and subsequently described species referred
to Chamaeleolis appear to form a clade (Hass et al.
1993; Jackman et al. 1999; Nicholson et al. 2005, 2012;
this study). The finding that Chamaeleolis is nested
within Anolis and Xiphosurus led authors operating
in the context of rank-based nomenclature (e.g., Hass
et al. 1993; Nicholson et al. 2012) to “synonymize”
the names (i.e., to treat them as if they refer to the
same taxon and therefore no longer use the younger
name Chamaeleolis); however, that rank-based practice
makes little sense phylogenetically. From a phylogenetic
perspective, Chamaeleolis is nested within both Anolis
and Xiphosurus, and those nested relationships can
be preserved by adopting appropriate phylogenetic
definitions of the names. We have emphasized the twig
(giant) ecomorph and the skull casque in our definition
of the name Chamaeleolis because those features give the
lizards in this clade a chameleon-like appearance, as
implied by the name.
Inferred composition : Anolis chamaeleonides Duméril
and Bibron 1837, A. porcus (Cope 1864), A. barbatus
(Garrido 1982), A. guamuhaya (Garrido et al. 1991), and
A. agueroi Diaz et al. (1998).
Etymology: Derived from the Greek chamae (on the
ground) and leo (lion), “que indica sus afinidades con
camaleones” (Cochran 1838: 72), plus the termination
-lis, “que trae[n] á la memoria la del nombre bárbaro de
la familia” (Cochran 1838:73).
Ctenocercus Fitzinger 1843 [nobis], converted clade
name
Synonyms: Anolis of Nicholson et al. (2012).
Definition : The most inclusive crown clade containing
Anolis carolinensis Voigt 1832 but not A. auratus Daudin
1802, A. bimaculatus (Sparrman 1784), A. armouri
(Cochran 1934), A. cuvieri Merrem 1820, A. semilineatus
Cope 1864, A. vermiculatus Duméril and Bibron 1837 and
A. punctatus Daudin 1802.
Reference phylogeny: Figure 2 of this study.
Comments: A clade corresponding to the core of
the one here named Ctenocercus, composed of the
carolinensis, argillaceus, and alutaceus species groups
of Williams (1976a) and A. sheplani (but not A. lucius),
was first inferred by Jackman et al. (1999). This clade
was partially corroborated by Poe (2004), who included
2017 POE ET AL.—EVOLUTION OF ANOLES 691
some additional species from those groups, except that
some of members of the alutaceus species group (A.
cyanopleurus, A. spectrum) were excluded . It was fully
corroborated by Nicholson et al. (2005; see also Alföld i
et al. 2011), who included still more species of the three
species groups. Finally, Nicholson et al. (2012; Fig. 5 but
not Fig. 4) and this study placed A. argenteolus and A.
lucius as sister to the above-described clade, although
with weak support. Nicholson et al. (2012) applied the
name Anolis to the larger clade (i.e., the smallest clade
containing both A. carolinensis and A. lucius). However,
as we have argued above, that name has a much
longer association with the clade of all anoles (lizards
descended from the first one possessing adhesive
toe pads and a dewlap synapomorphic with those in
A. carolinensis). Because a fundamental principle of
biological nomenclature is that a name is not to be used
for more than one taxon (clade), and because the name
Anolis has a much longer association with the clade of
all anoles, we apply the name Anolis to that clade and
resurrect the name Ctenocercus Fitzinger 1843 (based on
A. carolinensis) for the clade to which the name Anolis
was applied by Nicholson et al. (2012). Although there
is an older name, Acantholis Cocteau 1836, that is also
based on a member (A. loysianus) of the clade here
named Ctenocercus, that name is more appropriately
applied to a smaller clade including A. loysianus, such
as the argillaceus species group (see Varona 1985). Note
also that Anolis of Guyer and Savage (1986, 1992) and
Savage and Guyer (1989) is not equivalent to Anolis
of Nicholson et al. (2012) or to our Ctenocercus; in the
context of our inferred phylogeny, Anolis of those earlier
stud ies is a polyphyletic group.
Inferred composition : carolinensis subgroup ofWilliams
(1976a) = carolinensis group of Burnell and Hedges
(1990; see Garrido and Hedges [2001] for more recently
described species related to A. isolepis; the name
Pseudoequestris Varona (1985) is appropriate for the
largest crown clade containing A. isolepis but not A.
carolinensis), argillaceus species group of Williams
(1976a) = argillaceus series of Burnell and Hedges
(1990) = Acantholis sensu Varona (1985); see Navarro
et al. [2001], and Navarro and Garrido [2004], for more
recently described members of this group), angusticeps
subgroup of Williams (1976a) = angusticeps group of
Burnell and Hedges (1990) = Brevicaudata of Varona
(1985); see Estrada and Hedges [1995] and Diaz et al.
[1996] for more recently described members of this
group), sheplani series of Burnell and Hedges (1990),
alutaceus species group of Williams (1976a) = alutaceus
series of Burnell and Hedges (1990) = Macroleptura
of Garrido (1975); see Garrido and Hedges [1992] for
more recently described members of this group), lucius
species group of Williams (1976a) = lucius group of
Burnell and Hedges (1990) = Gekkoanolis of Varona
(1985).
Etymology: Derived from the Greek ktenos (a comb) and
kerkos (tail).
Ctenonotus Fitzinger 1843 [nobis], converted clade
name
Synonyms: cristatellus series (informal) of Gorman et al.
(1980).
Definition : The least inclusive crown clade containing
Anolis bimaculatus (Sparrman 1784), A. cristatellus
Duméril and Bibron 1837 and A. distichus Cope 1861.
Reference phylogeny: Figure 3 of this study.
Comments:A close relationship between the bimaculatus
series (including Anolis distichus) and the cristatellus
series was hypothesized by Etheridge (1959) and
corroborated by subsequent workers (e.g., Gorman et al.
1980; Jackman et al. 1999; Brandley and de Queiroz
2004; Poe 2004; Nicholson et al. 2005, 2012; Alföld i et al.
2011), who removed A. cybotes and its relatives from the
cristatellus series, placed A. distichus and its relatives in
their own series (because it was unclear whether they
were more closely related to A. bimaculatus versus A.
cristatellus), and transferred A. acutus, A. evermanni, and
A. stratulus from the bimaculatus series to the cristatellus
series. The name Ctenonotus (based on Lacerta bimaculata
Sparrman 1784) was originally applied by Fitzinger
(1843) to what now appears to be a polyphyletic group.
It was resurrected for a questionably monophyletic
group composed of the bimaculatus, cristatellus, and
cybotes series by Guyer and Savage (1986), and applied
explicitly, although informally, to the least inclusive
clade containing A. bimaculatus, A. wattsii, A. distichus,
A. cristatellus, and A. evermanni by Brandley and de
Queiroz (2004; a similar use was adopted by Nicholson
et al. [2012] in the context of rank-based nomenclature).
We have defined the name formally using a simplified
version of the informal definition given by Brandley
and de Queiroz (2004).
Inferred composition : bimaculatus series (Lazell 1972;
Gorman and Kim 1976; Schneider et al. 2001), cristatellus
series (Brandley and de Queiroz 2004), distichus series
Burnell and Hedges (1990).
Etymology: Derived from the Greek ktenos, comb, and
notos, back, presumably in reference to the dorsal crest
of lizards of the originally included species.
Norops Wagler 1830 [nobis], converted clade name
Synonyms: Beta section (informal) of Etheridge (1959),
Noropini and Noropina of Varona (1985).
Definition : The crown clade for which the beta type
of caudal vertebrae, in which the autotomic caudal
vertebrae bear long, anterolaterally d irected and distally
bifurcated transverse processes that originate posterior
to the autotomy septa (Etheridge 1959, Etheridge 1967),
692 SYSTEMATIC BIOLOGY VOL. 66
as inherited by Anolis auratus Daudin 1802, is an
apomorphy relative to other crown clades.
Reference phylogeny: Figure 3 of this study.
Comments: Monophyly of the beta anoles was inferred
by Etheridge (1959) on the basis of the unique and
derived morphology of their caudal vertebrae, and
this inference has been corroborated in numerous
subsequent stud ies (e.g., Guyer and Savage 1986;
Jackman et al. 1999; Poe 2004; Nicholson et al. 2005, 2012;
Alföld i et al. 2011; this study). Wagler (1830) proposed
the name Norops for the single species Anolis auratus.
He distinguished Norops from other anole “genera”
that he recognized (Dactyloa, Draconura) by, among
other things, weakly developed toepads. Boulenger
(1885) used Norops for anoles with a toepad in which
the distal phalanges are not raised above the pad ,
which is one component of weak pad development,
and included only A. auratus and A. ophiolepis (in
retrospect, a polyphyletic group). By the time of
Etheridge (1959), A. meridionalis had been added , but
Etheridge did not consider the included species to
be closely related and thus did not recognize Norops.
Savage and Talbot (1978) applied the name Norops to
Etheridge’s (1959) beta section, a use that has been
followed by some subsequent authors (e.g., Guyer and
Savage 1986, 1989; Savage and Guyer 1989; Nicholson
2002; Nicholson et al. 2012). Conveniently, the crown
clade diagnosed by the beta type of caudal vertebrae
is also the smallest crown clade containing the species
previously included in (the polyphyletic) Norops (e.g.,
by Boulenger [1885], and authors just prior to Etheridge
[1959]): A. ophiolepis, A. auratus and A. meridionalis.
Consequently, we apply the name Norops explicitly to
that clade by provid ing it with a phylogenetic definition
based on the derived morphology of the caudal
vertebrae.
Inferred composition : Trachypilus, Placopsis, and
Draconura (see below).
Etymology: Derived from the Greek norops (bright,
flashing, gleaming), presumably in reference to the
coloration of Anolis auratus.
Trachypilus Fitzinger 1843 [nobis], converted clade
name
Synonyms: sagrei species group (informal) of Williams
(1976a), sagrei series (informal) of Burnell and Hedges
(1990) and Nicholson (2002), Norops sagrei species group
of Nicholson et al. (2012). The sagrei series (informal)
of Etheridge (1959) refers to a paraphyletic group
originating in approximately the same ancestor (see
Comments).
Definition : The most inclusive crown clade containing
Anolis sagrei Duméril and Bibron 1837 but not A.
valencienni Duméril and Bibron 1837 and A. chrysolepis
Duméril and Bibron 1837.
Reference phylogeny: Figure 3 of this study.
Comments: Etheridge (1959) recognized the sagrei
series for Anolis sagrei and its close relatives with the
exception of A. ophiolepis, although he noted that A.
ophiolepis and A. valencienni were not d istinguishable
from the members of the sagrei series in terms of the
osteological characters that he stud ied . Williams (1976a)
assigned both A. ophiolepis and A. valencienni to the
sagrei series. Chromosomal evidence led Gorman and
Atkins (1968) to question the referral of A. valencienni
to the sagrei series, and Gorman (1973) later to remove
it (see Comments on Placopsis for additional details).
Subsequently, except for the addition of newly described
species, the composition of the sagrei series has been
stable (e.g., Burnell and Hedges 1990; Rodriguez-
Schettino 1999; Cádiz et al. 2013). Monophyly of the
sagrei series was inferred by Guyer and Savage 1986 from
then available karyological data, and it has subsequently
been corroborated repeated ly by DNA sequence data
sampled from increasing numbers of species and genes
(Hass et al. 1993; Jackman et al. 1999; Nicholson 2002;
Poe 2004; Nicholson et al. 2005, 2012; Alföld i et al. 2011;
this study). Varona (1985) applied the name Trachypilus
Fitzinger 1843 to the Cuban beta anoles (sagrei series
or species group) except A. ophiolepis—that is, to a
paraphyletic group originating in the same ancestor
as that of the sagrei series of more recent authors (e.g.,
Burnell and Hedges 1990; Nicholson 2002). Although
the name has rarely been used since then, because a
clade composed of the members of the sagrei series has
been inferred repeated ly, consistently and with strong
support, we here formalize the application of the name
Trachypilus to the sagrei series by provid ing it with a
phylogenetic definition.
Inferred composition : sagrei series (Nicholson 2002) or
species group (Rodriguez-Schettino 1999).
Etymology: Derived from the Greek trachys (rough)
and pilos (hair, cap, ball), presumably in reference to the
keeled scales in the parietal region of Anolis sagrei.
P lacopsis Gosse 1850 [nobis], converted clade name
Synonyms: grahami series (informal) of Shochat and
Dessauer (1981; see also Savage and Guyer 1989; Burnell
and Hedges 1990), Norops valencienni species group of
Nicholson et al. (2012).
Definition : The most inclusive crown clade containing
Anolis valencienni Duméril and Bibron 1837 but not
A. sagrei Duméril and Bibron 1837 and A. chrysolepis
Duméril and Bibron 1837.
Reference phylogeny: Figure 3 of this study.
2017 POE ET AL.—EVOLUTION OF ANOLES 693
Comments: Based on skeletal morphology, Etheridge
(1959) proposed the grahami series for the native beta
anoles of Jamaica (including Anolis conspersus of the
Cayman Islands but not including A. sagrei, a species
of Cuban origin) with the exception of A. valencienni,
which he considered more closely related to the sagrei
series (see also Williams 1976a). When A. valencienni was
found to have an identical karyotype to members of the
grahami series, Gorman and Atkins (1968) questioned
its relationship to the sagrei series, and Gorman (1973)
placed it in a series of its own. Based on immunological
d istances, Shochat and Dessauer (1981) transferred A.
valencienni to the grahami series, and monophyly of
the grahami series, including A. valencienni, has been
corroborated by numerous subsequent stud ies (Hedges
and Burnell 1990; Jackman et al. 1999, Jackman et al.
2002; Nicholson 2002; Poe 2004; Nicholson et al. 2005,
2012, this study). Because the monophyly of the grahami
series has been inferred in numerous studies, and
because the group does not have a formal name, we
here select such a name for the grahami series. There
are two preexisting names based on a member of the
grahami series: Xiphocercus Fitzinger 1843 and Placopsis
Gosse 1850 (both based on A. valencienni), neither of
which has been used for over 50 years and neither
of which has been applied previously to the grahami
series as a whole. Xiphocercus has been used more
recently, but it was applied to what is now considered
a polyphyletic group composed of A. valencienni and
either A. heterodermus Boulenger (1885) or A. darlingtoni
(Cochran 1935). Although Xiphocercus is older, it is
similar in both spelling and etymology to Xiphosurus,
the name of another anole clade. In order to avoid
confusion with that other clade, we have selected the
younger name Placopsis and applied it to the grahami
series with a phylogenetic definition.
Inferred composition : grahami series (Burnell and
Hedges 1990; Hedges and Burnell 1990; Jackman et al.
2002).
Etymology: Derived from the Greek plakos (a broad
plate) and opsis (the face), presumably in reference to the
large, flat scales in the frontal region of Anolis valencienni.
Draconura Wagler 1830 [nobis], converted clade name
Synonyms: Norops auratus species group of Nicholson
et al. (2012).
Definition : The most inclusive crown clade containing
Anolis chrysolepis Duméril and Bibron 1837 but not
A. sagrei Duméril and Bibron 1837 and A. valencienni
Duméril and Bibron 1837.
Reference phylogeny: Figure 4 of this study.
Comments: Monophyly of the subclade of Norops (beta
anoles) composed of all species except the members of
the (predominantly) Cuban Trachypilus and the Jamaican
Placopsis clades was inferred from a limited sample
by Jackman et al. (1999) and has been corroborated
subsequently with larger samples of both species and
characters (Poe 2004; Nicholson et al. 2005, 2012; Alföld i
et al. 2011; this study). The members of this species-rich
clade are predominantly mainland forms. The oldest
pre-existing higher taxon names based on members
of this clade are Draconura (for Anolis chrysolepis) and
Norops (for A. auratus), both proposed by Wagler (1830).
Norops has come to be associated with the (more
inclusive) clade of anoles with the beta type of caudal
vertebrae (see above). In contrast, Draconura has been
little used for over 100 years, since Boulenger (1885)
treated it as a synonym of Anolis (e.g., Schmidt 1919;
Barbour 1923; Dunn 1939; Etheridge 1959; Williams
1976a,b; Guyer and Savage 1986; Savage and Guyer
1989; Nicholson 2002; Poe 2004; Nicholson et al. 2012).
The lone exception was Varona (1985), who applied
the name Draconura to what now appears to be a
polyphyletic group composed of (at least) A. chrysolepis
and Cuban grass anoles of the alutaceus species group or
series (= Macroleptura; see Ctenocercus, above). Because
the “mainland” clade within the beta anoles has been
supported repeated ly but currently lacks a formal name,
and because the name DraconuraWagler 1830 is based on
a member of that clade and does not have a conflicting
trad itional use, we here establish that name for the
clade of all beta anoles that are more closely related
to A. chrysolepis than to A. sagrei and A. valencienni by
provid ing it with a phylogenetic definition.
Inferred composition : Norops auratus species group
(Nicholson et al. 2012).
Etymology:Derived from the Greek drakon (dragon) and
oura (tail).
REFERENCES
Alföld i J., Palma F.D., Grabherr M., Williams M., Kong L., Mauceli E.,
Russell P., Lowe C.B., Glor R.E., Jaffe J.D., Ray D.A., Boissinot S.,
Shedlock A.M., Botka C., Castoe T.A., Colbourne J.K., Fujita M.K.,
Moreno R.G., ten Hallers B.F., Haussler D., Heger A., Heiman D.,
Janes D.E., Johnson J., de Jong P.J., Koriabine M.Y., Lara M., Novick
P.A., Organ C.L., Peach S.E., Poe S., Pollock D.D., de Queiroz K.,
Sanger T., Searle S., Smith J.D., Smith Z., Swofford R., Turner-Maier
J.,Wade J., Young S., Zadissa A., Edwards S.V.,Glenn T.C., Schenider
C.J., Losos J.B., Lander E.S., Breen M., Ponting C.P., Lindblad-Toh
K. 2011. The genome of the green anole lizard and a comparative
analysis with birds and mammals. Nature 477:587–591.
Andrews R. 1971. Structural habitat and time budget of a tropical Anolis
lizard . Ecology 52:262–270.
Antonelli A., Nylander J.A., Persson C., Sanmartín I. 2009. Tracing the
impact of the Andean uplift on Neotropical plant evolution. Proc.
Natl Acad . Sci. 106:9749–9754.
Autumn K., Losos J.B. 1997. Notes on jumping ability and thermal
biology of the enigmatic anole Chamaelinorops barbouri. J. Herpetol.
31:442–444.
Bacon C.D., Silvestro D., Jaramillo C., Smith B.T., Chakrabarty P.,
Antonelli A. 2015. Biological evidence supports an early and
complex emergence of the Isthmus of Panama. Proc. Natl Acad . Sci.
112:6110–6115.
694 SYSTEMATIC BIOLOGY VOL. 66
Barbour T. 1920. A note on Xiphocercus. Proc. New Engl. Zool. Club
7:61–63.
Barbour T. 1923. West Indian investigations of 1922. Occ. Pap. Mus.
Zool. Univ. Mich. 132:1–9.
Barbour T., Ramsden C.T. 1919. The herpetology of Cuba. Mem. Mus.
Comp. Zool., Harvard Coll. 47:69–213.
Boulenger G.A. 1885. Catalogue of the Lizards in the British Museum
(Natural History). Vol. 2, 2nd edn. London: Stechert-Hafner Service
Agency.
Brandley M.C., de Queiroz K. 2004. Phylogeny, ecomorphological
evolution, and historical biogeography of the Anolis cristatellus
series. Herp. Monogr. 18:90–125.
Bundy D.A.P., Vogel P., Harris E.A. 1987. Helminth parasites of
Jamaican anoles (Reptilia: Iguanidae): a comparison of the helminth
fauna of 6 Anolis species. J. Helminthol. 61:77–83.
Burnell K.L., Hedges S.B. 1990. Relationships of West Indian Anolis
(Sauria: Iguanidae): an approach using slow evolving protein loci.
Caribb. J. Sci. 26:7–30.
Cádiz A., Nagata N., Katabuchi M., Díaz L.M., Echenique-Díaz L.M.,
Akashi H.D., Makino T., Kawata M. 2013. Relative importance
of habitat use, range expansion, and speciation in local species
d iversity of Anolislizards in Cuba. Ecosphere 4(7):78.
Cannatella D.C., de Queiroz K. 1989. Phylogenetic systematics of the
anoles: is a new taxonomy warranted? Syst. Zool. 38:57–69.
Cantino P.D., de Queiroz K. 2014. International code of
phylogenetic nomenclature. Version 5. Available from:
https:/ / www.ohio.edu/ phylocode/ .
Case S., Williams E.E. 1987. The cybotoid anoles and Chamaelinorops
lizards (Reptilia: Iguanidae): evidence of mosaic evolution. Zool. J.
Linn. Soc. Lond. 91:325–341.
Castañeda M. del R., de Queiroz K. 2011. Phylogenetic relationships
of the Dactyloa clade of Anolis lizards based on nuclear and
mitochondrial DNA sequence data. Mol. Phylogenet. Evol. 61:784–
800.
Castañeda M.R., de Queiroz K. 2013. Phylogeny of the Dactyloa clade
of Anolis lizards: new insights from combining morphological and
molecular data. Bull. Mus. Comp. Zool. 160:345–398.
Castañeda M.R., Sherratt E., Losos J.B. 2014. The Mexican amber anole,
Anolis electrum, within a phylogenetic context: implications for the
origins of Caribbean anoles. Zool. J. Linn. Soc. Lond. 172:133–144.
Castoe T.A., Daza J.M., Smith E.N., Sasa M.M., Kuch U., Campbell J.A.,
Chippindae P.T, Parkinson C.L. 2009. Comparative phylogeography
of pitvipers suggests a consensus of ancient Middle American
highland biogeography. J. Biogeogr. 36:88–103.
Castroviejo-Fisher S., Guayasamin J.M., Gonzalez-Voyer A., Vila
C. 2014. Neotropical d iversification seen through glassfrogs. J.
Biogeogr. 41:66–80.
Coates A.G., Obando J.A. 1996. The geological evolution of the Central
American Isthmus. In: Jackson J.B.S., Budd A.F., Coates A.G.,
ed itors. Evolution and Environment in Tropical America. Chicago:
University of Chicago Press.
Cochran D.M. 1934. Herpetological collections made in Hispaniola
by the Utowana Expedition, 1934. Occ. Pap. Boston Soc. Nat. Hist.
8:163–188.
Cochran D.M. 1935. New reptiles and amphibians collected in Haiti by
P.J. Darlington. Proc. Boston Soc. Nat. Hist. 40:367–376.
Cochran D.M. 1939. Diagnoses of three new lizards and a frog from
the Dominican Republic. Proc. New England Zoolog. Club 18:1–3.
Cochran D.M. 1941. The herpetology of Hispaniola. Bull. USNatl. Mus.
Vol. 177.
Cocteau J.T. 1838. (“1843”). Reptiles in R. de La Sagra, Historia Fisica,
Politica y Natural de la Isla de Cuba. Tomo IV. Paris, Arthus
Bertrand .
Collette B.B. 1961. Correlations between ecology and morphology in
anoline lizards from Havana, Cuba and southern Florida. Bull.Mus.
Comp. Zool. 125:137–162.
Conrad J.L., Rieppel O., Grande L. 2007. A Green River (Eocene)
polychrotid (Squamata: Reptilia) and a re-examination of iguanian
systematics. J. Paleontol. 81:1365–1373.
Cope E.D. 1864. Contributions to the herpetology of tropical America.
Proc. Acad . Natl Sci. Phila. 16:166–176.
Cope E.D. 1900. The crocodilians, lizards and snakes ofNorth America.
Ann. Rept. U.S. Natl Mus. 2:151–1270.
Creer D.A., de Queiroz K., Jackman T.R., Losos J.B., Larson A. 2001.
Systematics of the Anolis roquet series of the southern Lesser Antilles.
J. Herpetol. 35:428-441.
Cuvier G.L.C.F.D. 1817 (“1816”). Le régne animal d istribué d’après son
organization. Vol 2. Paris: Déterville.
Daudin F.M. 1802. Histoire naturelle, générale et particulière, des
reptiles. Tome IV. Paris, F. Dufart.
Daza J.M., Castoe T.A., Parkinson C.L. 2010. Using regional
comparative phylogeographic data from snake lineages to infer
historical processes in Middle America. Ecography 33:343–354.
de Queiroz K., Gauthier J. 1990. Phylogeny as a central principle
in taxonomy: phylogenetic definitions of taxon names. Syst. Zool.
39:307–322.
de Queiroz K., Gauthier J. 1992. Phylogenetic taxonomy. Annu. Rev.
Ecol. Syst. 23:449–480.
de Queiroz K., Chu L.R., Losos J.B. 1998. A second Anolis lizard in
Dominican amber and the systematics and ecological morphology
of Dominican amber anoles. Am. Mus. Novit. 3249:1–23.
Diaz L.M., Navarro N., Garrido O.H. 1998. Nueva especie de
Chamaeleolis (Sauria: Iguanidae) de la Meseta de Cabo Cruz,
Granma, Cuba. Avicennia 8/ 9:27–34.
Diaz L.M., Estrada A.R., Moreno L.V. 1996. A new species of Anolis
(Sauria: Iguanidae) from the Sierra de Trinidad , Sancti Spiritus,
Cuba. Carib. J. Sci. 32:54–58.
Dobson A.P., Pacala S.V., Roughgarden J.D., Carper E.R., Harris E.A.
1992. The parasites of Anolis lizards in the northern Lesser Antilles.
Oecologia 91:110–117.
Drummond A.J., Suchard M.A., Xie D., Rambaut A. 2012. Bayesian
phylogenetics with BEAUti and the BEAST 1.7. Mol. Biol. Evol.
29:1969–1973.
Duméril A.M.C., Bibron G. 1837. Erpétologie général ou histoire
naturelle complète des reptiles, Vol. 4. Paris: Roret.
Dunn E.R. 1939. Zoological results of the George Vanderbilt South
Pacific expedition of 1937. Part III, The lizards of Malpelo Island ,
Colombia. Acad . Nat. Sci. Philadelphia 4:1–3.
Estrada A.R., Hedges S.B. 1995. A new species of Anolis (Sauria:
Iguanidae) from eastern Cuba. Carib. J. Sci. 31:65–72.
Etheridge R.E. 1959. The relationships of the anoles (Reptilia:
Sauria: Iguanidae): an interpretation based on skeletal morphology
(Thesis). University of Michigan, U.S.A.
Etheridge R.E. 1967. Lizard caudal vertebrae. Copeia 4:699–721.
Farris D.W., Jaramillo C., Bayona G., Restrepo-Moreno S.A., Montes
C., Cardona A., Mora A., Speakman R.J., Glascock M.D., Valencia V.
2011. Fracturing of the Panamanian Isthmus during initial collision
with South America. Geology 39:1007–1010.
Fitch H.S., Echelle A.F, Echelle A.A. 1976. Field observations on rare or
little known mainland anoles. Univ. Kans. Sci. Bull. 51:91–128
Fitzinger L.J. 1826. Neue Classification der Reptilien nach ihren
natürlichen Verwandtschaften. Nebst einer Verwandtschafts-Tafel
und einem Verzeichnisse der Reptilien-Sammlung des K. K.
zoologischen Museum’s zu Wien. Vienna: J.G. Hübner.
Fitzinger L.J. 1843. Systema reptilium. Vienna: Braumüller and Seidel.
Fleishman L.J., Bowman M., Saunders D., Miller W.E., Rury M.J., Loew
E.R. 1997. The visual ecology of Puerto Rican anoline lizards: habitat
light and spectral sensitivity. J. Comp. Physiol. A 181:446–460.
Forsgaard K. 1983. The axial skeleton of Chamaelinorops. In Rhodin A.G.
and Miyata K. (eds.), Advances in Herpetology and Evolutionary
Biology: Essays in Honor of Ernest E. Williams. Cambridge: Mus.
Comp. Zool. Harvard Univ.
Gamble T., Geneva A.J., Glor R.E., Zarkower D. 2014. Anolis sex
chromosomes are derived from a single ancestral pair. Evolution
68:1027–41.
Garrido O.H., Pérez-Beato O., Moreno L.V. 1991. Nueva especie de
Chamaeleolis (Lacertilia: Iguanidae) para Cuba. Carib. J. Sci. 27
(3-4):162–168.
Garrido O.H. 1975. Distribución y variación del complejo Anolis
cyanopleurus (Lacertilia: Iguanidae) en Cuba. Poeyana 143:1–58.
Garrido O.H. 1982. Descripción de una nueva especie cubana
de Chamaeleolis (Lacertilia: Iguanidae), con notas sobre su
comportamiento. Poeyana 236:1–25.
Garrido O.H., Hedges S.B. 2001. A new anole from the northern slope
of the Sierra Maestra in eastern Cuba (Squamata: Iguanidae). J.
Herpetol. 35:378–383.
2017 POE ET AL.—EVOLUTION OF ANOLES 695
Garrido O.H., Schwartz A. 1968. Cuban lizards of the genus
Chamaeleolis. Quart. J. Florida Acad . Sci. 30:197–220.
Gauthier J.A. Kluge A.G., Rowe T. 1988. Amniote phylogeny and the
importance of fossils. Clad istics 4(2):105–209.
Glor R.E., Gifford M.E., Larson A., Losos J.B., Rodríguez Schettino L.,
Chamizo Lara A.R., Jackman T.R. 2004. Partial island submergence
and speciation in an adaptive rad iation: a multilocus analysis of the
Cuban green anoles. Proc. R. Soc. Lond. B 271:2257–2265.
Glor R.E., Losos J.B., Larson A. 2005. Out of Cuba: overwater d ispersal
and speciation among lizards in the Anolis carolinensis subgroup.
Mol. Ecol. 14:2419–2432.
Goldberg E.E., Roy K., Lande R., Jablonski D. 2005. Diversity,
endemism, and age distributions in macroevolutionary sources and
sinks. American Naturalist 165:623–633.
Gorman G.C. 1973. The chromosomes of the Reptilia, a cytotaxonomic
interpretation. In Chiarelli A.B., Capanna E., ed itors. Cytotaxonomy
and Vertebrate Evolution. London: Academic Press.
Gorman G.C., Atkins L. 1967. The relationships of Anolis of the roquet
group (Sauria: Iguanidae). II. Comparative chromosome cytology.
Syst. Zool. 16:137–143.
Gorman G.C., Atkins L. 1968. New karyotypic data for 16 species of
Anolis (Sauria: Iguanidae) from Cuba, Jamaica, and the Cayman
Islands. Herpetologica 24:13–21.
Gorman G.C., Kim Y.S. 1976. Anolis lizards of the eastern
Caribbean: a case study in evolution.II. Genetic relationships
and genetic variation of the bimaculatus group. Syst. Zool. 20:
167–185.
Gorman G.C., Buth D., Soule M., Yang SY. 1980. The relationships of
the Anolis cristatellus species group: electrophoretic analysis. J of
Herpetology 14:269–78.
Gorman G.C., Buth D.B., Soule M., Yang S. 1983. The relationships
of Puerto Rican Anolis: electrophoretic and karyotypic stud ies.
In Rhodin A., Miyata K., ed itors. Advances in herpetology
and evolutionary biology. Cambridge: Museum of Comparative
Zoology.
Gorman G.C., Lieb C.S., Harwood R.H. 1984. The relationships of
Anolis gadovi: albumin immunological evidence. Carib. J. Sci. 20:
145–152.
Gregory-Wodzicki K.M. 2000. Uplift history of the Central and
Northern Andes: a review. Geol. Soc. Am. Bull. 112:1091–1105.
Guyer C, Savage J.M. 1986. Clad istic relationships among anoles
(Sauria: Iguanidae). Syst. Zool. 35:509–531.
Guyer C., Savage J.M. 1992. Anole systematics revisited . Syst. Biol. 1:
89–110.
Hass C.A., Hedges S.B., Maxson L.R. 1993. Molecular insights into
the relationships and biogeography of West Indian anoline lizards.
Biochemical Systematics and Ecology 21:97–114.
Hedges S.B., Burnell K.L. 1990. The Jamaican rad iation of Anolis
(Sauria: Iguanidae): an analysis of relationships and biogeography
using sequential electrophoresis. Carib. J. Sci. 26:31–44.
Helmus M.R., Mahler D.L., Losos J.B. 2014. Island biogeography of the
Anthropocene. Nature 513:543–547.
Hertz P.E. 1976. Anolis alumina, new species of grass anole from the
Barahona Peninsula of Hispaniola. Breviora 437:1–19.
Hertz P.E. 1992. Temperature regulation in Puerto Rican Anolis lizards:
a field test using null hypotheses. Ecology 73:1405–1417.
Huelsenbeck J.P., Ronquist F. 2001. MRBAYES: Bayesian inference of
phylogenetic trees. Bioinformatics. 17:754–755.
Huelsenbeck J.P., Larget B., Alfaro M.E. 2004. Bayesian phylogenetic
model selection using reversible jump Markov chain Monte Carlo.
Mol. Biol. Evol. 2004:1123–1133.
Iturralde-Vinent M.A. 2006. Meso-Cenozoic Caribbean
paleogeography: implications for the historical biogeography
of the region. Int. Geol. Rev. 48:791–827.
Iturralde-Vinent M.A., MacPhee R.D. 1999. Paleogeography of the
Caribbean region: implications for Cenozoic biogeography. Bull.
Am. Mus. Nat. Hist. 238:1–95.
Jackman T.R., Larson A., de Queiroz K., Losos J.B. 1999. Phylogenetic
relationships and tempo of early d iversification in Anolis lizards.
Syst. Biol. 48:254–285.
Jackman T.R., Irschick D.J., de Queiroz K., Losos J.B., Larson A. 2002.
Molecular phylogenetic perspective on evolution of lizards of the
Anolis grahami series. J. Exp. Zool. 294:1–16.
Klutsch C.F.C., Misof B., Grosse W.R., Moritz R.F.A. 2007. Genetic
and morphometric d ifferentiation among island populations of two
Norops lizards (Reptilia: Sauria: Polychrotidae) on independently
colonized islands of the Islas de Bahia Honduras. J. Biogeogr.
34:1124–1135.
Knouft J.H., Losos J.B., Glor R.E., Kolbe J.J. 2006. Phylogenetic analysis
of the evolution of the niche in lizards of the Anolis sagrei group.
Ecology 87:S29–S38.
Köhler G. 2010. A revision of the Central American species related
to Anolis pentaprion with the resurrection of A. beckeri and the
description of a new species (Squamata: Polychrotidae). Zootaxa
2354:1–18.
Lanfear R., Calcott B., Ho S.Y.W., Guindon S. 2012. Partitionfinder:
combined selection of partitioning schemes and substitu tion models
for phylogenetic analyses. Mol. Biol. Evol. 29:1695–1701.
Lazell J.D. 1965. An Anolis (Sauria, Iguanidae) in amber. J. Paleontol.
39:379–382.
Lazell J.D. 1972. The anoles (Sauria, Iguanidae) of the Lesser Antilles.
Bull. Mus. Comp. Zool. 143:1–115.
Losos J.B. 1990. A phylogenetic analysis of character d isplacement in
Caribbean Anolis lizards. Evolution 44:558–569.
Losos J.B. 2011. Lizards in an evolutionary tree: ecology and adaptive
rad iation of anoles. Berkeley: University of California Press.
Losos J.B., Jackman T.R., Larson A., de Queiroz K.,Rodríguez-Schettino
L. 1998. Contingency and determinism in replicated adaptive
rad iations of island lizards. Science 279:2115–2118.
MacPhee, R.D., Iturralde-Vinent, M.A. 2005. The interpretation
of Caribbean paleogeography: reply to Hedges. In Proceedings
of the International Symposium ‘Insular Vertebrate Evolution: the
Palaeontological Approach (Vol. 12, pp. 175–184). Monagrafies de la
Societat d’Història Naturel de les Balears.
Magallón S.A. 2004. Dating lineages: molecular and paleontological
approaches to the temporal framework of clades. Int. J. Plant Sci.
165(S4), S7–S21.
Magallón S.A. 2010. Using fossils to break long branches in molecular
dating: a comparison of relaxed clocks applied to the origin of
angiosperms. Syst. Biol. 59: 384–399.
Matzke N.J. 2013a. Probabilistic historical biogeography: new models
for founder-event speciation, imperfect detection, and fossils allow
improved accuracy and model-testing. Front. Biogeogr. 5:242–248.
Matzke N.J. 2013b. BioGeoBEARS: BioGeography with Bayesian
(and Likelihood) Evolutionary Analysis in R Scripts. R package,
version 0.2.1. 2013. Available from: http:/ / CRAN.R-project.org/
package=BioGeoBEARS.
Matzke N.J. 2014. Model selection in historical biogeography reveals
that founder-event speciation is a crucial process in island clades.
Syst. Biol. 63:951–970.
Mello B., Schrago, C.G. 2014. Assignment of calibration information to
deeper phylogenetic nodes is more effective in obtaining precise
and accurate d ivergence time estimates. Evol. Bioinform. Online
10:79.
Merrem B. 1820. Versuch eines Systems der Amphibien. Marburg: J.C.
Kreiger.
Miller M.J., Bermingham E., Klicka J., Escalante P., do Amaral F.S.,Weir
J.T., Winker K. 2008. Out of Amazonia again and again: episodic
crossing of the Andes promotes d iversification in a lowland forest
flycatcher. Proc. Roy. Soc. B 275:1133–1142.
Moyle R.G., Andersen M.J., Oliveros C.H., Steinheimer F.D., Reddy
S. 2012. Phylogeny and biogeography of the core babblers (Aves:
Timaliidae). Syst. Biol. 61:631–651.
Mulcahy D.G., Noonan B.P.,Moss T., Townsend T.M., Reeder T.W., Sites
J.W., Wiens J.J. 2012. Estimating divergence dates and evaluating
dating methods using phylogenomic and mitochondrial data in
squamate reptiles. Mol. Phylogen. Evol. 65:974–991.
Myers C.M. 1971. Central American lizards related to Anolis pentaprion:
two new species from the Cordillera de Talamanca. Am.Mus. Novit.
2471:1–40.
Nicholson K.E. 2002. Phylogenetic analysis and a test of current
infrageneric classification of Norops (Beta Anolis). Herp. Mono.
16:93–120.
Nicholson K.E., Glor R.E., Kolbe J.J., Larson A., Hedges S.B., Losos
J.B. 2005. Mainland colonization by island lizards. J. Biogeogr. 32:
929–938.
696 SYSTEMATIC BIOLOGY VOL. 66
Nicholson K.E., Crother B.I., Guyer C., Savage J.M. 2012. It is time for
a new classification of anoles. Zootaxa 3477:1–108.
Nicholson K.E., Crother B.I., Guyer C., Savage J.M. 2014. Anole
classification: a response to Poe. Zootaxa 3814:109–120.
O’Shaughnessy A.W.E. 1875. List and revision of the species of
Anolidae in the British Museum collection,with descriptions of new
species. Ann. Mag. Nat. Hist. 15:270–281.
Ord T.J., Martins E.P. 2006. Tracing the origins of signal d iversity in
anole lizards: phylogenetic approaches to inferring the evolution of
complex behaviour. Anim. Behav. 71:1411–1429.
Pacala S.W., Roughgarden J. 1982. The evolution of resource
partitioning in a multid imensional resource space. Theor. Popul.
Biol. 22:127–145.
Pacheco N.N., Fernández A., Garrido O.H. 2001. Reconsideración
taxonómica de Anolis centralis litoralis y descripción de una
nueva especie del grupo argillaceus (Sauria: Iguanidae) para Cuba.
Solenodon 1:66–75.
Phillips J.G., Deitloff J., Guyer C., Huetteman S., Nicholson K.E. 2015.
Biogeography and evolution of a widespread Central American
lizard species complex: Norops humilis, (Squamata: Dactyloidae).
BMC Evol. Biol. 15:143–156.
Pinto G., Mahler D.L., Harmon L.J., Losos J.B. 2008. Testing the island
effect in adaptive rad iation: rates and patterns of morphological
d iversification in Caribbean and mainland Anolis lizards. Proc. R.
Soc. Lond. B 275:2749–2757.
Poe S. 2004. Phylogeny of anoles. Herp. Mono. 18:37–89.
Poe S. 2012. Anolis lizards of Parque Omar Torrijos (El Copé) Panama
2012.0. stevenpoe.net.
Poe S. 2013. 1986Redux:New genera of anoles (Squamata:Dactyloidae)
are unwarranted . Zootaxa 3626:295–299.
Poe S., Latella I.M., Ayala-Varela F., Yañez-Miranda C., Torres-Carvajal
O. 2015.A new species of phenacosaur Anolis (Squamata; Iguanidae)
from Peru and a comprehensive phylogeny of Dactyloa-clade Anolis
based on new DNA sequences and morphology. Copeia 103:
639–650.
Prates I., Rodrigues M.T., Melo-Sampaio P.R., Carnaval A.C. 2015.
Phylogenetic relationships of Amazonian anole lizards (Dactyloa):
taxonomic implications, new insights about phenotypic evolution
and the timing of d iversification. Mol. Phylogen. Evol. 82:
258–268.
Pyron R.A., Burbrink F.T., Wiens J.J. 2013. A phylogeny and revised
classification of Squamata, including 4161 species of lizards and
snakes. BMC Evol. Biol. 13:93.
Rambaut A., Drummond A.J. 2007. Tracer: MCMC trace analysis tool
v1.4.1. Available: http:/ / tree.bio.ed .ac.uk/ software.
Rambaut A., Suchard M.A., Xie D., Drummond A.J. 2014. Tracer v1.6,
Available from http:/ / beast.bio.ed .ac.uk/ Tracer.
Rand A.S. 1964. Ecological d istribution in Anoline lizards of Puerto
Rico. Ecology 45:745–752.
Rand A.S., Williams E.E. 1970. An estimation of redundancy and
information content of anole dewlaps. Am. Nat. 104:99–103.
Ree R.H., Sanmartín I. 2009. Prospects and challenges for parametric
models in historical biogeographical inference. J. Biogeogr. 36:
1211–1220.
Robinson D.R, Foulds L.R. 1981. Comparison of phylogenetic trees.
Math. Biosci. 53:131–147.
Rodríguez-Robles J., Jezkova T., Garíca M.A. 2007. Evolutionary
relationships and historical biogeography of Anolis desechensis and
Anolis monensis, two lizards endemic to small islands in the eastern
Caribbean Sea. J. Biogeogr. 34:1546–1558.
Rodríguez-Schettino L. 1999. The iguanid lizards of Cuba. Gainesville:
University Press of Florida.
Ronquist F., Teslenko M., Mark P., Ayres D.L., Darling A., Höhna
S., Larget B., Liu L., Suchard M.A., Huelsenbeck J.P. 2012.
MrBayes 3.2.2: efficient Bayesian phylogenetic inference and model
choice across a large model space. Syst. Biol. 61:539–542. http:/ /
mrbayes.sourceforge.net/ .
Rosen D.E. 1975. A vicariance model of Caribbean biogeography. Syst.
Zool. 24:431–464.
Ruibal R. 1961. Thermal relations of five species of tropical lizards.
Evolution 15:98–111.
Sanderson M.J., Shaffer H.B. 2002. Troubleshooting molecular
phylogenetic analyses. Ann. Rev. Ecol. Syst. 33:49–72.
Santos J.C., Coloma L.A., Summers K., Caldwell J.P., Ree R.,
Cannatella D.C. 2009. Amazonian amphibian diversity is primarily
derived from late Miocene Andean lineages. PLoS Biol. 7:
e1000056.
Savage J.M. 1980. A handlist with preliminary keys to the herpetofauna
of Costa Rica. Los Angeles: Alan Hancock Foundation (University
of Southern California).
Savage J.M. 1982. The enigma of the Central American herpetofauna:
d ispersal or vicariance? Ann. Mo. Bot. Gard . 69:464–547.
Savage J.M., Talbot J.J. 1978. The giant anoline lizards of Costa Rica and
Western Panama. Copeia 1978:480–492.
Savage J.M., Guyer C. 1989. Infrageneric classification and species
composition of the anole genera, Anolis, Ctenonotus, Dactyloa,
Norops and Semiurus (Sauria: Iguanidae). Amphibia- Reptilia
10:105–116.
Savage J.M. and Guyer C. 1991. Nomenclatural notes on anoles (Sauria:
Polychridae): stability over priority. Journal of Herpetology 25:365–
366.
Schmidt K.P. 1919. Descriptions of new amphibians and reptiles
from Santo Domingo and Navassa. Bull. Am. Mus. Nat. Hist. 41:
519–525.
Schneider C.J., Losos J.B., de Queiroz K. 2001. Evolutionary
relationships of the Anolis bimaculatus group from the Northern
Lesser Antilles. J. Herpetol. 35:1–12.
Schoener T.W. 1970. Size patterns in West Indian Anolis lizards.
II. Correlations with sizes of particular sympatric species-
d isplacement and convergence. Am. Nat. 104:155–174.
Schwartz A. 1974. An analysis of variation in the Hispaniolan giant
anole, Anolis ricordi Duméril and Bibron. Bull. Mus. Comp. Zool.
Harvard 146:86–146.
Schwartz A., Garrido O.H. 1972. The lizards of the Anolis equestris
complex in Cuba. Stud . Fauna Curaçao and Carib. Is. 39:1–86.
Schwartz A., Henderson R.W. 1991. Amphibians and reptiles of the
West Indies. Gainesville: University of Florida Press.
Schwartz A., Incháustegui S.J. 1980. The endemic Hispaniolan lizard
genus Chamaelinorops. J. Herpetol. 14:51–56.
Sedano R.E., Burns K.J. 2010. Are the Northern Andes a species pump
for Neotropical birds? Phylogenetics and biogeography of a clade
of Neotropical tanagers (Aves: Thraupini). J. Biogeogr. 37:325–343.
Shochat D., Dessauer H.C. 1981. Comparative immunological study of
albumins of Anolis lizards of the Caribbean islands.Comp.Biochem.
Physiol. 68:67–73.
Smith B.T., McCormack J.E., Cuervo A.M., Hickerson M.J., Aleixo A.,
Cadena C.D., Pérez-Emán J., Burney C.W., Xie X., Harvey M.G.,
Faircloth B.C., Glenn T.C., Derryberry E.P., Prejean J., Fields S.,
Brumfield R.T. 2014. The drivers of tropical speciation. Nature
515:406–409.
Sparrman A. 1784. Lacerta bimaculata, en ny Ödla från America.
Stockholm: Kongliga Vetenskaps Academiens Nya Handlingar.
Stamatakis A. 2006. RAxML-VI-HPC: maximum likelihood-based
phylogenetic analyses with thousands of taxa and mixed models.
Bioinformatics 22:2688–2690.
Suchard M.A., Huelsenbeck J.P. 2012. MrBayes 3.2.2: efficient Bayesian
phylogenetic inference and model choice across a large model space.
Syst. Biol. 61:539–542. http:/ / mrbayes.sourceforge.net/
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.
Thiele K. 1993. The holy grail of the perfect character: the clad istic
treatment of morphometric data. Clad istics 9: 268–275.
Thomas R. 1966.A reassessment of the herpetofauna ofNavassa Island .
J. Ohio Herp. Soc. 5:73–89.
Townsend T.M.,Mulcahy D.G.,Noonan B.P., Sites J.W.,Kuczynski C.A.,
Wiens J.J., Reeder T.W. 2011. Phylogeny of iguanian lizards inferred
from 29 nuclear loci, and a comparison of concatenated and species-
tree approaches for an ancient, rapid rad iation.Mol. Phylogen. Evol.
61:363–380.
Underwood G., Williams E.E. 1959. The Anoline lizards of Jamaica.
Bull. Inst. Jam. Sci. Ser. 9:1–48.
Vanhooydonck B., Herrel A.Y., Van Damme R., Irschick D.J. 2005.
Does dewlap size pred ict male bite performance in Jamaican Anolis
lizards? Funct. Ecol. 19:38–42.
2017 POE ET AL.—EVOLUTION OF ANOLES 697
Varona L.S. 1985. Sistematico de iguanidae sensu lato, y de anolinae en
Cuba (Reptilia; Sauria). Doñana Acta Vert. 12:21–39.
Velasco J.A., Martínez-Meyer E., Flores-Villela O., García A., Algar
A.C., Köhler G, Daza J.M. 2015. Climatic niche attributes and
diversification in Anolis lizards. J. Biogeogr. doi: 10.1111/ jbi.12627.
Vitt L.J., Cristina T., Avila-Pires S., Zani P.A., Espósito M.C. 2002. Life in
shade: the ecology of Anolis trachyderma (Squamata: Polychrotidae)
in Amazonian Ecuador and Brazil, with comparisons to ecologically
similar anoles. Copeia 2002:275–286.
Voigt F.S. 1832. Das Thierrich vom Baron von Cuvier. Vol. 2. Leipzig:
F.A. Brokhaus.
Wagler J.G. 1830. Natürliches System der Amphibien: mit
vorangehender classification der Säugethiere und Vögel : ein
Beitrag zur vergleichenden Zoologie. Munich: J. G. Cotta.
Wagler J.G. 1833. Deutung der in Seba’s Thesauro rerum naturalium
T. 1. et 2. enthaltenen Abbildungen von Lurchen, mit critischen
Bemerkungen. Isis von Oken 26:885–905.
Wiens J.J. 1995. Polymorphic characters in phylogenetic systematics.
Syst. Biol. 44:482–500.
Williams E.E. 1962. Notes on Hispaniolan herpetology 6. The giant
anoles. Breviora 155:1–15.
Williams E.E. 1965. The species of Hispaniolan green anoles (Sauria,
Iguanidae). Breviora 227:1–16.
Williams E.E. 1972. The origin of faunas. Evolution of lizard congeners
in a complex island fauna: a trial analysis. Evol. Biol. 47–89.
Williams E.E. 1974. South American Anolis: three new species
related to Anolis nigrolineatus and A. dissimilis. Breviora 422:
1–15.
Williams E.E. 1976a.West Indian anoles: a taxonomic and evolutionary
summary 1. Introduction and a species list. Breviora 440:
1–21.
Williams E.E. 1976b. South American anoles: the species groups. Pap.
Avul. Zool. 29:259–268.
Williams E.E. 1983. Ecomorphs, faunas, island size, and diverse
end points in island rad iations of Anolis. In Huey R.B., Pianka
E.R., Schoener T.W., ed itors. Lizard ecology. Cambridge: Harvard
University Press.
Williams E.E. 1989. A critique of Guyer and Savage (1986):
clad istic relationships among anoles (Sauria: Iguandidae): are the
data available to reclassify the anoles? In Woods C.A., ed itor.
Biogeography of the West Indies: past, present, and future.
Gainesville: Sandhill Crane Press.
Williams E.E., Rand A.S. 1969. Anolis insolitus, a new dwarf anole of
zoogeographic importance from the mountains of the Dominican
Republic. Breviora 326:1–21.
Williams E.E., Rivero J.A. 1965. A new anole (Sauria, Iguanidae) from
Puerto Rico. Part I. description. Breviora 231:1–9.
Wyles J.S., Gorman G. 1980. The classification of Anolis: conflict
between genetic and osteological interpretation as exemplified by
Anolis cybotes. J. Herpetol. 14:149–53.